Race 2011

RECENT ADVANCES IN
CIVIL ENGINEERING
EDITORS
S.B. Dwivedi
Rajesh Kumar
K.K. Pandey
Kesheo Prasad
Medha Jha
Anurag Ohri
P.B. Ramudu
By
Department of Civil Engineering,
Institute of Technology, Banaras Hindu University,
Varanasi-221005, India
Proceeding of the National Conference on Recent Advances in Civil Engineering
October 14-16, 2011, Varanasi

©Editors, RACE-2011
No part of the publication can be reproduced in any form or by any means without the prior
Written Permission of the publisher.
Due care has been taken to ensure that the information provided in this book is correct.
However, the editors bear no responsibility for any damage resulting from any inadvertent
omission or inaccuracy in this book.
ISBN 978-81-921121-0-7
Department of Civil Engineering,
Institute of Technology, Banaras Hindu University,
Varanasi-221005, India
EDITORIAL DESK
A consistent endeavor has been made by the Civil Engineering professionals since last few years
to solve the complex problems and challenges by employing certain advanced solution. In view
of fulfilling the research gap a National conference on “Recent Advances in Civil Engineering
(RACE-2011)” is being organized in Department of Civil Engineering, Institute of Technology,
Banaras Hindu University during 14th to 16th October-2011. In order to highlight the recent
advances in Civil Engineering presented papers of the conference related to Structural
Engineering, Hydraulic and Water Resources Engineering, Geotechnical Engineering,
Environmental Engineering, Transportation Engineering, GIS & Remote Sensing, Applied
Geology and Disaster Mitigation have been included in this edited volume of Proceeding RACE-
2011. This proceeding, we believe, shall be acknowledged in future as a land mark for acquiring
a good understanding of the potential role of Recent Advances in Civil Engineering to solve the
problems. Themes covered in the proceeding are not only quite relevant but also carries a lot of
weight in addressing Civil Engineering problems and seek the solutions using recent
research/innovations which took place till today. For the reasons mention above this proceeding
is more informative and worth full for researchers. It is exciting to know that recent problems in
Civil Engineering e.g. Earth quake resistant structures, Water Harvesting, Avalanches, Flood
Mitigation, Geo-environmental problem, Solid Waste Engineering, GIS & Remote Sensing,
Applied Geology etc. which call for greater attention for the country and has been addressed
among the various papers of the proceeding. This volume includes one hundred twenty research
papers which are peer reviewed. We highly appreciate the efforts of all researchers, scientists and
engineers from different parts of the country who readily accepted and responded to our request
and submitted the full papers well in time. Most of the papers of this proceeding are based on
original data base and organized properly.
We must acknowledge the financial support provided by Ministry of Earth Science, Govt. of
India, New Delhi, Head, SERC Division, Department of Science and Technology, New Delhi,
Director General, Council for Scientific and Industrial Research, New Delhi, Ministry of Water
Resources, New Delhi, Head, Department of Civil Engineering and Banaras Hindu University to
organize the National conference on “Recent Advances in Civil Engineering (RACE-2011)
during 14th to 16th October, 2011 and bringing out this proceeding in present shape.
This proceeding is a commendable attempt to highlight the Recent Advances in Civil
Engineering among the researchers of different disciplines of Civil Engineering. All papers are
properly arranged covering the varied aspects of Civil Engineering problems. However, still
there are many gaps in the relevant knowledge which can be fulfilled by bringing out more
volume on similar themes in future. We are pretty sure that this proceeding will be useful to the
post graduate students researchers working in the field of Civil Engineering.
Editors
RECENT ADVANCES IN CIVIL ENGINEERING-2011 i

CONTENTS OF THE PROCEEDINGS PAGE No
EDITORIAL DESK i
CONTENTS ii—vii
1 VARIABILITY OF MASS TRANSFER COEFFICIENT IN CONTROLLED CHAMBER 1-6

EVAPORATION EXPERIMENTS
C.S.P. Ojha, Hiroshi Yasuda, Surampalli Rao, Mohamed A. M. Abd Elbasit, Manoj Kumar
2 A SMOOTH DMS-FEM FORMULATION FOR MINDLIN PLATE BENDING 7-12
PROBLEMS
S. Narayan, D. Roy, R. M. Vasu
3 ENHANCING THE REACTIVITY OF FLY ASH FOR GEOTECHNICAL AND 13-21
GEOENVIRONMENTAL APPLICATIONS
P.V. Sivapullaiah
4 SORPTION BEHAVIOUR OF DEEP SEATED COAL: A GEOENGINEERING 22-26
APPROACH TO CARBON DIOXIDE SINKING
V. Vishal, T. N. Singh
5 CHARACTERISTICS OF BADLANDS MORPHOLOGY 27-30
Sanjay Tignath, Medha Jha
6 PERFORMANCE GRADING OF INDIAN BITUMINOUS BINDERS 31-36
G. Bharath, M. Amaranatha Reddy
7 RECENT DEVELOPMENTS IN NITRATE REMOVAL TECHNIQUES FROM WATER 37-47
P.K. Singh, Y.C. Sharma, A. L. Srivastav
8 SWELLING BEHAVIOUR OF EXPANSIVE SOIL MIXED WITH LIME AND FLY ASH 48-51
AS ADDITIVES
Dhirendra Kumar, Suresh Kumar, Bala Ramudu Paramkusam
9 USE OF MULTIPLICATIVE DECOMPOSITION METHOD FOR BUS TRAVEL TIME 52-57
PREDICTION UNDER HETEROGENEOUS TRAFFIC CONDITIONS
S.Vasantha Kumar, Lelitha Vanajakshi
10 INFLUENCE OF TYPE OF BINDER AND CRUMB RUBBER ON THE MORPHOLOGY 58-62
OF CRUMB RUBBER MODIFIED BITUMEN
P. S. Divya, J. Murali Krishnan
11 HYDROINFORMATICS IN GROUNDWATER EXPLORATION: A CASE STUDY 63-68
D.K. Pardeshi, A.O. Rathor, D. G. Regulwar
12 APPLICATION OF RS AND GIS FOR APPROXIMATING STREAM FLOW 69-73
Om Prakash Dubey
13 SEISMIC VULNERABILITY ASSESSMENT OF SKEW BRIDGES 74-80
S.P. Deepu, S. Ray-Chaudhuri
14 GEOCHRONOLOGICAL CONSTRAINTS OF EASTERN DHARWAR CRATON: A 81-87
REVIEW
D. Prakash, P. Chandra Singh, S. Tewari, I.N. Sharma
15 VELOCITIES IN NON-UNIFORM FLOW ON CLOSELY PACKED ROUGH BED 88-94
Kesheo Prasad, C.S.P. Ojha, K. M. Singh
16 ANALYSIS AND CONSTRUCTION OF A LOW COST FOLDABLE DOME 95-99
Rajesh Kumar, Aniruddh Vashisth, Vineet Singla
17 GROUNDWATER PROSPECTING OF URBAN AREA OF THE JABALPUR DISTRICT 100-102
USING REMOTE SENSING AND GIS
Medha Jha, Sanjay Tignath
18 PARAMETRIC STUDY OF PREDICTION OF SCOUR DEPTH IN SPILLWAYS - A 103-107
NEURAL NETWORK APPROACH
Arun Goel
19 URBAN INFRASTRUCTURE PLANNING USING GIS A CASE STUDY OF SURAT 108-111
Shri Chetan R. Patel, Nikunj B. Shah
RECENT ADVANCES IN CIVIL ENGINEERING-2011 ii

20 A STUDY ON LANDFILL LEACHING USING THREE DIMENSIONAL COLUMN 112-115
LEACH APPARATUS
P.J .Barman, S.A. Kartha, B. Pradhan
21 WATER SCARCITY: ISSUES, CONCERNS AND OPTION (A CASE STUDY OF 116-121
KOLKATA AND SUB- URBAN AREAS)
Sabita Madhvi Singh
22 SITE SELECTION FOR GROUND WATER RECHARGE USING TREATED 122-130
MUNICIPAL WASTEWATERS- A CASE STUDY OF VARANASI (INDIA)
P.K.Singh, Anurag Ohri, Abhishek Kumar Bhardwaj
23 ANALYSIS OF GAS DYNAMICS IN A FUNNEL DURING AIR SUCTION 131-136
K.K. Pandey, Aniruddh Vashisth
24 EVOLUTION OF SOME RESISTANCE LAW FOR NON-UNIFORM ACCELERATED 137-143
FLOW OVER CLOSELY PACKED ROUGH BED
Kesheo Prasad, C.S.P. Ojha, K. M. Singh
25 FACTORS AFFECTING BEHAVIOUR OF PILED RAFT FOUNDATION 144-150
Rajendra Singh Bisht, Baleshwar Singh
26 OPTIMISATION OF LOW RISE MULTI-STOREY BUILDING USING STEEL- 151-156
CONCRETE COMPOSITE SECTION & SEMI-RIGID BEAM TO COLUMN
CONNECTIONS
P. S. Buradkar
27 ARTIFICIAL NEURAL NETWORKS FOR REAL TIME RESERVOIR INFLOW 157-163
MODELING AND PREDICTION
A.K. Pardeshi, D.G. Regulwar
28 FLOOD FORECASTING SYSTEM IN KOSI BASIN 164-169
L. B. Roy, Sunil Kumar, Jay Prakash Kumar
29 PUSHOVER ANALYSIS OF RCC BUILDING WITH AND WITHOUT INFILL WALL 170-174
R. E. Dalvi
30 DEFORMATION AND STRENGTH CHARACTERISTICS OF FLY ASH ADDED SOILS 175-180
Tapas Das, Baleshwar Singh
31 GIS: AS A NECESSARY TOOL FOR URBAN PLANNING 181-186
Arati S. Petkar, Eshwar M. Bahirat
32 CANAL REGULATION AND SCADA 187-189
Kinnari R. Mishra, N.K.Sherasia
33 DELINEATION OF GROUNDWATER SATURATED FRACTURE ZONES IN HARD 190-194
ROCK TERRAINS USING INTEGRATED RESISTIVITY SURVEY APPROACH
N.P. Singh
34 VISCO-ELASTO-PLASTIC ANALYSIS OF LAYERED PLATE: A UNIFIED OVERLAY 195-200
APPROACH
P. Srimani, S. Roy Chowdhury, K. K. Ghosh,
35 AUTOMATED HIGHWAY SYSTEMS 201-204
S. Lakshminarasimhan, K.S. Yuvaraaj
36 A CASE STUDY OF MULA RESERVOIR SEDIMENTATION ASSESSMENT USING 205-209
BATHYMETRY ALONG WITH SATELLITEREMOTE SENSING TECHNOLOGY
D.D.Bhide,, P.R.Bhamare, S.N.Kulkarni
37 REVIEW ON USE OF RECYCLED AGGREGATES IN CONCRETE 210-214
G.D. Ransinchung R.N, Kumar Praveen, Chauhan Prakash Arun
38 EFFECT OF LOADING FREQUENCY ON PERFORMANCE OF RC BEAM-COLUMN 215-220
CONNECTIONS
C. Marthong, A. Dutta, S.K. Deb
39 STUDIES ON THE TREATMENT OF VEGETABLE WASTE WITH HIGHSOLID 221-226
ANAEROBIC DIGESTERS
S. Venkatesh
40 PROTOTYPE DEVELOPMENT FOR A BUS ARRIVAL TIME PREDICTION SYSTEM 227-233
UNDER INDIAN TRAFFIC CONDITIONS
Akhilesh Koppineni, Krishna Chaithanya, K. Sidharth, Lelitha Vanajakshi,
RECENT ADVANCES IN CIVIL ENGINEERING-2011 iii

41 CORROSION MONITORING TECHNIQUES FOR EVALUATION OF PERFORMANCE 234-238
OF STEEL IN CONCRETE STRUCTURE
Jitu Kujur, B. Bhattacharjee
42 ASSESSMENT OF CONCRETE STRENGTH USING FLYASH AND RICE HUSK ASH 239-245
Satish D. Kene, Pravin V. Domke, Sandesh D. Deshmukh, R.S. Deotale
43 PILE FOUNDATIONS DESIGN AND CONSTRUCTION 246-249
M. Sujan
44 MODELING QUEUING & DISCHARGE BEHAVIOUR OF HETEROGENEOUS 250-255
TRAFFIC AT MANUALLY CONTROLLED URBAN INTERSECTION
Ghadiyali G.A., Shah M.K., Katti B.K.
45 SENSITIVITY ANALYSIS OF SHALLOW FOUNDATION RESPONSES 256-261
Prishati Raychowdhury, Sumit Jindal
46 DUST DEPOSITION CAPACITY OF CERTAIN ROADSIDE PLANTS IN AIZAWL, 262-270
MIZORAM: IMPLICATIONS FOR ENVIRONMENTAL GEOMAGNETIC STUDIES
Prabhat Kumar Rai
47 DISASTER MITIGATION AND MANAGEMENT FOR FLOOD CONTROL USING 271-275
GEOINFORMATICAL TOOLS
M.B.Kumthekar, P.K.Deshpande, R.D.Padhye, S.B.Thalange
48 DYNAMIC ANALYSIS OF LAMINATED COMPOSITE FOLDED PLATE FOR 276-279
DIFFERENT BOUNDARY CONDITIONS
Sourav Chandra, Sreyashi Das, A. Guha Niyogi
49 CONCRETE BEAMS STRENGTHENED WITH NEAR SURFACE MOUNTED AFRP 280-282
RODS
V.G. Kalpana,K. Subramanian
50 DAMAGE ANALYSIS OF STEEL STRUCTURES 283-285
Pranav P. Pande
51 STREAM FLOW MODELLING OF BEAS RIVER AT MANALI, HIMACHAL PRADESH 286-289
S. K. Jain, S. P Rai, Rajeev S. Ahluwalia
52 RECENT DEVELOPMENTS IN THE MEASUREMENT OF WETTING SWCC 290-293
C. Malaya, S. Sreedeep
53 CHANNEL AND CORED UNIT FOR ROOFING\FLOORING FOR LOW COST 294-298
HOUSING: A REVEIW
Rajesh Kumar, Harshit Jain, S. B Dwivedi
54 A STUDY ON DUCTILITY OF CONCRETE AND MORTAR WITH THE ADDITION OF 299-305
RUBBER POWDER
C. B. K. Rao, R. Bhaskar,V. Rajendra Prasad
55 EFFECTS OF AGGRESSIVE CHLORIDE ENVIRONMENT ON HIGH PERFORMANCE 306-310
STEEL FIBRE REINFORCED CONCRETE
Dhirendra Singhal, Bal Krishan
56 THE INFLUENCE OF GEO-ENVIRONMENTAL PROPERTIES ON MUNICIPAL SOLID 311-316
WASTE
Sunita Kumari, A.K. Nema, K.B. Ladhane
57 CEMENT AND POLYPROPYLENE FIBRES TO IMPROVE THE LOAD BEARING 317-322
CAPACITY OF PEAT
Behzad Kalantari, Arun Prasad
58 EFFECT OF NaOH MOLARITY ON COMPRESSIVE STRENGTH OF GEOPOLYMER 323-327
CONCRETE IN AMBIENT CURING CONDITION
S. V. Joshi, M. S. Kadu
59 ECONOMICS AND STUDY UNDER BORDER IRRIGATION ON GROUNDNUT CROP 328-330
AT VARIOUS WATER APPLICATION LEVELS
A.B. Rathod, B.Jigar Parekh
60 PETROGRAPHIC EXAMINATION OF CORROSION INHIBITORS ADMIXED 331-335
CONCRETE
B.N. Singh, S.K. Singh, V. Kumar, M.A. Quraishi
RECENT ADVANCES IN CIVIL ENGINEERING-2011 iv

61 DETECTION OF ACTIVE LANDSLIDE AREAS IN HIMALAYAS USING SMALL 336-342
BASELINE SUBSET INTERFEROMETRY
A. Bhattacharya, Manoj. K. Arora, Mukat. L. Sharma
62 TIME PERIOD ANALYSIS OF LIQUID FILLED CYLINDRICAL CONTAINER DURING 343-348
SLOSHING
P. R. Maiti, Abhinav Srivastava
63 AVERAGE P-T ESTIMATES AND MINERAL CHEMISTRY OF THE TWO PYROXENE 349-354
BEARING GRANULITES FROM SONAPAHAR,WEST KHASI HILLS DISTRICT,
MEGHALAYA
S.B.Dwivedi, K.Theunuo
64 CRACK SENSING CAPABILITIES OF SMART CANTILEVER STRUCTURES 355-359
Nilanjan Mallik
65 ESTIMATION OF GUST EFFECTIVE FACTOR USING IS 875(PART 3) – 1987 AND 360-364
MODIFICATIONS PROPOSED BY PREMKRISHNA (2002)
Deepak Kumar, S. Mandal
66 ANALYSIS OF SKEW SLAB DECK USING ANALYTICAL METHODS 365-369
Rajesh Kumar, Veerendra Kumar, Bikram Kesharee Patra
67 REMOVAL MECHANISM OF Cr (VI) AND Fe (III) METAL IONS FROMEXPAN-SIVE 370-376
SOIL BY ELECTROKINETIC EXTRACTION
Ch. Ramakrishna, B. S. Nagendra Prakash, P.V. Sivapullaiah
68 APPLICATION OF ISOTOPIC TECHNIQUES IN HYDROLOGY OF MOUNTA-INOUS 377-381
REGION
S. P. Rai, Bhishn Kumar, Y.S. Rawat, P. Purushothaman
69 GROUND WATER ZONES AROUND JHANSI TOWN USING SATELLITE-REMOTE 382-387
SENSING DATA IN THE GRANITIC TERRAIN OFBUNDELKHAND
S.P. Singh, Yamini Singh
70 INFLUENCE OF CRUSHABILITY AND MIGRATION OF MOISTURE ONCBR 388-391
CHARACTERISTICS OF POND ASH
M.V.S. Sreedhar, G.Venkatappa Rao, R.Ramesh Reddy,
71 PERFORMANCE HISTORY OF WARM MIX ASPHALT 392-395
Rajan Choudhary, Eleena Gao, Ashok Julaganti
72 DETERIORATION OF GROUND QUALITY DUE TO OPENCASTCOAL MINES AND 396-399
OTHER RELATED INDUSTRIES IN SINGRAULIAREA
Raj K. Sharma, C.S. Singh,.
73 STABLE ISOTOPIC CHARACTERISATION OFGROUNDWATERS INBIST- DOAB 400-404
REGION, PUNJAB
P. Purushothaman, M.S. Rao, B. Kumar,Y.S. Rawat, Gopal Krishan
74 LINEAR AND NON LINEAR ANALYSIS OF SQUARE PLATE 405-407
Ashish Johri, Rajesh Kumar, Veerendra Kumar
75 LOCAL STRATEGIC PLANNING PROCESSES AND SUSTAINABILITY 408-416
TRANSITIONS IN INFRASTRUCTURE SECTORS
Rajesh Singh, P.K.S. Dikshit
76 ENVIRONMENTAL CAPACITY CONCEPT IN NOISE POLLUTION ANALYSIS OF A 417-420
ROADWAY
Kanakabandi Shalini, Brind Kumar
77 ESTIMATION OF SCOUR DEPTH AROUND BRIDGE PIERS BYUSING HEC-RAS 421-425
Dilip Kumar, Hira Lal Yadav, Sushil Kumar Himanshu
78 COMPARATIVE STUDY OF BUILDING WITH COMPOSITE ANDRCC STRUCTURE 426-432
A.K. Chitkeshwar, V.G. Meshram
79 SHEAR STRENGTH OF ORGANIC SOIL TREATED WITH CEMENTGROUT AND 433-437
MODIFIED WITH ADDITIVES
SinaKazemian, Arun Prasad
80 POTENTIAL EVAPOTRANSPIRATION ASSESSMENT BY USING 438-439
MWSWATMODELING ON SABARMATI RIVER
Shaikh M Zuned, P P Lodha
RECENT ADVANCES IN CIVIL ENGINEERING-2011 v

81 LOCATION OF WATER HARVESTING STRUCTURES IN SEETHARIVER BASIN OF 440-444
COASTAL KARNATAKA BY RS AND GISPERSPECTIVE – A CASE STUDY
Mohandas Chadaga, Ravindranath, Sreeniwas Poudel S., Varun Kadian
82 REVIEW ON SELF COMPACTING CONCRETE 445-449
Kumar Praveen, G.D. Ransinchung R.N, Jindal Abhishek
83 A STUDY ON SHOTCRETE TECHNOLOGY 450-456
Vipin Bansal, Abhishek Choudhary, Dipendu Bhunia, Ashutosh Parauha, Nikhil Shrivastava
84 SOME STUDIES ON THE INFLUENCE OF CLAY ON 457-460
CONSOLIDATIONCHARACTERISTICS OF COHESIVE SOILS
S.V. Rao
85 NONDIMENSIONAL ANALYSIS OF A CIRCULAR PLATE RESTING ONTHE 461-468
PASTERNAK FOUNDATION USING THE STRAIN ENERGY APPROACH
Ashish Gupta, Dilip Kumar
86 OPTICAL SENSOR BASED CRACK OPENING DISPLACEMNETS STUDIES IN 469-471
CONCRETE
K. Samrajyam, B.Sobha, T D Gunneswara Rao
87 DYNAMIC ANALYSIS OF NAILED OPEN CUTS- A CASE STUDY 472-479
Bhishm Singh Khati, Dilip Kumar, Swami Saran
88 COMPARATIVE STUDY OF OPTIMAL CROPPING PATTERN USING LINEAR 480-488
PROGRAMMING AND GENETIC ALGORITHM
Ingle Visha, D.G.Regulwar
89 A CRITICAL EVALUATION OF FOLLOWING PHENOMENA IN INDIAN MIXED 489-491
TRAFFIC CONDITIONS
K.V.R. Ravi Shankar, Tom V. Mathew
90 FEM FORMULATION OF DIRECT METHOD OF CONCRETE GRAVITY DAM - 492-496
NONLINEAR FOUNDATION INTERACTION
A Burman, D Maity, A Kumar, S Bhushan
91 DEVELOPMENT OF A STIFFNESS ANALYSIS FOR A STRUCTURAL ELEMENT IN A 497-501
BUILDING INFRASTRUCTURE
Pradeep Kumar, Swatantra K. Porwal
92 EMERGING TECHNOLOGIES FOR COLOUR REMOVAL 502-506
Shiva Shankar Y, Abhishek Kumar, Harshit Jain, Devendra Mohan
93 COMPARATIVE ANALYSIS OF BED LOAD FORMULAS FOR INTENSE BEDLOAD 507-512
CHANNELS
Sumit Talukdar, Bimlesh Kumar, Subashisa Dutta
94 ANALYSIS OF ROUGHNESS AND CRACKING ON SELECTED FLEXIBLE 513-518
PAVEMENT IN BANGALORE CITY
Prathima G
95 INVESTIGATIONS OF MIXING IN MECHANICALLY STIRRED TANK:COMPARISON 519-524
OF CD-6 IMPELLER AND RUSHTON TURBINE
Thiyam Tamphasana Devi
96 ESTIMATION OF HEAVY METAL CONTAMINATION FROM MUNICIPAL SOLID 525-532
WASTE LANDFILL AT KOLKATA USING, EPACMTP MODEL
Sanjukta Basak, Subhasish Chattopadhyay, Swapan Kumar Mukhopadhyay, Amit Dutta
97 BEHAVIOUR OF SUCTION CAISSON AS FOUNDATION SYSTEM FOR OFFSHORE 533-537
WIND TURBINE IN SANDY SOILS
Suchit Kumar Patel, Baleshwar Singh
98 DYNAMIC ANALYSIS OF MULTISTORIED BUILDING USING FLAT SLAB, 538-547
M.S.Janbandhu, V.G. Meshram
99 TECHNO-ECONOMIC STUDY OF REMOTE AREA SMALL HYDRO POWER PLANTS 548-550
Mon Prakash Upadhyay, Rahul Bhatt, Nitin Kumar Sahu
100 A PRAGMATIC APPROACH OF POT HOLES REPAIR UNDER INDIAN SERVICE 551-557
CONDITIONS
G.D. Ransinchung R.N, Praveen Kumar, Brind Kumar
RECENT ADVANCES IN CIVIL ENGINEERING-2011 vi

101 ANALYSIS OF THE MEAN MONTHLY TEMPERATURE SEVEN HOMOGENEO-US 558-562
REGIONS OF INDIA
R.K.S. Maurya, G. P. Singh
102 RELATE AVERAGE CONCENTRATION OF SEDIMENT TO LOCAL 563-565
CONCENTRATION FOR STEADY UNIFORM FLOW
N.D. Vernekar, A.R. Bhalerao
103 GEOTECHNICAL PROPERTIES OF SOIL CONTAMINATED BY TANNING 566-568
INDUSTRY EFFLUENTS
Ashwani Jain
104 WATER MANAGEMENT ON DRIP AND MICRO SPRINKLER IRRIGATION FOR 569-573
SUMMER GROUNDNUT CROP UNDER DEFICIT WATER SUPPLIES
A.B. Rathod, S. A. Trivedi
105 A REVIEW OF NOISE ANALYSIS FOR ROAD TRANSPORT SYSTEMS 577-583
Brind Kumar, G.D. Ransinchung R.N., Saurabh Gupta, Gajendra Kumar Yadav, Alok Kumar
106 BEARING CAPACITY OF POND ASH REINFORCED WITH A COIR GEOTEXTILE 584-587
M.V.S. Sreedhar,G. Venkatappa Rao,R. Ramesh Reddy
107 PERFORMANCE OF CONCRETE FILLED STEEL (CFT) HOLLOW BEAM SECTIONS 591-594
IN FLEXURE
Sunil Kute, T.Jayashree, Shantanu Pande
108 FORECASTING OF GROUND WATER LEVEL OF SIMGATEHSIL, RAIPUR 595-602
DISTRICT, CHHATTISGARH USING ARTIFICIAL NEURAL
NETWORK
Umank Mishra, K.K. Pandey, S B Dwiwedi
109 REDUCTION OF FLUORIDE WITH FREE JETS FOR INDUSTRIAL APPLICATIONS 598-601
Suresh Kumar N., Ravi Kumar, S.
110 COMPARATIVE BEHAVIOUR OF TWO STOREY BRICK MASONRY BUILDING 602-609
MODELS OF CONFINED, UN-CONFINED AND MODEL AS PER IS CODE
Vikash Khatri, P. K. Singh, P.R. Maiti
111 DAMAGE ESTIMATION OF R.C.C STRUCTURAL MEMBER BY DISPLACEMENT 610-613
BASED ANALYSIS TECHNICS
A.Vimala, Ramancharla Pradeep Kumar
112 LONG-TERM STATISTICAL EXPRESSION OF TEMPERATURE OVER INDIA 614-617
R. K. S. Maurya, G. P. Singh
113 EFFECTS OF CORROSION INHIBITOR ON THE PERFORMANCE OF MILD STEEL 618-624
AND PRESTRESSING STEEL
Vasugi Jegan , Vandhana Mary Jacob, Radhakrishna G. Pillai
114 REMEDIAL MEASURES FOR RESTORATION OF DISTRESSES INPART PARALLEL 625-634
TAXIWAY AND APRON – A CASE STUDY OF LAL
BAHADUR SHASTRI AIRPORT AT BABATPUR, VARANASI
Brind Kumar, Pramod Kumar Singh
115 AN EXACT SOLUTION OF MHD FLOW DUE TO ACCELERATED MOTION OF 635-637
INFINITE POROUS PLATE
J. Singh, Munna Lal, Dheeraj Agrawal, Mona Goyal
116 GPS DERIVED STRAIN ANALYSIS AND CRUSTAL DEFORMATION IN INDIA 638-642
Abhay P. Singh Abhishek Rai
117 MODELLING HETEROGENEOUS TRAFFIC USING VISSIM 643-647
A.K. Maurya, S. Sharma, S. Biswas
118 RECENT ADVANCES IN HYDROLOGICAL ANALYSIS, MODELING, AND 648-655
DECISION - MAKING
R.D. Singh, M. Goel
119 STATIC AND DYNAMIC STABILITY OF PLATES AND SHELLS 656-662
L.S. Ramachandra
120 EFFECT OF BRIDGE PIER’S SHAPE ON SCOURING 663-666
Manvendra Singh Chauhan, U. K. Choudhary, P. K. Singh Dikshit
AUTHOR INDEX 667-669
RECENT ADVANCES IN CIVIL ENGINEERING-2011 vii

ISBN 978-81-921121-0-7
VARIABILITY OF MASS TRANSFER COEFFICIENT IN

CONTROLLED CHAMBER EVAPORATION EXPERIMENTS
C.S.P. Ojha
Dept. of Civil Engineering, Indian Institute of Technology, Roorkee, India
Hiroshi Yasuda
Arid Land Research Center, Tottori University, Japan
Surampalli Rao
USEPA, Kansas, USA
Mohamed A. M. Abd Elbasit
Desertification Research Institute, National Center for Research, Khartoum, Sudan
Manoj Kumar
Dept. of Civil Engineering, Indian Institute of Technology, Roorkee, India
ABSTRACT: Evaporation is estimated using mass transfer, energy or a combination approach which includes
mass transfer as well as energy based approach. In the literature, there are a large number of studies which use
mass transfer approach as the basis of estimating evaporation. In this study, the potential of mass transfer
approach are evaluated using data from controlled chamber experiments in which radiation and wind velocity
were kept constant and temperature and relative humidity profiles were varied in different patterns. Currently,
FAO procedure lists three approaches to compute air vapour pressure based on temperature and relative
humidity profiles. In this study, the impact of using different procedures of estimating air vapour pressure is also
examined to assess the use of mass transfer approach for estimating evaporation. To achieve this, a part of the
data is used to calibrate mass transfer coefficient which is subsequently used to project evaporation for future
states. Accordingly, strategies are ranked for their potential in estimating evaporation.
Key Words: Mass transfer coefficient, Vapour pressure, Evaporation rate, Variability, Relative humidity and
Temperature.
INTRODUCTION
Evaporation is the phenomenon by which a substance is Evaporation research was started with the pioneer
converted from liquid phase to vapour. Evaporation is work of Sir Dalton in 1802 who observed that
relevant in many disciplines from small to very large evaporation rate is closely related to vapour
scale. It is important in determining the water balance pressure difference between evaporating surface
of watersheds, allowing prediction and estimation of and the surrounding air. This led to the classical
runoff and ground water recharge. Evaporation data are Dalton equation, also called aerodynamic equation.
required in managing both irrigation and dry land Tanner and Sincalir (1983) reported that the vapour
farming operations. The amount and rate of evaporation pressure deficit is a dominant factor influencing the
from water surfaces are the parameters essential for evaporation. Vapour pressure denotes the partial
designers to plan storage reservoirs of waste water and pressure exerted by the water vapour present in the
brine ponds. Such information is also needed in air. Vapour pressure at equilibrium state is called
determining the evaporation from natural lakes, the saturation vapour pressure (Ps). It is a function
irrigation scheduling, condenser cooling water and of temperature only; it increases with increase in
water requirements for hydroelectric power, as well as temperature (Dingman, 1994). Evaporation rate
settling of various water-right disputes, etc. from an open water surface depends on the
In fact, if the man is to survive and improve living difference between the vapour pressure of water
standards on earth, something must be done to increase and air (Pa) (called vapour pressure deficit), vapour
the supply of fresh water. One method of increasing the pressure of water being the saturation vapour
available water supply could be to decrease or prevent pressure for the temperature of water. Thus, the
evaporation from water surfaces. Evaporation not only vapour pressure deficit (Ps-Pa) is an indicator of
reduces the amount of available water, but it also actual evaporative demand of air (Yoder et al
decreases the quality of water and increases impurity 2005). Evaporation continues as long as vapour
concentration because only pure water is lost by pressure of water surface exceeds the vapour
evaporation. For all the above-mentioned reasons, there pressure of the air (Trewartha and Horn, 1980).
is a need for better understanding of the evaporation Singh and Xu (1997) reported that the mass
process. It is therefore necessary to examine the transfer equations to determine the evaporation are
evaporation process and contributing factors and to sensitive to vapour pressure gradient. Gianniou and
develop a method of accurately predicting evaporation Antonopoulos (2007) observed that the particularly
rates from water body. lower evaporation rates are sensitive to air vapour
pressure. Winter (1981) reported the error in
RECENT ADVANCES IN CIVIL ENGINEERING-2011 1

determining the evaporation rates which also depends developing several strategies for computing
on vapour pressure deficit. It has been further reported mass transfer coefficient.
that the errors of 25% could be introduced in 4. To evaluate the different strategies of mass
calculating the u (Ps -Pa) term (u being the wind transfer coefficients computations in
velocity) in the mass transfer equation if the estimation of evaporation for future states and
corresponding errors in air and water temperatures is to rank them on the basis of their ability to
1%. Ojha et al (2010) reported the errors in evaporation forecast future states.
and compared the errors with three different models.
DESCRIPTION OF EXPERIMENTAL SETUP
Air vapour pressures are not directly measured by any To achieve the above objectives, the experiments
instrument. However its value depends on the were conducted in controlled programmable
measurement of temperature and relative humidity chamber with base area equal to 7.5 m2. The
which is most commonly used in practice. A number of chamber is located at Arid Land research centre,
methods are available for the estimation of vapour Tottori, Japan. The radiation and wind velocity
pressure and vapour pressure deficit. Jansen et al (1990) were kept constant throughout the experiments.
reported procedure to estimate the vapour pressure. The average value of radiation and wind velocity
Allen et al. (1998) reported three models for estimating was 75.1 W m-2 (the average light intensity was
vapour pressure from temperature and relative 50,000 lux) and 0.9 m s-1, respectively. The
humidity. chamber temperature (T) and relative humidity
(RH) have been programmed to follow four T/RH
Chuck and Sparrow (1987) investigated experimentally patterns; namely: increasing T decreasing RH
the evaporation characteristics of a rectangular pan with (T/RH1), decreasing T increasing RH (T/RH2),
water at the bottom panel of the duct and turbulent air increasing T increasing RH (T/RH3), and
was passed through the duct and over the surface of the decreasing T decreasing RH (T/RH4) shown in fig
water. The thermal conditions of the air and water and 1. The maximum, average, and minimum
evaporation rate were measured. The height of water in temperature was 30, 20, and 10°C, respectively. On
the pan was varied and two pan lengths of 12.5 and the other hand the maximum, average, minimum
27.9 cm were used. Adams et al. (1990) studied the relative humidity was 75, 50, and 25%,
evaporation from heated water bodies. The discharge respectively. The starting temperature and humidity
coming out of the nuclear plant was taken for study at the four T/RH patterns was the average. The rate
where the temperature of water was approximately 700 of change in temperature and relative humidity was
C. Theoretical analysis had been used to examine 2°C h-1, and 5% h-1, respectively. Each pattern has
various evaporation regimes and found that at heated tack 24 hours to complete its cycle. To measure the
ponds as wind speed increases, the thermal loading water evaporation, a steel evaporation pan with
decreases. Pauken et al. (1999, 1995 and 1993) 19.8 cm diameter was placed on electronic scale
performed experiments by evaporating heated water (Sartorius LP-2200P, Goettingen, Germany) with
from a circular pan in a low speed wind tunnel. The 0.01 g accuracy. The scale was programmed to
evaporation boundary layer that resulted was a transfer the weight data every 10 minute to a
combined turbulent forced and turbulent free personal computer shown in fig 2. The air
convection boundary layer, where the forced temperature and relative humidity was also
convection was dominated by the air velocity and the measured in 10 minute time interval using thermo
free convection was a result of the density difference recorder sensor (RS-12, ESPEC MIC Corp., Osaka,
between the air at the surface of the water and the Japan). The radiation flux density in the chamber
ambient air. It was found that 30% of the evaporation was measured using pyranometer in W m-2 (PYR,
rate was due to free convection and only 10% by forced Decagon Devices, Washington, USA). The
convection and the rest was by combined free and force pyranometer was fixed on the chamber floor and
convection. connected to a data logger (Em50, Decagon
With the above background, the following objectives Devices, Washington. USA).
are set for the present study.
1. To perform controlled chamber experiments in
which radiation and wind velocity are kept
constant.
2. To produce different variations of temperature and
relative humidity to look into variability of air
vapour pressure as it depends on temperature and
relative humidity of air as per FAO (Food and
Agriculture Organization) procedure.
3. To estimate air vapour pressure using different Fig 1 Temperature and humidity patterns applied to
approaches suggested by FAO and to use this in the chamber

RH max
Pa   Ps (Tmin )
100 (8)
In eq. (6) Tmin is minimum temperature, Tmax is
maximum temperature and RHmean is the average
relative humidity which is computed as
RH mean  ( RH max  RH min ) / 2 (9)
where, RHmin is minimum relative humidity and
RHmax is maximum relative humidity
Based on use of eqs (6), (7) and (8) for computing
Pa, seven strategies are considered to calculate the
value of mass transfer coefficient K in eq. (2).
These strategies are listed below:
Strategy 1
Ps (transient) and Pa (transient)
In strategy (1), Ps and Pa are taken transient at any
given time corresponding to measured temperature
and relative humidity.
Strategy 2
Ps (transient) and Pa (constant)
In strategy (2), Pa is computed from eq. (7)
Fig 2 Computer-controlled chamber (a) general view of Strategy 3
the chamber and (b) experimental setup Ps (average) and Pa (constant)
In strategy (3), Ps is obtained by considering all the
THEORETICAL BACKGROUND values of Ps in a given time period (one day) and Pa
The mass transfer coefficient depends on the vapour is computed from eq. (7)
pressure difference (∆Pv) between the air and surface of Strategy 4
water, evaporation rate (me) and is also a function of Ps (average) and Pa (average)
wind velocity f (u). The evaporation rate (me) is directly In strategy (4), Ps and Pa are obtained considering
proportional to pressure difference of vapour, i.e. all the value of Ps and Pa in a given time period.
.
me  (Pv ) Strategy 5
(1) In strategy (5), average value of Ps and Pa are used.
Invoking dependence of mass transfer coefficient on Ps (average) is computed as
wind velocity through a function f (u), one can also Ps (average) = (Ps (max) + Ps (min))/2 (10)
express eq. (1) as and Pa (average) = RH (average) *Ps (average)
.
m e  Kf (u )( Pv ) (11)
(2) where, RH (average) is computed as (RHmax +
where, K is the proportionality constant and is denoted RHmin)/2 (12)
as mass transfer coefficient. Therefore, Strategy 6
.
me Pa = Ps (Tmin) RHmax,
K and Ps (transient) is taken similar to strategy 1.
f (u ) Pv (3) Strategy 7
In equation (3) me is taken as the evaporation rate in Pa = Ps (Tmin) RHmax and Ps (Average)
g/m2hr, u as velocity in m/sec and (∆Pv) is the pressure
difference of vapour in kPa, expressed as ANALYSIS OF EVAPORATION DATA
Pv  ( P s  Pa ) (4) Under different variations of temperature and
where, Ps the vapour pressure at the surface of water relative humidity, evaporation in the pan showed
and Pa is the vapour pressure of air. Ps can be computed the following variation, as given in the fig.3. As per
using steam tables or approximate relationships .One the strategy (1), Ps and Pa are considered as
such relationship (Subramanya 2007) is transient at any given time. Using eq. (5), Ps can be
17.27T (5) computed as a function of time. Similarly,
P  P (T )  4.584 exp(
s sat s ) multiplying Ps at any time with the corresponding
237.3  T
where, T is the temperature in degree C relative humidity, Pa can be obtained as a function
To estimate Pa, FAO recommends three approaches. of time. For example, corresponding to data of
According to first approach temperature and relative humidity, the variation of
Pa  RH mean [ Ps (Tmax )  Ps (Tmin )] / 2 Ps and Pa with time is shown for day one in fig 4
(6)
and 5. The values of evaporation rate is shown in
According to second approach
fig.3 and from the eq.(2) , the mass transfer
Pa  [ Ps (Tmin )  RH (max)  Ps (Tmax )  RH (min) )] / 2
(7) coefficient is computed and is shown in fig. 6. Fig.
According to third approach

6 can be also used to compute average values of mass strategy (4) gives the best agreement between
transfer coefficient for day one. computed and observed evaporation for all the
days. In real life, it is very difficult to have
300 continuous monitoring of temperature and RH; thus
250 use of strategy (4) may not be always feasible.
Evaporationrate(g/m2 hr)
200 Thus, to screen out the next best performing

150 strategy, fig. 7 to 10 favour the use of strategy (5).
100 It can be seen that the strategies considered in this
50 work can be classified into three categories as per
0
0 2 4 6 8 10 12 14 16 18 20 22 24
the data availability. Strategy (1) and (2) pertain to
Time (hr) the category when the data availability is not a
concern and transient variation of temperature and
Fig.3 Variation of evaporation rate with time
RH are available. However in many situations,
such type of data availability lacks. In view of this,
5000
4500
one can think of using FAO procedure in which the
4000 data availability is restricted to Tmin (minimum
3500
temperature), Tmax (maximum temperature) , RHmin
Ps (kPa)
3000
2500
2000
(minimum relative humidity), RHmax(maximum
1500
1000
relative humidity) and evaporation in a given time
500 period (typically one day). Strategies 5, 6 and 7 fall
0
0 5 10 15 20 25 30 into this category. The FAO approach also suggests
Time (hr) computing Pa using when the data on RHmin is not
Fig.4 Variation of Ps with time available, as in strategy (7)
It is interesting to see that strategies 3, 4 and 5

1600
1400
produced lower estimates of evaporation than the
1200 observed one. On the contrary, strategies 1, 2, 6
1000 and 7 yield higher estimates of evaporation than the
Pa (kPa)
800
observed one.
600
400
200 In this work, the considered strategies have a
0 potential of being used in different situations of
0 5 10 15 20 25 30
Time (hr)
data availability. The experimental data in
controlled chamber experiment has offered
Fig.5 Variation of Pa with time simultaneous evaluation of these strategies which
otherwise would have been very difficult to
Using the data of fig.3, mass transfer coefficient is achieve. From the analysis of data, it is apparent
computed using different strategies and the variation of that there is no substitute for the lack of precise
the same is shown in the fig. 6 data base on temperature, RH and evaporation.
However, under data limited conditions, the
RESULTS AND DISCUSSION strategies 5 to 7 offer better potential for
Fig.3 shows the transient variation of evaporation rate. application, as per the analysis of present
It can be seen that between 0 hrs to 10 hrs, evaporation experiments. Further details on these strategies are
rate increases sharply while variation is small from 11 available in Ojha et al. (2011 b).
hrs to 24 hrs. It shows the evaporation rate in an
interval of 10 minutes during the experimental period.
1600
From the analysis of evaporation data, the variation K (Strategy 1)
1400
between the mass transfer coefficients are obtained 1200
K (Strategy 2)
K (Strategy 3)
using different strategies and all these is shown in fig. 1000 K (Strategy 4)
K (Strategy 5)
6. To study the impact of using these mass transfer
K
800 K (Strategy 6)
coefficients evaporation has been estimated for day 2, 600 K (Strategy 7)
3, 4 and 5 of controlled experiments. Day one is used to 400
calibrate mass coefficient and this calibrated mass 200
transfer coefficient is used to compute evaporation for 0

0 2 4 6 8 10 12 14 16 18 20 22 24
other days. The objective of this is to identify the best Time (hr)
performing strategies.
Figs. 7 to 10 show the computed evaporation using Fig 6 Variation of constant K with different
different strategies along with observed evaporation for approaches
day two to five. From comparison of performance of
different strategies in fig. 7 to 10, it can be seen that the

300
CONCLUSIONS
250
Evaporation rate (g/m hr)

1. Mass transfer approach has the potential of
2 200
being used as an estimator of evaporation.
Observed
150
Computed
2. There can be a large number of strategies to
100 evaluate mass transfer coefficients and these
50
strategies vary depending on the availability of
0
data. In this work, the use of seven strategies is
1 2 3 4 5 6 7 demonstrated to estimate K.
Strategies 3. Strategies are found to fall into two categories
in terms of their underestimation or
Fig. 7 Comparison of observed and computed overestimation of evaporation. Strategies give
evaporation rate for second day the similar performance for all the days of
forecasts.
4. Strategy based on average saturation vapor
300 pressure and average air vapor pressure, where
250
average is based on the consideration of
complete data in calibration range, is found to

2
200
Observed
outperform other strategies.
150
Computed
100 NOMENCLATURE
50
ACRONYMS
0
1 2 3 4 5 6 7 RH Relative humidity
Strategies ENGLISH SYMBOL
As area of evaporating surface (m2)
Fig. 8 Comparison of observed and computed cm centimetre
evaporation rate for third day Dh hydraulic diameter of pan (m)
D Pan diameter (m)
300
g Acceleration due to gravity (m/sec2)
K mass transfer coefficient
250
m metre
2
200 .
150
Observed m mass flow rate of air (kg/sec)
Com puted
.
100
me mass flow rate of water vapour from pan
50
(kg/sec)
0 Ps
1 2 3 4 5 6 7 partial pressure of water vapour at the
Strategies surface of water
Fig. 9 Comparison of observed and computed

Psat saturated water vapour pressure (Pa)
evaporation rate for fourth day Pa vapour pressure of air
P pressure (Pa)
T temperaure (K)
300
u velocity of air (m/sec)
250
E vap oration rate (g/m hr)
GREEK SYMBOLS
2
200
µ dynamic viscosity (Ns/m2)
Observed
150
Computed
 kinematic viscosity (m2/sec)
100
SUBSCRIPTS
50 a air
0 Avg average
1 2 3 4 5 6 7 min minimum
Strategies max maximum
s saturation
Fig. 10 Comparison of observed and computed v water vapour
evaporation rate for fifth day

REFERENCE Winter T C. 1981, Uncertainties in estimating the
Adams E E, Cosler, D J & Helfrich, K R.1990, water balance of lakes. Water Res. Bul. 17: 82-
Evaporation from heated water bodies: Predicting 112.
combined forced plus free convection. Water Yoder R E, Odhiambo L O and Wright W C. 2005
Resour. Res., 26: 425-435 . Effects of vapour pressure deficit and net
Allen R G, Pareira L S, Raes D, & Smith, M. 1998, irradiance calculation methods on accuracy of
Crop evapotranspiration: guidelines for computing standardized Penman-Monteith equation in a
crop water requirements. Irrigation and Drainage Humid climate. Journal of Irrigation Drain.
Paper. No. 56, F.A.O., Rome, Italy. Engg., ASCE 131: 228-237. (2005)
Chuck W & Sparrow E M.1987, Evaporative mass
transfer in turbulent forced convection duct flows.
International Journal of Heat and Mass Transfer.
30: 215-222.
Dingman S l. 1994, Evapotranspiration. Physics
Hydrology. McMillan Pub. Co., New York, 256-
265.
Gianniou S K and Antopoulos V Z.2007, Evaporation
and energy budget in Lake Vegoritis. Journal of
Hydrology. 345: 212-223.
Incropera F P & DeWitt D P 2002, Fundamentals of
Heat and Mass Transfer. Fifth ed., John Wiley &
Sons, New York, 465-531.
Jensen M E, Burman R D and Allen R G. 1990,
Evapotranspiration and irrigation water
requirements. ASCE manuals and reports on
engineering practice No. 70, ASCE, New York.
Ojha C. S. P., Khobragade S. D. and Adeloye A. J.
2011a, Estimating air vapour pressure in a semi-
arid region using FAO-56. Journal of Irrigation
and Drainage Engineering. ASCE.
Ojha C. S. P., Yasuda H., Rao S., Abd Elbasit M. A. M.
and Kumar M. 2011b, Evaporation in relation to
CO2 concentration: Analysis of mass transfer
coefficient Atmospheric Environment (in press)
Pauken M T. 1999, An experimental investigation
of combined turbulent free and forced evaporation.
Experimental Thermal and Fluid Science 18: 334-
340.
Pauken M T, Farley B, Jeter S M & Abdel-Khalik.
1995, An experimental investigation of water
evaporation into low-velocity air currents.
ASHRAE Trans 101: 90-96.
Pauken M T, Tang T D & Jeter S M. 1993, A novel
method for measuring water evaporation into still
air. ASHRAE Trans. 99: 297-300.
Singh V P and Xu C Y.1997, Sensitivity of mass
transfer based evaporation equations to errors in
daily and monthly input data. Hydrological
Process, 11: 1465-1473.
Subramanya K. 2007, Engineering Hydrology. Tata
McGraw-Hill, New Delhi, 74.
Tanner C B and Sincalir T R. 1983, Efficient water use
in crop production: research or research. American
Society of Agronomy, Inc., Madision, Wisc.1-27.
Trewartha G T and Horn L H. 1980, Atmospheric and
precipitation. An Introduction to Climate, 5th
edition, McGraw Hills Book Co., New York., 41-
53.

ISBN 978-81-921121-0-7
A SMOOTH DMS-FEM FORMULATION FOR MINDLIN PLATE

BENDING PROBLEMS
S. Narayan
Computational Mechanics Lab, Department of Civil Engineering, Indian Institute of Science Bangalore, India
D. Roy
Corresponding Author, Computational Mechanics Lab, Department of Civil Engineering, Indian Institute of Science
Bangalore, India, Email: royd@civil.iisc.ernet.in
R. M. Vasu
Department of Instrumentation and Applied Physics, Indian Institute of Science Bangalore, India
ABSTRACT: The main purpose of this article is to introduce a strictly higher order continuous and
polynomial reproducing functional approximation within an FE-like domain discretization for the numerical
solution of Mindlin plate bending problems. While the classical C0 shape functions, typically used with the
FEM, is known to suffice to tackle the weak form corresponding to such problems, it is demonstrated through
several numerical illustrations that the higher order inter-element continuity, possessed by the presently adopted
approximation scheme (referred to as the DMS-FEM), indeed plays a remarkably significant role in improving
the numerical accuracy, convergence characteristics and locking-resistant nature of the computed response.
INTRODUCTION
Plate bending constitutes an important class of Total DOF-s required for complete polynomial
problems in engineering applications of solid 1
mechanics. Of the two well known plate theories, function of order n is (n+1)(n+2). Fig. 1 shows
Kirchhoff plate and Mindlin–Reissner, the former has 2
seen wide applications to solve bending of thin plates nodes and corresponding DOF-s for TUBA-6 (n =
with negligible shear deformation. A major 5) and TUBA-15 (n = 7) elements. A problem with
complication in the finite element (FE) implementation the TUBA family is the use of normal vectors to
of this theory is the requirement of uniformly C1 define the DOF-s, especially given that a vector to
continuity of shape functions for approximating the be normal to a surface is not preserved by an affine
displacement field. Using deflection (w) and rotation transformation. Thus a triangle described in natural
(θ) as independent field variables, however, partly co-ordinate is not affine equivalent to one in
eases the requirement of C1 continuity and this has led Cartesian co-ordinate. Bell (1969) has proposed a
to development of the discrete Kirchhoff triangle modified TUBA element by expressing the normal
(DKT) element. Here, the condition of zero transverse derivative at each mid-point as a linear
shear strain is satisfied at some discrete points in the combination of first and second derivatives of the
element along the edges. The basic idea is to compute function at the corresponding end points.
the shear strains not directly from the derivatives of the
displacements but at a few discrete collocation points Mindlin-Reissner plate theory accounts for shear
using the displacements alone. This allows for a deformation. An advantage of the Mindlin-Reissner
continuous representation through their interpolation model over the biharmonic plate model is that the
over the element(s) via appropriate shape functions. energy involves only first derivatives of the
unknowns and so the so-called conforming FE
approximations require, in principle, only C0 shape
functions instead of their C1 counterparts necessary
for the biharmonic model. However Mindlin-
Reissner plate elements exhibit shear locking as the
thickness of the plate approaches zero. Shear
locking is characterized by the numerically
generated incorrect transverse forces under
bending. Early methods have tried to overcome
shear locking by reduced integration or a selective
Fig. 1 TUBA 6 and TUBA 15 elements reduced integration. The idea has been to split the
strain energy into two parts, one due to bending
Efforts at enforcing C1 continuity of shape function and the other due to shear, and use different
across inter element boundary have been made by integration rules for the terms involving bending
taking higher order derivatives of displacement field as and shear strains. Hughes (1982) has developed a
degrees of freedom (DOF-s). The TUBA family of linear triangle (i.e. a 3-noded triangle using linear
plate elements (Argyris et al.1968) is an example of shape functions) element using the so-called one
such elements. These elements are based on complete point centroidal quadrature for integration.
polynomial displacement functions of order n ≥ 5. Quadratic triangular elements have also been

explored to overcome this problem. Sze et al. (1997) tetrahedralization for domain discretization (as in a
have developed T6/3 * changing the quadrature rule for class of mesh-free methods) whilst adopting
integration over original T6/3(Sze et al. 1997). polynomial reproduction for deriving the shape
Zhongnian (1992) has reported a 3-noded triangular functions (as in the FEM), have been shown to
element with 3 basic DOF-s per node and 2 internal possess superior convergence and locking-resistant
rotational DOF-s (11 DOF-s) using selective reduced characteristics vis-à-vis the FEM for a fairly
integration on the shear energy term. Reduced general class of linear and nonlinear elasticity
integration results in zero-energy modes and instability problems. The aim of this work is to extend the
due to rank deficiency. The enhanced assumed strain DMS-FEM for Mindlin plate bending problems
(EAS) method can be used to avoid locking and assess its performance for different orders of
phenomena as well (Simo and Rifai 1990). Sze et al. polynomial reproduction. DMS-Splines (DMS
(1997) have developed an AST-6 element by being an acronym for Dahmen, Micchelli and
optimizing sampled strains of T6/3*. The element Seidel, authors who introduced the spline; Dahmen
T6/3-B3 (Zienkiewicz and Lefebvre 1988) has 3 et al. 1992) are essentially weighted sums of
bubble heterosis nodes as rotational DOF-s and 6 simplex splines over triangles. DMS-Splines of
boundary nodes. TRI-6 is an assumed strain quadratic order n are Cn-1 continuous across the inter-
element that does not have a bubble node. MITC7 triangular boundaries. In developing the plate
(Bathe et al. 1989) is another quadratic triangular bending element, no numerical artifices, typically
strain element having a bubble heterosis node. Wang et employed in the FE literature, have been used.
al. (2004) have traced shear locking in Mindlin- Numerical results are compared with those via
Reissner plate formulations to the inability of the AST-6 and MITC7, two prominently used plate
approximation functions to reproduce the Kirchhoff bending elements allowing for shear deformation.
mode, and to the inability of the numerical method to The resulting element, which is practically free
achieve pure bending exactness (BE) in the Galerkin from thickness locking and spurious zero energy
approximation. In their study, the Kirchhoff mode modes, also shows conspicuously better accuracy
reproducing condition (KMRC) is ensured for than its FE counterparts.
Mindlin-Reissner plates. The approximation functions
for displacements and rotations are constructed to meet DMS-FEM Shape Functions
the KMRC. They have derived integration constraints DMS-splines are weighted sums of simplex splines,
for achieving BE and proposed a curvature smoothing which are a multivariate generalization of the well-
method (CSM) to meet the bending integration known univariate B-splines. A degree n simplex
constraints. spline consists of a family of smooth, degree n
piecewise polynomial functions each defined over
Mesh-free methods developed more recently allow a set of n  N dim  1 points, an element
adequate flexibility in customizing the approximation
Ndim
functions for the desired smoothness. In mesh-free xR of which is called a knot and the set of
methods with stabilized conforming nodal integration knots, the knotset. Here N dim denotes the domain
(SCNI) (Chen et al. 2001), the domain is discretized
into cells that in a sense define the field of nodes (such dimension (presently we have N dim  2 ). Simplex
as the cells of a Voronoi diagram). Integration is n 1
splines have C continuity if no three knots are
performed along the edges of each cell. Although
mesh-free methods such as the element-free Galerkin collinear in the knotset. The domain of DMS-
(EFG) obtain good accuracy and high convergence splines is a proper triangulation. A degree n DMS-
rates, the non-polynomial approximation space spline basis function at a point x  R 2 is defined
increases the computational overhead while evaluating I
over the control points c and is given by
the shape functions and the weak forms. Recently, the
so called smoothed finite element method (SFEM), an
interface of the SCNI with the FEM, has been applied   I
det V  M  x | VI  ,
to two dimensional elasticity problems (Liu et al.
2007). It is shown that the SFEM is stable, accurate
where a M x | VI   is simplex splines of degree n
and effective. I
and V , a triangle consisting of the end knots of

Other recent attempts at establishing such handshake the knotclouds of the three vertices of the triangle I,
strategies that combine the FEM and mesh-free corresponding to a control point cI , i.e.
methods include the NURBS-based smooth parametric
methods (Shaw and Roy 2008, Shaw et al. 2008) and a
smooth FEM based on triangular B-splines or DMS-
I

V   v0I
0
v I
1
vI
, 1 , 2
2  . The number of control
splines (Sunilkumar and Roy 2010a, 2010b; ( n  1)( n  2)

Sunilkumar et al. 2011). The latter class of methods points is equal to in R2 .
(DMS-FEM), which typically use triangulation or 2

Depending on the knot length which is typically a Eq. 6 can be written as M ( x, y )b ( x, y )  H (0)
small fraction of the smallest triangular side, a DMS- where,
spline is nearly supported over a triangle (more M ( x, y ) 
precisely, it is supported over the polygon formed by Nnd
connecting the knots located close to the vertices of the H T

( x  xi , y  yi ) ( x  xi , y  yi ) H ( x  xi , y  yi ) is
i 1
triangle). Every DMS-spline is constructed
corresponding to a nodal point x in the triangle to the the moment matrix and the coefficient vector is
half-open convex hull of which x belongs. In general, thus given by:
DMS-splines satisfy the partition of unity and their
derivatives (including the splines themselves) are The global shape function in 2D is finally given by:
globally smooth. Nevertheless, it has been  i ( x, y ) 
conclusively demonstrated (Sunilkumar and Roy H T ( x  xi , y  yi )M 1 ( x, y ) H (0) ( x  xi , y  yi ) (5)
2010a) that functional approximations directly based
on these splines could be inaccurate owing to their MINDILIN PLATE
sensitive dependence on the knot locations. Such Consistent with the basic postulates of the
sensitivity is largely overcome by introducing a Reissner-Mindlin plate theory, one describes the
polynomial reproduction strategy while constructing
the shape functions, with DMS-splines acting as
displacement fields, in-plane  u,v  and
kernels in the integral functional representation. transverse  w  , purely through rotations ( x ,  y ) ,
Specifically, let the shape functions be required to
reproduce a complete set of polynomials up to and the transverse displacement of the mid-layer as
 (Fig. 2):
degree p  n . Now, the approximant u  x, y  for a
u  z x ( x, y); v  z y ( x, y); w  w( x, y) (6)
target function u  x , y  in a bounded domain Ω is
given in the standard form:
N nd
u ( x, y)    i ( x, y )ui (1)
i 1
where N nd represents the number of nodes in Ω,

ui  u  xi , yi  is the function value at particle i and
Fig. 2 Kinematics of Mindlin plate deformation
 i  x, y  are the globally smooth shape functions
Strains on planes parallel to the middle layer are
given by: thus given by:
i (x, y)  HT (x  xi , y  yi )b(x, y)(x  xi , y  yi ) (2)      
 x  z x ;  y  z y ;  xy  z  x  y  (7)
T
H  x, y  is a set of monomials defined as x y  y x 
x y   
|   | p
, b ( x, y ) are coefficient functions and those in the transverse direction are obtained
as:
determined below and   x  xi , y  yi  is the DMS-
 w   w 
spline based at  xi , yi  acting as the kernel function.  xz   x   ;  yz   y   (8)
 x   y 
The coefficients b ( x, y ) are obtained based on the For isotropic, homogeneous materials, bending
following reproduction conditions: moment resultants are thus derivable as:
Nnd M  DL (9)
 ( x, y)  1
i
(3) T
i 1 Where M T   M x , M y , M xy  ;    x ,  y  ;
Nnd
 
 ( x, y )( x y  )  x y 
i i i |    | p (4) 1  0  
0

i 1 Et 3  ; 
 x .
From Eq. 4 D 2 
 1 0 
12(1  )   
Nnd
1   L   0 
y 
 (x, y)  x  x   y  y      
 
|    | p (5) 0 0  
i1
i i i | || |,0
 2   
 
From Eqs. 2 and 5  y x 
N nd Here E and  are Young’s modulus and Poisson’s
T
H
i 1
( x  xi , y  yi )b( x , y ) ( x  xi , y  yi ) H ( x  xi , y  yi ) ratio, t the plate thickness. Accounting for similar
(6) material constitution, the transverse shear force
 H (0)
resultants are given by:

S     w ;    Gt . (10)
Here G is the shear modulus of the material and  the
shear correction factor (value of  is presently taken
to be 5 6 ). From the moment-shear and linear
momentum balance we get:
LT DLθ +  (w + θ )  0 (11) (a) DMS order 3 (b) DMS order 2
T  (w   )   q  0 (12) Fig. 3 Control points and knotcloud
subject to appropriate prescriptions of boundary Thus taking variation in Eq. (15) and substituting
conditions. This is an irreducible system corresponding Eq. 17, we obtain discretized form as:
to minimization of total potential energy.
1  t3 T T 
 ( , w)   (L )T D(L )d    B b DBb d     B s B s d    e
2   12 
  (w   )T  (w   )d  (13)  NT qd    NT Sd  (16)
  
where q is applied load vector and S the applied
  wqd    bt
 traction vector fields. This can be written as:
(15)
K e  f (17)
DISCRETIZATION AND NUMERICAL where K is the stiffness matrix given by:
IMPLEMENTATION
t3
Proper domain triangulation is presently ensured by K= BTb DBb d     BTs B s d 
12 
(18)
Delaunay triangulation. Number of control points 
3 In this formulation K is integrated using seven
 pn  in R required for a triangle corresponding to
point gauss quadrature rule. Given the non-
(n  1)(n  2) polynomial rational nature of the DMS-FEM shape
order n DMS-spline is . Their functions, use of such larger number of quadrature
2 points is only expected and an optimal number of
locations define where the control points influence
such points must be found a-priori through some
more. Projections of control points coincide with the
numerical experiments.
nodes of pn noded Lagrange triangle. Consistent with
the mesh-free literature, projections of control points NUMERICAL RESULTS
on triangles may be called particles. Their functionality In the first example, a clamped square plate of side
is the same as that of nodes in the FEM. For order 2 L and with varying L to t ratio is subjected to a
DMS-splines, the total number of particles required is central point load. Given the symmetry, only a
6 and for order 3 it is 10. Particles and knots for DMS- quadrant of the plate is modeled by 8 elements, see
spline of order 3 are shown in Fig. 3. Fig. 4. The central deflections, normalization by
The field variables are interpolated as thin plate solution (Timoshenko 1959) as predicted
 e1  by the DMS-FEM and compared with the AST-6
 .  element for varying length to thickness ratio, are
 w   presented in Table 1. The solutions via the DMS-
  (14)
x   1I3 . . .  pn I3   .   Ne FEM  n  3 not only show higher numerical
   . 
 y   accuracy (in terms of their closeness to the solution
corresponding to the thin plate limit), but also
epn  exhibit significantly reduced deterioration as L/t
where, I3 is third order identity matrix, and increases.
T
 ei   wi  xi  yi  is the vector of kinematic
degree of freedoms at the ith node.
Substituting Eq. 16 in Eq.9 and Eq.10 we get
x 
    xz  (15)
  x   z B b e    B s e
    yz 
 xy 
Fig. 4 Triangulation of a Quadrant of square

Table 1 Normalized central deflection of clamped
square plate subjected to central point load
L/t ratio AST-6 DMS-FEM
(n=3)
102 0.872 0.961
103 0.712 0.923
104 0.707 0.913
105 0.707 0.913
For the second set of examples, again consider a

square plate of L t  1000 . As before, only a
quadrant is meshed using Delaunay triangulation (Fig.
3). In the first case, the plate is assumed to be
subjected to a central point load and, in the second,
uniform pressure. Both simply supported  w = 0  and Fig. 6 Simply supported square plate with
uniformly distributed loaded
fully clamped boundary
conditions  w = 0; θ x = 0; θ y = 0  are considered. Table 2 Deflection of a square plate
For numerical work with the DMS-FEM, we have used ( L t  1000 ) via DMS-FEM  n  3
shape functions corresponding to both n  2 and no of simply supported Fully clamped
elements udl point udl point
n  3 . The predicted deflections at the centre, load load
normalized with respect to the thin plate limit as 18 1.0000423 0.979437 0.923741 0.916524
before, are listed in Tables 2 and 3. While the DMS- 36 1.0004906 0.988949 0.985732 0.970960
FEM  n  2  shows a slightly poorer performance 62 1.0006939 0.998245 0.996485 0.994884
106 1.0004430 0.996421 0.998643 0.993420
via-a-vis the MITC7 or AST-6 elements (Fig. 5),
236 1.0004265 1.000017 0.999874 1.001465
results via DMS-FEM  n  3 shows significantly
Table 3 Deflection of a square plate ( L t  1000 )
improved accuracy in comparison with both of these
elements (Fig. 6). via DMS-FEM  n  2 
no of simply supported Fully clamped
For the third and final example, we adopt Morley’s elements udl point load udl point
skew plate model, i.e. a skew rhombic plate with skew load
26 0.966640 0.915473 0.222531 0.24941
angle 30 , all sides simply supported and subjected 36 0.968799 0.918116 0.384413 0.40163
62 0.985270 0.960369 0.664072 0.71643
uniform pressure. The same is analyzed using DMS- 106 0.994097 0.974748 0.873208 0.88119
FEM  n  3 . Central deflections, as normalized by 236 0.997920 0.990156 0.967651 0.96645
thin plate series solution of Morley (1963) and

compared with the solutions based on the AST-6
element, are shown in Fig. 7. DMS-FEM clearly
outperforms the AST-6 scheme once again.
Fig. 7 Morley’s skew plate
CONCLUDING REMARKS
The smooth DMS-FEM, which enables C1 or still
Fig. 5 Simply supported square plate with a central higher order continuity uniformly over a bounded
point load geometrical domain of interest, is extended and
explored for Mindlin plate bending problems.
While the domain discretization is via
triangulation, bearing close similarity with the one

employed within the standard FE framework, the Wang, D. & Chen, J. S. (2004) “Locking-free
generation of the polynomial reproducing smooth stabilized conforming nodal integration for
shape functions requires the additional usage of knots, meshfree Mindlin-Reissner plate formulation”
whose placement must satisfy a non-collinearity Computer Methods in Applied Mechanics and
constraint, marks a departure from the usual FE mesh. Engineering, 193(12-14), 1065 - 108
While computing the DMS-FEM shape functions Nguyen-Xuan, H., Rabczuk, T., Bordas, S. &
(typically needed at the quadrature points) is Debongnie, J. (2008) “A smoothed finite
computationally involved, the extra numerical work is element method for plate analysis” Computer
amply compensated through a significant improvement Methods in Applied Mechanics and
in convergence and accuracy of solutions via-a-vis Engineering, 197(13-16), 1184 - 1203
those via some of the best known finite elements, e.g. Sunilkumar, N. & Roy, D. (2010) “A Smooth
AST6 or MIT7 for Mindlin plates. As evidenced Finite Element Method Based on Reproducing
through the presented numerical work, this Kernel DMS-Splines ” CMES: Computer
distinguishing feature of the DMS-FEM is more Modeling in Engineering & Sciences, 65(2),
pronounced as the thin plate limit is approached. Work 107-153
is currently under way to extend to method to Morley, L.S.D. (1963) “Skew Plates and
nonlinear plates and shells, including formulations Structures”, Pergamon Press, Oxford.
involving finite rotations and geometrically exact Timoshenko, S. & Woinowsky-Krieger, S (1959)
kinematics. “Theory of Plates and Shells” , McGraw-Hill,
New York
REFERENCES Zienkiewicz, O. C. & Taylor R. L. (2000) “The
Batoz, J.L., Bathe, K.J. & Ho, L.-W.( 1980) “A study Finite Element Method volume 2: Solid
of three-node triangular plate bending elements” Mechanics” Butterworth-Heinemann, Oxford
International Journal for Numerical Methods in
Engineering, John Wiley & Sons, Ltd, , 15(12),
1771-1812
Lee, S. W. & Zhang, J. C. (1985) “A six-node finite
element for plate bending”
Engineering, John Wiley & Sons, Ltd, 21, 131-143
Papadopoulos, P. & Taylor, R. L. (1990) “A triangular
element based on Reissner-Mindlin plate theory”
Engineering, John Wiley & Sons, Ltd, 30(5),
1029-1049
Sze, K. Y., Zhu, D. & Chen, D.P. (1997), “Quadratic
Triangular C0 Plate Bending Element”
Engineering, John Wiley & Sons, Ltd, 40(5), 937-
951
Zienkiewicz, O., Taylor, R., Papadopoulos, P. &
Oñate, E. (1990) “Plate bending elements with
discrete constraints: New triangular elements”
Computers & Structures, 35(4), 505 - 522
Argyris, J. H., Fried, I. & Scharpf, D. W. (1968) “The
TUBA family of plate elements for the matrix
displacement method” Aeronautical Journal, 72,
701-709
Zhongnian, X. (1992) “A thick-thin triangular plate
element” International Journal for Numerical
Methods in Engineering, John Wiley & Sons, Ltd,
33(5), 963-973
Zienkiewicz, O. C. & Lefebvre, D. (1988) “A robust
triangular plate bending element of the Reissner-
Mindlin type” International Journal for
Numerical Methods in Engineering, John Wiley &
Sons, Ltd, 26(5), 1169-1184
Bathe, K. J., Brezzi, F. & Cho, S. W.(1989) “The
MITC7 and MITC9 Plate bending elements”
Computers & Structures, 32(3-4), 797 - 814

ISBN 978-81-921121-0-7
ENHANCING THE REACTIVITY OF FLY ASH FOR

GEOTECHNICAL AND GEOENVIRONMENTAL APPLICATIONS
P.V. Sivapullaiah
Professor, Department of Civil Engineering, Indian Institute of Science, Bangalore – 560012 Email: siva@civil.iisc.ernet.in
ABSTRACT: The use of abundantly available fly ash needs to be promoted to avoid disposal problems and
associated environmental hazards. Most bulk uses of fly ashes depend on its pozzolanic nature. Enhancing the
pozzolanic reactivity can enhance its use for bulk geotechnical and Geoenvironmental applications. The use of
gypsum, apart from lime, can decrease not only its lime leachabiity to sustain and enhance its pozzolanic
reactivity which inturn enhances its strength and volume change behaviour. This can also reduce its hydraulic
conductivity one of the main limitation for its application as barrier material for waste disposal facilities.
Amending gypsum to fly ash reduces the leachability of trace elements present in fly ash and improves the
retention capacity for different contaminants in leachates.
PRODUCTION OF FLY ASH

Current energy policies worldwide consider thermal their dams or bunding will be rapid and on a massive
power generation as a major source of power for scale. Some trace elements in coal are naturally
industrial development. Fly ash generated from the radioactive. These radioactive elements include
combustion of coal presently comprises about 900 uranium (U), thorium (Th), and their numerous decay
million tons every year. With the commencing of products, including radium (Ra) and radon (Rn) etc.
super power thermal power plants and the increasing However, the vast majority of coals and most of the
use of low grade coals with high ash contents, the fly ashes are not significantly enriched in radioactive
production of ash currently is about one thousand elements, or in associated radioactivity.
million tons per annum in India alone.
TYPES OF ASH
ENVIRONMENTAL CONCERN OF FLY ASHES
Management and disposal of such huge quantities of Fly Ash
fly ash are bound to create tremendous environmental It consists of inorganic, incombustible matter present
stress unless the utilization rate of fly ash is not in the coal that has been fused during combustion into
increased by a significant order of magnitude. It can a glassy, amorphous structure. Fly ash material gets
cause all the three environmental risks - air, surface solidified while suspended in the exhaust gases and is
water and ground water pollution. Air pollution is collected in the electrostatic precipitators. The lighter
caused by direct emission of toxic gases from the one, which goes up the chimney and gets collected
power plants as well as wind-blown ash dust from ash either in mechanical or electrostatic precipitators, is
mound/pond. The air-borne dust can fall on surface known as fly ash. Some portion of fly ash escapes
water system or soil and may contaminate the along with hot gases through chimney. Since the
water/soil system. The wet system of disposal in most particles are solidified while suspended in the exhaust,
power plants causes discharge of particulate ash they are generally spherical in shape and range in
directly into the nearby surface water system. The sizes from 0.5 micron to 100 micron. Fly ash is such a
long storage of ash in ponds under wet condition and complex material that its characterization in detail is
humid climate can cause leaching of toxic metals quite difficult. With the type of fly ash its physical,
from ash and contaminate the underlying soil and chemical and engineering properties vary due to the
ultimately the groundwater system. As fly is fine and following factors.
light, it can easily spread by wind into the regions  Type of coal used: Fly ashes, produced from
leeward sides of the ash dumps. In these regions fly burning anthracite or bituminous coals have
ash will be everywhere on the beds, dining tables and pozzolanic properties. Fly ashes produced from
of course in human eyes, ears, and lunges. lignitic or sub bituminous coals have
The health hazards and environmental impacts cementitious properties in addition to the
result from the mobilization of toxic elements from pozzolanic ones.
ash. Presence of fly ash in lunges may cause serious  The treatment to which the coal has been
lung diseases like silicosis, fibroses, bronchitis, subjected prior to combustion.
pneumonitis, etc. Mobilization of various elements  The method of combustion.
from the ash into the environment depends on climate,  Furnace temperature
soils, indigenous vegetation and agricultural practices.  Amount of air circulation
Bulk quantities of fly ash are usually stored wet  Collection and storage places adopted
rather than dry so that fugitive dust is minimized. The
resulting impoundments (ponds) are to be typically
large and stable for long periods, but any breach of

Bottom Ash of original coals and combustion temperature
Bottom ash refers to the non-combustible constituents (Hubbard and Dhir, 1985). The mineralogy of fly ash
of coal with traces of combustibles embedded in also depends on the combustion methods. The boiler
forming clinkers and sticking to the hot side walls of a operating at high temperatures tends to have a large
coal-burning furnace during its operation. The amount of crystalline phase. Thus the dry bottom fly
clinkers fall by themselves into the water or ash has higher content of crystalline phase than that of
sometimes by manual poking, and get cooled. the wet bottom one. This also results in lower
pozzolanic activity of the dry bottom fly ash than that
Pond Ash of the wet bottom fly ash comparing at the same
Fly ash and bottom ash are often sluiced into ponds particle size. Fly ash can be mineralogically divided
and lagoons. Such heterogeneous materials collected into three major categories.
in pond are referred to as pond ash. Bottom ash 1. Amorphous Glass
accounts for nearly 20% and fly ash for 80% of the 2. Mullite (3Al2O3.2SiO2) – Quartz (SiO2) and
total ash produced. Magnetite Spinel (Includes magnetite (Fe3O4),
hematite (Fe2O3), ferrite, and γ-Fe2O3 (El-Mogazi et
PROPERTIES OF FLY ASH al., 1988; Liem et al., 1983). The range of
Ash properties may also vary within the same boiler mineralogical composition and the glass content
at various times in response to varying demands. The found are summarized in Table 2.
chemical and mineralogical compositions and
physical and morphological properties of fly ashes are Table 2 Mineralogical composition of fly ashes
summarized in the following sections. Mineral Range, weight percent
Quartz 2-10
Chemical Composition Mullite 5-15
The elemental composition of fly ash is highly
Hematite 1-3
variable and is directly related to the source of coal,
its pretreatment and the operation of the plant while Magnetite 0.5-3
burning the coal. Fly ash is an alumino silicate glass Glass content 50-90
consisting of the oxides of Si, Al, Fe and Ca with
minor amounts of Mg, Na, K, Zn and S and various Category two and three are crystalline phases.
trace elements. The principal constituents of fly ashes They collectively account for 25 to 30% of
are silicon dioxide (SiO2), aluminum oxide (AI2O3), composition (Liem et al., 1983). Amorphous glass
iron oxide (Fe2O3), calcium oxide (CaO) and carbon material predominates in most of the fly ash samples
content. Ash also contains smaller amounts of MgO, (Tazaki et al., 1989), and it gives fly ash its
TiO2, Na2O and K2O; and very small quantities of pozzolanic properties. The finest fraction is the richest
other 20 to 50 elements. The concentration associated in glass. The proportions of crystalline minerals -
with the ash may be either adsorbed on the surface of quartz, mullite, magnesite and hematite in high
particle or incorporated into matrix (Natusch et al., calcium (CaO > 15%) fly ashes are much smaller than
1974). The range of chemical composition of Indian in low (CaO < 5%) or medium calcium (CaO, 5 -
coal ashes is reported in Table 1. 15%) fly ashes.
Table 1 Range of chemical composition of Indian Morphology

coal ashes The morphology of a fly ash particle is defined as the
Compounds Fly Ash Pond Ash Bottom Ash form or structure of the particle. Knowledge of
SiO2 38-63 37.7-75.1 23-73 particle morphology can be quite helpful in predicting
Al2O3 27-44 11.7-53.3 13-26.7 the physical properties and leaching behavior of the
TiO2 0.4-1.8 0.2-1.4 0.2-1.8 fly ashes.
Fe2O3 3.3-6.4 3.5-34.6 4-10.9
MnO 0-0.5 bd-0.6 bd-0.3
MgO 0.01-0.5 0.1-0.8 0.1-0.7
CaO 0.2-8 0.2-0.6 0.1-0.8
K2O 0.04-0.9 0.1-0.7 bd-0.56
Na2O 0.07-0.43 0.05-0.31 bd-0.3
LOI 0.2-3.4 0.01-20.9 0.61-51.6
bd: below detection; LOI: loss on ignition
Mineralogy
The mineralogy of fly ash depends to a great extent on
its chemical composition. The phases of minerals Fig. 1 SEM Image of Fly Ash Particles
which exist in fly ash are governed by the impurities Fly ash particles are classified into eight different

morphological groups based on the microscopic predominantly of sand-size fraction with some silt-
examination and chemical analysis. size fraction. Based on the grain-size distribution, the
Fly ash is made up of clear glass spheres and rounded coal ashes can be classified as sandy silt to silty sand.
spongy glass particles. The clear glass spheres are They are poorly graded with coefficients of curvatures
generally smaller than other particles. Some particles ranging between 0.61 and 3.70. The coefficient of
have both small and large bubbles and range from uniformity is in the range of 1.59–14.0.
foam to cenospheres. Cenospheres are hollow spheres
filled with nitrogen and carbon dioxide. Opacity in Fineness
spheres is due to the presence of magnetite and ASTM C 618-08 (2008) uses the fineness for
hematite. Fly ashes may also contain plerospheres specifying ash as Class F and C. It specifies that not
which are hallow spheres filled with smaller spheres more than 34% of fly ash particles shall retain on 45
(Fig. 1). m sieve for class F and C fly ashes. However, more
information is given by particle size distribution
PHYSICAL PROPERTIES curve.
Specific Gravity
Specific gravity values of fly ashes vary considerably Surface Area
from 1.7 to 2.9 unlike soils. The variation in specific The surface area of particles is important because it
gravity values arise due to the changes in boiler load, controls the total adsorption capacity to some extent.
firing conditions, differential settlement in lagoons in The surface areas of fly ash particles generally vary
lagoon ash, degree of pulverization of coal, design of inversely with the particle size (i.e. the smaller the
the furnace etc. Specific gravity of fly ash depends particle, the larger the surface area). Particles over 74
considerably upon its iron content, which increases µm the shape deviates from the spherical one and
the specific gravity and carbon content, which takes irregular sponge like shape which in turn
decreases the specific gravity. Because of the increases the surface area. The larger particle sizes are
generally low values for the specific gravity of coal usually carbonaceous which indicates they are likely
ash compared to soils, ash fills tend to result in low result of incomplete combustion.
dry densities. The reduction in unit weight is of
advantage in the case of its use as a backfill material Pozzolanic Reactivity
for retaining walls since the pressure exerted on the In almost all geotechnical applications pozzolanic
retaining structure as well as the foundation will be property of ashes plays an important role. It is the
less. The other areas of application include main property of the agent, which differentiates it
embankments especially on weak foundation soils, from silt.
reclamation of low-lying areas, etc. The variation of ASTM C 618-08 (2008a) has classified fly ash for use
specific gravity of the coal ash is the result of a as a mineral admixture in portland cement concrete as
combination of many factors such as gradation, Class F and Class C fly ashes. Class F fly ashes shall
particle shape and chemical composition (Gray and be essentially pozzolanic fly ashes. Class C fly ashes
Lin, 1972). Coal ash comprises mostly of glassy shall be essentially self pozzolanic fly ashes.
cenospheres and some solid spheres (Prashanth, Pozzolanic reactivity is influenced by - particle size
1998). The reason for a low specific gravity could distribution, fineness, particle morphologies,
either be due to the presence of large number of mineralogical composition and chemical composition.
hollow cenospheres from which the entrapped air Based on pozzolanic property, fly ashes can be
cannot be removed, or the variation in the chemical classified as self pozzolanic, pozzolanic and non
composition, in particular the iron content, or both. pozzolanic. Self pozzolanic or self hardening fly ashes
are those which produce cementitious compounds and
Colour thus harden with time in the presence of water,
Fly ash spans a colour range of light tan to grey to without external addition of lime. Pozzolanic fly ashes
black. Increased carbon content causes a darker grey - are those, which produce cementitious compounds on
black tone, while increased iron content imparts tan the addition of lime. It is the lime content which
colour to ash. differentiates between self pozzolanic and pozzolanic
Particle Size Distribution fly ashes. Non pozzolanic fly ashes do not produce
80 to 99% of fly ash sample by weight passes 74 µm sufficient cementitious compounds even on addition
screen size. Fly ash is generally fine grained and falls of lime (Sivapullaiah et al. 1995). Reactive fly ashes
within the silt range (2 µm to 74 µm). Fly ashes are those, which react with lime to give significant
disposed off by wet ash disposal system and deposited amount of strength. Non reactive fly ashes do not give
near the discharge area, show more than 50% fine sufficient strength even on the addition of lime. Self
sand range of particles. Fly ashes, which are deposited pozzolanic and pozzolanic fly ashes are reactive fly
away from the discharge unit show more than 50% ashes whereas non pozzolanic fly ashes are mere non
silt range particles. In pond ashes silt-size fraction reactive fly ashes Sivapullaiah et al., 1998).
predominates with some sand-size fraction. The Particle size distribution plays an important role
bottom ashes are coarser particles consisting in determining the activity of the fly ash. The particles

in the size range of about 5 to 30 m play a decisive
role in determining the activity of fly ash. Smooth Compaction Characteristics
spheres of sizes ranging from 2 to 25 m are The density of coal ashes is an important parameter
considered to be more reactive. For low calcium fly since it controls the strength, compressibility and
ashes, the reactivity is found to be directly permeability. In general, fly ashes show considerable
proportional to the number of particles smaller than variation in compaction due to variation in the nature
10 m, and inversely proportional to that greater than of fly ash produced from the same power plant over
45 m. However, sole evaluation of fraction passing time (Yudhbir and Honjo, 1991; Winter and Clarke,
through a 45 m sieve does not by itself appear to be 2002). The variation of dry density with moisture
fully appropriate. Particle size distribution alone may content for fly ashes is less compared to that for a
not reflect the potential reactivity of a fly ash though well-graded soil, both having the same medium grain
the finer the fly ash the higher is the pozzolanic size (Moulton, 1978). Relatively higher void ratios
reactivity. Linear relationship exists between specific tend to limit the buildup of pore pressures during
surface area of fly ash and its pozzolanic reactivity. compaction, allowing the fly ashes to be compacted
But the relation is found to be linear only in a limited over a larger range of water content (Toth et al.,
range. 1988). The dry density values of fly ashes are
Fly ashes which contain mainly spherical particles, influenced by particle shape, particle distribution and
ranging between 2 to 25 m and predominantly the low unit weight of solid particles (De Santayana
smooth glassy surface without any deposition but with and Mazo, 1994). For fly ashes having a wide range
only adherence, are highly reactive fly ashes. Fly of specific gravities compaction curves expressed on a
ashes which contain mainly spherical particles having volume basis, viz., volume of solids per unit volume
of compacted sample or porosity or void ratio and
size of 2 to 35 m and surface deposition and some
volume of water per unit volume of solids, can better
sintering are considered to have moderate reactivity.
serve to explain the compaction behavior (Prashanth,
Fly ashes which contain mainly irregular shaped
1998). The dry densities in coal ashes are relatively
particles (being relatively coarser) along with small
less sensitive to compaction water content compared
amount of spherical particles, are considered to have
to soils, hence the field compaction can be carried out
poor reactivity. But the rough textured porous glass
at dry side of the optimum taking care of the dust
spheres possess larger surface areas for reactivity than
problem.
smooth spheres and the particle size distribution alone
may not reflect the potential of reactivity of a fly ash.
Shear Strength Characteristics
The crystalline minerals - quartz, mullite, magnesite
The shear strength of fly ash is affected by the density
and hematite present in fly ash are non reactive at
and moisture content of the test sample, with
ordinary temperatures. The crystalline minerals like
maximum shear strength exhibited at the optimum
CS, C3A, C4A3S etc. present in fly ash are highly
moisture content (Lamb, 1978). Pozzolanic fly ash
reactive. High calcium fly ashes generally contain
has been observed to have a friction angle that is
minerals like CS, C3A, C4A3S. The amount of SiO2 or
usually in the range of 26° to 42. Most of the shear
SiO2 + Al2O3 in fly ash influences the pozzolanic
strength is due to internal friction (Singh and Panda,
activity for long periods of curing (Sivapullaiah et al.,
1996). For fly ash specimens complete loss of
1998). Higher ultimate strengths are obtained when
cohesion occurs on full saturation whereas the angle
the silica or silica plus alumina content is high. The
of shearing resistance remains unchanged (Indraratna
reactivity of a particular fly ash can be quite a
et al. 1991). However pozzolanic fly ashes improved
complex phenomenon that cannot be related to only
strength and do not loose strength due to saturation
its SiO2 + Al2O3 + Fe2O 3 content and lime plays a
after curing.
very important role (Sivapullaiah et al., 1998).
Hydraulic Conductivity Behavior
ENGINEERING PROPERTIES OF FLY ASHES
Hydraulic conductivity is an important parameter in
Among the various uses of fly ash, its bulk utilization
the design of liners to contain leachate migration,
is possible only in geotechnical engineering
dykes to predict the loss of water, embankments as
applications. The challenge for the scientific
well as the stability of slopes and as a sub-base
community is to exploit these applications, as low
material. The hydraulic conductivity of well-
tech materials such as fly ashes help in solving large-
compacted fly ash has been found to range from 10-4
volume societal-environmental needs. The various
to 10-6 cm/s, which is roughly equivalent to the
engineering properties that necessitate the
normal range of permeability of silty sand to silty clay
applicability of fly ashes in many civil engineering
soils. The hydraulic conductivity of fly ash is affected
applications in general and in geotechnical
by its density or degree of compaction, its grain size
applications in particular are: compaction
distribution, pozzolanic activity and its internal pore
characteristics, shear strength, hydraulic conductivity
structure (Rajasekhar, 1995; Pandian and
and compressibility characteristics.
Balasubramonian, 1999).
A brief review of these properties is presented in the
Compressibility Characteristics
following sections.

Compressibility characteristics of coal ash depend on
its initial density, degree of saturation, self-hardening Fly ash as Valley Fill
characteristics and pozzolanic activity. Consolidation Fly ash can be used extensively with the use of geo-
occurs more rapidly in compacted fly ash than in silty textiles to allay environmental concern.
clay soil because partially saturated fly ashes are less
compressible compared to fully saturated ones Fly ash Fly ash in Dams
has a higher void ratio and greater permeability than Fly ash can form the downstream shell of the dam
the soil. The age-hardening can reduce the time rate of supporting an impervious core. Fly ash is protected by
consolidation, as well as the magnitude of the filter and drainage layers to ensure dry condition. The
compressibility. upstream shell is of a rock fill to provide stability
under rapid draw down condition. Fly ash occupies
FLY ASH UTILIZATION almost 50% of the gross volume of the embankment.
The beneficial applications of fly ashes based on their Downstream slope of fly ash is protected by soil
favorable properties are: cover.
 Fly ash has been used as a fill material in
embankments, behind retaining walls and as a Stabilisation of Soils
reclamation material. Fly ash has been extensively used for stabilization of
 Fly ash is capable of adsorbing various compounds soils such as expansive soils. Pozzolanic fly ashes
present in leachates (Fly ash is used to stabilize cement soil particles and improve their strength
heavy metal contaminated soils (Due to its light considerable. Generally the strength of soil increases
weight it is used for the production of light weight with fly ash content and curing period. While self
aggregates, pozzolanic fly ash develops strength without lime
 It is used as sub-base material in road construction addition pozzolanic fly ashes improve the strength of
It is used as a decolorizing agent for dye effluents. soil in presence of lime. Initial moulding water
 It is used in the manufacture of clay flooring / content plays an important role.
terracing.
 It is used in the manufacture of sintered glass Fly ash as Interposing Layer
ceramics. CNS soil is used as interposing layer between
 Due to its pozzolanic nature, it has been used for expansive soil and foundation to reduce the swelling
the partial replacement of cement in concrete; in potential of the foundation soil. Recently attempts
the manufacture of high strength and low drying have been made to use fly ash with minimum amount
shrinkage concrete; in the manufacture of workable of expansive soil as an interposing layer.
and abrasion resistant concrete and chloride
resistant concrete Stabilisation of Contaminated Soils
 It is used in mitigation of acid mine drainage It is known that contamination of soils can alter the
caused by the reactive mine tailings. engineering behaviour of soils to various extents
 It is used along with lime in stabilizing soil bocks depending on the type of soil, concentration and
which are used in building construction. duration of interaction. It has been shown that an
 It is used as an ameliorant in increasing the crop increase in swelling and a decrease in compressibility
productivity. of both swelling and non swelling soils can occur due
to contamination with NaOH or acid sulphate. This
GEOTECHNICAL ENGINEERING has been attributed to free energy changes due
APPLICATIONS mineralogical alterations due soil alkali/acid
About 65% of ash generated is used for geotechnical interaction. Detailed studies have shown that fly ash is
applications. The uses are: very effective to reduce contaminant effects (Fig. 2,
Sivapullaiah and Hari, 2009).
Fly Ash in Embankment and Sub-Base 30
To minimize the potential for environmental problems

25
with structural embankments, it is necessary to restrict
the rate of water permeation of fills. For this purpose 20
% Swell - H/H *100
granular lift berm should b provided. The berm and

the top soil also prevents the erosion of fly ash due to 15
wind. Preventing ground water intrusion is equally BC soil + 2N NaOH

BC soil + 25%NFA + 2N NaOH
important. Capillary break layer should be provided. It 10 BC soil + 50%NFA + 2N NaOH
NFA + 2N NaOH
is also necessary to prevent shifting of fly ash into
5
coarse material with a filter cloth geo-textile.
0
Building Foundation using Fly Ash 0 20 40 60 80
Time (days)
100 120 140
To minimize deeper percolation, fly ash is used along

Fig. 2 Effect of Fly Ash on the alkali induced swell in
with an earthen or synthetic barrier.
Soil

Fly ash as Landfill Liner Ca  2  Al2 O3  H 2 O  x CaO. y Al2 O3 . z H 2 O
Fly ash because of some unburnt carbon posseses Eq. 2
sorption capacity. Fly ash is non swelling, non Al O
shrinking and non compressible material Ca2  2 3 Si O H2 O  xCaO. y Al2 O3. z SiO2 .wH2 O
2 3
(Sivapullaiah and Lakshmikanth, 2004). The high Eq. 3
permeability of fly ash can reduced with addition of The hydroxide derived calcium silicates
low permeability soil. Fly ash with about 20% of contribute to additional strength to the stabilised fly
expansive soil can be used as landfill liner. Fly ash ashes. The surfaces of nucleating fly ash particles
bentonite mixture is stable in alkaline solutions.
provide sites for Ca (OH )2 , C-S-H, and other
FACTORS LIMITING THE USE OF FLY ASH hydration products to precipitate, thereby leading to a
Fly ash can be either an industrial waste material and more homogeneous distribution of hydration products
an ecological nuisance, or a valuable raw material. (Xu and Sarkar, 1994). The effect of formation of
The various factors which limit the usage of fly ashes these compounds with curing at different lime
in various civil engineering and geotechnical contents on the lime leachability behavior with
applications are listed below: continuous passing of leaching fluid (millipore grade
 The problem of leaching of trace elements from water) is presented in the subsequent paragraphs.
fly ashes into nearby aquifers which in turn The surfaces of nucleating fly ash particles
contaminate the ground water. Hence, the use of provide sites for Ca (OH )2 , C-S-H, and other
fly ash as landfill or road bed material is hydration products to precipitate, thereby leading to a
considered as a serious ecological concern. more homogeneous distribution of hydration products.
 Most of the fly ashes generated in India often do Formation of these compounds which increase with
not meet the requirements for effective the curing period reduces lime leachability ratio which
performance in various geotechnical applications is taken as the ratio of lime leached to the total lime
as they are not self cementing even though they added. The strength gain is sustainable only if the
are pozzolanic in nature. Hence, in order to lime leachability is minimized. Lime leachability from
increase their pozzolanic activity certain fly ashes is due to their sustained leaching under the
cementing agents are added to them. But often, it prevailing hydrological field conditions and their
is observed that the cementing agent itself is continuous exposure to fresh water. Lime is sparingly
prone to leaching making the stabilized matrix soluble in water and dissociates into calcium and
vulnerable. hydroxyl ions and the concentration of hydroxyl ions
 All the potential fly ash handling resources have can only give a maximum pH value of 12.6. More and
now reached their maximum level of utilization more lime is dissociated only if the produced calcium
viz., in cement industry as a pozzolanic and or hydroxyl ions are used in pozzolanic reactions
admixture, stockpiling near thermal power or these ions are diffused away into the water. Thus if
stations etc. Hence, there is a need to explore the the ions are diffused away before they react with silica
other possible uses of fly ash. of fly ashes to produce calcium silicate hydrate gel,
The reuse of fly ash as an engineering material lime is wasted and the pozzolanic reactivity gets
primarily stems from its pozzolanic nature, spherical inhibited affecting the performance of fly ashes in
shape, and relative uniformity. many applications.
Hence, the lime leachability studies on fly ashes
ENHANCEMENT OF PROPERTIES OF FLY amended with lime contents in the range of 1 to 10%
ASH are conducted. The addition of lime triggers the onset
The strength development in class F fly ashes occurs of pozzolanic reaction early, by increasing the
significantly with the addition of lime. solubility of silica as it breaks the Si-O bonds in the
Ca (OH )2 is consumed by the pozzolanic reactions silica rich glassy phases of fly ash. The hydration of
resulting in the formation of various gel forms of fly ash begins immediately with the depolymerization
calcium silicates, aluminates and alumino silicates of glassy phases releasing alumina and silica. This in
which gradually crystallize with time. The hydroxide turn produces cementing compounds such as calcium
derived calcium silicates contribute to the additional silicate gel which gradually crystallizes with time
strength of the stabilised fly ashes. attributing strength to fly ash particles. The nature
With curing, Ca (OH )2 is consumed by the and amount of compounds formed vary both with the
amount of lime added and duration of curing period.
pozzolanic reactions resulting in the precipitation of It has been seen that the lime reaches
various calcium silicates, aluminates and equilibrium value at any given lime content after
aluminosilicates. The addition of lime activates the passing of water for two days. Further it is observed
silica and alumina phases in the following manner: that for both the fly ashes, the equilibrium
concentration of calcium in the leachate was found to
Ca  2  SiO2  H 2 O  x CaO. y SiO2 . z H 2 O increase with increase in lime content. This leaching
Eq. 1 concentration of lime is not in proportion to the

amount of lime added. Hence the amount of hardens the matrix. The amount of lime reached is
equilibrium concentration of lime leached is relatively more for NFA compared to that for MFA.
expressed as ratio of lime in the leachate to the total This might be due to the higher free lime content
amount of lime added. This ratio is observed to initially present in NFA than in MFA. Also coarser
decrease with increase in the lime content. This NFA particles leach relatively more amount of
indicates that lime is converted from more soluble calcium ions.
form to less soluble forms of calcium silicate It has been shown from the leachability of
hydrates. This confirms that relatively more lime is calcium from fly ashes with different lime contents
actually utilized for pozzolanic reactions with silica that the ratio of calcium leached to the total calcium
content of fly ash to produce cementitious added is an indication of the conversion of free lime
compounds. into the less soluble forms. The ratios of calcium
leached to the total amount of calcium added are
Role of Gypsum calculated for both the fly ashes with different lime
Attempts have hence been made to reduce the contents and are compared with the addition of
leachability of lime by incorporating gypsum. The gypsum at different curing periods. The ratio
role of gypsum in lime-silica reactions has been drastically reduces with the inclusion of gypsum. Also
analysed in detail and based on these studies the there is a noteworthy reduction in the ratio at higher
mechanism of change is explained as follows: lime and gypsum contents. This marked reduction in
Gypsum is known to accelerate the gain in strength by the ratio in the presence of gypsum further confirms
altering the course of hydration of calcium silicate the alteration of hydration products formed, compared
which predominantly occurs in the early hydration to those formed with the addition of lime alone.
stages (Sivapullaiah and Baig, 2011). Apart from Qualitatively similar trends occur for both the fly
calcium silicate hydrate, calcium aluminate hydrate is ashes at different lime and gypsum contents.
also formed favoring the hydration of fly ash. Addition of gypsum aids in the release of sulphate
Addition of gypsum aids in the release of sulphate ions which react with alumina phase of fly ash to
ions which react with alumina phase of fly ash as produce x CaO. y Al2 O3 . z Ca SO4 . w H 2 O . In the
shown in equation 4 below: presence of gypsum the enhanced pozzolanic activity
of lime stabilised fly ashes results in a material denser
Ca2SO4 2Al2O3H2O  xCaO. yAl2O3. zCaSO4.wH2O with higher strength. It has been shown that the
Eq. 4 formation of calcium silicate hydrate along with the
formation of calcium aluminate hydrate occurs and
Hence, in the presence of gypsum the enhanced this reduces the lime leachability of fly ashes. The
pozzolanic activity of lime stabilised fly ashes results leachability of lime from fly ashes reduces
in denser, higher strength material. The manner in significantly in the presence of gypsum. The steep
which the formation of calcium silicate hydrates along reduction in the presence of gypsum is due to the
with the formation of calcium aluminate hydrate formation of calcium aluminate silicate hydrate apart
affects the lime leachability of fly ashes has been from the calcium silicate hydrate. With increase in
studied and the effect of addition of 1 and 2.5% curing period, the amount of lime leaching from
gypsum contents on the lime leachability values of gypsum treated lime stabilized fly ashes reduces to a
specimens cured for 7 and 14 days have been considerable extent. The effects of the reduction in the
discussed below. lime leachability on the strength, CBR values,
It has been shown that the concentrations of calcium compressibility, and hydraulic conductivity have been
in the leachate for the lime stabilized fly ash samples elaborated.
at the end of 7 days of curing period amended with It is made clear that the addition of gypsum increases
1% and 2.5% gypsum contents reach equilibrium the strength of fly ashes at any lime content. At lower
within two days of passing water through the samples curing periods the increase in strength with gypsum
in the mould. It is seen that fly ashes with gypsum and low lime contents is quite significant. Increase in
leach out lower equilibrium amounts of lime strength is observed after a considerable period at
compared to fly ashes without any gypsum at any higher lime contents. Thus the increase in strength at
given lime content, even though the amount of higher lime and gypsum contents after curing for
calcium present in the matrix increases. The lime more than 60 days is found to increase by 40 times.
leachability decreases with increase in gypsum Fly ash which responds readily to lime stabilization
content. The reduction in the leaching of calcium is shows accelerated gain in strength due to the addition
due to the alteration of the nature of pozzolanic of gypsum at early curing periods. Further the
reaction compounds producing calcium aluminate increase in strength achieved with gypsum is least
silicate hydrate which minimizes further leaching of susceptible to the effects of alternate wet and dry
calcium from the matrix. With increase in curing cycles and to soaking in different solutions containing
period to 14 days, the lime leachability values are heavy metals.
further reduced, which confirms that, the cementation
products formed at higher curing period further

heavy metals. Further to study the role of pH
particularly at acidic levels, which can become
important when they are leached continuously due to
the prevailing hydrological conditions, the retention
capacities of fly ashes have been studied by batch
equilibrium tests. The retention by precipitation is far
more than adsorption and hence retention of heavy
metals at higher pHs is more than that at lower levels
(Fig. 4).
100.000
pH 2
pH 4
10.000 pH 6
Metal Extraction (%)

pH 8
pH 10
1.000
Fig. 3 Effect of Lime and Gypsum on the Hydraulic

Conductivity of Fly Ash 0.100
Addition of gypsum is found to be much more

beneficial than increasing lime content in gaining
0.010
unconfined compressive strength. Though the strength Pb Cd Zn Mn Cu
improvement with additives is one of the important Fig. 4 Effect of Gypsum on the Leachability of Trace
factors in reducing the compressibility, other factors Elements from Fly Ash
such as particle size distribution also control the
compressibility behaviour leading to a large scatter in Along with the increase in pH as the retention
the relationship between the reduction in the transforms predominantly from surface adsorption to
compressibility and reduction in lime leachability complex metal precipitation, the kinetic rate constants
ratio. increase rapidly. At every pH the equilibrium sorption
of metal ions is best described by Langmuir isotherm.
Increased additions of lime reduced the hydraulic Since the amount of metal ion retained increases
conductivity of fly ashes but gypsum along with lime proportionately with the increase in initial
is more effective than lime alone in reducing the concentration of the metal ion in solution, the
hydraulic conductivity of fly ashes due to the percentage removal remained constant. The retention
enhanced hydration and accelerated formation of order for all metal ions studied is the same for both
cementitious compounds. Addition of more than 1% the fly ashes at all the pH values indicating that the
of gypsum is effective only with the higher lime adsorption is predominantly by silica surfaces than by
contents. The reduction in the hydraulic conductivity alumina or iron oxide surfaces.
increases with increase in curing period. Significant Along with very favourable improvement in strength,
reduction in hydraulic conductivity values that occurs reduction in compressibility and hydraulic
in the presence of gypsum at high lime contents of 5 conductivity with the addition of gypsum to fly ashes,
and 10 percent is due to the combined actions of their low leachability of trace elements and significant
enhanced hydration, formation of cementitious adsorption capacity at different environmental
compounds and the precipitation of unreacted lime conditions make them important barrier materials.
(Fig. 3). Even though the hydraulic conductivity of fly It has been broiught out the role of gypsum in
ashes is reduced many folds with the addition of lime modifying the nature of the pozzolanic compounds in
and gypsum, a general correlation is still observed to fly ash lime reactions, reducing greatly the lime
exist between the hydraulic conductivity values and leachability and at the same time enhancing most of
the lime leachability ratios (Sivapullaiah and Baig, the geotechnical properties along with their durability
2010). which enlarges their potential for bulk applications.
Of the many industrial leachates heavy metal
contaminants are the most important ones. Thus the SUMMARY
retention capacity of fly ashes with respect to heavy Many uses of fly ashes in geotechnical engineering
metals needs to be established in their application as such as reclamation of low lying area, construction of
barrier materials. It is well known that two dominant road, dikes and dams are mainly based on pozzolanic
mechanisms responsible for the retention of metal reactivity, which is due to its reactive silica. There is
ions by fly ashes are precipitation and adsorption. Fly an optimum lime content for each fly ash for
ashes studied had enough of lime to keep the pH of maximum reactivity. More applications can be found
the fly ash system sufficiently high to precipitate by enhancing the reactivity of fly ashes. Gypsum

which enhances the reactivity of fly ashes improves Sivapullaiah P.V. and Arif Ali Baig M. (2011). Role
its strength, CBR value, volume stability is also found of Gypsum in the Strength Development of F Fly
to reduce its hydraulic conductivity. Gypsum Ashes with Lime, ASCE Journal of Materials in
amended fly ash can be used as barrier material for Civil Engineering, Vol. 23, 197-206.
waste impound materials. Gypsum is found to reduce Sivapullaiah PV, Lakshmikantha H (2004) Properties
the lime leachability of fly ashes and enhance the of Fly Ash as Hydraulic Barrier. Soil and Sediment
retention capacity of various trace elements present in Contamination, 13(5), 489-504.
the lechates. Sivapullaiah PV, Prashanth JP and Sridharan A
(1995) Optimization of Lime Content for Fly Ash.
REFERENCES Journal of Testing and Evaluation, ASTM, 23(3),
De Santayana FP and Mazo CO (1994) Behavior of 222-227.
Fly Ash in Experimental Embankments. Sivapullaiah PV, Prashanth JP, Sridharan A and
Proceedings of 13th International Conference on Narayana BV (1998) Reactive silica and strength
Soil Mechanics and Foundation Engineering, New of fly ashes. Journal of Geotechnical and
Delhi, 1603-1606. Geological Engineering, 16, 239-250.
El-Mogazi D, Lisk DJ, and Weinstein LH, (1988) A Sivapullaiah, P. V. and Hari Prasad Reddy, P. (2009),
Review of Physical, Chemical and Biological “Fly Ash to Control Alkali Induced Volume
Properties of Fly Ash and Effects on Agricultural Changes In Soils” Ground Improvement, 162,
Ecosystems. The Science of the Total pp.167-173.
Environment, 74, 1-37. Tazaki K, Fyfe WS, Sahu KC and Powell M (1989)
Gray DH and Lin YK (1972) Engineering Properties Observations on the nature of fly ash particles.
of Compacted Fly Ash. Journal of the Soil Fuel, 68(6), 727-734.
Mechanics and Foundations Division, ASCE, 98, Toth PS, Chan HT and Cragg CB (1988) Coal ash as
361–380. structural fill with special reference to Ontario
Hubbard FH and Dhir RK (1985) A Compositional experience. Canadian Geotechnical Journal, 25,
Index of the Pozzolanic Potential of Pulverized- 694–704.
fuel Ash. Journal of Materials Science Letters, 3, Winter MG and Clarke BG (2002) Improved use of
958-960. pulverised fuel ash as general fill. Proceedings of
Indraratna B, Nutalaya P, Koo KS and Kuganenthira the Institution of Civil Engineers, Geotechnical
N (1991) Engineering behavior of low carbon, Engineering, 155(2), 133-141.
pozzolanic fly ash and its potential as a Xu A and Sarkar SL (1991) Microstructural study of
construction fill. Canadian Geotechnical Journal, gypsum activated fly ash hydration in cement
28, 542-555. paste. Cement Concrete Research, 21, 6, 1137–
Lamb DW (1978) Ash disposal in dams, mounds, 1149.
structural fills, and retaining walls. Proceedings of Yudhbir and Honjo Y (1991) Application of
the Third International Ash Utilization geotechnical engineering to environmental control.
Symposium, Pittsburgh, PA, 262-280. Proceedings of the 9th Asian Regional Conference
Natusch DFS, Wallace JR and Evans Jr. CA (1974) on Soil Mechanics and Foundation Engineering,
Toxic trace elements: preferential concentration in Bangkok, 2, 431–469.
respirable particles. Science, 183 (4121), 202-204.
Pandian NS and Balasubramonian S (1999)
Permeability and consolidation behaviour of fly
ashes. Journal of Testing and Evaluation, ASTM,
27, 337–342.
Prashanth JP (1998) Evaluation of the properties of
fly ash for its use in geotechnical applications. Ph.
D. Thesis, Indian Institute of Science, Bangalore.
Prashanth JP, Sivapullaiah PV, Sridharan A (2000)

Pozzolanic fly ash as a hydraulic barrier in
landfills. Engineering Geology, 60, 245-252.
Rajasekhar C (1995) Retention and permeability
characteristics of clays and clay–fly ash systems
subjected to flow of contaminants. Ph.D. Thesis,
Indian Institute of Science, Bangalore.
Sivapullaiah P.V., and Arif Ali Baig M.
(2010).Gypsum Treated Fly Ash As A Liner For
Waste Disposal Facilities, , Waste Management 31,
359-369.

ISBN 978-81-921121-0-7
SORPTION BEHAVIOUR OF DEEP SEATED COAL: A GEO-

ENGINEERING APPROACH TO CARBON DIOXIDE SINKING
V. Vishal
IITB-Monash Research Academy, Indian Institute of Technology Bombay, Mumbai – 400076, India
T. N. Singh
Department of Earth Sciences, Indian Institute of Technology Bombay, Mumbai – 400076, India, Email: v.vishal@iitb.ac.in
ABSTRACT: The rate of generation of greenhouse gases has increased in the past century with the rapid rate of
industrialization. This has eventually led to an increase in the global average surface temperature. With a still higher
rate of industrialization and subsequent release of carbon dioxide, the global temperature is expected to rise further
in the current century. It has been identified by the Intergovernmental Panel on Climate Change (IPCC), that apart
from switching to low carbon energy with advanced technologies, it is important to consider the geological
sequestration of greenhouse gases into deep reservoirs for preventing the release of these gases into the atmosphere
as an attempt to greenhouse remediation. There are various probable sink spaces in the underground for storing
CO2. Of other sequestration spaces, coal beds offer a value added product of methane, the recovery of which can be
enhanced by injecting CO2. Understanding coal characteristics in terms of its sorption capacity is vital to ensure that
the process of storing is safe and to quantify the volume of gases involved. In this paper, an overview of the sorption
behaviour of coal is discussed with the perspective of sequestration of greenhouse gases in it along with enhanced
recovery of clean, low carbon, energy i.e. methane.
Keywords: CO2 sequestration, Coal Bed Methane (CBM), coal, sorption
INTRODUCTION
The advent of geoengineering to meet with the growing as Coal Bed Methane (CBM) is a low carbon, green
challenges in several aspects for sustainable energy, when burnt converts to less potent
development has come a long way in the recent years. greenhouse gas, CO2. CBM has witnessed a phase
One of the major questions ahead of the scientists, when it was done away with to prevent any fire
researchers and policy makers is that of the gradual problems during coal mining to the current state of
increase in the mean global surface temperature of the development to the level of a mainstream fuel. Coal
Earth in the past two centuries. At the same time, the also offers a safe sinking space for the waste CO2 for
demand of energy for large scale development of long term storage underneath and this has been opted
various nations is also growing at a rapid rate. This for prevention of CO2 release in the air. Thus, a dual
growth will be directly related to the release of benefit in utilizing deep seated coals, for producing
pollutants, especially green house gases, as a methane and in place storing CO2, makes greenhouse
continuation of the recent trend of development. IPCC gas sinking in these, the most preferred option. Indian
(2005, 2007) recognized that the global warming has coals at greater depths, especially from the
taken place with a significant increase in the Gondwana coalfields have substantial amounts of
concentration of greenhouse gases like carbon dioxide methane and may be utilized for the geological
in past century and the two are correlated. There comes sequestration of CO2 for enhanced recovery of CBM
the question of prevention of release of these (ECBM).
greenhouse gases into the atmosphere to let the Earth
maintain its temperature and prevent such a dramatic Coal behaviour in CO2 – ECBM Process
rise. Methane is generated as a byproduct during the
Coal exists in huge quantities in India. It not only offers coalification process of physic-chemical alteration of
huge energy base to India but also adds to the peat as a consequence of bacterial decay, heat action
maximum pollution due to release of enormous and in due course of time. Coal is formed and the
quantities of CO2 especially from the coal based power process and degree of maturation produces coals of
plants. In this context, coal also contains significant various ranks. The conversion of peat to coal is
quantities of methane depending on several factors like associated with generation of hydrocarbons as a
rank of coal, temperature, pressure conditions of byproduct, namely methane. Coal seam gas (CSG)
horizon, moisture content, etc. The methane, referred to typically contains more than 95% methane other than
fewer quantities of higher hydrocarbons or some
CO2. Methane is produced from these coal reservoirs

ISBN 978-81-921121-0-7
if the content is in good volumes. However, production system and enables gas molecules to stay in free state
behaviour of coal reservoir has to be properly than in adsorbed state. The gas adsorption also varies
understood before the attempt to tap the fuel. Coal as the pH of the coal seam system. Azmi et al. (2006)
characteristics are quite different from the conventional experimented on Malaysian low rank coals and
geological reservoirs. Coal acts as the source as well as obtained that acidic treated samples have a higher
the reservoir for the hydrocarbon. Coal exhibits a dual CO2 gas adsorption rate followed by alkali and finally
porosity model with the fine spaces within the matrix neutral conditions, although under in-situ conditions,
blocks representing the micropores while the fractures if CO2 sequestration is applied, the high range of
or cleats system constituting the macroporous structure variations in pH values make it inconclusive to
of coal. The flow of gases in micropores is different incorporate the pH dependent variability in sorption
from that in the cleats. The gases in the micropores in coal. Their research also brought forth the fact that
flow by diffusion, something like that described by adsorption rate depends on particle size of adsorbent.
Fick’s Law while they follow a Darcian model during Fine grained solids have a larger surface area and
flow across the cleats. Coal seams are usually saturated hence enable more adsorption than coarse matrix.
with water content and when a well is dug, dewatering
starts and water continues to be pumped out from the Geo-engineering of CO2 – ECBM process
seam. Excessive dewatering causes depressurization Sorption of gases
which leads to the release of methane from coal. The Production of gas from coal seams will involve
gases are not stored in the pores spaces by means of excessive dewatering that causes depressurization of
compression as in case of conventional reservoirs. They the seams. This eventually induces the release
are stored on to the porous surface of the coal by methane from the micropores that flow through the
mechanism of adsorption. Coal has affinity towards cleat system to reach the production wells. Slowly as
CO2 over CH4, the degree of affinity varying from 3 to the pressure declines, production of methane from the
10 times depending on several parameters. This is an coal also decreases. In such a case, CO2 may be
advantage in storing CO2 inside the coal seams which injected into the reservoir to maintain the reservoir
may lead to enhanced CBM recovery. pressure that would lead to further release of
methane. CO2 is preferentially adsorbed by the coal
FACTORS CONTROLLING SORPTION and being denser, it settles into the porous spaces and
CHARACTERISTICS OF COAL finally methane production is enhanced from the
The gas content that is stored in the adsorbed state in block. This process seems advantageous for
coal is affected by various control parameters. As the production of fuel and in place sequesters the
hydrocarbons are a product of coalification, the hazardous greenhouse gas to be stored for long
characteristics are highly affected by the degree of periods of time. This geoengineering approach for
maturation. Low rank coals are formed under lower P-T green house remediation has gained attention
range and usually contain lesser gas content. worldwide and needs to be worked out for Indian
Experimental works by several researchers has scenario. It is important to understand the
established that the adsorption capacity increases with petrophysical characteristics along with the
increasing rank for moisture saturated coals while it geotechnical properties of the reservoir rocks, the
decreases for dry coals (Briggs and Sinha, 1933, relative permeability of the fluid phases, sorption
Reucroft and Sethuraman, 1987, Yee et al., 1993, Levy, characteristics of coal, insitu P-T conditions, etc. The
1997, Busch et al., 2003, Vishal, 2007). The content of sorption characteristics of coal for various gases are
macerals and their types also control the adsorption expressed as a function of pressure at a constant
behaviour of coal. More commonly accepted theory is temperature, called as adsorption isotherms.
that there is a significant increase in the gas sorption Adsorption isotherms are used to study the gas
capacity with increase in vitrinite content of coal, storage capacity of coal with respect to gas pressures
particularly in high rank coals (Levine, 1993, (or concentrations) at a particular temperature.
Lamberson and Bustin, 1993, Crosdale et al., 1998 and The adsorption isotherms of methane on coal are
many others). However, Ettinger et al. (1966) had most commonly described using Langmuir equation
shown the inertinite-rich coals having the greatest (1916) which is based on the dynamic equilibrium
methane adsorption capacity. The gas adsorption between the adsorbent and adsorbate as a function of
capacity of coal will depend much upon the insitu pressure or concentration of the adsorbate. Langmuir
pressure- temperature conditions of coal seams (Yee et isotherm is produced when the extent of adsorbing
al, 1993, Levy et al., 1997, Bustin and Clarkson, 1998, molecules coverage is believed to be limited to one
Schroeder et al., 2001, Azmi et al., 2006). While higher molecular layer (Type I).
pressure helps more gas to remain adsorbed in place, The equation for the Langmuir isotherm for a single
higher temperatures increase the kinetic energy of the gas phase is given as:

ISBN 978-81-921121-0-7
p Jharkhand area is discussed in this article. The input

V  VL (1)
pL  p parameters are taken mostly from the field and
laboratory studies while a few standard curves like
where V is the volume of gas adsorbed per unit mass in
that of relative permeability model have been
the coal matrix and p is the pressure of the system. VL is
assumed. The Langmuir constants have been taken
the Langmuir volume constant while pL is Langmuir from the published works from Central Institute of
pressure constant. VL represents the maximum volume Mining and Fuel Research. These parameters have
of gas that a unit mass of coal can adsorb onto its been input to model the scenario of a 800m deep coal
matrix surface at saturation pressure. pL is the pressure seam chosen for simultaneous release of methane
at which the storage capacity of the coal is equal to one- along with carbon-dioxide sequestration using
half of the Langmuir volume VL. These constants are
COMET3 simulator.
material specific and are determined by best fitting of
The disposition of gases has been monitored and the
desorption data. spatial distribution of both CO2 and CH4 is shown in
However, in case of gas mixtures, the equation is Figure 1 and 2, respectively after 200 days and 2000
modified as individual gases are said to compete for the days. The adsorption pattern of both the gases,
same sorption sites. Arri et al (1992) established the
namely CO2 and CH4 is monitored throughout the
extended Langmuir equation for gas mixtures as:
study period. However, the two diagrams are simply
ri
p (2) a representation of the spatial distribution of the two
 p L i
V i  V L i gas phases. The CO2, injected from the centrally
rj
1 p located well is adsorbed by the coal instantly and the
j p L j rate of spread of gas all along is slower than the
migration of methane. The injected CO2 pushes
NUMERICAL APPROACH methane by displacing it from the sorption sites and
Modeling of coal beds for CO2 driven ECBM recovery finally, the gas is driven towards the production well
is carried out after understanding the complex reservoir which enhances the recovery of methane. This is a
phenomena. Key features in such a model should study to quantify the volume of gas produced in turn
include: heterogeneity and anisotropy in coals, dual of the volume of CO2 sequestered in it. Further study
porosity (micropores and macropores), transport of establishes that over a period of 6000 days, the
gases by diffusion (governed by Fick’s law) and flow in chosen rock strata will adsorb 73.83 bcf of CO2 and
cleats (governed by Darcy’s law), interaction of CO2 in place it produces a total of 2.71 bcf of CBM. The
and CBM, coal matrix shrinkage due to desorption of results indicate that the rate of water production in
gas and swelling due to adsorption of gas, stress the initial period of study is very high (around 43
dependency of the whole system, variations in the BPD) while it starts declining at the end of 400 days
permeability of coals throughout the process and the and finally it wanes at the rate of 1.2 BPD).
utility of extended Langmuir’s isotherm for estimation
of volumes of gases involved. Several reservoir CONCLUSIONS
simulators have been designed to model the flow Characteristics of coal are highly variable from one
characteristics of CBM reservoirs. The simulator should basin to the other and also within the same basin.
be multi-component, multi-phase, multi-porosity It is vital to understand the sorption behavior of
simulator capable of modeling underground fluid-rock the reservoir for sinking CO2 alongside
environment in coal capable of encompassing the coal- production of methane. Gas desorption rate is
cleat matrix represented by a well defined dual porosity controlled mainly by the pressure conditions and
system as described by Warren and Root (1963). the relative saturation of gases. Experimental as
Numerical study provides solves complex equations well as simulation studies are required at various
that describe some physical phenomena using iterations scales to accurately predict the rock response in
and approximations. The approach is important to context to sequestration. CO2 sinking in coal can
investigate the several probabilities indifferent be visualized as a “value added” sequestration
scenarios and solve complex problems using numerical where in the waste gases are utilized to tap the
models. However, the setting up of model and its valid fuel. Sorption/desorption induced
establishment to the real scenario is very important and swelling/shrinkage in coal for sorptive gases like
extremely critical. It should reflect the real conditions CO2 and CH4 needs to be worked out for Indian
with minimum limitations. Hence, in case of coal scenario. Moreover, the gas molecules compete
reservoir, this approach is highly important to monitor for the same sorptive sites and it is vital to
and quantify the flow of gases and its response thereof understand the degree of preference of coal for
onto the surrounding. A brief study on the coal either of the gases, which will eventually control
sorption behaviour for Gondwana coal seams in

ISBN 978-81-921121-0-7
PW PW
IW IW
Fig. 1a PW Fig. 1b PW
Fig. 1: Matrix methane concentration in the area over a period of 200days and 2000days respectively (IW=Injection
Well, PW = Production well)
PW PW
IW IW
Fig. 2a PW Fig. 2b PW
Fig. 2: Matrix carbon dioxide concentration in the area over a period of 200days and 2000days respectively
(IW=Injection Well, PW = Production well)

ISBN 978-81-921121-0-7
the storage capacity of the reservoir. Therefore northern British Columbia: effect of maceral
sorption of gases onto coal surface is a critical composition. AAPG Bullelin Vol. 77, pp. 2062–
phenomenon and it controls the entire 2076.
storage/production of gases from coal seams. Langmuir, I., 1916. The constitution and fundamental
properties of solids and liquids. Journal of the
REFERENCES American Chemical Society, Vol. 38, pp. 2221-
Azmi A.S., Yusup, S., Muhamad, S., 2006 The 2295.
influence of temperature on adsorption capacity Levine, J.R., 1993. Coalification: the evolution of
of Malaysian coal; Chemical Engineering coal as source rock and reservoir. In: Law, B.E.,
Processing, Vol. 45, pp. 392-396. Rice, D.D. Eds.., Hydrocarbons from Coal.
Arri, L. E., Yee, D., Morgan, W. D. and Jeansonne, American Association of Petroleum Geologists,
M. W., 1992. Modelling coalbed methane AAPG Studies in Geology, Vol. 38, pp. 39–77.
production with binary gas sorption. SPE 24363, Levy, J.H., Day, S.J., Killingley, J.S., 1997. Methane
SPE Rocky Mountain Regional Meeting, Casper capacities of Bowen Basin coals related to coal
Wyoming. properties. Fuel, Vol. 76, pp. 813–819.
Briggs, H. and Sinha, R. P., 1933. Expansion and Reucroft P.J. and Sethuraman A.R., 1987. Effect of
contraction of coal caused respectively by the Pressure on Carbon Dioxide Induced Coal
sorption and discharge of gas, Proceedings of the Swelling. Energy & Fuels, Vol. 1, No. 1, pp. 72-
royal society of Edinburgh, Vol. 53, pp. 48-53. 75.
Busch, A., Gensterblum, Y., Siemons, N., and Vishal, V., 2007. Coal Bed Methane: An
Krooss, B.M., 2003. Investigation of preferential Introduction, Indian Journal of Earth Sciences,
sorption behaviour of CO2 and CH4 on coals by Vol. 33, Nos. 1-4, pp. 76-79.
high pressure adsorption/desorption experiments Warren, J. E., Root, P. J., 1963. “The Behavior of
with gas mixtures. International Coalbed Naturally Fractured Reservoirs,” Society of
Methane Symposium, University of Alabama, Petroleum Engineers’ Journal, pp. 245-255.
Tuscaloosa, Alabama, 5-9 May, paper 0350. Yee, D., Seidle, J.P., Hanson, W.B., 1993. Gas
Bustin, R. M., and Clarkson, C. R., 1998. Geological sorption on coal and measurement of gas content
controls on coalbed methane reservoir capacity in hydrocarbons from coal. AAPG Studies in
and gas content. International Journal of Coal Geology, Vol. 38, pp. 203–218.
Geology, Vol. 38, pp. 3-26.
Crosdale, P.J., Beamish, B.B. and Valix, M., 1998.
Coalbed methane sorption related to coal
composition. International Journal of Coal
Geology, 35, 147–158.
Ettinger, I., Eremin, B., Zimakov, Yavovskaya, M.,
1966. Natural factors influencing coal sorption
properties: I. Petrography and the sorption
properties of coals. Fuel 45, 267–275.
IPCC, 2005. IPCC Special Report on Carbon Dioxide
Capture and Storage. Prepared by the Working
Group III of the Intergovernmental Panel on
Climate Change [Metz, B., O. Davidson, H. C.de
Conninck, M. Loos, and L.A. Meyer (eds.)].
Cambridge University Press, Cambridge, UK
and New York , USA , pp . 442.
IPCC, 2007. Climate Change 2007, The physical
science basis. Contribution of Working Group I
to the Fourth Assessment Report of the
Intergovernmental Panel on Climate Change
[Solomon, S., D. Qin., M. Manning., Z. Chen.,
M.Marquis., K.B. Averyt., M. Tignor and H.L.
miller (eds)]. Cambridge Univ. Press,
Cambridge, UK and NY, N.Y. USA.
Lamberson, M.N. and Bustin, R.M., 1993. Coalbed
methane characteristics of gates formation coals,

ISBN 978-81-921121-0-7
CHARACTERISTICS OF BADLANDS MORPHOLOGY

Sanjay Tignath
Professor,Govt. Autonomous Science College, Jabalpur-482001, M. P , India. tignathsanjay@rediffmail.com
Medha Jha
Assistant Professor, Institute of Technology, Banaras Hindu University, Varanasi, 221005 U. P., India
.jha_medha@rediffmail.com
ABSTRACT: Badlands manifest a systematic morphology laterally and vertically. An investigation of

alluvial badlands of Narmada in Narsinghpur district of Madhya Pradesh (India) led to a rational description of
their formation and an understanding of their preference in natural choices of angle and spacing. Badlands
represent a storey-like morphology. Proper understanding of this morphology may be best utilized for
developing riparian water ponds and on slope agriculture practices.
INTRODUCTION
The development of badland topography has been Despite the above, there still exists less than complete
attributed to headward erosion (Blong, 1966) of rills understanding on the subject of badland formation
and gullies (Tuckfield, 1964; Yi-Fu Tan, 1966), and and its pattern development.
creep on slopes (Horton, 1945) propped up by bank
collapse (Schumm, 1956a, 1956b). Topographic MORPHOLOGY
hollows are common in the sandy-silty alluvial plains Study was carried out along the Narmada River of
with gentle slopes. The erosive phenomenon central India through Narsinghpur district. The
gradually develops a network of ravines by cutting badland forms were surveyed along the main river, its
and deepening below the general ground level of the major tributaries like Sher, Omar and Balurewa, and
original accretional plain. The common slope forms its local tributary system generating badlands on the
of ravines constitute a small convex upper slope and Narmada alluviums.
a straight slope below it, leading down to channel
(Hack, 1960), and where concave slopes exist, these Storey-forms of Badlands
are often channelized seasonally. Badlands exhibit a typical storeyed morphology.
The present day concepts of gullying are primarily Simplest badlands show only single storey with
founded on the researches carried out in various parts respect to the channel bed of the master river and the
of the world during 60’s and 70’s. The works of latter may be designated as zero level/floor. Like-
Blong (1970), Berry (1970), Deshmukh, D.S. et.al wise upper floors may be designated first floor,
(2009),Lusby (1970), Gregory (1977a), Butzer second floor and so on. Each floor is an active level
(1971), Harris & Vita-Finzi (1968), and Yi-Fu Tan of gullying. However, it is noticeable fact that the
(1966) are worth citing among several others. Natural isolated mounds of alluviums rising above the levels
causes of badlands formation include climatic approach at the extreme to be encircled by hexagonal
changes in rainfall and variation in temperature along pattern of gullies-cum-channels. In this regard, the
with alluvial type. Anthropogenic causes are more floors are different from an erosional bench which
severe and include the change in land use and land may be recognizable along the master river. Each of
cover, including changes in agricultural practices, the successive floors of a badland may not be
deforestation, overgrazing, burning for vegetation represented in a single profile across the banks.
clearance, etc. Floored morphology of the badlands often collects
Using the simplest configuration of convex and rainfall for some time during the monsoon period or
concave segments, Smith and Bretherton (1972) rainy season. Shallow pools form on the floors due to
proposed a landscape forming transport law also scouring action of run-off water on the floor from the
applicable to badlands. The magnitude of sediment immediate mounds. It may form multilevel short-
discharge per unit width at any point is a function of lived pools. However, this typical form of badlands
the magnitude of the local gradient, S, and the processes can be positively utilized in land and soil
magnitude of discharge of water per unit width, q. conservation with added advantage of improved and
Expressed mathematically, qs = f (S, q), where f sustained discharge in the stream.
represents the ‘function of’ and qs is the magnitude of
discharge of sediment per unit width. They also Patterns and spacing
presented the concept of critical distance, Xc, from The most common type of pattern and spatial
top to inflection point along the profile, where arrangement that channels develop in the badlands by
perturbations grow into a rill like feature, and as headward erosion (Fig. 1) exhibits a regular spacing
these deepen, the intervening ridges, if less than Xc in of channel (as discussed later) with frequent
width, are left without formation of rills. occurrence of confluence angles of 300, 600, 900, and
1200, as shown in Fig. 2 (derived from Fig. 1)

depicting the frequency of angles for badlands of the
quaternary-recent alluvial plains of Narmada in the 60o
90o
Narsinghpur district of Madhya Pradesh in India.
From observations and as suggested by the natural
choice of angles, these channels appear to grow in
such a fashion that leads to a nearly hexagonal
pattern, which, however, is generally obscured by
rounding up of corners as the system approaches the 120o
culmination stage of headward-joining of channels. Scale:
In fact, this is the headward erosion at the aforesaid
30o = 10 confluences
angles which inevitably and eventually leads to the
formation of near hexagonal polygons(Fig 3).
Fig. 2 Frequency of confluence angles of Dhamani

and Saras badlands.
Fig. 1 Badlands of rivers Dhamani and Saras in

Narsinghpur district, M.P. India
To describe the formation and development of

channel networks, the distances of all the off-shoot
channels from the downstream confluences were
measured from Fig. 1 in various classes of segment
lengths, as shown in Table 1. Fig.3 Schematic representation of pattern
This table shows the frequency of various segment development of channels in badlands. X = minimum
lengths. Based on their high frequency of occurrence, spacing. G1, G2, G3 are the first, second, third
five classes appear to dominate the pattern: 100 - generation off-shoots.
150m, 200 - 250m, 350 - 400m, 500 - 550m, and 700 Small badlands are generally seen to develop
- 750m. Notably, the 200 - 250m class shows the intensively on the concave side of a stream or convex
highest frequency of occurrence and in this class, side of a bank or ambilateral to an individual stream
most segments were measured close to 250 m. Here, or between the interfluvial regions of small streams.
gullies of less than 100m though occurring frequently Sometimes, the channels extend from adjacent
were not considered as dominating for these were convex bends of a meander and converge so as to
under-grown offshoots in the overall pattern of enclose the part of land in the inner side of a concave
Narsinghpur badland. However, the segments of 125 bend.
m exhibited fully developed off-shoots. Thus, the Generally, the gradient of central channel bed is
variation in segment lengths of channels can be extremely less, usually of the order of 1: 1000,
described in multiples of the minimum spacing of whereas in case of exterior links (Shreve, 1966) or
125 m between two off-shoot channels of the sources (Werritty, 1972), it commonly ranges from 1:
network. Table 1 indicates that 70% of off-shoot 300 to 1: 500. An idealized profile of such a channel
channels and gullies belong to classes 200 - 250m, system is shown in Fig. 4. The figure depicts that in
100 -150m, and less than 100m, the first of which the badlands of nearly flat or very gently sloping
constitutes more than 30%. These facts are consistent regions, the lateral channels remained below the level
with the concept of selective minimum area for of headscarp of the main channel and submergence of
gullying held by several researchers.

headscarp would lead to submergence of the whole Gregory K.J.(1977a),Context of river channel
system. changes. In: Gregory KJ (ed.) River channel
With the above initial conditions of channel bed changes, John Wiley &Sons,Chichester,1-
slope of badlands, which are nearly representative of 11,(11.1).
real world ravine characteristics as found in the study Hack, J.T. (1960), Interpretation of erosional
area, two opposite problems of land and water topography in humid temperate regions, Amer.
resources development, namely erosion and J. Sci., 258A, 80-97.
deposition (silting), may be combined to recover the Harris,D.R.&VitaFinzi,C.(1968),Kokkinopolis- A
land and improve water resources in such badlands. Greek badland. Geog. Journal, 134, 537-545.
To this end, for favouring sediment deposition in the Horton, R.E. (1945), Erosional development of
space of ravine network upstream, the water velocity streams and their drainage basins: Hydrological
can be greatly attenuated by constructing a check approach to quantitative analysis,’ Geol. Soc.
dam on a suitable site at the end of badland belt. The Amer. Bull., 56, 275- 370.
impounded water behind the check dam is likely to Lusby,G.C.(1970) , Hydrologic and biotic effects of
stay for long period, of course, depending on the rate grazing and nongrazing near Grand Junction,
of water discharge through the check dam, volume of Colorado. U.S. Geol. Survey Prof. Paper, 700B,
water reaching the badland and its spread area in the 232-236.
reservoir, infiltration rates, and climatic conditions. Schumm, S.A. (1956a),Evolution of drainage systems
Thus, the reservoir will act as a site for sub-aqueous and slopes in badlands in Perth Amboy, Geol.
aggradational leveling of the ravines of badlands. The Soc. Amer. Bull.67, N.J., 597-646.
stored water in the whole network will immediately Schumm, S.A (1956b), The role of creep and rain-
result in advantages. wash on the retreat of badland slopes, Amer. J.
Sci., 254, 693-706.
SUMMARY AND CONCLUSIONS Shreve, R.L. (1966),Statistical law of stream
This study brings to attention the storey-nature of numbers, Jour. Geol., 74, 17-37.
badlands formation. There is a natural selection of Smith, T.R. & F.P. Bretherton (1972),Stability and
angle of offshoots of rills-channels. The geometric the conservation of mass in drainage basin
form of channel network in an ideal case tends to evolution, Water Resources Research, Vol. 8,
assume a hexagonal pattern causing development of No.6, 1507-1529.
mounds. The spacing of channels also has preferred Tuckfield, C.G.(1964) Gully erosion in the New
minimum distance or its multiples. Badlands are Forest, Hampshire. Amer. J. Sci.,262, 795-807.
degenerative geomorphic systems resulting in loss of Werrity A. (1972), Topology of stream networks, In:
soil, soil quality, moisture and decline of Spatial Analysis in Geomorphology, Chorley,
groundwater table. Understanding of the morphology R.J. (Ed.), Harper and Row Pub., London, 167-
of badlands may be helpful in prioritization and 196.
planning of their watersheds. Yi-Fu Tan (1966), New Mexican gullies: A critical
review and some recent observations. Annals.
ACKNOWLEDGEMENT Ass. Amer. Geog.,56, 573-597.
The first author is thankful to Professor U.C. Chaube
for supervising the Post Doctrol research project at
IIT Roorkee and to the UGC Delhi for funding the
same.
REFERENCES
Berry,L (1970), Some erosional features due to
piping and sub-surface wash with special
reference to Sudan. Geogfafiska annaler, 52A,
113-119.
Blong,R.J.(1970),The development of discontinuous
gullies in a pumice catchment. Amer. J. Sci. 268,
369-383.
Butzer,K.W.(1971) , Fine alluvial fills in the Oranfe
and Vaal basins of South Africa. Proc. Ass. Am.
Geog.,3, 41-48.
Deshmukh, D.S., U.C. Chaube, S. Tignath, Medha
Jha & S.K. Tripathi (2009), Geomorphological
analysis and Distribution of Badland around the
Confluence of Narmada and Sher Rivers,
Scifronts: J. mult.sc., Vol:III (3), Dec. Pp 47-55

Table 1: Segment length of ravineous offshoot channels
Range of segment Frequency Cumulative Relative cumulative Relative frequency
length frequency frequency (%) in proportion (%)
(m)
25-50 8 8 7.54 7.54
50-100 17 25 23.58 16.03
100-150 15 40 37.73 14.15
150-200 9 49 46.22 8.49
200-250 33 82 77.41 31.19
250-300 2 84 79.29 1.88
300-350 1 85 80.19 0.9
350-400 4 89 83.96 3.77
400-450 3 92 86.79 2.83
450-500 1 93 87.69 0.90
500-550 4 97 91.46 3.77
550-600 0 97 91.46 0.00
600-650 3 100 94.29 2.83
650-700 0 100 99.95 0.00
700-750 6 106 99.95 5.66
Fig: 4 Simplified profile of a badland channel.

ISBN 978-81-921121-0-7
PERFORMANCE GRADING OF INDIAN BITUMINOUS BINDERS

G. Bharath
Research Scholar, Deptt., of Civil Engg., IIT Kharagpur, email: bharat.varma.020@gmail.com
M. Amaranatha Reddy
Associate Professor, Dett., of Civil Engg., IIT Kharagpur, email: manreddy@iitkgp.ac.in
ABSTRACT: Bituminous binders play an important role in the performance of the pavements and therefore selection of
quality and quantity of binder is crucial. For a given location, pavement temperature and expected traffic are the general
parameters considered for choosing type of binder and mix design for amount of binder. Normal, polymer, crumb rubber
modified binders are being currently used in India to build high volume roads and these binders are classified based on the
parameters such as (i) viscosity at 60 oC (ii) penetration value and (iii) softening point value of the binders respectively. All
binders need to satisfy the requirements as per Indian standards. Though binders are being classified based on different
criteria and corresponding tests, the expected performance of the pavements cannot be guaranteed as no performance
related tests/specifications are included in the evaluation of binders. On the other hand SUPERPAVE specifications in USA
has adopted performance based grading of binders (PG), combining all binders, to meet expected climatic conditions and
significant improvement in the pavement performance has been observed since its adoption. It is therefore necessary to
initiate the development of perfromncee based specifications in india. As a first step towards this direction, Indian binders
need to be classified in to performance grading system. Keeping the above in view, an attempt has been made in the
present study to classify Indian binders based on performance based tests and to suggest type of binders to be used in
different regions of India. In the absence of performance data of India pavements, SUPERPAVE specifications have been
considered while arriving at equivalent PG grade. By adopting suitable model, pavement temperature was estimated from
maximum air temperature data of different regions. Binders considered in the study were evaluated using dynamic shear
rheometer (DSR) and results were compared with SUPERPAVE specifications to arrive at appropriate PG grading.. From
the maximum pavement temperature data of different regions and PG grading of Indian binders, an appropriate binder to be
used in different regions was suggested.
INTRODUCTION
Indian road infrastructure has been expanding at a faster maximum air temperature for selecting binder and
rate with ambitious development of road networks under recommends rutting parameter (complex modulus- G*
National Highways Development Programme (NHDP), and phase angle- δ; G*/ sin δ) as one of the requirements
State Highways Improvement Programmes (SHIPs), Bharat of the binder. SUPERPAVE specifications adopted in
Nirman, Pradhan Mantri Gram Sadak Yojana (PMGSY) to USA are based on field performance where the binder is
cater the needs of the demand for road transportation net designated as Performance Grade (PG) bitumen and is
work. Also modernization and expansion of airports are classified mainly based on application pavement
being undertaken to meet the increasing air travel demand. temperature. The PG system uses a common set of tests
Large portions of these roads have flexible pavements with to measure physical properties of the binder that can be
bituminous layer at the surface. Higher pavement directly related to field performance (fatigue, rutting and
temperature, overloading and higher tyre pressures are the cracking) of the pavement at extreme temperatures.
major problems of the Indian highways and causing Kandhal, 2005 has emphasized the need of similar types
premature failures. of performance based specifications in India and
suggested an appropriate binder (viscosity grade) to be
For a given aggregate gradation, type binder and its content used for different applications.
are very crucial as they affect the performance of the
pavement significantly. Amount of binder to be used is Classification of normal binders in India has been
arrived at based on mix design where as type of binder to gradually changed and a number of binder tests are
be selected based on a criterion that will result in expected incorporated to judge the quality of binders. At present,
pavement performance. Selection of type of binder in India viscosity based grade of bitumen is in practice (IS: 72-
is largely based on experience that takes in to account the 2006) for normal binders, penetration and softening point
maximum air temperature of the location. MoRTH is the basis for modified binders (IS: 15462, 2004). Most
guidelines (2001) specifies use of 60/70 (VG-30) or of the tests are empirical in nature and no performance
modified binders for high volume roads and the criterion related tests are included in evaluation of binders.
for selection of binder is temperature (hot, medium and However Strategic Highway Research Program (SHRP)
cold region) and traffic (high, medium and light). Recent introduced performance based binder specification
specifications on modified binders (IRC: SP-53:2010) uses Performance Grade (PG) based on the shear

susceptibility parameters to address rutting, fatigue and based tests include long-term aging, determination of
thermal cracking failures of the pavements. Selection of design temperatures, low temperature stiffness and
particular type of binder is based on traffic and temperature tensile tests, high temperature complex modulus (and
of the particular location. For example PG 68-28 is phase angle). The SUPERPAVE protocol for selecting a
expected to serve in a region where seven day average PG binder grade is determination of the pavement design
maximum pavement temperature with some reliability is 68 temperatures for proposed construction site and choosing
0
C and minimum is - 28 0C. PG grade binder should satisfy the appropriate PG grade binder based on the design
the performance tests as per AASHTO MP-1, 1998 temperatures. The selected binder should satisfy all test
including rutting parameter complex modulus (G*) and requirements as per AASHTO MP-1.
phase angle (δ) i.e G* / sin δ.
PG BINDER SPECIFICATION TESTS:
In order to improve the pavement performance, it is PG binder testing includes tests performed on unaged or
desirable to adopt performance based evaluation of binders original, short-term aged, and long-term aged asphalt
and therefore classification of binders in to PG grade is a binder and confirm to AASHTO MP-1(1998). . Short
first step towards this direction. Among the different and long term aging represents the loss of volatiles and
distresses, rutting is the major concern of the Indian oxidation that occurs during construction and pavement
highways and effort should be made to improve rutting after 5 to 10 years of service life respectively. Rolling
resistance of the bituminous mixes. Complex modulus thin film oven test (AASHTO T 240) and pressure aging
(G*) and phase angle (δ) and G* / sin δ parameter explains vessel test (AASHTO PP1, 1998) simulates these aging.
the rutting behavior of the binder. Keeping this in view, Table 1 shows the binder, equipment used for sample
G*/ sin δ value of the binder is considered as as the main preparation and test procedure for binder
parameter to classify the prevailing binders in to equivalent characterization.
PG grade.
In the present paper, an attempt has been made to Rotational viscosity testing (ASTM D 4402) is included
categorize typical Indian binders in to PG grade and to ensure that the binder can be pumped at the asphalt
appropriate binder for different regions has been suggested. plant (ASTM, 1998) and complex shear modulus (G*)
For this, six binders including three modified binders were and the phase angle (δ) and G*/sin δ (linear visco-elastic
considered and these binders were evaluated as per IS: 72- range) is an indicator of the stiffness or resistance of the
2006 and IS: 15462, 2004 guidelines. Rheological asphalt binder to deformation under load and G*sin δ for
parameters such as complex modulus and phase angle were fatigue resistance. Binder condition, parameters and its
measured at different temperatures using dynamic shear specification are given below
rheometer. Maximum air temperatures in some locations of Binder Condition Parameter Specification
India were collected and expected pavement temperatures Unaged binder G*/sinδ ≥ 1.0 kPa
were estimated from SUPERPAVE air –pavement
temperature model. Max. temperature at which G*/ sin δ RTFO residue G*/sinδ ≥ 2.2 kPa
values of the a particular binder satisfy Superpave PAV residue G*sinδ ≤ 5000 kPa
specifications is considered for classifying the binders into
equivalent PG grade. Models for estimation of Pavement Temperature
In Superpave PG grade binder classification, estimation
SUPERPAVE PG BINDER SPECIFICATION high pavement design temperature at a depth 20 mm
Performance-related binder parameters are used in the below the pavement surface for any given project
SUPERPAVE Performance grade (PG) binder specification location based on the following relationship
and these parameters are related to service pavement T20mm = (Tair - 0.00618 Lat2 + 0.2289 Lat + 42.2)(0.9545)
temperatures to which binders will be exposed during their - 17.78 (1)
service life. PG XX-YY is the designation of the binder where T20mm= high pavement design temperature at a
where XX refers to highest seven-day average pavement depth of 20 mm
temperature and -YY to lowest one day pavement Tair = seven-day average high air temperature; Lat = the
temperature For example, a PG 58-28 would meet the PG project’s location in degrees latitude
Binder specification for a design high pavement
temperature up to 58°C and design low pavement Evaluation of Binders
temperature down to -28°C. The PG grade binder needs to Three normal binders( VG-10, VG-30 and VG-40) and
satisfy several test requirements including some three modified binders ( PMB 40, PMB 70 and CRMB
performance based tests. These include the viscosity at 55) were evaluated as per relevant Indian standards
135°C, short term aging of binder by Rolling Thin Film (IS:73-2006 and IS:15462-2004) . Results of tests
Oven Test (RTFOT), and Flash Point. Other performance conducted on normal and modified binders are given in

Table 2 and 3. From the above, it is observed that all the
binders evaluated for different test parameters satisfy the
requirements as per standards.
Evaluation of Rheological Parameters Using Dynaimc

Shear Rheometer
Bitumen binders are visco-elastic in nature and its
behaviour is influenced by temperature and loading
frequency. As binder are subjected to pavement
temperature during its service life, it is necessary to
characterize binders over the range of pavement service
temperatures. As binder contribute towards resisting rutting
and fatigue failures, it is also necessary to identify Photo-1: Dynamic Shear Rheometer used for Evaluating
appropriate characteristics of binders that can be used to Rhelogical Parameters of Binder
explain the fatigue and rutting performance of binders. A
number of researchers have identified complex modulus Rheological parameters of VG 30 unaged and aged
(G*) and phase angle (δ) parameters would explain the (RTFO and PAV) binder at different temperature is
both rutting and fatigue behavior of the binder. Complex given in Table 4 to Table 6. From the above, VG 30
modulus is a measure of the total resistance of a material to binder is expected to performance better up to 640C
deformation when repeatedly sheared. The phase angle (higher temperature –G*/ sin δ- 1 kPa for unaged
indicates the relative magnitude of recoverable and non- binder) and 190C ( intermediated temperature- G* sin δ-
recoverable deformation of binder. Dynamic shear 5000 kPa for PAV binder) and -24 0C (low temperature-
rheometer (DSR) is used extensively for evaluating the extrapolated from intermediate temperatures). Therefore
rheological properties such as G* and phase angle of VG 30 binder can be assigned as equivalent to PG 64-24.
bituminous binders (AASHTO TP5, 1988). Similarly for other binders G * and sin δ values at
different temperatures and angular frequency of 10
DSR is used to evaluate the binder response by subjecting a rads/sec were evaluated. The maximum service
thin bitumen film to oscillatory (sinusoidal) stresses at temperature of different unaged binders satisfying the
constant temperature and frequency. The Principle of DSR G*/sin value of 1.00 kPa as per Superpave specifications
is illustrated in Fig.1. Binder film (1 or 2 mm as case may were noted from this data and presented in Table 7. The
be) is sandwiched between two parallel circular plates. intermediate and low temperature values of the different
Bottom plate is fixed and top plate oscillates at a given binders were obtained are also presented in Table 7.
frequency. Fig.2 shows typical variation of shear stress and From these data, equivalent PG grade of the binders
shear strain during a DSR test. For a given frequency and considered in this study were obtained and reported in
temperature, two main parameters, complex shear modulus the Table 7.
(G*) and phase angle (δ) are determined from shear Maximum and minimum air temperatures of these
stresses and corresponding strains and log between them. regions were obtained from meterological department
Complex modulus is computed using equation 2. and are shon in Photograph 2.. From these temperatures,
Complex modulus, pavement temperatures were estimated using Superpave
G* = (max - min) / (max - min) (2) pavement temperature prediction model (eauation:1).
Table 8 gives the details of air temperature, average
Where, max and min are maximum and minimum shear latitude of different regions and corresponding pavement
stresses respectively, max and min are maximum and temperature obtained from the model. Also presents the
minimum shear strains respectively. equivalent PG grade for each region and actual grade of
Phase angle (δ) is the lag between the applied shear stress binder to be used that includes bumping ( one grade
and the resulting shear strain (see Fig.2). For typical paving higher) for higher traffic. Superpave recommends
binders, the complex modulus (G*) value ranges from 500 increasing the high temperature grade by one grade when
to 6000Pa and phase angle ranges from 50 to 900. For the design traffic level exceeds 30 million ESALs (see
evaluation of rheological parameters of binder i.e complex Table 8). Bituminous binders are classified by viscosity
modulus and phase angle, unpaged and RTFO aged binders grading in India and a number of empirical tests are
were tested in DSR at different temperatures and angular specified as per Indian standards. No proper guidelines
frequency of 10 rad/s (corresponds to a traffic speed of are available for selection of binder for pavement
about 90 km/hr). Photograph 1 shows the Dynamic Shear construction at a given location. Superpave recommends
Rheometer used for evaluating rhelogical parameters of performance based grading of binder with number of
binder. performance tests on binder to control different

pavement distresses and uses pavement temperature and Association of State Highway & Transportation
traffic parameters for selecting binder for a region. Officials Washington, DC.
AASHTO PP1 (1988). Standard Practice for Accelerated
Aging of Asphalt Binder Using a Pressurized Aging
Vessel (PAV). American Association of State
Highway & Transportation Officials Washington,
DC.
AASHTO TP5 (1988). Standard Test Method for
Determining the Rheological Properties of Asphalt
Binder Using a Dynamic Shear Rheometer,
American Association of State Highway &
Transportation Officials Washington, DC.
AASHTO TP3 (2000). Standard Test Method for
Determining the Fracture Properties of Asphalt
Binder in Direct Tension (DT), American
Association of State Highway & Transportation
Officials Washington, DC.
ASTM D 4402 – 06, Standard Test Method for Viscosity
Determination of Asphalt at Elevated Temperatures
Using a Rotational Viscometer.
ASTM D2872 (2004). Standard Test Method for Effect
of Heat and Air on a Moving Film of Asphalt
(Rolling Thin-Film Oven Test.)
BIS: 73 (2006). Specifications for Viscosity Grade
Photo-2:Regional temperature distribution for India
Paving Bitumen, Bureau of Indian standards, New
Delhi.
From the present study, normal and modified binders used
BIS: 15462 (2004). Specification for Polymer and
in India, which are presently classified based on viscosity
Rubber Modified Bitumen, Bureau of Indian
and other methods, are converted in terms of performance
standards, New Delhi.
grade (PG) binders. Rheological parameters of the binders
IRC: SP: 53 (2010), Specification for Modified Bitumen,
and SHRP guidelines are the main criteria for obtaining
Indian Roads Congress, New Delhi.
equivalent PG grade. Based on altitude and air temperature
Kandhal, P.S (2005). Selection of Bitumen for Paving
of different regions of India, expected pavement
Highways in India, Vol. 33, Indian Highways,
temperatures were estimated and based on this appropriate
pp.17-27.
grade of binder to be used was suggested for better
MoRTH (2001): Specifications for Roads and Bridge
performance.
Works, 4th edition, New Delhi.
SP-1 (2003), Superpave Series No.1, Superpave
REFERENCES
Performance Graded Asphalt Binder Specification
AASHTO MP1 (1998). Standard Specification for
and Testing, Asphalt Institute, 3rd Edition, Research
Performance Graded Asphalt Binder. American
Park Drive, Lexington, KY.
Table 1 Binder sample preparation and test procedure for binder characterization
Binder Binder characterisation
Condition of the Equipment used for sample
Parameter Equipment Test Procedure
Binder preparation
Viscosity Rotational Viscometer ASTM D 4402
Unaged/ original -------- Flash Point Cleveland Open Cup AASHTO T 48
G* & sin δ G*/ sin δ Dynamic shear Rheometer AASHTO TP 5
G* and sin δ
Short tem aged Rolling Thin Film oven Dynamic shear Rheometer AASHTO TP 5
G*/ sin δ
G* and sin δ
Dynamic shear Rheometer AASHTO TP 5
G*sin δ
Long term aged Pressure aging Vessel
Creep stiffness Bending Beam Rheometer AASHTO TP1
Direct Tension Direct Tension Tester AASHTO TP 3

Table 2: Results of tests conducted on normal binders (IS: 73, 2006)
VG-40 VG-30 VG-10
Parameter Test Method
Result Spec. Result Spec. Result Spec.
Absolute Viscosity at 600C, P IS: 1206 (Part 2) (1978) 4250 min 3200 3228 min 2400 1296 min 800
Kinematic Viscosity at 1350C, cSt IS: 1206 (Part 3) (1978) 721 min 400 528 min 350 312 min 250
Penetration at 250C, 100g, 5s,
IS: 1203 (1978) 43 40-60 62 50-70 90 80-100
0.1mm
Softening Point (R&B), 0C IS: 1205 (1978) 65 min 50 57 min 47 48 min 40
Table 3: Results of tests conducted on modified binders (IS: 15462, 2004)

PMB-40 PMB-70 CRMB-55
Parameter Test Method
Result Spec. Result Spec. Result Spec.
Penetration at 250C, 100g, 5s, 0.1mm IS:1203 (1978) 41 30-50 59 50-90 47 Max 50
Softening point (R&B), 0C IS: 1205 (1978) 62 Min 60 54 Min 55 59 Min 60
Elastic recovery of half thread in ductilometer at 150C, % IS:15462, Annex-A (2004) 72 Min 70 75 Min 70 55 Min 50
0
Separation, difference in softening point (R&B), C IS:15462, Annex-B (2004) 2 Max 3 2 Max 3 3 Max 4
Viscosity at 1500C, P IS:1206 (Part I) (1978) 7.2 3-9 5.9 2-6 5.6 Max 4
Table 4 Rheological parameters of VG 30 binder at different temperatures (unaged)

Temperature 0C Complex modulus (G*) kPa Phase angle (δ) G*/ sin δ (kPa)
46 17.80 84.15 17.89
52 6.52 85.67 6.53
58 2.74 86.46 2.74
64 0.98 86.75 0.98
70 0.68 86.52 0.68
Table 5 Rheological parameters of VG 30 binder at different temperatures (RTFO aged)

Temperature 0C Complex modulus (G*) kPa Phase angle (δ) G*/ sin δ (kPa)
46 39.53 81.14 40.01

52 14.80 84.07 14.88
58 5.55 85.89 5.57
64 2.07 87.18 2.07
70 1.06 87.74 1.06
Table 6 Rheological parameters of VG 30 binder at different temperatures PAV aged)

Temperature Phase angle
0 Complex modulus (G*) kPa G*sin δ (kPa)
C (δ)
22 3652 56.41 3040
19 5775 52.92 4600
16 9290 48.96 7015
13 13900 44.95 9817
10 21808 40.52 14205
Table 7 Service temperatures of different binders and Performance Grade

Temperature(0C)
Maximum Performance grade
Viscosity grade (VG) Intermediate Low (PG)
unaged RTFO aged (PAV aged) (SP-2)
VG-10 56 57 13 -30 PG58-30
VG-30 63 62 19 -24 PG64-24
VG-40 76 82 25 -18 PG82-18
PMB-40 83 83 34 -12 PG82-12
PMB-70 73 72 25 -18 PG76-18
CRMB-55 82 83 34 -12 PG82-12
RECENT ADVANCES IN CIVIL ENGINEERING-2011
35
Table 8: PG Binder selection guide table
Region Average Average Average Equivalent PG grade for Binder for high traffic >(3*10^7)ESAL
Latitude max. air pavement normal traffic <(10^7 (Bumping)
temp(oC) temp(oC) ESAL)
PG Grade Equivalent
Indian Grade
1 23.5o 38 50 PG 52-24 PG 64-24 VG-30
2 28o 42 54 PG 58-18 PG 70-18 PMB-70
3 20o 40 52 PG 58-18 PG70-18 PMB-70
4 140 39 51 PG 52-24 PG 64-24 VG-30
5 32o 32 44 PG 46-30 PG 58-30 VG-10
Applied shear Rotating plate

stress or strain
Bitumen film
Fixed plate
Fig.1: Principle of Dynamic Shear Rheometer
Phase angle, δ = Δt.

max
= angular frequency
Shear stress,
time
min
Time lag, Δt
max
Shear strain, time

min
Fig.2: Typical Stress vs Strain Response in DSR Test

36
ISBN 978-81-921121-0-7
RECENT DEVELOPMENTS IN NITRATE REMOVAL TECHNIQUES

FROM WATER
P.K. Singh
Corresponding author, Professor, Department of Civil Engineering, I.T., B.H.U. Varanasi-221005, email:
dr_pksingh1@rediffmail.com
Y.C. Sharma
Associate Professor, Department of Applied Chemistry, I.T., B.H.U., Varanasi-221005.
A. L. Srivastav
Research Scholar, Department of Applied Chemistry, I.T., B.H.U., Varanasi-221005
ABSTRACT: Nitrate contamination of the groundwater is a cause of concern throughout the world, and India
is not an exception to it. More than 95% of the rural population and about 30 to 40% of the urban population in
India depends on ground water for their domestic requirements. Use of nitrogenous fertilizers in agricultural
practices and seepages from domestic wastewaters appear the major sources of nitrate contamination of ground
water. The desirable and permissible limits for nitrate in drinking water are 45 mg NO3-/L in India. Nitrate
contamination of drinking water is problematic due to its harmful biological effects. High concentrations can
cause methemoglobinemia, and have been cited as a risk factor in developing gastric an intestinal cancer. Due to
these heath risks, a great deal of emphasis has been placed on finding effective treatment processes to reduce
nitrate concentrations to safe levels. This paper attempts to summarize the nitrate contamination status of ground
waters in India, reported health risks associated with consuming high nitrate containing waters, and updated
stage of nitrate treatment technologies for drinking water. Widely accepted nitrate removal methods including
chemical reduction, biological denitrification, adsorption, ion exchange, and reverse osmosis have been
discussed and new developments in the area using electro dialysis, combined processes and inorganic ion
exchange materials for point of entry (POE) and point of use (POU) treatment systems have been reported.
Keywords: Nitrate removal technologies, PRB, RO, Inorganic ion exchange, Hydrous metal oxides (HMOs).
INTRODUCTION
More than 95% of the rural and about 30 to 40% of Due to these heath risks, a great deal of emphasis
the urban population in India depend on ground water has been placed on finding effective treatment
for their domestic requirements. Ground water processes to reduce nitrate concentrations to safe
contamination is a worldwide problem that has levels. Reducing the amount of chemical fertilizers
economic and human health impacts (WHO, 2002). in agriculture will help alleviate the problem, and
Inorganic anions of nitrate, arsenic and fluoride have may not hurt crop yields. With these preventive
been found in potentially harmful concentrations in methods the concentration of nitrate in the
drinking water sources (Smith et al., 2002; Petrovic et groundwater may be reduced over time. Biological
al., 2003; Velizarov et al., 2004). Nitrate, due to its denitrification of various descriptions including
high water solubility, is possibly the most widespread based on electrochemical approaches and catalytic
groundwater contaminant in the world, imposing a support are found suitable for in situ applications.
serious threat to drinking water supplies and Among the physico-chemical methods, whereas
promoting eutrophication (Majumdar and Gupta, reverse osmosis, nanofiltration, and electro dialysis
2000). Point and non-point sources of nitrate are membrane based separation processes for nitrate
contamination include agricultural and urban runoff, removal, ion exchange has been one among the best
disposal of untreated sanitary and industrial wastes in available technologies to treat nitrate contaminated
unsafe manner, leakage from septic systems, landfill water in ex-situ conditions such as Point of Use
leachate, animal manure, NOx air stripping waste (POU) and/or Point of Entry (POE) treatments.
from air pollution control devices (Bhatnagar and
Sillanpa, 2011). High concentrations of nitrate in NITRATE CONTAMINATION STATUS OF
drinking water may cause methemoglobinemia, and GROUND WATERS IN INDIA
have been cited as a risk factor in developing gastric Among the three basic physiogeographic geological
an intestinal cancer. Nitrate ion is a precursor of condition the intervening Indo-Gangetic plain has
carcinogen and also causes diabetes (Kostraba et al., the greatest risk for nitrate pollution (Lunkad,
1992).The epidemiologic evidence of linkages 1993). Central Ground Water Board (CGWB),
between drinking water nitrate and risk of specific Ministry of Water Resources, Government of India
cancers, adverse reproductive outcomes, and other has enlisted the districts of 18 states where nitrate
health outcomes in the context of the current level in groundwater exceeds 45 mg NO3-/L. In
regulatory limit for this anion in drinking water have Uttar Pradesh, the nitrate affected districts are as
been reviewed recently (Ward et al., 2005). given in Table 1.

ADVERSE HUMAN HEALTH EFFECTS OF levels for drinking purpose. Firstly, there are non-
HIGH NITRATE CONCENTRATIONS IN treatment techniques that consist of blending
DRINKING WATER drinking waters, or changing water sources. The
Nitrate in drinking water is associated with a number second option is the use of treatment techniques,
of health problems in human beings such as such as ion exchange, reverse osmosis, biological
methemoglobinemia, oral cancer, cancer of colon, denitrification and chemical reduction etc to
rectum and other gastrointestinal cancers, Alzheimer’s achieve the objectives. Excellent reviews have been
disease, Non-Hodgkin’s lymphoma etc. (Rao and published for nitrate removal from drinking water
Puttanna, 2006). In India, methemoglobinemia has (Green and Shelef, 1994; Kapoor and Virarghavan,
been found prevalent in all age groups in areas of 1997; Shrimali and Singh, 2001). Nitrate removal
Rajasthan with high nitrate concentrations in drinking methods in water treatment are basically of two
water (Gupta et al. 1999). Recurrent acute respiratory types: (i) Elimination Processes, and (ii) Separation
tract infections in some areas of Rajasthan have been Processes. Whereas chemical and biological/
attributed to high nitrate concentrations in drinking biochemical methods remove nitrate from water by
water (Gupta et al., 2000). Nitrate in drinking water reductive elimination, reverse osmosis, electro
has been recognized as a factor for enhanced goiter dialysis, and ion exchange are physicochemical
incidence in many countries including Bulgaria methods using separation technique for nitrate
(Gatseva et al., 1998). Increased nitrate intake could removal from water (Singh, 1999). It has been
also affect the thyroid function. The reason is that the established that among these processes ion
nitrate-ion (NO3-) alone or in mixture with thiocyanate exchange and biological denitrification are more
(SCN-) and perchlorate (ClO4-) shares the same acceptable than other technologies. Ion exchange
transport mechanism at the sodium iodide (I-) process is more viable for ground water nitrate
symporter of thyroid gland and competitively inhibits removal while biological denitrification is more
the iodide uptake (Tonacchera et al., 2004). This suitable for surface water flows (Kapoor and
inhibition could lead to a decrease in thyroid hormone Virarghavan, 1997).
secretion, followed by an increase in the thyroid-
stimulating hormone. In the end thyroid gland Chemical nitrate reduction
enlargement (goiter) could occur (Eskiocak et al., In-situ chemical reduction of nitrate is achievable
2005). Nitrate toxicity to human is manifested by using zero-valent iron (ZVI). It has been shown that
increased levels of methemoglobin and formation of zero-valent iron (Fe0) can serve as an electron donor
carcinogenic compounds (WHO, 2007). Excess of to reduce nitrate. Two possible mechanisms are
nitrate in water can cause several problems. The effect direct reduction of nitrate by Fe0 and its indirect
of nitrate itself is called primary toxicity. Its high reduction by hydrogen generated by iron corrosion.
intake in drinking water causes abdominal pain, Nitrate reduction by Fe0 is a spontaneous process
diarrhea, vomiting, hypertension, increased infant under acidic conditions:
mortality, central nervous systems defects, birth
defects, diabetes, spontaneous abortions, respiratory
tract infections and changes to the immune system For the above reaction to proceed at a significant
(Samatya et al., 2006). Secondary toxicity of nitrate is rate the solution pH must be low or hydrogen must
due to microbially reduced reactive nitrite ion in the be supplied. However, this process has some
intestine. It has been correlated with disadvantages regarding release ammonium which
methemoglobinemia, especially in infants of age must be successively removed. The process does
under 6-7 years. MetHB (methemoglobin) is formed not evolve rapidly at neutral pH, and formate,
when nitrite oxidizes the ferrous ion in hemoglobin to oxalate, citrate, chloride, sulphate, borate and
the ferric form. The condition of methemoglobinemia phosphate acting as ligands on ZVI significantly
is characterized by cyanosis, stupor and cerebral reduce nitrate reduction rate (Su and Puls, 2004).
anoxia. Symptoms include bluish grey or brownish Catalytic reduction process for nitrate removal from
grey skin colour, irritability, and excessive crying in ground water involves use of some specific
children (Samatya et al., 2006). In tertiary toxicity, the catalysts like Pd-Cu/TiO2. Centi et al., (2002)
reaction between nitrite and secondary or tertiary reported works on ground water nitrate elimination
amine in acidic medium can result in the formation of using catalysis which offers various advantages
N- Nitroso compounds, some of which are known as with respect to conventional technologies. Using a
carcinogenic, teratogenic, and mutagenic (Samatya et Pd-Cu-based catalytic membrane, a reaction
al., 2006). temperature below 15◦C, a mixed 4:1 CO2:H2 feed
and controlling bulk solution pH by HCl addition, it
NITRATE REMOVAL METHODS FROM was found possible to obtain a nitrate conversion
WATER FOR DRINKING PURPOSE higher than 80% even with ammonium ion
In dealing with the nitrate problem in subsurface formation below 0.5 ppm, i.e. the maximum
waters, there are two options for achieving safe nitrate

concentration allowable to meet the requirements for microorganisms, carbon source residuals, organic
drinking purposes. However this technology seems to matters resulting from bacterial activity, nitrites and
be effective on small scale application and cannot be no dissolved oxygen. High biomass concentration in
used effectively at a commercial level. Ammonium the denitrification reactor is usually obtained by the
ions form as by-product of the reduction in an amount biological sludge attachment to solid media, which
higher than the allowed limit (0.5 mg/L). Use of a is either stationary (fixed bed), moving (rotating
catalyst in the form of suspended powder causes discs), or fluidized (fluidized bed). Schipper et al.,
problems in catalyst recovery, and water (2004) worked on the denitrification of ground
contamination by suspended particles containing water by making denitrifying walls. They are
noble metals. constructed by digging a trench perpendicular to
groundwater flow and mixing the aquifer material
Biological Denitrification process offers the with organic matter, such as sawdust, which acts as
possibility of selective reduction of nitrate to nitrogen. a carbon source to stimulate denitrification. For
However, the nitrate treated water may be efficient functioning, walls need to be permeable to
contaminated with microorganisms and metabolic groundwater flow.
substances; hence a final treatment is necessary before
human consumption. Permeable Reactive Barrier (PRB) for in situ
Denitrification
Autotrophic denitrification processes utilize PRB is one of the in situ methods for contaminated
autotrophic denitrifiers, to reduce nitrate or nitrite to ground water treatment. The technology was
nitrogen gas. The energy source of the autotrophic developed as a less costly, more effective
denitrifying microorganisms is derived from inorganic alternative to the standard pump-and-treat
oxidation-reduction reactions with elements such as methods used to clean up aquifers and soils
hydrogen or various reduced-sulfur compounds as the contaminated with chlorinated solvents, fuel
electron donor, while the carbon sources are inorganic hydrocarbons, explosives, nitrates, and toxic
carbon compounds (e.g. CO2, HCO3-, etc.). metals. Laboratory studies have shown, however,
Autotrophic denitrification using sulfur or sulfur that many other contaminants, including nitrate and
derivatives has been studied by several investigators nitrite can also be treated using PRB technology
in laboratory and pilot-scale experiments. The with the selection of the proper reactant (EPA,
following stoichiometric relationship has been 1998). Nitrate has been reported to be readily
reported (Green and Shelef, 1994). reduced by metallic iron and has been suggested as
(i) Sulphur a promising candidate for treatment by iron PRBs
55S+50NO3- +38H2O+20CO2 +4NH4+ (Moore and Young, 2005). Although most of the
4C2H7O 2N+25N2 +64H++ + 55 SO42- Fe0 based PRB rely on chemical reduction;
(ii) Sulfides heterotrophic denitrification (HD), autotrophic
5S2- +8NO3- +8H+ denitrification (AD), and a combination of
5SO42- + 4N2 +4H2O simultaneous heterotrophic and autotrophic
denitrification (HAD) can also be used in PRB
(iii) Thiosulfate process.
5S2O32- +8NO3- +H2O
4N2 + 10SO42- +2H- Autotrophic Denitrification in PRB
Biological process using autotrophic denitrification
(iv) Hydrogen (AD) requires no external carbon source and
5H2 +2NO3 - produces low amounts of biomass. Autotrophic
N2 + 4H2O + 2OH- denitrification using S0 as an electron donor occurs
by sulfur-oxidizing bacteria such as Thiobacillus
Hydrogen, being inherently clean and only very denitrificans and Thiomicrospira denitrificans, and
slightly soluble in water (1.6 mg/L at 20°C) may be a stoichiometic equation can be expressed as follows:
good reactant and hence many investigators have 55S + 20CO2 + 50NO3- + 38H2O + 4NH4+
studied autotrophic denitrification using hydrogen.
Stoichiometrically, 0.35 mg of hydrogen is required 4C5H7O2N + 25N2 + 55SO42- + 64H+
for complete denitrification of 1.0 mg N (Green and A negative outcome of sulfur-supported
Shelef, 1994). denitrification is sulfate release and consumption of
Heterotrophic denitrification process uses an alkalinity. Lately, many studies have focused on
organic carbon source as the electron donor. Ethanol preventing the inhibition effect of alkalinity
is the most common carbon source used followed by decrease using limestone and sodium bicarbonate.
methanol and acetic acid (Green and Shelef. 1994). Alternatively, various workers have used hydrogen
Post treatment is always required since the effluents gas as electron donor for autotrophic denitrification.
from the denitrification reactor contain Hydrogen-supported AD is a well known and

developed approach. Till et al. (1998) proposed the Rocca et al., (2005) proposed a new approach
use of zero- valent iron, which indirectly produces combining the heterotrophic and autotrophic
hydrogen to support autotrophic denitrification and denitrification (HAD) together. HAD is based on
showed that increasing the hydraulic retention time the use of ZVI coupled with cotton as solid carbon
increased the nitrate removal efficiency. Table 4 gives source. In the system, ZVI reduces dissolved
some of the system descriptions of autotrophic oxygen (DO) in water and produces cathodic
denitrification. hydrogen, while cotton acts both as a source of
organic carbon for HD and as solid-carrier for
Heterotrophic Denitrification in PRB biofilm formation. At the same time, carbon dioxide
Heterotrophic denitrification is the biological process generated by HD can be used as inorganic carbon
in which nitrate is anaerobically converted to nitrogen source by the microorganisms involved in AD
gas. Heterotrophic bacteria need an external source of process simultaneously:
organic carbon to develop their metabolism. Many
typical simple carbon compounds have been used to
support groundwater denitrification processes, such as
ethanol, methanol and acetate. Recently various
This coupled process provided higher volumetric
studies have been conducted to evaluate the potential
nitrate removal ratio (VNR) than HD using cotton
use of solid carbon (SC) as energy source in
as carbon source and solid support. The process is
denitrification processes, such as cellulose-based
shows potentials to remove simultaneously nitrate
material and biodegradable polymers, which can act as
and other priority pollutants such as Cr (VI) and
permeable reactive barriers (PRB) for in situ
TCE via ZVI reduction (Rocca et al. 2007). For the
application. Cellulose is the most abundant renewable
case of its in-situ application of HAD, a double-
resource in the world being a basic component of all
layer barrier could be suitable as illustrated in Fig.
plant materials. Cotton is the purest form of naturally
1. The first layer would be filled with an iron/sand
occurring cellulose with the highest specific external
mix followed by a cotton/sand layer. Mixing with
surface to support the heterotrophic denitrification
sand or, if necessary, with gravel assures wall
(HD) process. Rocca et al. (2005) reported that the use
permeability higher than the aquifer one, avoiding a
of cotton provided very good nitrate removal
wall by-pass. In this application a critical detention
performance and the nitrate removal ratio increased
time of 0.6 to 1.0 h in the ZVI zone is required to
with temperature till an optimum of 28°C. However, it
assure low (<1 mg) ammonium production. A ratio
was coupled with an increased release of total organic
between nitrate and phosphorous of 7.5 was found
carbon (TOC) and bacteria and the performance
to be sufficient for no-growth limiting conditions
decreased with time due to compression of cotton and
(Rocca et al., 2008).
reduced permeability.
Adsorption
A field level PRB for in situ denitrification was built
In many cases, wells serve small local communities
at Bardowei farm, Cambridge, North Island (New
which cannot be connected to centralized water
Zealand) in January 1996 using saw dust as reactive
treatment units. Comparatively, adsorption seems to
filling material. A 35m long, 1.5m wide and 1.5 m
be a more attractive method for the removal of
deep trench (volume 78.75 m3) was filled with 40 m3
nitrate in terms of cost, simplicity of design and
of sawdust. During 5 years of observation, although
operation. Different adsorbents have been tested for
there was a large decline in denitrifying enzyme
the removal of nitrate from water. Selection of a
activity, denitrification rates were still sufficient to
suitable adsorbent media for NO3 - removal from
account for nitrate removal. Subsequently, due to
water generally depends on several factors
increased biomass and reduced permeability, the
including,
ground waters were found to bypass under the wall
(1) The range of initial NO3- concentrations,
(Schipper et al. 2004). The results suggested the
(2) Other competing ions and their concentration in
precautions for the use of suspended wall type PRB in
water,
aquifers with coarse textured subsoils. Table 5 shows
(3) Optimization of adsorbent dose,
various systems using solid carbon supported
(4) Adjustment of pH in water, and
denitrification systems.
(5) Proper operation and maintenance.
Bhatnagar and Sillanpa (2011) compiled a list of
Gibert et al. (2008) studied seven natural organic
several materials which have been explored as
substrate for their possible application in PRB for
adsorbent for NO3- removal from water (Fig.2). It
denitrification and found that softwood could be one
included double layered hydroxides/hydrotalcite-
of the potential filling material for the purpose.
type compounds and modified chitosan.
Agricultural wastes, after surface modification,
Heterotrophic and autotrophic denitrification (HAD)
have also been explored for NO3- removal and, in

some cases, these wastes show an appreciable replacement of nitrate, sulfate and bicarbonate ions
potential for NO3- removal. However, during chemical (Richard, 1989). Vaaramaa and Lehto (2003)
treatment of adsorbent, one should not ignore the cost worked on the ion exchange process and evaluated
factor. Low production cost with higher removal five organic and three inorganic resins for efficient
efficiency would make the process economical and removal of transition metals and other anions The
efficient. Strong Base Anion (SBA) exchanger was found to
remove nitrate from feed ground water. Samatya et
Thus, selection of a suitable adsorbent for effective al., (2006) used nitrate selective ion exchange
NO3- removal is a complex task. A particular sorbent resins, Purolite A 520 E and presented experimental
which shows higher uptake of NO3- in the laboratory, results showing nitrate removal from water.
may fail in field conditions. Chabani et al. (2006) studied the kinetic modeling
Conventionally, nitrate removal by different of nitrate removal from ground water by ion
adsorbents has been employed using micron-sized exchange resin (Amberlite IRA 400) and observed
particles, however, in recent years; nanotechnology that diffusion of nitrates into pores of the resin was
has emerged as one of the attractive technologies for not the only rate-controlling step and results of the
water treatment. The benefits of using nano-materials intra-particle diffusion and the film diffusion
may derive from their self-assembly, large surface models showing the film diffusion as the main rate-
area and enhanced reactivity (Hristovski et al., 2007) limiting step. Freundlich isotherms indicated
and can be potentially utilized for water remediation maximum sorption capacity of the resin to be
(Zhao et al., 2010). Islam and Patel (2009) 769.2mg/g at 25 C.
investigated the ability of Mg–Al–Cl hydrotalcite-like
compound for the removal of nitrate from synthetic Reverse Osmosis
nitrate solution. At an initial nitrate concentration of Reverse Osmosis (RO) a process for the removal of
10 mg N/L using 0.3 g of adsorbent in 100mL of dissolved ions from water in which pressure is used
nitrate solution, a removal of 87.6% under neutral to force the water through a semi permeable
condition was observed. membrane element which will pass the water but
reject most of the dissolved materials. This method
Ion Exchange Process has better economy, large automation possibilities,
Ion exchange involves the passage of nitrate loaded lower level in feed and control parameter and there
water through a resin base containing strong base is no need for post treatment. However this
anion exchanger resins on which nitrate ions are technology is relatively expansive and usually
exchanged for chloride until the resins exchange applicable for medium and small scale application.
capacity exhausted. It merely displace nitrate into brine solution having
Available technical data, experience and economies disposal problem. Schoeman et al., (2003) have
indicate that ion exchange process is most suitable studied the nitrate removal by reverse osmosis in
method for ground water supply for its simplicity rural areas of South Africa. It was demonstrated that
effectiveness, recovery and relatively low cost . the RO process could be very effectively applied for
Peter et al., (1987) had used three strong base anion water denitrification and water desalination in a
exchangers Duolite A 161, Duolite A 165 and rural area. Nitrate-nitrogen was reduced from 42.5
Amberlite IRA 996 for removal of nitrate from mg/L in the RO feed to only 0.9 mg/L in the RO
drinking water and found them to be more nitrate product water. The TDS of the RO feed was
selective than sulphate in treating high nitrate reduced from 1292 mg/L to 24 mg/L in the RO
concentration in potable water. Biological permeate. Therefore, excellent quality water could
denitrification reactor was used for removal of nitrate be produced for potable purposes.
from regenerant, reducing nitrate to N2 gas. In this RO and NF are known as high pressure membrane
combination of ion exchange and biological processes whereas UF and MF are known as low
denitrification, regeneration was possible in 6 hr at a pressure membrane processes (Xu and Chellam,
flow rate of 11 bed volumes/ hr. Ion exchange 2005). NF falls between RO and UF, and its
processes suffers from three difficulties. The first is separation characteristics are based on sieve effect.
the production of large amounts of waste with high But, as most commercial NF membranes are
concentration of nitrate, sulfate and chloride or charged, so the rejection of ions by NF membranes
bicarbonate concentrations, having disposal problems is the consequence of the combination of
and are found unfit from both financial as well as electrostatic and steric interaction associated with
environmental point of view. The second problem in charge shielding, Donnan exclusion and ion
ion exchange process is of selectivity. Other problem hydration (Peeters et al. 1998) . Moros et al. (2005)
associated with the use of ion exchange technique for reported performance of three commercial
the removal of nitrates is the increase in both nanofiltration membranes (NF90 and NF 270, Dow-
corrosiveness (due to chloride ions) and FilmTec; ESNA1-LF, Hydranautics) in the removal
aggressiveness (low alkalinity) of water due to the of nitrate ions from model solution and ground

water in Spain. NF90 membrane is designed to system with respect to emphasizing enhanced
remove high percentage of nitrate ions, NF270 is membrane selectivity alone. Kabay et al. (2006)
commonly used for hardness removal, and ESNA1-LF reported the effects of process parameters on
is a new membrane characterized by its low energy separation performance of nitrate by electro dialysis
consumption. NF90 and ESNA1-LF showed good and found that the percent removal of nitrate with
nitrate rejection results (over 90%) with model electro dialysis method decreased at low initial
solution in nitrate concentration range of 50- 150 mg concentration of nitrate and it increased when the
NO3-/L, but important differences were observed concentration in the feed solution increased. The
while working with ground water. Thus, pilot plant operation time became short when a high potential
studies were felt essential before selecting the best was applied. Menkkouchi Sahli et al. (2006)
suited membrane (Moros et al., 2005). reported technical optimization of nitrate removal
Other Upcoming Processes from ground water from Morocco using an ED pilot
plant of 24 m3/h flow. CMX-Sb cation exchange
Electro-dialysis membrane (64 no.) and ACS anion exchange
USEPA has identified reverse osmosis and electro membranes (60 no.) were used in the pilot plant.
dialysis along with ion- exchange as the "best Menkkouchi Sahli et al. (2008) compared the nitrate
available technologies" for nitrate removal, and water removal from brackish ground water by chemical
treatment costs for both of reverse osmosis and electro adsorption using chitosan and electro dialysis. The
dialysis were reported to be equivalent. Nitrate results indicated that the chitosan adsorbent
removal from ground water using reverse osmosis removed 12 mg NO3-/g in a retention time of 10
ranges from $ 1.55/m3 treated in small systems to $ minutes and from electro dialysis different quality
0.40/m3 treated in large systems, when nitrate can be obtained with changes in driving force or ED
concentration is reduced from 100 mg N/L to 10 mg time. A water safe with respect to nitrate could be
N/L. Cost could be expected to be 20 -50% lower if obtained at a demineralization rate of 50%, but
nitrate levels in raw water is about 50 mg N/L only other parameters still exceeded the permissible
(USEPA, 1991). The ED combines the advantages of limits for potable water. Overall good potable water
selectivity and low chemical demand (Hell et al. could be obtained at a demineralization rate of 70%.
1998). RO and ED have large automation possibilities.
Lower nitrate levels in feed, ease of controls for Combined Processes
process parameters and not needing excessive post Some researchers have tried to use a combination of
treatment are advantages of RO and ED (Elmidaoui two basic approaches to optimize the process.
and Elhannouni, 2002). Sahli et al., (2005) conducted Ayyasamy et al., (2007) used a two-stage nitrate
experiments to optimize condition of nitrate removal removal system- one biological and the other a
from ground water employing electro dialysis process, chemical treatment using a coagulant. For the
using a pilot plant having a capacity 24 m3/d. The biological system, the effect of carbon sources on
optimization was carried out on the basis of nitrate removal was first investigated using mineral
correlation between demineralization-denitrification, salt medium (MSM) containing 500 mg/L nitrate to
recovery rates, and power and reagent consumptions. select the most effective carbon source. Starch was
Electro dialysis operation was conducted continuously used as a representative carbon source and nitrate
during an extended period, under optimized conditions removal was observed in MSM solution and
to control the constancy of the electro dialysis plant groundwater samples containing 500 mg/L and 460
performances. Optimized technical and economical mg/L nitrate respectively. About 86% and 89% of
conditions were proposed to reduce nitrate from nitrate were removed from the MSM solution and
groundwater. However this process is relatively groundwater samples respectively at 72 h. Chemical
expansive compared to biological denitrification and coagulants such as alum, lime and poly aluminum
ion exchange. One drawback is the saturation of chloride were tested for the removal of nitrate
membrane surface with charge ion and it is to be remaining in the samples. Among the coagulants,
cleaned intermittently by reversing the polarity. It is lime at 150 mg l–1 exhibited the highest nitrate
mainly suitable for ultra pure water. Elyanow (2005) removal efficiency with complete disappearance
discussed Electro dialysis reversal (EDR) process and from the MSM solutions. Thus, a combined system
described advances in EDR membranes, systems and of biological and chemical treatments was found
performance in nitrate removal. The EDR system effective in complete removal of nitrate from
periodically reverses the polarity of the electrical field, groundwater. Das and Chaudhary (2005) have
and consequently the dilute and concentrates studied the electrochemical denitrification of
compartments, to help flush scale forming ions off the ground water with an objective to maximize the
membrane surface and minimize membrane cleaning. nitrate transformation to nitrogen gas. They used
It was noted that the improvements to the spacer that aluminium, graphite, iron and titanium as an
allow better transport of nitrate to the membrane electrode. While aluminum, iron and titanium
surface have more of an economic impact on the electrodes showed 70–97% nitrate reduction, with

graphite electrode the removal was only 8%. Nitrate et al (2010) studied the adsorption feasibility of
was transformed to ammonia with iron and aluminum nano-alumina for nitrate removal from aqueous
electrodes but with titanium electrodes, nitrogen was solutions. The maximum nitrate sorption capacity of
apparently the major end product. Shin and Cha nano-alumina was found 4.0 mg/g at 25±2 ◦C at pH
(2008) have evaluated the nitrate removal efficiency 4.4. Singh et al. (2011) reported the nitrate removal
employing microbial reduction in the presence of potentials of hydrous bismuth oxides (HBOs) and
Nanoscale Zero Valent Iron (NZVI) to assess the by XRD, FTIR and pHPZC analyses of materials,
feasibility of employing NZVI in the biological nitrate delineated the roles of hydroxide and chloride
treatment. Nitrate was completely removed or reduced present in the system.
within 3 days in a Nanoscale Fe(0)-cell reactor , while
only 50 % of the nitrate was abiotically reduced over 7 Concluding Remarks
days at 25◦C. The removal rate of nitrate was Nitrate contamination of the groundwater is a cause
unaffected by the presence of high amount of of concern throughout the world. Nitrate, due to its
sulphate. It has been indicated that Fe (II), might act high water solubility is possibly the most
as an electron donor which is produced during widespread groundwater contaminant in the world,
anaerobic corrosion in the Fe(0) cell system, imposing a serious threat to drinking water supplies
supported by anaerobic culture in 14 day. This study and promoting eutrophication. Widely accepted
demonstrated the potential applicability of employing nitrate removal technologies include chemical and
NZVI iron as a source of electrons for biological catalytic reduction, biological denitrification,
nitrate reduction. For microbial nitrate reduction adsorption, ion exchange, reverse osmosis and
associated with one disadvantages is that it relies on electro dialysis. Ion exchange, RO, and ED are
organic substrate or explosive hydrogen gas. approved as Best Available Technologies (BAT) to
treat NO3− contaminated water by USEPA (Canter,
Inorganic Ion Exchange 1997; Haugen, 2002). Removal of nitrate using
Ball and Harries (1988) reported that anion exchange resin based ion exchange materials is technically
resins are less stable than cation exchange resins and feasible but in the area of drinking water treatment
undergo a degradation of the exchange group in it faces many difficulties. Several workers reported
service. Strongly basic groups of anion exchange organic fouling and consequent capacity loss of ion
resins undergo thermal degradation over 20-80oC, exchange materials. Thermal instability and
with a consequent loss of capacity. They are also more bacteriological degradation potentials of resin based
susceptible to foulants. Gonder et al (2006) reported filters are the major problems in drinking water
capacity loss in an organically fouled anion exchanger treatment application. Hence search for inorganic
and observed that anion exchange resins are more materials for the purpose is of high importance.
sensitive to fouling by organic materials. Various inorganic materials including hydrous
Under such situations, a need to search for inorganic metal oxides (HMOs) have been reported to remove
ion exchange materials for possible application in nitrate removal from water. Nano alumina and
drinking water treatment was felt. The available hydrous bismuth oxide (HBO) powders are being
literature indicates that there are very few inorganic evaluated for the purpose. As rightly observed by
materials which show nitrate removal properties by Habuda-Stanic et al. (2008), in water treatment, an
ion exchange. Fritche (1993) reported nitrate and other adsorption technology includes simultaneous
anions removal from aqueous solutions using yellow processes of adsorption and filtration. Various
bismuth hydroxide precipitate and inferred ion hydrous oxides of metals have been found to have
exchange as the possible mechanism. Kodama (1993) excellent adsorptive properties, but, they are
developed a new inorganic anion exchanger from available only as fine powders or are generated in
thermal decomposition of basic bismuth nitrate and situ as gels, which causes difficulties when it comes
used it to remove iodide for conc. <2.0 moles/L. Singh to their separation in water treatment process. Thus,
(1999) made a detailed study on the possibility of finding an optimal carrier that will enable the
using various bismuth compounds for nitrate removal contaminant removal by column adsorption is an
from water. The study indicated that hydrous bismuth objective of many investigations. This remains the
oxides (HBOs) have definite potentials for nitrate challenge in the contemporary ground water
removal from water and ion exchange of nitrate with contaminants treatment for drinking purpose at
embedded chloride might be the possible mechanism point of entry (POE) and point of use (POU) scale
of nitrate removal. Singh and Ghosh (2000) reported applications.
nitrate removal potentials of various bismuth
compounds and showed that the polymeric forms of References
hydrous bismuth oxide are more efficient in nitrate Ayyasami, P. M., Shanthib, K., (2007). Two-stage
removal. Singh et al. (2010) studied different hydrous removal of nitrate from ground water using
metal oxides (HMOs) for their potentials in removal biological and chemical denitrification. Journal
of arsenic, fluoride and nitrate from water. Bhatnagar

of Bioscience and Bioengineering. 104-2, 129- permeable reactive barrier (PRB), Bioresource
134. Technology 90, 7587-7596.
Ball, M., Harries., (1988). Ion exchange resin Gonder, B. Z., Kaya, Y., Vergili I., Barlas H.,
assessment. In Michael. S. (Ed.), Ion Exchange (2006). Regeneration of industrial district
For Industry. Ellis Horwood Publishers, wastewater using a combination of Fenton
Chichester, West Sussex. England. process and ion exchange—a case study.
Bhatnagar, A., Kumar, E., Sillanpa, M., (2010). Journal of Resources, Conservation and
Nitrate removal from water by nano-alumina: Recycling. Vol. 52-2, 425-440.
Characterization and sorption studies, Chemical Green, M., Shelef, G., (1994). Treatment of nitrate
Engineering Journal, 1-37. contaminated ground water for drinking
Bhatnagar, A., Sillanpa, M., (2011). A review of purposes. In Zoller, U. (Ed.), Ground Water
emerging adsorbents for nitrate removal from Contamination and Control. Marcel Deckker,
water. PII: S1385-8947(11)00168-9 DOI: Inc. New York, 587-606.
doi:10.1016/j.cej.2011.01.103. Gupta, S.K., Gupta, R.C., Seth, A.K., Bassin, J.R.,
Canter, L.W., (1997). Nitrates in Groundwater, CRC Gupta D.K., Sharma, S., (1999). American J.
Press, Boca Raton. Gastroenterol., 94, 1808-1812.
Centi, G., Siglinda, P., (2002). Department of Gupta, S.K., Gupta, R.C., Seth, A.K., Bassin, J.R.,
Industrial Chemistry and Engineering of Gupta D.K., Sharma, S., (2000). Environ.
Materials, University of Messina, and INSTM. Health Prospect, 108, 363-366.
Consortium for the Science and Technology of Habuda- Stanic, M., Kalajdzic, B., Kules, M.,
Materials, Salita Sperone 31, 98166 Messina, Natalija, V., (2008). Arsenic and arsenate
Italy. sorption by hydrous ferric oxide / polymeric
CGWB (2010). Central Ground Water Board, material”. Journal of desalination. 229 (1-3), 1-
Ministry of Water Resources; Government of 9.
India. Haugen, K.S., Semmens, M.J., Novak, P.J., (2002).
Chabani, M., Amrane, A., Bensmali, A., (2006). A novel in-situ technology for the treatment of
Kinetic modeling of the adsorption of nitrate by nitrate contaminated groundwater, Water Res.
ion exchange resins. Journal of Chemical 36, 3497– 3506.
Engineering. 125-2, 111-117. Hell, F., Lahnsteiner, J., Frischherz, H.,
Das, B., P., Chaudhary, S., (2005). Electrochemical Baumgartner, G., (1998). Experience with
denitrification of simulated ground water. Journal fullscale electrodialysis for nitrate and hardness
of Water Research. 39-17, 4065-4072. removal, Desalination 117, 173–180.
Elmidaoui, A., Elhanouni, F., Taky, M., Chay, M.A., Hristovski, K., Baumgardner, A., Westerhoff, P.,
Menkouchi Sahli, M.A., Echihabi, L., Hafsi, M. (2007). Selecting metal oxide nanomaterials for
(2002). Optimization of nitrate removal operation arsenic removal in fixed bed columns: from
from ground water by electro dialysis. J. Sep. nanopowders to aggregated nanoparticle media,
Purif. Technol., 29:235-244. J. Hazard. Mater. 147, 265–274.
Elyanow, D. Persechino, J., (2005). Advances in Islam, M., Patel, R., (2009). Nitrate sorption by
Nitrate Removal. GE Water and Process thermally activated Mg/Al chloride
Technolog, Technical Paper. hydrotalcite-like compound, J. Hazard. Mater.
EPA, (1998). Permeable reactive barrier technologies 169, 524–531.
for contaminant remediation. United States Kabay, N., Yuksel, M, Samatya, S.,Arar, O.,
Environmental Protection Agency, Washington, Yuksel, U., (2006). Effect of process
DC. EPA/600/R-98/125. parameters on separation performance of nitrate
Eskiocak, S., Dundar, C., Basoglu, T., Altaner, S., by electrodialysis. Separation Science and
(2005). The effects of taking chronic nitrate by Technology 41, 3201-3211.
drinking water on thyroid function and Kapoor, A., Virarghavan, T., (1997). Nitrate
morphology. Clin. Exp. Med. 5, 66–71. removal from drinking water-Review. J.
Fritsche, U., (1993). Removal of nitrates and other Environmental Engineering. 123(4): 371-380.
anions from water by yellow bismuth hydroxide. Kodama, H., (1993). The removal and solidification
J. Environmental Science, health, A28 (9): 1903- of radioactive iodide ions using a new organic
1913. anion exchanger. Special Publication, Royal
Gatseva, P., Vladeva, S., Pavlov, K., (1998). Society of Chemists. 122, 55-62.
Incidence of goiter among children in a village Kostraba, T.N., Gay, E.C., Rewers, M., Hamman,
with nitrate contamination of drinking water. R.F., (1992). Nitrate levels in community
Folia Med. (Plovdiv, Bulgaria) 40, 19–23. drinking waters and risk of IDDM. An
Gibert, O.,Pomierny, S., Rowe, I., Kalin, R.B., (2008). Ecological analysis, Diab. Care 15, 1505–1508.
Selection of organic substrates as potential Lunkad, S.K., (1993). Rising nitrate levels in
reactive materials for use in a denitrification ground water and increasing N-fertilizer

consumption. A menace for future generation. Schipper, L.A., Barkle, G.F., Hadfield, J.C.,
Sankalp, ES (2), 1-13. Vojvodić-Vuković, M., Burgess, C.P., (2004).
Majumdar, D., Gupta, N., (2000). Nitrate pollution of Hydraulic constraints on the performance of a
groundwater and associated human health groundwater denitrification wall for nitrate
disorders, Ind. J. Environ. Hlth. 42, 28–39. removal from shallow groundwater. J.
Menkouchi Sahli, M.A., Annouar S., Mountadar, M., Contaminant Hydrol., 69 , 263–279.
Soufiane, A., Elmidaoui, A., (2008). Nitrate Schipper, L., A., Barkle, G., F., Vukovic, M. V.,
removal of brackish underground water by (2005). Maximum rates of nitrate removal in
chemical adsorption and by electrodialysis. denitrification wall. Journal of Environmental
Desalination 227, 327-333. Quality 34, 1270-1276.
Menkouchi Sahli, M.A., Tahaikt, M., Achary, I., Schoeman, J. J., Steyn, A., (2003). Waste water
Taky, M., Elhanouni, F., Hafsi, M., Elmghari, M., and water pollution. Desalination 155, 15.
Elmidaoui, A., (2006). Technical optimization of Shin, Y., K., Cha, D., K., (2008). Microbial
nitrate removal for groundwater by ED using a reduction of nitrate in presence of nano scale
pilot plant. Desalination 227, 200-2008. zero valent iron. Journal of Chemosphere. 72-2,
Moore, A.M., Young, T.M., (2005). Chloride 257-262.
interactions with iron surfaces; implications for Shrimali, M., Singh, K. P., (2001). New methods of
perchlorate and nitrate remediation using nitrate removal from water. Journal of
permeable reactive barriers. J. ASCE, 131, 924- Environmental pollution 112, 351-359.
933. Singh, P. K., (1999). Nitrate removal from water by
Peeters, J.M.M., Boom, J.P., Mulder, M.H.V., bismuth based media. Ph.D. Thesis, Department
Strathmann, H., (1998). J. Membr. Sci., 145, 199- of Civil Engineering. I.I.T. Kanpur.
209. Singh, P. K., Ghosh, D. K., (2000). Nitrate removal
Peter, V., Wim, F., Van der Hoek., Klapwijk, A., from water by bismuth based media. In. Water
(1987). Nitrate removal from ground water-use of Recycling and Resource Management in the
a nitrate selective resin and a low concentrated Developing World, 459-456.
regenerant. Water, Air, & Soil Pollution. Singh, P. K., Sharma, Y.C., Srivastav, A.L., (2010).
Petrovic, M., Gonzalez, S., Barcelo, D., (2003). Hydrous metal oxides for contaminants removal
Analysis and removal of emerging contaminants from ground water. Journal of Multiple Science
in wastewater and drinking water, TrAC Trends (IV).
Anal. Chem. 22, 685–696. Singh, P. K., Sharma, Y.C., Srivastav, A.L., Shukla,
Rao, E. V. S. P. and Puttanna, K. (2006). Strategies A., (2011). Hydrous bismuth oxides (HBOs) for
for combating nitrate pollution. Current Science, nitrate removal from water, National
91 (10), 1335-1339. Conference on Ground Water for Drinking
Richard, Y.R., (1989). Operating experiences of full (GWD-2011).
scale biological and ion exchange denitrification Smith, A.H., Lopipero, P.A., Bates, M.N.,
plants in France. Journal of Institution of Water Steinmaus, C.M., (2002). Arsenic epidemiology
and Environment Management. 13, 154-167. and drinking water standards, Science 296,
Rocca, C. D., Belgiorno, V., Meriç, S., (2008). 2145–2146.
Heterotrophic/autotrophic denitrification (HAD) Su, C., Puls, R.W., (2001). Arsenate and arsenite
of drinking water: prospective use for permeable removal by zero valent iron: effect of
reactive barrier. Desalination. phosphate, silicate, carbonate, borate, sulfate,
Rocca, C. D., Belgiorno, V., Meric, S., (2005). Cotton chromate, molybdate, and nitrate relative to
supported heterotrophic denitrification of nitrate- chloride. Env. Sci. Technol. 35, 4562-4568.
rich drinking water with a sand filtration post- Till, B.A., Weathers L.J., Alvarez, P.J.J., (1998).
treatment. Water SA, 31(2) 229–236. Fe(0)- supported autotrophic denitrification.
Rocca, C.D., Belgiorno, V., Meric, S., (2007). Environ. Sci. Technol., 32(5), 634-639.
Overview of in situ applicable nitrate removal Tonacchera, M., Pinchera, A., Dimida, A.,
processes. Desalination 204, 46-62. Ferrarini, E., Agretti, P., Vitti, P., Santini, F.,
Sahli, A. M. M., Hafsi, M., Echihabi, L., Chay, L., Crump, K., Gibbs, J., (2004). Relative potencies
(2005). Electro-Dialysis process for nitrate and additivity of perchlorate, thiocyanate,
removal using pilot plant study, development and nitrate, and iodide on the inhibition of
optimization of techniques. 227- 3713. radioactive iodide uptake by the human sodium
Samatya, S., Kabay, N., Yuksel, U., Arda, M.,Yuksel, iodide symporter. Thyroid. 14, 1012–1019.
M., (2006). Removal of nitrate from aqueous USEPA (1991). Federal Register, 56(20): 3552,
solution by nitrate selective ion exchange resins. United States Environmental Protection
J. reactive and functional polymers 66, 1206- Agency, Cinicinnati, Ohio. (Cited from Green
1214. and Shelef, 1994).

Vaaramaa, K., Lehto, J., (2003). Removal of metals WHO (2007). Nitrate and nitrite in drinking-water:
and anions from water by ion exchange process. Background document for development of
Journal of desalination. 155, 157-170. WHO guidelines for drinking water quality.
Velizarov, S., Crespo, J.G., Reis, M.A., (2004). /http: //www.
Removal of inorganic anions from drinking water who.int/water_sanitation_health/dwq/chemicals
supplies by membrane bio/processes, Rev. /nitratenitrite 2ndadd.pdfS.
Environ. Sci. Bio/Technol. 3, 361–380. Xu, W. and Chellam, S. (2005). Initial stages of
Ward, M.H., deKok, T.M., Levallois, P., Brender, J., bacterial fouling during dead end
Gulis, G., Nolan, B.T., VanDerslice, J.,(2005). microfiltration. Environ. Sci. Technol. 39,
Workgroup report: drinking-water nitrate and 6470-6476.
health recent findings and research needs, Zhao, X., Wang, J., Wu, F., Wang, T., Cai, Y., Shi,
Environ. Hlth. Perspect. 113, 1607–1614. Y., Jiang, G., (2010). Removal of fluoride from
WHO (2002). Guidelines for Drinking Water Quality. aqueous media by Fe3O4 @Al(OH) 3 magnetic
Recommendations, 2nd Edition. World Health nanoparticles, J. Hazard. Mater. 173, 102–109.
Organization, Geneva.
Table 1: Districts of Uttar Pradesh having more than 45 mg NO3-/L in ground waters
State Districts
Uttar Pradesh Agra, Aligarh, Allahabad, Ambedkar Nagar, Auraiya, Badaun, Baghpat, Balrampur, Banda,
Barabanki, Bareilly, Basti, Bijnor,Bulandshahr, Chitrakoot, Etah, Etawa, Fatehpur, Firozabad,
GB Nagar, Ghaziabad, Ghazipur, Hamirpur, Hardoi, Jaunpur, Jhansi, Kannauj, Kanpur Dehat,
Lakhimpur, Mahoba, Mathura, Meerut,Moradabad, Muzaffarnagar, Raebareli, Rampur, Sant
Ravidas Nagar, Shahjahanpur, Sitapur, Sonbhadra, Sultanpur, Unnao
(Source: CGWB, 2010)
Table 2: Important features of different techniques for nitrate removal from water
S.N. Nitrate Removal Important features
Technologies
1. Chemical methods No waste disposal is required.
pH and temperature effects are important.
Post treatment is required due to production of by products.
Maximum reported efficiency >60-70%.
High operational cost.
2. Biological methods Requires biomass waste disposal.
Temperature effect is important.
Post-treatment is required due to microorganisms.
>99 % efficiency can be achieved.
Medium operational cost.
3. Adsorption Requires saturated/spent adsorbent disposal.
pH and temperature effects are important.
Post treatment is often not required.
Removal efficiency varies with different adsorbent.
4. Ion exchange Requires waste brine disposal.
pH and temperature effects are not important.
Post treatment is required due to corrosivity of product water.
Approx. 90 % efficiency can be achieved.
5. Reverse osmosis Requires high TDS disposal
pH and temperature effects are not important.
Post treatment is required due to corrosivity of product water.
>95 % efficiency can be achieved.
High operational cost.
(Source: Bhatnagar and Sillanpa, 2011)
Table 3: A comparative view of nitrate removal efficiencies from ground water

Methods Nitrate Removal Efficiency
(%)
Chemical methods >60-70
Biological methods >99
Adsorption Varies with the different adsorbents
Ion-exchange ~90
Reverse osmosis >95
(Source: Bhatnagar and Sillanpa, 2011)

Table 4: Different autotrophic denitrification systems
a
VNR: Volumetric nitrate removal ratio (Source: Rocca et al. 2007)
Table 5: Different solid carbon supported denitrification systems (Rocca et al., 2007)
Fig 1: Potential permeable reactive barrier application of the HAD (Rocca et al. 2007)
Fig.2: Origin of different adsorbents of nitrate removal from water (Source: Bhatnagar and Sillanpa, 2011)

ISBN 978-81-921121-0-7
SWELLING BEHAVIOUR OF EXPANSIVE SOIL MIXED WITH LIME

AND FLY ASH AS ADDITIVES
Dhirendra Kumar
Former Post Graduate student, Deptt. of Civil Engg, Institute of Technology, Banaras Hindu University, Varanasi.
Suresh Kumar
Assistant Professor, Deptt. of Civil Engg, Institute of Technology, Banaras Hindu University, Varanasi
Bala Ramudu Paramkusam
Assistant Professor, Deptt. of Civil Engg, Institute of Technology, Banaras Hindu University, Varanasi
ABSTRACT: It is well known that expansive soils undergo large amounts of heaving and shrinking due to seasonal
moisture changes. The main aim of this research is to determine the optimum lime content (OLC), optimum fly ash content
(OFAC) to reduce the swelling pressure of the expansive soil. For this in the present study the evaluation of the swelling
behavior of expansive soil (black cotton soil) with the addition of lime (1%, 2%, 3%, 4%, 5%) and fly ash (5%, 10%, 15%,
20%, 25%) at different percentages at maximum dry density (MDD) and optimum moisture content (OMC) state as well as
at  2% OMC. The swelling pressure of mixed soil at each percentage of lime, and fly ash, and finally the optimum content
of lime, fly ash, determine on the basis of minimum swelling pressure of mixed soil. It was found that both lime and fly ash
reduced the swelling pressure, however, the addition of lime reduced the swelling pressure to a greater degree than the fly
ash. Even though it takes much less percentage of lime than percentage of fly ash to reduce the swelling pressure of a highly
expansive soil, it may be less expensive to utilize fly ash, which is a waste product of electric power production plants.
INTRODUCTION
Expansive soils also called swelling soils undergo (1994) studied the influence of fly ash on the
harmful volume changes corresponding to alter in characteristics of expansive soil and he conclude that
moisture content. The alternate swelling and addition of fly ash is observed to increase the
shrinkage of expansive soils in alternate wet and dry maximum dry unit weight and decrease the optimum
seasons cause severe cracking in lightly loaded moisture content up to certain fly ash content called
structures founded on them such as foundations, ‘optimum fly ash content’ and the trend is reversed for
pavements, canal beds and linings. According to fly ash content exceeding the optimum fly ash
Jones and Holtz 1973, the damage caused due to these content. Bell (1996) presented a study using the linear
soils is more than any other natural hazards, including shrinkage test with montmorillonite and different
earthquakes and floods. percentages of additive lime and showed that the
shrinkage decreased with additional lime addition, but
The expansive soils, which are spread over extensive that it was not a linear decrease. Erdal Cokca (2001)
areas of India in states, like Rajasthan, Madhya studied the effect of fly-ash on expansive soil and
Pradesh, Gujarat, Andhra Pradesh, Karnataka and experimental findings confirmed that the plasticity
Tamilnadu posed serious problems for buildings and index, activity and swelling potential of the samples
roads. The lightweight structures are severely affected decreased with increasing percent stabilizer and
due to high swelling pressure exerted by these soils. curing time and the optimum content of fly-ash in
Such type of large scale distress, due to expansive and decreasing the swell potential was found to be 20%.
shrinking nature of expansive soil, can be prevented Pandian et.al. (2002) studied the effect of two types of
by either obstructing the soil movement and reducing fly ashes Raichur fly ash (class F) and Neyveli fly ash
the swelling pressure of soil or making the structure (class C) on the CBR characteristics of the black
sufficiently resistant to damage from soil movement. cotton soil. Phanikumar and Sharma (2004) a similar
There are so many additives such as lime, cement, fly study was carried out and the effect of fly ash on
ash, rice husk, gypsum; geosynthetics etc. are engineering properties of expansive soil through an
available to improve the physicochemical properties experimental programme. Al-Rawas et al. (2005)
of clay soils in order to permanently stabilize them. study the effect of lime, cement, Combinations of
But fly ash and lime are easily available and more lime and cement, Sarooj (artificial pozzolan) and heat
economical so it is used in large extent as additive to treatment on the swelling potential of Al-Khod
improve the strength of expansive soil. Some of the expansive soil. In the present research the efforts have
researchers worked on the stabilization of the been put on to study the effect of lime and fly ash
expansive soil and discussed in following. Katti, addition on swelling behavior of an expansive soil
(1979) worked on the swelling soil causes due to its with an objective to understand the comparative
expansive nature due to water, cracks are formed in efficiency of these two additives.
buildings, lining etc. He gave the various suggestions
MATERIALS AND METHOD
to overcome the problem due to expansive soils using An expansive soil (S) obtained from NTPC Mouda,
cohesive non-swelling layer (CNS) layer. Choudhary Nagpur was used and fly ash (FA) obtained from

Anpara thermal power plant. The physical properties Table 3: Mix designation and description
of the expansive soil and fly ash are shown in the Mix Description
table 1 & 2. The grain size distribution curves of both designation
expansive soil and fly ash shown in fig. 1 &2. The M0L Expansive Soil at OMC + 0 % lime
lime used in this investigation was purchased from the M1L Expansive Soil at OMC + 1 % lime
local market. The lime (L) was quick lime (CaO) M2L Expansive Soil at OMC + 2 % lime
M3L Expansive Soil at OMC + 3 % lime
which was allowed to slake with addition of water M4L Expansive Soil at OMC + 4 % lime
produced hydrated lime (Ca (OH) 2). Expansive soil M5L Expansive Soil at OMC + 5 % lime
has been mixed with different proportion of lime, and M0FA Expansive Soil at OMC + 0 % fly ash
fly ash at OMC (23%), at dry condition i.e. 2% less M5FA Expansive Soil at OMC + 5 % fly ash
than OMC (21%), at wet condition i.e. 2% more than M10FA Expansive Soil at OMC + 10 % fly ash
OMC (25%) to evaluate its swelling pressure. The M15FA Expansive Soil at OMC + 15 % fly ash
preparation of mix and designation shown in table 3. M20FA Expansive Soil at OMC + 20 % fly ash
M25FA Expansive Soil at OMC + 25 % fly ash
D0L Expansive Soil at -2% OMC + 0 % lime
Table 1: Physical properties of expansive soil
D4L Expansive Soil at -2% OMC + 4 % lime
Properties Values D0FA Expansive Soil at -2% OMC + 0% fly ash
Liquid limit (%) 68 D20FA Expansive Soil at -2% OMC + 20 % fly ash
Plastic limit (%) 20 W0L Expansive Soil at +2% OMC + 0 % lime
Shrinkage limit (%) 7.05 W4L Expansive Soil at +2% OMC + 4 % lime
Gravel (%) 0 W0FA Expansive Soil at +2% OMC + 0 % fly ash
Sand (%) 6 W20FA Expansive Soil at +2% OMC + 20 % fly ash
Silt (%) 35
Clay (%) 59 RESULTS AND DISCUSSION
Maximum dry density (g/cc) 1.52 At OMC, the swelling pressure of black cotton soil
Optimum Moisture content (%) 23 with lime at 0%, 1%, 2%, 3%, 4%, and 5% by dry
Specific gravity 2.70 weight of soil are 48.3, 38.35, 25.3, 10.35, 6.9 and
UCS (kg/cm2) 4.0 19.55 kN/m2 respectively, which are shown in the
Shear stress or cohesion (kPa) 60.20 table 4. The swelling pressure of black cotton soil
Angle of internal friction (o ) 21.5 with fly ash at 0%, 5%, 10%, 15%, 20%, and 25% by
Coefficient of permeability (cm/sec) 1.5x10-8 dry weight of soil are 48.3, 44.27, 38.35, 33.35, 27.6
Differential Free Swell (%) 64 and 52.9 KN/m2 respectively, which are shown in the
Swelling Pressure (kPa) 48.2 table 4.
soil classification (IS) CH
At dry condition, the swelling pressure of soil at
Table 2: Physical properties of fly ash without additive, 4% lime, and 20% fly ash are 70.15,
Properties Value 11.5, and 40.25 kN/m2 respectively which are shown
Specific gravity 2.14 in table 4. At wet condition, the swelling pressure of
Sand particle (%) 22 soil at without additive, 4% lime, and 20% fly ash are
Silt particle (%) 78 20.7, 5.75, 36.8 kN/m2 respectively which are shown
Clay particle (%) 0.0 in table 4. Fig 3-5 show the variations in swelling
Maximum dry density (g/cc) 1.104 pressure values of expansive soil, without additives
Optimum Moisture content (%) 28% and with additives as per the mix designations.
For the preparation of sample first expansive soil and From the above result it is found that the swelling
water added at OMC at different percentage of pressure of soil generally decreases when the additive
additives (lime and fly ash) and mix them in tray content increases. It is also observed that lime is more
properly and compact in a swelling pressure mould as effective than fly ash. The decrease in the swelling
per IS:2720:2002 Part XLI. Later prepared sample pressure may be explained due to the pozzolanic and
with assembly were kept in water drum and tested cation exchange reaction occurred between the soil
using swelling pressure apparatus (constant volume and additive. The reactions in lime and soil may
method) and recorded the initial and final proving ring create the new minerals as the cementatious
readings after the proving ring reading become component CSH (calcium silicate hydrate), and CAH
constant. In the case of expansive soil the swelling (calcium aluminate hydrate). These new products
pressure was evaluated based on reading constant increase the strength, reduce the swell potential and
after 10 days. For other tests in case of lime mixed change the soil classifications.
soil and fly ash mixed soils the reading become
constant after 4 days and corresponding swelling CONCLUSIONS
pressure values reported. It has been found that from the experimental
programme the swelling pressure is decreasing up to
an optimum content of additives further increase in

100 80
70
90
Swelling Pressure in KN/m2 --

Percentage Finner in ( % ) -----
60
Clay=60.0 %
Silt=34.0 % AT DRY CONDITION
80
Sand=6.0 % 50
40
70
30 AT OMC
60
20
50 10
AT WET CONDITION
0
40 0 2 4 6 8 10
1E-3 0.01 0.1 1 10
Time in days ---
Grain Size in( mm) ----- Fig. 3 Comparison of swelling pressure of expansive
Fig. 1 Grain size distribution curve of soil soil with time at dry side of OMC, at OMC, and at
wet side of OMC
60
SP=48.3
Swelling Pressure (KN/m2) 50
SP=38.35
40
30
SP=25.3
SP=19.55
20
SP=10.35
10 SP=6.9
0
0 1 2 3 4 5
Percentage of lime
Fig. 4 Variation of swelling pressure of soil with lime
at OMC
Fig. 2 Grain size distribution curve of fly ash
70
60
Table 4: Swelling pressure values of soil at OMC SP=52.9
(23%) and  OMC with lime and Fly ash SP=48.3
Swelling Pressure (KN/m 2)
50
Swelling Swelling SP=44.27
Mix Pressure Mix Pressure SP=38.35

40
( kN/m2) ( kN/m2) SP=33.35
M0L 48.3 M20FA 27.6
M1L 38.35 M25FA 52.9 30 SP=27.6
M2L 25.3 D0L 70.15

M3L 10.35 D4L 11.5 20
M4L 6.9 D0FA 70.15
M5L 19.55 D20FA 40.25 10
M0FA 48.3 W0L 20.7
M5FA 44.27 W4L 5.75 0
M10FA 38.35 W0FA 20.7 0 5 10 15 20 25
M15FA 33.35 W20FA 36.8 Percentage of fly ash
Fig 5. Variation of swelling pressure of soil with fly
ash at OMC

additives it leads to increase in swelling pressure. The
following are the conclusions
1. On increasing the lime content the swelling
pressure of soil decreases steadily to a lowest
value at 4% and then it increases slightly.
Therefore the optimum Lime content (OLC) is
equal to 4%.
2. On increasing the fly ash content the swelling
pressure of soil decreases steadily to a lowest
value at 20% and then it increases rapidly.
Therefore the optimum fly ash content (OFAC) is
equal to 20%.
3. The swelling pressure of the soil in case of dry
condition is more than the wet condition i.e. the
swelling pressure of soil is decreases with
increase in the moisture content.
4. At optimum lime content, lime retards the
swelling pressure of soil in dry as well as in wet
condition.
5. At optimum fly ash content, fly ash has less role
to retards the swelling pressure of soil in dry and
wet condition.
REFERENCES
Al-Rawas, A.A., Hago, A.W., and Al-Sarmi, H.
(2005). Effect of lime, cement and sarooj
(artificial pozzolona) on the swelling potential of
an expansive soil from Oman. Building and
Environment, Elsevier, Vol.40, 681-687.
Bell, F.G. (1996). Lime stabilization of clay minerals
and soils. Engineering Geology. Vol. 42, 223-
237.
Choudhary, A.K. (1994) Influence of Fly Ash on the
Characteristics if Expansive Soil. Indian
Geotechnical Confrence-1994 on Developments
in Geotechnical Engineering, Warangal, India,
215-218.
Erdal Cokca (2001). Use of class C fly ashes for the
stabilization of an expansive soil. ASCE, Journal
of Geotechnical and Geoenvironmental
Engineering. Vol. 127, 568-573.
Jones, D.E. and Holtz, W.G. (1973). Expansive Soils -
the hidden disaster. Civil Engineering. Vol. 43
No. 8. 49-51.
Katti, R.K. (1979). Search for solutions to problems in
black cotton soils. Indian Geotechnical Journal,
Vol. 9, No. 1.
IS: 2720- 1977 (Reaffirmed 2002) Part XLI.
Measurement of Swelling Pressure of Soil,
Bureau of Indian Standards (BIS), New Delhi.
Pandian,N.S. Krishna,K.C. and Leelavathamma B.
(2002). Effect of Fly Ash on the CBR Behaviour
of Soils , Indian Geotechnical Conference,
Allahabad,Vol.1, 183-186.
Phanikumar B.R., & Radhey S.Sharma (2004). Effect
of flyash on engineering properties of expansive
Soil. Journal of Geotechnical and
Geoenvironmental Engineering, Vol. 130, No.7,
764-767.

ISBN 978-81-921121-0-7
USE OF MULTIPLICATIVE DECOMPOSITION METHOD FOR BUS

TRAVEL TIME PREDICTION UNDER HETEROGENEOUS TRAFFIC
CONDITIONS
S.Vasantha Kumar
Ph.D Research Scholar, Dept. of Civil Engineering, IIT Madras, Chennai-36. Email: vasanth_research@yahoo.co.in
Lelitha Vanajakshi
Assistant Professor, Dept. of Civil Engineering, IIT Madras, Chennai-36. Email: lelitha@iitm.ac.in.
ABSTRACT: The provision of expected arrival time of the next bus which requires the prediction of bus travel
time is one of the key areas of advanced public transportation which aims for a congestion free traffic on the
urban roads by increasing the public transport usage. One of the popular tools used for such prediction is using
time series methods. However, most of the reported studies explored the use of advanced time series tools such
as Box-Jenkins Autoregressive (AR), Moving average (MA) or combination of AR and MA models (ARMA,
ARIMA etc.). But there are classical methods in time series analysis like multiplicative decomposition, additive
decomposition etc. developed long before the Box-Jenkins models, which are powerful and even now popular in
many areas such as financial forecasts. However, there are not many reported studies which explored the
usefulness of the decomposition methods for the problem of bus travel time prediction. The present study tries
to explore the usefulness of multiplicative decomposition method for bus travel time prediction under
heterogeneous traffic conditions. The developed model is tested for its accuracy using 41 actual bus trips data in
a particular transit route representing a typical heterogeneous traffic environment in India. The results are found
to be encouraging and the developed model is transferable to a similar environment.
INTRODUCTION increased comfort and convenience will definitely

The problem of traffic congestion and ways to tackle attract more users to public transport buses. Provision
it is a question of serious concern which needs to be of expected arrival time information of buses through
answered in most of the metropolitan cities around the display boards at bus stops/ web, providing real time
world and similar situation exists in most urban areas information on passenger loading, etc. provided either
of India. The use of personal vehicles without any pre-trip or en-route can lead to more reliability thus
restriction during procurement stage or usage stage is making public transit a hassle-free mode for urban
the major reason for traffic congestion especially in commuters.
India. The total private vehicle population in Chennai,
India is increased from 10 lakhs in 1999 to almost 32
lakhs in 2011 as shown in Fig.1. (THE HINDU dated
March 2, 2010 & Times of India dated July 11, 2011).
This account for about 300 percent rise in private
vehicle population in the last 12 years and everyday
1700 new vehicles are being registered in Chennai
(Times of India dated July 11, 2011). The increase of
private vehicles on the urban roads not only causes
traffic congestion but leads to more accidents. For
example, around 700 people die in road accidents
every year in Chennai and another 5000 gets involved
in mishaps across the city (Deccan Chronicle dated
December 25, 2009). As shown in Fig.1, during the
same period, the public transit buses were not
increased much and the numbers were still reduced
from 4235 in 2009 to 3421 in 2011. The
transportation planners suggests a minimum of 5000-
6000 buses in order to meet the demand of 5.2 million
passengers using public transit services everyday in
Chennai (Times of India dated July 11, 2011). In
Bangalore, 6100 buses are being used every day for a
population of atleast 7 lakhs less than Chennai.
Though it is important to increase the fleet size which
helps to reduce overcrowding during peak hours, this
is not the only solution to make public transport a
more attractive option. More reliable services with

Fig.1. Growth of private vehicles and public transport prediction. The data collection is detailed next
buses in Chennai during 1999 and 2011 followed by the detailed methodology and results and
The experience in developed countries shows that, discussion.
with the introduction of services of these kinds, the
increase in public transport ridership will be evident. LITERATURE REVIEW
For example, after the introduction of “Washington A brief review of the studies in the area of bus travel
Metro” and “Transport for London”, the public time prediction using data driven techniques and
transport ridership went up by 18 percent (THE model based approaches are given below.
HINDU dated October 21, 2010). Even in Chennai,
after the introduction of Air-Conditioned buses, more The data driven techniques includes regression, time
than 24 percent had shifted from cars and two- series and machine learning methods such as neural
wheelers (THE HINDU dated May 10, 2011). If one network and the model based approach mainly using
wants to develop a bus arrival time prediction based Kalman filtering technique. Bo et al. used one month
on only the time chart/schedule list provided by the weekdays bus trip data for developing a linear
Metropolitan Transport Corporation (MTC) of regression model for bus arrival time prediction in
Chennai, a prediction error of 30-40 minutes is Dalian City, China. Jeong developed five regression
observed which is unacceptable from a user models using distance and schedule adherence as
perspective (THE HINDU dated July 26, 2011). The independent variables. The GPS data collected over
Google transit, a website which aims to promote six months from Houston metro buses of Texas were
public transport has been recently launched for used as the input data. Patnaik et al. used six months
Chennai and Hyderabad. It provides the list of bus weekday data from Automatic Passenger Counting
services between the user-defined origin and [APC] systems, divided into eight categories based on
destination and approximate travel time by public different time periods of the day as input and
transit in that route. The major drawback of Google developed the regression model. The time period is
transit is that, the travel times are not based on real used as a dummy variable in the regression model.
time data and so the reliability of the travel time Ramakrishna et al. used only 25 trips as the input for
information is a question. In order to make the travel developing the regression and artificial neural
time/arrival time information more reliable, the network (ANN) model under heterogeneous traffic
prediction model used behind should be accurate and conditions. Bhandari used seven months of AVL data
should take the real time data as input. The present of transit buses in New Jersey for developing an Auto
study tries to focus on these issues and try to develop Regressive (AR) time series model for bus arrival
prediction model using one of the classical time series time prediction. Similar to Patnaik et al., the input
methods namely, multiplicative decomposition data of bus trips were classified based on seven
technique for bus travel time prediction using real categories of time periods using starting time of each
time Global Positioning System (GPS) data from trip. Chien et al. used the simulated data from the
public transit buses operating in a particular transit morning peak to develop an ANN based bus arrival
route of Chennai. The objective is to find an accurate time prediction model. Vanajakshi et al. used only
prediction methodology which can overcome the preceding two buses data for predicting the next
disadvantages of the popularly used prediction vehicle travel time in a model based approach using
methods such as Box-Jenkins ARIMA model, Kalman filtering technique (KFT) under
machine learning techniques etc. For example, Box- heterogeneous traffic conditions. Dailey et al. used
Jenkins ARIMA model has many disadvantages such KFT with input of bus trips data collected on different
as stationarity and specialized software requirement, days at the same time of the day.
transferability issues, difficulty in understanding etc.
Machine learning techniques are data driven and From the above studies, it can be seen that, most of
require huge data base for better performance, which the prediction models were developed for
may not be available in many cases such as the homogeneous traffic conditions and only few studies
present study. The present study overcomes these were concentrated on heterogeneous traffic
difficulties by using the multiplicative decomposition conditions. Also, the use of classical time series
method, which is easy to understand, simple to methods like multiplicative decomposition was not
execute using just spreadsheet program, and is not attempted yet for bus arrival time prediction. This
site-specific. Moreover, the input time series need not clear gap leads to the motivation for the present study
be stationary. Hence, the present study tries to which aims to develop a prediction model using
concentrate on whether the advantages of multiplicative decomposition technique which can be
multiplicative decomposition technique can be used for bus travel time prediction under
utilized for bus travel time prediction under heterogeneous traffic conditions.
heterogeneous traffic conditions.
DATA COLLECTION AND EXTRACTION
The following section gives a brief review of The study stretch selected for the present study was
important studies in the area of bus arrival time route number 5C, which connects the Parrys bus

depot in the northern part of Chennai, and the period t, Tt is the trend cycle component at period t,
Taramani bus depot in the southern part of Chennai
city. The average time headway between the buses St is the seasonal component (index) at period t, and
was 15-30 minutes. The total route length is 15 km et is the irregular component at period t. For
and the approximate travel time to cover the total
stretch is 60 minutes during peak hour and 40 minutes multiplicative decomposition method, Eq. (1) takes
during off-peak hour. There are 21 bus stops and 14 the form,
signalized intersections in this route. The selected
road stretch is a typical representation of an urban Yt  Tt  St  et (2)
road in India. The total route comprises road links of
different categories like major arterials, and collector The first step in multiplicative decomposition method
streets with varying volume levels. is the calculation of seasonal indexes through the use
of simple moving average (MA) and then centered
Seven public transit buses in route number 5C are moving averages (CMA). The idea behind the moving
fitted with GPS units and the data reporting is at every averages is that the average eliminates some of the
5 sec. Data from a total of 41 trips in the month of randomness in the data, and leaves a smooth trend-
September and December 2010 were used in this cycle component, i.e., deseasonalized time series data.
study. One of the previous studies (Kumar et al. 2011) The formula for calculating simple moving average of
showed that, for the selected study stretch, the most order k, or MA (k) is given as,
significant trips for predicting a current trip are same
day same time trips from one week before (W1), and m
two weeks before (W2) and the previous two trips 1
(PV1, PV2) of the same day. Hence, the data
MA(k ) 
k
Y
j  m
t j (3)
collection also involved the GPS data of all the trips
during corresponding week1 and week 2 of test dates.
Where, m  ( k  1) / 2 on either side of the time
The study stretch of 15 km is divided into 150 series. If the order of MA is odd, then CMA is
sections of each 100m length and the validation of the calculated further by taking average of the two
proposed methodology involved the comparison of consecutive moving averages, which will produce a
observed bus travel time with the predicted bus travel smooth data free from any seasonal effects. Thus, the
time in each of these subsections, for all the trips. CMA’s represent a perfect deseasonalized data.
Hence, the data extraction involved the extraction of
100m section travel times of 41 trips during test dates, The degree of seasonality, called seasonal factor (SF)
35 trips during week 1 and 38 trips during week 2 can be obtained by taking the ratio of the actual value
after checking for any outliers in the raw GPS data. to the deseasonalized value. That is,
METHODOLOGY Yt (4)
The methodology proposed here for bus travel time SFt 
prediction uses one of the classical time series method
CMAt
called multiplicative decomposition technique. The
steps involved in decomposing a time series data Now, by averaging seasonal factors of same time
using multiplicative decomposition technique to period, which will remove any irregular
predict the test vehicle (TV) travel time, is explained components (et ) , seasonal index ( St ) can be
below.
obtained for each k for Eq. (2). It is to be noted here
Steps in Multiplicative Decomposition that, the seasonal index provide a useful insight into
the time series being analyzed. The next step is to
The decomposition methods usually try to identify
two separate components of the basic underlying calculate the trend cycle component, Tt . The trend
pattern that tend to characterize the given time series, cycle component requires the deseasonalized data
namely the trend cycle and seasonal factors. The trend which can be obtained by dividing the raw data or the
cycle represents the long term changes in the series. actual value (Yt ) , by the seasonal index ( St ) . The
The seasonal factor is the periodic fluctuations of
constant length. The mathematical representation of method of least squares can then be used to fit a
the decomposition approach is, regression line using deseasonalized data as the
dependent variable and ‘t’ as the independent
Yt  f (Tt , St , et ) (1)variable, where t varies from 1 to n, and n is the total
number of data points in a time series. Once Tt and
Where, Yt is the time series value (actual data) at St were calculated, Eq.2 can be used to find the
forecasts for (n+1), (n+2) etc.

using Eq. 4 for a sample data is given in Fig.2. The
The ‘time series’ in the present study comprises purpose is to explore the usefulness of seasonal index
section travel times of W1, W2, PV1 and PV2 for for a transit authority.
forecasting the section travel times of test vehicle
(TV). Hence, for predicting a particular trip, the input
is a time series consisting of 600 section travel time
values (i.e, 150/trip × 4 trips of W1, W2, PV1 and
PV2). Thus for predicting a total of 41 trips in 2
weekdays, 24600 data points will be available. Once
the 100m section travel times of TV are computed,
they were summed up for 500m sections in order to
facilitate the mean absolute percentage error (MAPE)
calculation as explained below.
The validation of the methodology proposed above

involves the comparison of observed bus travel time
with the predicted bus travel time in each of the 30 Fig.2. Seasonal index calculated for a sample data
sections (each of length 500m) of the route, for all the
trips. The Mean Absolute Percentage Error (MAPE) is The seasonal index in Fig.2 (Y-axis) is a
used as a measure of estimation accuracy and is dimensionless value calculated for each 100m
calculated using subsections for the entire study stretch of 15 km long.
The higher the index value, higher the section travel
time of the public transit buses in that section. By
1 n tt pre  ttobs (5)
MAPETR   tt
n i 1
 100 using the index values, the transit authority can
explore the areas spatially, where the public transit
obs
buses take a longer travel time to pass through that
100m section. If dwell time is not the cause, the
where, MAPETR is the mean absolute percentage reason for higher travel time may be possibly the
error between observed and predicted travel time for signal delays. So, the transit authority may justify for
each trip (‘TR’ ranges from 1 to 41). The term ‘ tt pre ’ a bus priority system if the signal delays causes
relatively high seasonal index values or the Bus Rapid
and ‘ ttobs ’ are the predicted and observed travel times Transit (BRT) System. Thus, the intermediate results
of the bus in each subsection of the study route of multiplicative decomposition technique are also of
considered and ‘n’ is equivalent to 30, the total great significance besides the final bus travel time
number of 500m sections in the test route selected. prediction results.
The MAPE result of multiplicative decomposition The final results of bus travel time prediction using
method was compared with the real-time application multiplicative decomposition technique is presented
method, described as follows. In real-time application here. The average MAPE obtained for December 22
method, the traffic conditions when PV2 passes is and September 30, 2010 was 27 and 23 respectively.
assumed to remain the same till TV passes. The In practical sense, the MAPE can be interpreted as
assumption can be justified since the time headway follows. If the transit bus is travelling at a uniform
between PV2 and TV is in the range of 15-30 speed of 25 km/hr, it takes 72 sec. for traversing a
minutes. Thus the 500m section travel time of PV2 500m section. If the average MAPE is 25, then the
was assumed to be the predicted 500m section travel predicted bus travel time in that section would be, 72+
time of test vehicle (TV) and MAPE was calculated (25% of 72 sec.) or 72- (25% of 72 sec.). Thus the
between observed and predicted test vehicle travel predicted travel time may be 90 sec. or 54 sec. The
time using Eq. (5) for all the 41 trips. Afterwards, the final purpose of predicted section travel times is to
MAPE result of real-time application method was extract the bus arrival time at a particular bus stop. In
compared with the multiplicative decomposition order to evaluate the accuracy of the arrival time, a
method to check the performance of decomposition hypothetical bus stop is assumed to be located at a
method. cumulative distance of 10 km from the origin. A
sample of two trips that started at off-peak and peak
RESULTS AND DISCUSSION hour was taken in order to evaluate, whether the
While applying the multiplicative decomposition predicted bus arrival time at the hypothetical bus stop
technique, the moving average of order 150 is taken is reliable or not. The 500m section travel times of
(‘k’ value in Eq.3) since there are 150 section travel predicted and observed were summed up till 10 km
times available in each trip of the series. Before distance in order to arrive the predicted and observed
presenting the MAPE results of bus travel time arrival times at the hypothetical bus stop. It was found
prediction, the results of seasonal index calculated that, the absolute difference for off-peak and peak

hour trip was 105 and 312 sec. That is, for an off-peak We can observe from Fig.3-5 that, the predicted travel
trip, the difference between the actual and predicted time reasonably follows the trend of observed travel
arrival time at the hypothetical bus stop was time and hence the time series method using
approximately 2 minutes and for the peak hour trip, multiplicative decomposition technique can be
the difference is approximately 5 minutes. It is considered for bus travel/arrival time prediction under
reported that from a user perspective, a prediction heterogeneous traffic conditions.
error of 5 minutes is acceptable for a 60 minutes trip
(Schweiger et al.). A plot of observed vs. predicted The MAPE plot of multiplicative decomposition and
section travel times for sample trips during the two real time application method is shown in Fig.6 for all
test dates of December 22 and September 30, 2010 are the 41 trips of test dates. In 28 out of 41 trips,
shown in Fig.3, 4 and 5. The sample plots shows multiplicative decomposition performs better than real
predicted and observed travel times for trips started time application method.
during an off-peak hour (Fig.3), morning peak hour
(Fig.4) and evening peak hour (Fig.5). The
corresponding MAPE are 17.106 (Fig.3), 16.583
(Fig.4) and 20.725 (Fig.5).
Fig.6. MAPE plot of multiplicative decomposition

and real-time application method
CONCLUSION
Fig.3. Plot of observed vs. predicted bus travel time In India, under Jawaharlal Nehru National Urban
(trip started during an off-peak hour) Renewal Mission (JNNURM), public transport buses
in most of the urban cities are slowly being equipped
with GPS to get real time positioning of transit buses.
In Chennai, 600 buses were equipped with GPS out of
the total fleet size of 3400. Process is underway to
install GPS in more number of buses. This real time
positioning can effectively be used to develop a bus
arrival time information system which can be
deployed at bus stops through variable messaging
signs (VMS) or through dedicated websites. The en-
route information on what time the next service will
arrive will help the commuters to decide upon
whether to wait or not. If the information is provided
pre-trip through websites, the users can reduce their
Fig.4. Plot of observed vs. predicted bus travel time waiting times at bus stops. Deploying such an
(trip started during morning peak hour) Advanced Public Transportation Systems (APTS) will
not only reduce the number of personal vehicles on
the road but will increase the revenue of the public
transport services by more number of transit users.
The present study tries to focus on this problem of
accurate prediction of bus arrival time using one of
the classical time series methods namely the
multiplicative decomposition technique. The results
showed that, the average MAPE for the two test dates
is around 25%. It was found that, the decomposition
method is able to predict the arrival time within 2-5
minutes of the actual arrival time for a hypothetical
bus stop. The performance of decomposition method
Fig.5. Plot of observed vs. predicted bus travel time
was compared with a base line method called real
(trip started during evening peak hour)
time method and the decomposition method was
found to be performing better in 28 out of 41 cases.

Thus under heterogeneous traffic conditions for bus Vanajakshi, L., Subramanian, S.C. and Sivanandan,
arrival time prediction, multiplicative decomposition R. (2009). Travel time prediction under
method can be considered. It is to be noted here that, heterogeneous traffic conditions using global
when compared to ARIMA models, the positioning system data from buses. IET Intell.
decomposition technique is easy to understand and Transp. Syst, 3(1), 1 – 9.
can be implemented using a simple spreadsheet
program. Also, there is no site-specific issue, because
there is no specific model involved for the selected
study stretch. Hence, there is no transferability issue
with decomposition technique when compared to
ARIMA models.
ACKNOWLEDGMENT
The authors acknowledge the support for this study as
part of the project CIE/10-11/168/IITM/LELI by
Ministry of Urban Development, Government of
India, through letter No. N-11025/30/2008-UCD.
REFERENCES
Bhandari, R.R. (2005). Bus arrival time prediction
using stochastic time series and Markov chains.
Doctoral Thesis, Dept. Civil Eng., New Jersey
Institute of Technology, Newark.
Bo, Y., Jing, L., Bin, Y. and Zhongzhen, Y. (2009).
An adaptive bus arrival time prediction model.
Proceedings of the Eastern Asia Society for
Transportation Studies, Vol.7, 2009.
Chien, S.I., Ding, Y. and Wei, C. (2002). Dynamic
bus arrival time prediction with artificial neural
networks. Journal of Transportation
Engineering, 128(5), 429 – 438.
Dailey, D.J. (1999). An algorithm for predicting the
arrival time of mass transit vehicles using
automatic vehicle location data. Proceedings of
the 70th Annual Meeting, CD-ROM,
Transportation Research Board of the National
Academics, Washington, D.C.
Jeong, R.H., and Rilett, L.R. (2005). Bus arrival time
prediction model for real time applications.
Proceedings of the 84th Annual Meeting, CD-
ROM, Transportation Research Board of the
National Academics, Washington, D.C.
Kumar, S.V., Chaitanya, K., Koppineni, A. and
Vanajakshi, L. (2011). Travel Time Pattern
Identification under Heterogeneous Traffic
Conditions for Bus Arrival Time Prediction. 2nd
National Conference on Intelligent
Transportation Systems, Mumbai, India.
Patnaik, J., Chien, S. and Bladikas. A. (2004).
Estimation of bus arrival times using APC data.
Journal of Public Transportation, 7(1), 1 – 20.
Ramakrishna, Y., Ramakrishna, P., Laxshmanan, V.
and Sivanandan, R. (2006). Bus travel time
prediction using GPS data. Proceedings of the
Map India 2006, New Delhi.
Schweiger, C.L. (2003). Real-time Bus Arrival
Information Systems - A Synthesis of Transit
Practice. TCRP 395 Synthesis 48, Transportation
Research Board, Washington, D.C.

ISBN 978-81-921121-0-7
INFLUENCE OF TYPE OF BINDER AND CRUMB RUBBER ON THE

MORPHOLOGY OF CRUMB RUBBER MODIFIED BITUMEN
P. S. Divya
Department of Civil Engineering, IIT Madras, Chennai - 600036.
J. Murali Krishnan
Associate Professor, Department of Civil Engineering, IIT Madras, Chennai - 600036.
ABSTRACT: Crumb rubber modified bitumen (CRMB) provides enhanced performance properties when
compared to unmodified bitumen. To understand the influence of various factors such as crumb rubber type,
dosage rate and bitumen type on the improved performance, morphological investigations were conducted at IIT
Madras. In this study, the microstructure of CRMB was observed using Environmental Scanning Electron
Microscope. Coarse and fine graded rubber particles were blended with air blown and blended bitumen. Three
dosage rates for rubber particles (8, 10 and 12%) were used. ESEM images revealed the network fibril like
structure of CRMB and their change with the type of binder, dosage rate and gradation of rubber particles. In
order to study the chemical composition analysis of the network fibril of CRMB, Energy Dispersive X-ray
Spectrometry (EDX) analysis was carried out on ESEM images. EDX analysis showed that the network fibrils
for CRMB were different from the unmodified binders due to the presence of rubber particles.
Keywords: Environmental Scanning Electron Microscopy, Crumb Rubber Modified Bitumen, Energy
Dispersive X-ray Spectrometry, Image Processing, Air blown bitumen, Blended bitumen.
INTRODUCTION Heitzman (1992), the rate of interaction depends on the

The increased applications of Crumb Rubber Modified interaction temperature and duration, size and surface
Bitumen (CRMB) in pavements are due to its improved profile of CRM, amount of work done during mixing
performance characteristics. The property enhancement and the amount of lighter aromatic fractions in bitumen.
due to the addition of modifiers to bitumen depends The amount of lighter fractions in bitumen depends
upon the type of modifier. CRMB provides primarily upon its crude source and processing method
improvement against rutting, fatigue cracking and low (Corbett, 1984). Processing methods such as air
temperature cracking (Bahia et al., 2001). blowing significantly affect the percentage of lighter
According to Whiteoak (1990) bitumen is a colloidal fractions available in bitumen (Glozman and
system consisting of asphaltenes, resins and oil. The Akhmetova, 1970; Patwardhan and Khade, 1982).
high molecular weight asphaltenes are surrounded by Buckly and Berger (1962) found that the time required
resins and this micelle structure is dispersed in oil. The for swelling is proportional to the square of the particle
asphaltene/resin structures changes with the processing size which means that while finer particles interact
conditions, it also changes when the binder is modified. much faster, it may not be the same with coarser
The rheological properties of the modified binders are a particles. Abdelrahman and Carpenter (1999)
consequence of the arrangement and association of investigated that fine rubber particles affect the liquid
asphaltene particle + modifier in the oily phase. phase of the binder whereas the coarse rubber particles
In order to study the effect of Crumb Rubber Modifier affect the binder matrix phase.. With the increase in
(CRM) in binders the interaction mechanisms have to interaction temperature, the rate of swelling of rubber
be understood. Addition of rubber particles to bitumen increases but the extent of swelling decreases (Green,
results in complex interaction mechanisms between 1977). From these studies, it can be understood that
rubber particles and bitumen. The interaction interaction mechanisms results in the change in
mechanisms are normally classified as swelling of morphology of the rubber-bitumen blends. Hence it is
rubber molecules, dissolution of rubber in bitumen and necessary to understand the change in microscopic
devolatilization and cross-linking in rubber (Shen et al., structure of CRMB.
2009). When crumb rubber is added to bitumen, the Various microscopic techniques such as Optical
lighter aromatic fractions are absorbed by the rubber Microscopy, Atomic Force Microscopy (AFM),
particles (Heitzman, 1992; Frantzis, 2004). The bitumen Transmission Electron Microscopy (TEM) and
– rubber interaction is not chemical in nature; but, fluorescence microscopy are used to study the internal
rubber particles swell due to the absorption of bitumen's structure of modified binders. These techniques are able
oily phase at very high temperature (above 160 °C) to find only the dispersion of modifiers in the binders
leading to a gel-like structure (Heitzman, 1992). and are not able to capture the internal network
Swelling of rubber particles depends upon various structure of the binders (Airey et al. 2002; Chen et al.
factors related to bitumen and rubber. According to 2002; Morale et al. 2005; Masson et al. 2005). Rozeveld

et al. (1997) used Environmental Scanning Electron
Microscopy (ESEM) to study the change in internal
structure of unmodified binders as well as Polymer
Modified Bitumen (PMB). ESEM technique was
considered as the best technique for studying the change
in network structure for oil bearing materials like
bitumen. From their study it was understood that
internal structure of unmodified binders as well as PMB
consists of network fibrils. In the case of PMB these
network fibrils were strongly aligned in a particular
direction whereas in the case of unmodified binders the
network fibrils were randomly arranged. It was also (a) (b)
found that this network structure changes with dosage Fig. 1 ESEM image for blended bitumen with 8 %
rate of polymer as well as with the aging of binders. coarse CRM at 250 X
From the above discussions, it is clear that to develop
an understanding of CRMB, it is necessary to carefully With initial trials, it was found that an exposure time of
understand and quantify the change in internal structure 8 minutes is required to observe noticeable changes in
of CRMB. The aim of this investigation is to use ESEM the microstructure and hence all the images were
and explore the morphology of a wide variety of captured for 8 minutes of exposure time. The network
CRMB. structure obtained for blended bitumen with 8% coarse
CRM at a magnification of 250 X is shown in figure 1
MATERIALS AND METHODS (b). Figure 2(a) shows the ESEM image for CRMB
In total, 12 CRMB samples, prepared from two different made with blended bitumen and 2(b) shows a similar
type of binder (PDA blended and air blown), two image with blown bitumen containing 12 % fine CRM.
gradation (fine and coarse) and three dosage rates (8, 10 The difference in the microstructure developed for
and 12 %) of CRM were used. The unmodified binder blended and blown bitumen can be clearly seen.
used corresponds to VG 30 grade as per Indian
standards (IS73:2006, 2006). All the blends were
prepared in the pilot batching plant at R&D facility of
Bharat Petroleum Corporation Limited, Greater Noida,
U.P, India. Table 1 gives the details of gradation of
CRM used in this investigation.
Table 1 Gradation of CRM

Mesh size Fine CRM Coarse CRM
(µm) (% retained) (% retained)
600 0.48 0.2
425 3.46 1.77
250 50.02 34.76 (a) (b)
150 40.95 54.82 Fig. 2 ESEM image for blended and blown bitumen
75 4.17 6.26 with 12% fine CRM
Pan 0.92 2.19
Figure 3(a) and (b) shows the ESEM for blended
bitumen with 12 % coarse and fine CRM. The
The internal structure characterization was done using difference in the network structure between fine and
ESEM, Quanta 200, available at the Department of coarse CRM are clearly observed.
Metallurgical and Materials Engineering, IIT Madras.
All the samples were preconditioned by heating it to
150 ºC and were mixed to get a uniform blend. These
samples were spread on glass slides as thin films of
thickness about 1 mm and were observed under electron
beam under four different magnification scales- 250,
500, 1000 and 2000 X.
ESEM RESULTS
Initially when the sample was exposed to electron
beam, the sample surface was featureless as shown in
figure 1(a).
(a) (b)
Fig. 3 ESEM image for blended bitumen with 12%
coarse and fine CRM

Figure 4(a), (b) and (c) shows the ESEM images for determine the chemical composition. It is to be noted
blended bitumen with 8, 10 and 12 % fine CRM. that the chemical composition analysis with EDX is less
precise when compared to other methods. It is also to be
noted that EDX cannot detect lighter elements typically
below the atomic number of Na for detectors which are
equipped with Be window (Shindo and Oikawa, 2002,
Dragnevesk et al., 2009).
EDX spectrum is a plot of X-ray count vs Energy (in
keV). The energy peaks corresponding to the different
elements present in the samples is also shown. Figure 6
(a) (b) (c) shows the EDX spectrum for crumb rubber.
Fig. 4 ESEM image for blended bitumen with 8, 10 and
12 % fine CRM
The ESEM images for unmodified blended and air

blown bitumen were also captured. Figure 5(a) and 5(b)
shows the ESEM images for unmodified blended and
air blown bitumen. The network structure of unmodified
blended and air blown bitumen does not show any
difference in terms of fibril density and diameter but it
varies due to the addition of rubber.
Fig. 6 EDX Spectrum for Crumb rubber
(a) (b)
Fig. 5 ESEM image for unmodified blended and blown
bitumen
According to the ESEM studies carried out on

unmodified binder and PMB by Rozeveld et al. (1997),
the network structure of unmodified binder consists of Fig. 7 EDX Spectrum for CRMB
asphaltene and resins but the network structure of PMB
consists of asphaltene, resins and to some extent the
constituents of the modifier.
ENERGY DISPERSIVE X-RAY

SPECTROMETRY (EDX) ANALYSIS
In order to understand the elemental composition of the
network fibril of the CRMB, Energy Dispersive X-ray
Spectrometry (EDX) analysis was carried out on the
ESEM images obtained after 8 minutes for all the
samples as well as on the crumb rubber particle. EDX is
normally used to find the chemical composition of the Fig. 8 EDX Spectrum for VG 30
materials and create elemental composition maps. The
EDX system available at IIT Madras used a software Figure 7 and 8 shows the EDX spectrum for CRMB and
called “EDX-Genesis” to collect and analyze energy VG 30 respectively. The difference between unmodified
spectra. When EDX is used in spot mode, the full binder, crumb rubber and CRMB are seen. It can be
elemental spectrum can be recorded in very little time. noted that the elemental peak heights between VG 30
The software is used to identify the peaks and this and CRMB are different and this may be hypothesized
makes EDX a comprehensive survey tool to identify due to the presence of rubber particles in the network
unknown constituents prior to an accurate quantitative fibrils. This also provides an extra proof that the
analysis. Typically the peak height ratios are used to network fibrils are not beam induced artifact.

IMAGE PROCESSING bitumen difference exists between the dosage rate of
The ESEM images were analyzed using GIMP- image CRM.
processing software. This analysis was done by Table 3 Diameter for different samples
assuming isotropic morphology of the material (Stangl Crumb Dosage Diameter
et al., 2006). Initially the junction points were identified rubber type rate (%)
and the centre line was traced. Using the centre line, the Blown Blended
inter-junction length and the diameter was determined. Unmodified 0 Mean 11.5 36.2
Determination of number of junction points and tracing S.D. 13.7 14.2
the centre line was purely based on eye-judgment Coarse 8 Mean 34.5 35.9
(Figure 8). S.D. 6.8 19.4
10 Mean 23.6 37.8
S.D. 8.9 14.7
12 Mean 20.6 30.2
S.D. 5.9 2.5
Fine 8 Mean 30.2 15.4
S.D. 3.4 3.6
10 Mean 23.7 16.0
S.D. 3.2 5.0
12 Mean 21.4 13.8
S.D. 6.5 2.0
Fig. 8 Image Processing
Fig. 9 shows the variation of diameter of different
The diameter of the network strands varies especially
CRMB.
near the junctions and hence the diameter was measured
at different locations and the average was taken As the
inter-junction length and diameter varied over a wide
range of values, these parameters were studied in
conjunction with standard deviation of the observations
(Table 2 and 3)
Table 2 Inter-junction length for different samples
Crumb Dosage Inter-junction
rubber type rate (%) Length
Blown Blended
Unmodified 0 Mean 38.8 36.2
S.D. 13.7 14.2
Coarse 8 Mean 34.5 35.9 Fig. 9 Diameter for different CRMB
S.D. 6.8 19.4
10 Mean 23.6 37.8
S.D. 8.9 14.7
12 Mean 20.6 30.2
S.D. 5.9 2.5
Fine 8 Mean 30.2 15.4
S.D. 3.4 3.6
10 Mean 23.7 16.0
S.D. 3.2 5.0
12 Mean 21.4 13.8
S.D. 6.5 2.0
Image processing results revealed the influence of type Fig. 10 Inter-junction length for different CRMB.
of bitumen, gradation and dosage rate of CRM on the
number of junctions, inter-junction length and diameter Figure 10 shows the inter-junction length for different
of network strands. samples. Due to the addition of rubber the inter-junction
The red coloured lines correspond to the error bar to a length of the binders decreased. Inter-junction length
standard deviation of 2 µm and the green line also depends on gradation of CRM for blended bitumen
corresponds to the fibril diameter for unmodified and on dosage rate of CRM for blown bitumen.
blended and blown bitumen. It is clear from the figure
that due to rubber addition the fibril diameter had SUMMARY AND CONCLUSIONS
increased. For blended bitumen, considerable difference ESEM studies on the different types of CRMB revealed
exists between the gradation of CRM whereas for blown the fact that the morphology of CRMB varied with the
type of binder, dosage rate and gradation of CRM. It

was found that the binder surface was initially plain and Corbett, W., L. (1984). “Refinery processing of asphalt
the network strands were observed after some minutes cement." Transportation Research Record, 999,1-6.
of beam exposure. EDX analysis done on ESEM images Dragnevski, K. I., Donalda, A. M., Clarkeb, S. M., and
showed that the difference in the network fibrils of Maltby, A. (2009). “Novel applications of ESEM
CRMB from unmodified binders exists due to the and EDX for the study of molecularly thin amide
addition of crumb rubber. The analysis of the ESEM monolayers on polymer films”. Colloids and
images by image processing provided an useful Surfaces A-Physicochemical and Engineering
indication of change in network structure with the Aspects, 337(1-3), 47-51.
processing and modification of binders. Frantzis, P. (2004). “Crumb rubber-bitumen
Based on the results of this study, the following interactions: Diffusion of bitumen into rubber.”
conclusions can be drawn: Journal of Materials in Civil Engineering, 16(4),
 ESEM is a powerful technique and when 38-390.
coupled with EDX, can unravel the internal Glozman, E. P. and Akhmetova, R. S. (1970).
structure of modified binders. “Chemical composition of bitumen components."
 EDX analysis showed that the network fibrils Chemistry and Technology of Fuels and Oils, 6(5),
obtained are not beam induced artifact and the 364-367
difference in network structure of CRMB from Green, E. and Tolonen, W. (1977). “The chemical and
unmodified binders are due to the addition of physical properties of asphalt-rubber mixtures, part
crumb rubber. I basic material behavior." Report No. FHWA-AZ-
 Image processing technique showed that the HPR14-162, Arizona Department of
internal structure varied with type of binder, Transportation.
gradation and dosage rate of CRM. Heitzman, M. (1992). “State of the Practice- Design and
 The fibril diameter and inter-junction length construction of asphalt paving materials with
depended more on gradation of CRM for crumb rubber modifier." FHWA Technical Report:
blended bitumen and for air blown bitumen A-SA-92-022, Federal Highway Administration,
these parameters changed with dosage rate of Washington D. C.
CRM Masson, J. F., Leblond, V., and Margeson, J. (2005).
“Bitumen morphologies by phase detection atomic
ACKNOWLEDGMENTS force microscopy”, Journal of Microscopy, 221(1),
The authors would like to acknowledge here Ms. Sonal 17-29.
Maheswari, Senior Scientist, R&D, M/s Bharat Morales, G. M., Partal, P., Navarro, F. J., Boza, M. F.,
Petroleum Corporation Limited for manufacturing these Gallegos, C., Gonzalez, N., Gonzalez, O., and
binders and sharing the same with us. Our thanks are Munoz, M. E. (2004). “Viscous properties and
due to Ms. Shanthi, Material Testing Facility at the microstructure of recycled eva modified bitumen."
Department of Metallurgy, IIT Madras for help with the Fuel, 83(1), 31-38.
use of ESEM. Patwardhan, S. R. and Khade, S. B. (1982). “Kinetics of
air-blowing of petroleum bitumen." Industrial &
REFERENCES Engineering Chemistry Process Design and
Abdelrahman, M. A. and Carpenter, S. H. (1999). Development, 21(1), 154-157.
“Mechanism of interaction of asphalt cement with Rozeveld, S. J., Shin, E. E., Bhurke, A., France, L., and
crumb rubber modifier." Transportation Research Drzal, L. T. (1997). “Network morphology of
Record, 1661, 106-113. straight and polymer modified asphalt cements”,
Airey, G. D., Singleton, T. M., and Collop, A. C. (2002). Microscopy Research and Technique, 38(5), 529-
“Properties of polymer modified bitumen after 543.
rubber-bitumen interaction”, Journal of Materials Shen, J., Amirkhanian, S., Xiao, F., and Tang, B.
in Civil Engineering, 14(4), 344-354. (2009). “Influence of surface area and size of
Bahia, H. U., Hanson, D. I., Zeng, M., Zhai, H., Khatri, crumb rubber on high temperature properties of
M. A., and Anderson, R. M. (2001). crumb rubber modified binders”, Construction and
“Characterization of modified asphalt binders in Building Materials, 23(1), 304-310.
superpave mix design.” NCHRP Report No. 459, Shindo, D., and Oikawa, T. (2002). Analytical Electron
Transportation Research Board, National Research Microscopy for Materials Science, Springer,
Council, Washington, D.C. Tokyo.
Buckly, D. and Berger, M. (1962). “The swelling of Stangl, K., Jager, A., and Lackner R. (2006).
polymer system in solvents." Journal of Polymer “Microstructure-based identification of bitumen
Science, 56(163), 163-174. performance”, Road Materials and Pavement
Chen, J. S., Liao, M.C., and Shiah, M. S. (2002). Design, 7, 111 - 142.
“Asphalt modified by styrene-butadiene-styrene Whiteoak, D. (1990). Shell Bitumen Handbook. Shell
triblock copolymer: Morphology and model.” Bitumen, UK, Surrey.
Journal of Materials in Civil Engineering, 14(3),
224-229.

ISBN 978-81-921121-0-7
HYDROINFORMATICS IN GROUNDWATER EXPLORATION: A

CASE STUDY
D.K. Pardeshi
Post graduate student, Department of Civil Engineering, Government College of Engineering, Aurangabad 431005,
Email: erdipeekapardeshi@gmail.com
A.O. Rathor
Asst.Professor, Department of Civil Engineering, Rajarshi Shahu College of Engineering, Pune 411045,
Email: ashutosh_rathor@hotmail.com
D. G. Regulwar
Associate Professor, Department of Civil Engineering, Government College of Engineering, Aurangabad 431005,
Email:dgregulwar@rediffmail.com
ABSTRACT: GIS is a tool for storing, manipulating, retrieving and presenting both spatial and non-spatial data
in a quick, efficient and organized way. Preparation and maintenance of data in the form of maps and referenced
tabular files itself can be considered as a primitive form of Geographic Information System (GIS). In the present
investigation Remote Sensing and GIS techniques have been in order to reveal the mode of occurrence,
movement and potentiality of ground water. In this study an integrated remote sensing and GIS based
methodology is developed and worked for the exploration of ground water resources. The different thematic
maps have prepared by using Quantum GIS software. The analysis is carried out by using weighted index
overlay method. It has been followed to explore groundwater potential zone map. The information of all layers
were generated and integrated to prepare groundwater potential map. This output map is identified with the help
of hydrology, geology and geomorphic character. The output values are generated with the help of groundwater
potential zone map and classified in four types such as poor, moderate, good and excellent. Also the ground
water prospect zones gives output values such as excellent 21%, good 23%, moderate 26% and poor 30%.
Keywords: Groundwater potential zone, Geographic Information System, Remote Sensing, Weighted index
overlay method.
INTRODUCTION
Groundwater is one of the most valuable resources, dependent features such as lithology, geomorphology,
which supports human health, economic development soil, land use/land cover, structures, slope, lineament,
and ecological diversity. The importance of drainage pattern and hydrology using remotely sensed
groundwater for the existence of human society satellite data along with GIS. It is also used to prepare
cannot be overemphasized. Groundwater is the major database in the layers, an integrate map preparation
source of drinking water in the area of rural and and analysis of combined layer. The field data is
urban. Also, it is an important source of water for the collected for quantifying various lithological and
agricultural and the industrial sector. Being an geomorphological units with reference to their
important and integral part of the hydrological cycle, potential for groundwater occurrence.
their availability are depends on the rainfall and
recharge conditions. Till recently it had been STUDY AREA
considered a dependable source of water. India is The study area is bounded by North latitudes
heading towards a freshwater crisis mainly due to 19016’30” to 19021’0” and East longitudes 74 013’30”
improper management of water resources and to 74016’30” falls in the Sangamner Taluka of
environmental degradation, which has lead to a lack Ahmednagar district in Maharashtra and covered by
of access to safe water supply to millions of people. Survey of India toposheet no.47 I/7. In the hard rock
This freshwater crisis is already evident in many parts terrain of Eastern part, the groundwater occurs in an
of India, varying in scale and intensity unconfined condition within the fractured hard rocks
In recent years, the use of remote sensed data and and its weathered mantle. The potential aquifer in this
Geographic Information System (GIS) application has part comprises of fractured hard rock’s with
been found increasing in a wide range of resources secondary porosity that extends up to 50 from the
inventory, mapping analysis, monitoring and surface. In the laterites and the alluvial areas
environmental management. Remote Sensing groundwater is in unconfined state in shallow aquifers
provides very useful methods of survey, and in confined state below a blanket of clay of
identification, classification and monitoring several varying thickness in deeper aquifers. The availability
forms of earth resources and also helps in acquisition of groundwater is problematic because of the annual
of data in a short time at periodic intervals. The rainfall in the watershed is relatively moderate, the
process of GIS is to input sets of raw data to produce low recharge rate, generally high runoff mainly from
useful output information. The study includes western part of the watershed, unsuitable aquifer
demarcation of groundwater potential zones which is conditions over the parts of hard rock areas, long and
consider ground water as a function of the various hot dry season, excess withdrawals of groundwater in
parts of the alluvial tract for irrigation purposes have

caused continuous lowering of the groundwater table lineament, soil, contour, slope, drainage density
over the years. whose suitability analysis has been performed purely
May is the hottest month of the year with the mean from a groundwater point of view and does not
daily maximum temperature at 38.9° C and the mean include geotechnical considerations.
daily minimum at 22.4° C. On individual days during
DATA AND METHODOLOGY
the hot weather period temperature occasionally goes
Survey of India Toposheet (SOI) No. 47 I/7 using to
up to 43° or 44° C. With the onset of South-West
prepare base map, contour map and drainage map.
monsoon in the district there is an appreciable drop in
Linear Image Self Scanning sensor IV data of the
temperature and weather becomes pleasant. With the
Indian Remote Sensing Satellite P6 (IRS-P6 LISS IV)
withdrawal of the monsoon by about the first week of
with 5.8m resolution satellite data used and it helps to
October day temperatures increase slightly and a
prepare Digital elevation model (DEM), FCC,
secondary maximum day temperatures is recorded.
Lineament density, observable terrain features like
However the night temperatures decrease steadily
geological, geomorphologic, structural features and
after the withdrawal of the monsoon. The highest
their hydrological characteristics. The visual
maximum temperature recorded at region was 43.7° C
interpretation of IRS P6 data provided information
on 9th May 1960 and the lowest minimum 2.2° C on
pertaining to hydrogeomorphic features was very
7th January 1945.
useful in understanding the nature and water
HYDROLOGY potentiality of different landforms. With the help of
Intensive fieldwork has been carried out in the study Geographic Resources Analysis Support System
of structural features of flows, joint and dykes. The (GRASS), Quantum GIS software the study area
water, which occurs below the surface of the land is georeference and digitized, also help to prepare
invisible and not readily accessible and consequently thematic map such as Soil texture, slope, land used
it always possesses on element for its occurrence. In and land cover, geology, geomorphology and
order to understand the nature of occurrence of groundwater potential map.
ground water, it is necessary to study the hydrological The prime objective of the study was to evaluate the
framework and hydrogeomorphological settings groundwater potential of the regions, the approach
controlling the sub-surface and aerial distribution of was to evaluate all the relevant factors viz.,
ground water. The hydrogeomorphic unit of study geological, geomorphologic, metrological,
area is shown in fig 1. The ground water is not geophysical and hydrological details. Remote sensing
universally present everywhere therefore, it is data was used to derive some of the input parameters
necessary to evaluate the geological horizons, and GIS was used as an analytical tool for integrated
geomorphic landforms governing the ground water analysis. Integrating the information on these multiple
occurrence and their yielding capacities. aspects lead to identification and delineation of the
Hydrological conditions control the occurrence of zone, which are favorable for groundwater recharges,
water and the sub-surface rocks govern its distribution storage and exploration.
and movement while the hydro geomorphic landforms The literature survey covered collection of existing
govern the surface flow and infiltration. In this literatures on ground prospect zones, artificial sites
approach, all the variables controlling ground regime and collection of existing data from different
such as geology, geomorphology, groundwater Government organization. In the field work
occurrence, artificial recharge of groundwater, surface geographical location of the road intersection and nala
water and drainage. cutting center was determined by a Global Positioning
Alluvium is the most important water bearing Systems (GPS) at an accuracy of < + 5 in horizontal.
formation cover lesser portion of the study area. The Rock samples were collected across the nala cutting,
heavy monsoon rainfall results in quick recharge of hillside and dykes for laboratory analysis. Broken
the alluvium aquifers and water levels attain their samples of surface rock (as weathered as possible)
highest elevation during July – August period, they was collected for geological description. Properties of
start declining from the October with the rate the alluvium, soil and under laying parent material
increasing after November. The saturated thickness was recorded.
of alluvium aquifer is very little during the period of Laboratory works include the generation of thematic
March to May. Almost all the dug wells inventories in maps from the Survey of India (SOI) topographical
the study area penetrating in alluvium (which from the sheets and (IRS) data. A perspective view of this map
main aquifer). Recharge ponds and check dams has been generated to enhance the understanding of
provide a good measure of groundwater potential the role of topography in reservoir site selection.
zone in hard rock terrains by collecting surface runoff Various thematic maps were prepared on 1:50000
and increasing the surface area of infiltration. scales and integrated using Open source image
Suitability of these structures depends on various processing in GRASS. The important steps in the
factors, which can be identifying using GIS methodology are the various thematic maps such as
techniques. Weighted indexing has been adopted to base map, geology, geomorphology, soil, land
suggest the ideal locations for groundwater using 7 use/cover, lineament, drainage, and contour maps are
parameters namely geology, geomorphology, generated through conventional filed methods using

the Survey of India (SOI) toposheet and IRSP6-LISS for to evaluate groundwater potential zone. Map no. 1,
IV imagery hard copy and IRS PAN Imagery and 2, 3, 4 are gives details of Geology map,
digital data, the thematic maps were converted into Geomorphology map, Soil map, Lineament map
the vector format using digitization in QGIS software, respectively and these were used for weightage
the Weightage overlay assignment work in GRASS analysis.
Fig. 1 Hydrogeomorphic unit
GIS ANALYSIS AND MODELING

The layered information obtained for study areas from basis for categorization. Digital image processing
various sources viz. satellite data, topographic map involves the manipulation and interpretation of digital
and other thematic map need to be integrated for the images with the aid of computer. In general satellite
prediction of an area referenced groundwater potential data can be interpreted in two ways; the first one is
zones. In this study a weightage overlay model is visual interpretation by using hard copy of satellite
developed for this purpose. The modeling concept data printed in a photographic paper, the second
used in this study is based on the fact that any method is digital image processing, where
thematic map pertaining to a given study provides interpretation, mapping is done with the help of
information on different aspects of groundwater computer. Digital image processing may involve
potential. numerous procedures including formatting and
correcting of the data, digital enhancement to
Digital Image Processing facilitate better visual interpretation, or even
In order to make use of digital data available from automated classification of target and features entirely
satellite imagery, it must be processed in a manner by computer. In order to process remote sensing
that is suitable for the end user. For many projects this imagery digitally, the data must be recorded and
processing includes categorizing the land into its available in a digital form suitable for storage on a
various use functions. The classification process computer tape or disk obviously, the other
categorizes all pixels in a digital image into one of requirement for digital image processing is a
several land cover classes, or "themes". This computer system, with the appropriate hardware and
categorized data may then be used to produce software to process the data. In the present study,
thematic maps of the land cover present in an image. initially, the IRS satellite data and the image
Normally, multispectral data are used to perform the processing facilities available with GRASS GIS
classification and indeed, the spectral pattern present software was used.
within the data for each pixel is used as the numerical
Supervised Classification

Supervised classification involves training of the area is underlain by hard rock, the occurrence of
computer to identify patterns on the basis of ground ground water is greatly controlled by the nature of
truth, formulates decision rule for categorization and bedrock, geomorphology soil, slope, land use,
then use the above decision rules to classify the rest of lineament and drainage density. The weightage have
the image data. In supervised classification, we need been given each maps such as 20 for geology, 20 for
to identify classes and extract the information, i.e. geomorphology, 15 for soil, 15 for slope, 15 for
Land covers types of interest in the satellite image. lineament density respectively.
Extraction of unmixed spectral value from The classes (categories) in each map were given class
homogeneous land unit is called training site this values in such a way that the class, which is highly
stage is called training stage. The ERADAS 8.6 / favorable for ground water potential, was given the
GRASS GIS software is used to develop a statistical highest value i.e. of the attribute itself. The weightage
characterization of the reflectance for each given for the attributes and the values corresponding
information class. This stage is often called signature to classes are shown in table 1. The integration
analysis. Once a statistical characterization has been analysis was carried out by superimposing all the nine
achieved for each pixel and making a decision about maps (attributes) by multiplying the weightage for
which of the signatures it resembles most. There are each map with corresponding weightage for each
several techniques available for making these decision class. In this analysis geology and geomorphology are
and are often termed as classifiers. GRASS for considered to be the most important parameter for
windows offer different classification methods for the ground water potential and hence value 20 was
present study. Maximum likelihood classifier is used assigned shown in table no.1.
to classify different land cover units.
Nk
Reclassification and Grading of Maps GWP = [S (Wk Vjk)]/100
Creating a new raster model by classification is often (1)
referred to as reclassification, reading or transforming K=1
through look-up tables. Two methods of data
Where nk = number of attributes, Wk = weightage of
reclassification may be used. The first method is a one
attribute k, Vjk = value assigned to class j of attribute
to one change, meaning a cell value in the input grid
k. Eq. (1) gives the groundwater potential of the
is assigned a new of 1 in the output grid. In
products of weightage factors for all the maps and the
reclassification three maps were used. While
class values of each pixel are summed up and 100
reclassification these maps the standard scheme for
divided the sum.
reclassifications followed some parameters while
knowledge based ranking was followed for other Table 1 Weightage assignments of various thematic
parameters. While reclassification of the different maps
maps, ranking values ranging from 1 to 5 were Thematic map Class Category Rank Weightage
assigned for different classes within each map. Prior Geomorphology Denudation Hill Poor 1
to reclassification, all the 7 thematic maps in the Plateau Good 3 20
vector format have to be rasterised. All the maps are Water Body Excellent 5
Geology Compact Unjointed Basalt Poor 1
in vector polygons. Rasterisation of vector polygon Dyke Good 3 20
can be done using GRASS. After rasterisation, Jointed & Fractured Basalt Excellent 5
reclassification of Drainage, Lineament density map Soil Gravelly Clay Loam Poor 2
Gravelly Loam Good 3 15
is done using reclassify. Gravelly Sandy Loam Excellent 5
Slope Moderately steep Slope Poor 2
Weightage Assignment Moderate Slope Moderate 3
15
To delineate the groundwater potential zones Gentle Slope Good 4
Very Gentle Excellent 5
integrated analysis was carried out, also in process
Lineament density Low Poor 1
digitized thematic maps such as geology, Medium Moderate 2
15
geomorphology, soil, land use/cover, slope, drainage, High Good 3
lineament and depth of water table were used for this Very High Excellent 4
analysis. The attributes are assigned weightage (Wk) Spatial Overlays and Analysis
depending on their relative influence on ground water The analysis work was carried out with after
potential and recharge sites. Similarly the categories weightage assignment and reclassified of all maps.
of each attribute were also assigned numerical values Analysis is performed for generating ground water
(Vjk) based on their relative influence. This enabled prospect map with the help of GIS analysis software.
in performing numerical integrated analysis and also The image calculator available in GRASS is used for
semi-quantitative evaluation. A similar numerical this analysis. Mathematical operation was used where
weighted approach was developed during the present the rank assigned to different classes within each
study called the “Numerical analysis of multi- parameter was multiplied with the weightage and
thematic information (NAMTI)”.The weightage finally all of the seven parameters used for analysis
values for attributes and classes were assigned based were added up together with the help of image
on relative influence on ground water potential. Since calculator. The final output shows the map of ground

water potential map no.5.The output values for all the Spatial Value Output Ground Water Prospect Zones
four classes varied from poor, moderate, good and 15 – 18 Excellent
excellent. Table no.2 gives the spatial output values 11 – 14 Good
for their classes. 8 – 10 Moderate
Table 2 Spatial overlays and analysis 4–8 Poor
Map no.3 Soil texture map
Map no. 1 Geology map
Map no. 4 Lineament map
Map no. 2 Geomorphology map

ACKNOLEDGEMENTS
We are grateful to WOTR (Watershed Organization
Trust, Pune) for their valuable support in extraction of
various maps and the facilities they have provided.
REFRENCES
Sener E., Davraz A. and Ozcelik M. (2005),
groundwater investigations, Hydrogeology
Journal, pp. 826–834.
Akram J. and Mushtaq H. W. (2009), delineation of
Groundwater Potential Zones, Journal Geological
Society of India, Vol.73, pp. 229-236.
Rao Y. S. and Jugran D. K.(2003), delineation of
groundwater potential zones and zones of
groundwater quality suitable for domestic
purposes, Hydrological Sciences Journal, pp. 821-
833.
Mlisa, A. and Hartnady, C. U. (2003), geo-informatics
for groundwater exploration, Proceedings of the
21st International Cartographic Conference
(ICC), pp. 540-546.
Richard M. T. (1995), a low cast geographical
information system, Hydrogeology journal, Vol.3,
Map no.5 Geomorphology map no.3, pp. 21-31.
Meijerink A. J.(1996), International Institute of
The groundwater potential zones map generated Aerospace Survey and Earth Sciences,
through this model was verified with the yield data to Hydrological sciences-Journal-des science
ascertain the validity of the model developed and Hydrologiques, 4l, 4 August 1996, pp. 549-56.
found that it is in agreement with the bore wells yield Per S.(1996), groundwater assessment in a Rural
data. The integration of various thematic maps groundwater project in Ghana, Hydrogeology
prepared from remote sensing, toposheet and field journal, Vol.4, no. 3, pp. 40-50.
information was carried out based on empirical Kazemi R., Porhemmat J. & Khirkhah (2009),
modeling technique, where each of the thematic unit investigation of Lineaments Related to
of various thematic maps has been assigned grades Groundwater occurance, Research Journal of
ranging from 1 to 5, with 1 representing lowest and 5 Environmnt science 3(3), pp. 367-375.
representing highest groundwater potential value. Sudarsana R. G., Raghu K., S. Reddy G. and M.
These maps were digitizing and raster encoded with Ravikumar(2009), identification of groundwater
the respective grade values. The integration of potential zones, Indian Journal of environment
geological and lineament information are important in protection, Vol.29, no, pp. 690-704.
preparing hydro geomorphic potential map Goyal R., Arora A.N. (2003), ground water modeling,
CONCLUSION Journals of water resources, Map India
In the present study GIS based methodology and Conference.
integrated remote sensing has been developed and Choubey V. K.(1996), assessment of waterlogged
demonstrated for evaluation of groundwater area in IGNP Stage I, Hydrology Journal, Vol.
resources. Weightage assignment has been applied XIX (2), pp. 81-90.
and very useful in the selection of ground water Ahmadi H.(2003), mass movements hazard zonation
potential zone. Change in land use is mainly due to in Germichay watershed, Ardebil, Iran”, journal of
the hydrological factors as is clear from the change disaster management, Map India conference.
image derived by subtraction of the land use maps. Balachandar. D, Alaguraja. P., Sundaraj .P,
Moderately high-resolution remote sensing data (IRS- Rutharvelmurthy. K , Kumaraswamy. K., artificial
PAN and LISS-IV) provide details of the terrain, as recharge zone, International Journal of Geomatics
well as visualize the general groundwater condition and Geosciences, Vol.1, no.1, 2010, pp. 84-97.
indirectly. GIS works as storing information about Gert A. S. (1997), Design and Operation of Water
world as collection of thematic layers that can be Resources Systems”, IAHS Publ. no. 242, pp. 3-
linked together by geography to solve many real 15.
world problems. The application of remote sensing Schumann A. H. (1993), Development of conceptual
and GIS have been found to be effective techniques semi-distributed hydrological models and
for generating data and integrating the different estimation of their parameters, Hydrological
layers, with detail of groundwater assessment which Sciences Journal, pp. 519-528.
reduce costs, time and effort.

ISBN 978-81-921121-0-7
APPLICAION OF RS AND GIS FOR APPROXIMATING STREAM

FLOW
Om Prakash Dubey
Professor, WALMI, Water and land management Institute (WALMI), Lucknow 226025,India
ABSTRACT: A watershed remains in dynamic equilibrium and responses to the changes in the land
characteristics. If developmental activities are not planned it may result in to floods, droughts and, or
erosion in the watershed. In the present study response of a Siwalik watershed on stream flow due to
changes in land cover has been studied. The data used for the study includes aerial photographs,
satellite imagery and ancillary data. The data has been analyzed visually as well as digitally. A brief
review of the available rainfall-runoff techniques has been made. The rational approach and SCS
curve number approach has been used for estimating the runoff. This model does not account the land
slope hence its use in assessing the impacts is limited. In this study a watershed model has been
proposed that takes in to account land slope in addition to watershed land cover and soil
characteristics. The proposed watershed model is simple and its input parameters can easily be
extracted from satellite data products. This model may be used to study the effect of developmental
activities and changes in land cover on stream flow. It has been found that in Ratmau watershed an
increase in forest cover induces reduction in stream flow. However detailed study is required before
finalizing a watershed management scheme.
BACKGROUND answer to the effect of land use changes. Attention is,

The rainfall runoff relationship has been represented therefore, being focused on physically based
by many empirical relationships. In 1608, Pierre watershed models. These models have potential to
Perrault developed rainfall runoff relationship for overcome many of the deficiencies associated with
Seine river basin France (Linsley, 1967). Since then conventional models. On the other hand, such models
several researchers have developed various are complex and considerable resources in human
relationships for example Zieglar (1880), Kuichling expertise and computing capability are needed for
(1889), Justin (1915), Mead (1919), Gregory and their development and applications. In the light of
Arnold (1932). In India too a lot of work has been these concerns, System Hydrologic European or SHE
carried out by several organizations e.g. Central model was developed. This model requires about 45
Water Commission, Irrigation Research Institute, parameters for calibration and testing. Hence its
Indian Council of Agricultural Research, various State application to data deficit watershed is difficult.
Irrigation Departments, Universities and individuals
for example, Binnie, Barlow, Khosla, Lacey, INFERENCES FOR REVIEW OF LITRATURE
Majumdar, Panchang (1954), Sehgal (1967), Rao and Though a very brief review of the literature is done,
Pentaih (1971) etc. etc. These relations are based on even then it is evident that, a variety of factors
several assumptions, for example, runoff is maximum influence stream flow from a watershed. These can be
when rainfall duration is equal to or greater than time grouped as (i) Storm characteristics: Type, nature,
of concentration, runoff is linearly related to rainfall, season, intensity, duration, aerial extent, frequency,
frequency of peak discharge is same as that of rainfall, direction of storm, antecedent precipitation etc. (ii)
and the coefficient of runoff is same for all frequency Meteorological characteristics: Temperature,
and for all storms. Humidity, Wind velocity, Pressure variation,
Evapotranspiration etc. (iii) Basin characteristics:
Many organizations and individuals have used Size, Shape, Altitude, Topography, Geology, Land
statistical techniques to develop rainfall-runoff cover, Orientation drainage network, etc. and (iv)
relationships (Krishnaswamy 1976, Majumdar 1965, Storage characteristics: Depression, Pools and Ponds,
Bhalerao et.al. 1977, Subba Rao 1979). Most of the Stream Control, Flood plains, Ground water and
proposed relationships are linear in nature. However, Channel storage. Most of the available methods for
now non linear and multivariate relationships have studying the rainfall – runoff are generally not
also been developed (Mockus 1949, Das et.al. 1971, accommodating the catchments physical
Chow 1964). Sol Conservation Society (SCS) characteristics.
proposed following formula to compute runoff,
popularly known as SCS model. The model involves, The conventional method of collecting the watershed
rainfall (P), initial abstraction (S), Curve number characteristics has many limitations such as: (i)
(CN), and runoff (Q). Accessibility to remote areas (ii) Providing spatial
It has been noticed that many conventional rainfall data (iii) Frequency of observation (iv) Detectability
runoff models are notable to provide satisfactory of very slow and fast processes. (v) Storage, retrieval

and processing of the data (vi) Cost and representative basic logic in detecting the land cover changes using
data. As a result observation sites are kept minimum remote sensing technique is that changes in earth
possible and in easily accessible areas. Hence feature results in changes in radiance value. Selecting
possibility of missing the important data cannot be appropriate data may reduce the effect of atmospheric
ignored. Probably this is the reason that catchments conditions, sun’s azimuth and altitude, and soil
parameters were not given proper weightage in moisture on the data product. Changes in land use can
rainfall runoff studies. be detected by comparative analysis or simultaneous
analysis of multi temporal data. Remote Sensing
Applications of Remote Sensing applications to study some of the land cover changes
Remote sensing techniques have several advantages are listed in Table 1. In the present investigation,
over the conventional techniques of data collection. remotely sensed data has been used to detect the land
These techniques are capable of providing point as cover changes and hydrologic modeling.
well as spatial data in near time through out day and
night and in all weather conditions for a large area. STUDY AREA AND SCOPE OF STUDY
Digital nature of the data helps in automatic The area under investigation Ratmau watershed,
processing. Remotely Sensed data can be analyzed covering about 500 Sq.km.(Fig 1), is bounded
visually as well as digitally or a combination of both. between latitudes 290 55’ N to 300 10’ N and
Through remote sensing required parameters are longitude 770 55’ to 780 05’ E. The area is drained by
extracted by cause and effect analogy. Remote Ratmau stream system. In general area is slopping
sensing techniques may be applied basically in three towards south. Summer season begins from the end of
ways, such as, input to conventional models, April continues up to middle of June. The temperature
synergistic use of conventional and remote sensing variation during summer months is between 130 C and
techniques and use of model developed specifically to 450C. During winter it is (-) 20 C to 340 C. Hills are
remote sensing. Only remote sensing techniques are generally covered with Sal, Shisham, Khair and
not sufficient and efficient for providing the Bambo. Ground elevation in the area varies between
representative data. It has been proved by many 810 m to 280 m above mean sea level. Ground slope
institutions (NASA, NRSA etc.) and individuals that varies between 25m/Km to 0.3m/Km.. The present
remote sensing techniques through synergistic use of study aims to study the response of Ratmau watershed
multi sensor, multilane and multi season data makes by formulating rainfall-runoff relationship.
the data extraction sufficient, accurate and useful in Synergistic use of remote sensing techniques and
evaluating the physiographic, geologic, soil, conventional techniques been utilized to extract the
hydrologic and climatic data (c.f. Verdant, 1977; required data.
Hitchock, et.al. 1975; Kristof, 1971; Montgomery,
1976; Rango et.al. 1973; Aldrich, 1971; Gupta, 1980; INVESTIGATION PROGRAM
Ragan, 1976 and Foody 1999). Step by step procedure of the investigation is being
described below.
From the previous discussions, it is clear that for
studying and assessing the impact of land cover Definition of the objective and data collection
changes on stream flow one needs varied and large Objective of the study is to assess the stream flow of
data set. Remote Sensing techniques have Ratmau watershed. In order to achieve the objectives
multidimensional applications to such studies. The of the study topographic map from Survey of India at
synoptic view, almost instantaneous survey and 1: 250,000 and 1: 50,000 scales, Aerial Photographs
repetitive coverage are some of the unique capabilities at 1: 50,000 scales, Satellite imagery F.C.C. at 1:
of remote sensing. Remote sensing can provide spatial 250,000 scales and 1: 50,000 scales have been
and temporal data related to landscape. Satellite analyzed. The remotely sensed data pertaining to the
remote sensing now enables easy, accurate, time and years 1972, 1973, 1974 and 1990 have been selected
cost effective mapping and updating watershed data keeping in view the season, and spectral
base (Chakraborty 1993). characteristics. Observed Rainfall data is tabulated in
Table 2. Analysis has been carried out for the months
Detection of land cover changes requires multi June, July, August and September. Other months have
temporal data. Previously, land cover data were not been considered as rainfall during this period is
acquired through ground surveys, topographic maps insignificant to produce runoff.
and other sources. Data acquisition from these
methods are laborious, time consuming and Table.2. Rainfall in Ratmau Watershed
uneconomical. In addition to the several well-known Rainfall in mm during
advantages, remote sensing provides the necessary Months
repetitive information at short interval of time with Years June July Aug. Sept
reasonable cost (Jenson 1986). The repetitive 1972 67.0 117.0 281.0 124.0
information can be extracted either from aerial 1973 303.0 507.0 205.0 90.0
photographs, satellite imagery or digital data. The 1990 28.0 354.0 335.4 218.8

Analysis of Remotely sensed data. 1% of the total area. The overall accuracy of the
Remotely sensed data was interpreted at 1: 250,000 interpreted map was found to be about 80%. Land
scales to delineate landscape features (land cover and cover statistics in Ratmau watershed is shown in
soil) using image characteristics. The minimum area Table 3. An analysis of Table 3 reveals that during the
of 2mm x 2mm on the paper was delineated. As result period 1972 forest cover in the watershed was of the
of data analysis Land covers Map and Soil Map at 1: order of 56 % where as during 1990 it reduced to 48
250,000 scales were prepared. In order to assess the %. Areal extent of human settlement is showing an
over all accuracy of the interpreted map a two-tier increasing pattern from 1.54 % to 2.78 %.
approach has been adopted. First of all about 10% of Agricultural land, Bare land, Barren land are showing
the total study area was analyzed on Aerial year-to-year fluctuation depending upon season,
photographs and Satellite imagery at 1: 50,000 scale. agricultural practice.
Then after, Field traversing was conducted in about
Table 3. Land Cover Statistics

Land Cover Type Land Cover Area Coverage in Hectare (Based on Image Analysis)
1972 1973 1974 1977 1990
Thick Forest 17032.25 13437.50 17656.25 16093.75 9687.50
Thick Forest 11718.75 6718.75 4531.25 4531.25 14062.50
Agricultural Land 18593.75 26250.00 25625.00 25781.25 8437.50
Bare Land 1718.25 23442.00 14061.00 957.95 14687.95
Barren Land 781.25 937.50 781.25 2167.50 2343.75
Urban Area 781.25 937.50 625.00 1093.75 1406.25
Total Area 50625.45 50625.45 50625.45 50625.45 50625.45
Estimation of Stream Flow using Rational Formula first of all runoff

coefficient for different types of land cover soil-
Stream flow has been estimated using Rational slope complex has been determined. For this a
Formula and SCS model. SCS model look up table (Table.4 & Fig 1) has been
accommodates land cover and soil characteristics prepared using Analytical Hierarchy Decision
including rainfall as input. Therefore, SCS model Process (Satty 1988). For this data pertaining to
is suitable for studying the effect of land cover the year 1972 and 1973 have been used and
changes only. Developmental projects induce decision has been carried out through a decision
changes in land cover, land slope, and soil cascade modeling. The problem was decomposed
characteristics such as compaction, field in to three components. Weights corresponding
permeability etc. Therefore SCS model may not to each subcomponent were calculated using a
find a vide scope in assessing the impacts. While Factor Analytic Model.
Decision Variable - Weight Score
10
9
0.8691
y = 0.9328x
8 2
R = 0.9938
7
y = 9.4722x-0.9665
Weight Score
6 R2 = 0.9857 Slope
5 Permeability
4 Land Cover(%/10)
-0.5533 Power (Land Cover(%/10))
3 y = 3.9121x
2 Power (Slope)
R = 0.9478
2 Power (Permeability)
1
0
0 2 4 6 8 10 12 14
Decision Variable
Fig. 1 Relationship between Terrain Parameter and Score

Table. 4 Catchments Characteristics and Weight
Details
Catchment’s Characteristics Weight
Range Description
> 10 High Relief 6.0
Land Slope (%) 5-2 Medium Relief 4.5
<2 Low Relief 1.0
< 1.5 C, D 3.0
Hydrologic Soil Group
1.5 – 10.0 B 1.5
Permeability (cm / hr.)
> 10.0 A 1.0
> 60 Forest 1.0
Land Cover
60 – 40 Agricultural 2.0
(%) Ground cover)
< 40 Barren, Open 3.0
Score - Runoff coeficent
1.2
1
y = 0.0952x - 0.1236
2
R = 0.9972
Runoff Coefficent
0.8
Series1
Theoritical
0.6
y = 0.0408x1.2902 Power (Theoritical)
R2 = 0.9997 Linear (Series1)
0.4
0.2
0
0 2 4 6 8 10 12 14
Total Score
Fig. 2 Relationship between Total Score and Runoff Coefficient
For a particular land unit, based on its characteristics

weights corresponding to land slope, Hydrologic Soil Where, Q is stream flow, m is total number of rain
group and Land Cover can be assigned using Table 4. gauges, n is total number of land cover, Ai,j is area of
These weights can be totaled to get total score (Ts). Ith type land cover in jth rain gauge station, Ci is
Relationship between total score, Ts, and runoff runoff coefficient for the ith type of the land cover, Cj
coefficient, C, in least squares sense (Coefficient of has been evaluated by inversion of the system
correlation = 0.9972, Fig. 2) has been found to be: response. In order to implement this model the
watershed was decomposed in to smaller cells. Runoff
C = 0.0952 - .126 Ts coefficient for each cell was evaluated using the look
up table (Table 4).
Once rainfall and runoff coefficient is known, stream
flow can be easily computed through watershed RESULTS
modeling. Physical system representing runoff from a
watershed has been mathematically expressed as: Stream flow from the Ratmau watershed has been
computed for the years, 1972 and 1990 and results are
Q = ∑ Ci Ij Ai,j shown in Table 5.

Table 5. Computed Stream Flow
1972 1977
Runoff Runoff
Rainfall
Month Watershed SCS Rainfall Watershed SCS
(mm)
Model Model (mm) Model Model
(mm) (mm) (mm) (mm)
Jun 67.00 75.10 36.0 20.81
July 117.00 60.84 45.78 494.50 237.36 223.49
Aug. 281.00 146.12 115.37 452.00 216.96 306.58
Sep. 124.00 64.48 28.43 126.90 60.91 46.79
CONCLUSIONS Dawdy, D.R., R.W. Litchy and J.M. Bergman, 1972.

From this study following conclusions have been A Rainfall-Runoff Simulation Model of
drawn. Estimation of flood Peaks for Small Drainage
(i) Remotely sensed data has to be properly Basins, U.S.G.S. Prof. Paper 28, Washington
selected for getting the optimal results. It is D.C.
useful in data collection, storage, retrieval Dubey, O.P., 1990 Impact of landcover charges on
and processing. Sediment Yield Proc. VIEC PP 40-48.
(ii) SCS model and the proposed watershed Holton, N.H., G.J. Stilner, W.H. Hensen and N.C.
model are yielding more or less similar Lowpez. 1975. USDA HL-74 Revised Model of
results. Water Hydrology. Tech. Bull. 1518, Agriculture
(iii) Watershed model that incorporates Research Service, USDA, Washington D.C.
watershed characteristics namely, land slope, Hydraulic Engineering Centre, 1982. Hydrologic
Soil, and Land cover, and hence can be used Analysis of Unganged Watershed using HEC-1.
to study the effect of developmental Trg. Document No. 15, US Army Corp. of
activities including land cover changes on Engineers, Davis, California.
stream flow. McCuen, R.H. A Guide to Hydrologic Analysis using
(iv) Land cover changes influences the water SCS Methods. Prentice Hall, New Jersey.
yield from a watershed. In Ratmau watershed Ragan, R.M. and T.J. Jackson. 1976. Hydrograph
it has been observed that an increase in forest Synthesis using Landsat Remotesensing and
cover (% ground cover) induces reduction in S.C.S.Models. NASA X-913-76-161 Goddard
stream flow. If land cover doubles it reduces Space Flight Centre, Green Belt., M.D.
stream flow to half. However this requires a Saaty, T.L. 1988. The Analytical Hierarchy Process.
detailed study before finalizing a watershed University of Piltsburgh, Pa
management scheme. Satapathy K.K., R.K. Panda, R.N.Prasad, 1977, An
Experimental Investigation of Runoff Production
REFERENCES on Hill Slopes Under various Land Uses. JIWRS,
Barrett, E.C. and Curtis, L.F., 1976, Introduction to 17(3) pp 34-42.
Environmental Remotesensing, London, Shahlace, A.K., Nuter W.L., Buroughs ER and
Champman and Hall, New York. Morris, L.A. (1991) Runoff and Sediment
Chakraborti, A.K., 1993, Strategies For Watershed Production from burned forest sites in the
Management and Planning Using Remote Georgia Piedmont, water Resources Bulletin,
Sensing Technique. Photonishwachak vol.21, No. 27(3) : 485-493.
2 , 1993.
Chandra, S. and Sharma, K.P. 1978, Application of
Remotesensing to Hydrology of Rivers with
Small and Medium Catchments, CBIP, Vol.II.
Crow, F.R., Ghermaziew, T, Pathak, C.S. (1983). The
effect of land use parameters on runoff simulation
by the USDAHL Hydrology Model, Transactions
of ASAE, 26(1) 148-152.
Crowford, N.H. and R.K. Linsley, 1966, Digital
Simulation in Hydrology, Stanford Watershed
Model IV, Technical Report 39, Deptt. Of Civil
Engg. Stanford Univ. Stanford, California.

ISBN 978-81-921121-0-7
SEISMIC VULNERABILITY ASSESSMENT OF SKEW BRIDGES

S.P. Deepu
Former Graduate Student, Department of Civil Engineering, IIT Kanpur, Kanpur - 208016, Email: deepusp@gmail.com
S. Ray-Chaudhuri
Assistant Professor, Department of Civil Engineering, IIT Kanpur, Kanpur - 208016, Email:samitrc@iitk.ac.in
ABSTRACT: Bridges are the most important component of a surface transportation system. Past earthquakes
have demonstrated that bridges are highly vulnerable to earthquake induced failure with skew bridges being
more susceptible to damage during earthquakes than a normal bridge. This work focuses on seismic
vulnerability assessment of skew bridges. Representative numerical models of four skewed bridges of skew
angle 15°, 30°, 45° and 60° along with a normal bridge, all with similar span and deck width, are developed
using a widely used software to study the effects of varying skew angles. A suite of ground motions with
varying strong motion properties and representing different hazard levels are used. Seismic fragility curves are
generated for assessing the vulnerability of these bridges. Damage measures considered for fragility analysis are
the deck unseating and damage of columns. Results indicate that as the skew angle increases, there is an
increase in the probability of failure for a given ground motion intensity. The results of this study provide a
better understanding of behavior of bridges with different skew angle under seismic loading.
INTRODUCTION NUMERICAL MODELLING

Bridges are one of the most important elements of any A representative normal bridge along with four
land transportation network. In order to cater the skewed bridges of skew angles 15°, 30°, 45° and 60°
growing industries and population, transportation but with similar span and deck width (i.e., two equal
network in a few countries is undergoing a rapid spans of 20 m each and an overall width of 14.18 m
expansion, which in turn requires construction of a measured along the skew direction) are considered.
vast number of bridges and over-crossings for For each of these bridges, at first, a preliminary design
providing grade separation and thereby ensuring fast is performed based on the examples from Barker and
and uninterrupted traffic flow. Many of these bridges Puckett (1997) followed by a seismic design. For the
are skewed or curved in plan as per the requirements seismic design, the AASHTO-LRFD (2007) bridge
of network layout. design specifications are used along with IBC-2006
IBC (2006) design spectrum for San Francisco area
Over the past few decades, it has been observed that (ZIP- 94101). The superstructure of the bridges
many bridges and highway over-crossings have been consists of a concrete T- beam girder deck section
severely damaged during earthquakes. It has also been with 250 mm thick deck slab and six 1.5 m deep T-
found that skew and curved bridges are the most beam girders. The girders are placed on a column bent
vulnerable when compared to straight bridges. So far consisting of three reinforced concrete columns all
many researchers have investigated the static and with similar dimensions. Each end of the bridge is
dynamic behaviour of skew bridges with/without supported on abutments. The stiffness of the
considering the flexibility of the bridge deck. To save abutments is assumed to be negligible for modeling
space, these studies are not listed here but can be purpose. The base of each column is assumed to be
found in Deepu (2011). In a recent work, Pottatheere fixed (i.e., no soil-structure interaction is considered).
and Renault (2008) studied the seismic vulnerability For concrete, the compressive strength is assumed to
of skew bridges by developing fragility curves for be 3 ksi (20 MPa) while for steel, it is assumed to be
bridges with skew angle of 0°, 12:23° and 45°, 40 ksi (275 MPa).
considering bridge column failure. Although it has
been found that skew bridges are prone to deck Bridge Superstructure
unseating damage, no systematic studies is conducted The superstructure consists of a reinforced concrete
to understand and evaluate the probability of bridge continuous deck slab of 40 m overall length with 250
failure under ground motion with given intensity and mm thickness in the supported parts and 260 mm
considering deck unseating damage. In this work, a thickness in the overhanging part. It is supported on
systematic study is conducted to assess sesimic six reinforced concrete T- beam girders having a cross
vulnerability of skew bridges with varying skew section dimension of 1.5 m x 0.45 m. The centre to
angles and when subjected to an ensemble ground centre spacing
motions of different hazard levels. Vulnerability between any two T-beams is 2.44 m .The total span of
assessment is done by means of fragility curves the bridge is 40 m with an intermediate bent cap
assuming it as a two-parameter log-normal supporting the 6 girders at the mid span. The cross
distribution function with unknown parameters. sectional area of the T-girder deck slab is 6.9353 m2
and moments of inertia about different axes are

J=0.3237 m4, I33=1.4313 m4 and I22=118.7519 m4. A
schematic diagram of the cross section of the
superstructure is shown in Figure 1.
Columns, Cap beam and Bearings

The cross sections of all the three columns of all the
models are assumed to be circular in shape. All the
columns are designed for meeting the imposed
seismic load based on the design spectrum. It is found
that the force demand increases with an increase in Fig. 1 Cross section of the deck slab
skew angle. This is because as the as skew angle
increases, the structure becomes stiffer and thus EIGENVALUE AND M-φ ANALYSIS
attracts more force. As a result, the columns of the It has been found that the first mode in all these
normal bridge and the 15° and 30° skewed bridges are models is in-plane rotation. The second and third
1.7 m in diameter, whereas the columns of the 45° and modes are longitudinal and transverse translations,
60° bridges are of 1.8 m and 1.9 m in diameter, respectively. The time periods obtained from analysis
respectively. For longitudinal reinforcements, #18 are given in Table 1. It can be observed from this
bars are used and for confining reinforcement, #11 table that for all models, the fundamental and second
bars with 6 inches (150 mm) spacing are used as the periods decrease as the skew angle increases thereby
stirrups for each column and for all the five bridge implying the increase in stiffness with an increase in
models. The height of these columns from the centre skewness. However, for the third mode, the period
line of the cap beam to the bottom is assumed to be 7 increases with the increase in skew angle. This
m. All the columns are equally spaced in the implies that the transverse stiffness of all these models
transverse direction with the spacing being 5 m for the reduces with the increase in skew angle.
normal bridge model and in all other skew models, the
spacing is such that the projected spacing along the Table 1 Eigenvalue analysis results
direction perpendicular to the longitudinal direction is Modes Normal 15° 30° 45° 60°
5 m. Thus the actual transverse spacing between the First* 2.03s 1.98 s 1.37s 1.04s 0.79s
columns in all the skew models depends on the skew Second# 0.78s 0.78s 0.76s 0.71s 0.67s
angle. The cap beam is assumed to be of dimension Third** 0.30s 0.31s 0.36s 0.40s 0.44s
1m x 2m and is modeled as a reinforced concrete *In-plane rotation, #Longitudinal displacement,
beam with 3 ksi concrete. **Transverse displacement
The cross sections of the columns are modeled in SAP Generally, for dynamic analysis involving concrete
2000, v14.0.0- Advanced (2009). In all the columns, sections, the Moment-curvature (M-φ) relation is
the percentage of the longitudinal steel used are less idealized with an elastic perfectly plastic behavior for
than 4% of the gross cross sectional area. In the simplicity of analysis. In this study, M-φ curves are
longitudinal direction, a column acts as a cantilever idealized as per clause 3.3.1, Caltrans SDC (2010).
beam with base being fixed. However, in the Table 2.3 gives the plastic moment capacity (Mp) and
transverse direction, both ends of a column can resist yield curvature (φ Y) obtained from the moment-
rotation. curvature analyses for all the bridges models. It can be
observed from this table that as the skew angle
To model these bridges, two types of bearings are increases, φY decreases but Mp increases.
used. They are friction isolator bearing, which is used
at the abutment and the rocker bearing, which is used Table 2 Salient values from M-φ analysis
at the bent. The friction isolator bearings have a Modes Normal 15° 30° 45° 60°
stiffness of 123 kN/mm and a friction coefficient of φY(×10-
0.2 in the longitudinal direction. In other two 3 2.36 2.35 2.34 2.207 2.086
/m)
translational directions, the bearing is assumed to col
Mp
provide rigid connection. 15202 15551 15978 18644 22962
(kNm)
The girders are supported on the cap beam via rocker PLASTIC HINGE DETAILS
bearing, which is modeled as a nonlinear element The analytical plastic hinge length is the equivalent
(Wen) with the vertical and transverse directions length of column over which the plastic curvature is
fixed. In the longitudinal direction, the bearing is assumed to be constant for estimation of plastic
modeled with an effective stiffness, ke=14 kN/mm, rotation. According to Caltrans SDC (2010), the
yield strength fyield being approximately equal to 4% length of the plastic hinge, Lp is calculated as follows
of the normal reaction, which is equal to 34kN, and (clause 7.6.2(a)):
the post-yield stiffness ratio β of 0.018. (Pan et al.,
2007).

Lp = 0.08L + 0.022 fye dbl (mm,MPa)
(1)
where L=length from the point of contra-flexure to the

section with maximum moment and dbl=diameter of
the longitudinal reinforcing bar and fye=the effective
yield strength of the longitudinal reinforcing bars.
Based on these calculations, the length of plastic
hinge Lp is assigned as 0.62 m for all the five models
considered.
GROUND MOTION DETAILS

For the non-linear dynamic analysis, 60 ground
motions with different hazard levels and strong
motion characteristics are used in this study. These
ground motions are taken from the SAC steel project
for the Los Angeles area (SAC). Each model is
subjected to three sets of ground motions representing
hazard levels 2%, 10% and 50% in 50 years. Note that
each set contains 10 ground motions with two
orthogonal horizontal components for each ground
motion. The three sets of ground motions are named
as LA01- LA20, LA21- LA40 and LA41- LA60.
These acceleration time histories have been derived
from historical recordings or from physical
simulations and have been altered so that their mean Fig. 2 Longitudinal displacement time history of node
response spectrum matches with the 1997 NEHRP 531 of different models under load case LA21_LA22
design spectrum, modified from soil type of SB--SC
to soil type SD and having a hazard specified by the
1997 USGS maps.
NONLINEAR TIME HISTORY ANALYSIS

Nonlinear time history analysis is done with each bi-
directional ground motion after performing a dead
load analysis and holding the stress and deformation
states so that the P-delta effects can be properly taken
into consideration. The analysis is carried out with
Newmark’s constant acceleration direct integration
method. The models are assigned with mass and
stiffness proportional Rayleigh damping with 5%
damping in first and second modes.
RESULTS
Longitudinal and Transverse Displacement
To demonstrate the response of the bridge models,
LA21_22 has been considered. Figure 2 shows the
longitudinal displacement time histories and Figure 3
shows the transverse displacement time histories of
different models for corner node of slab. Figure 4
gives the salient points of the deck slab model.
It is seen that the longitudinal displacement of Fig. 3 Transverse displacement time history of node
different models under the two ground motion cases 531 of different models under load case LA21_LA22
have minimal effect on their skewness. Investigation
of the maximum longitudinal displacement under
LA21_L22 shows that the both 45° and 60° model has
longitudinally displaced through 0.39 m where as it is
0.37 m for normal model. The permanent deformation
of the all the models were very small with maximum Fig. 4 Salient points of the deck slab model
occurring for 30° skew model with 0.023 m.

It is seen that as the skewness of the model increases,
the amount of transverse displacement increases
during the ground shaking. It is also observed that all
the skew models show a permanent transverse
deformation with maximum being around 6cm for
model with 60° skew angle when subjected to
LA21_LA22 ground motion whereas there is little
permanent deformation for normal model. The
maximum transverse displacement is 0.134 m and
0.101 m under LA21_LA22 and LA35_LA36
respectively and it occurs in the case of 60° model.
In-Plane Rotational Response

The rotation of the bridge deck is calculated using the
relative longitudinal displacement of the two corner
points on the right. i.e., node numbers 531 & 553. The
rotation (R) in degrees is calculated using the
expression,
( 1531 − 1553 )
=
cos
360
× (2)
2π Fig. 6 Rotational time history of the deck
where U1531 is the longitudinal displacement of node

531, U1553 is the longitudinal displacement of node Curvature ductility
553, L is Under strong dynamic loads, the response of the
the width of the slab (14.18 m) and θ is the skew column can surpass its elastic limit forming plastic
angle. hinges at the ends of the columns. The damage of a
column is generally quantified in terms of ductility
demands (displacement, rotational or curvature). The
curvature ductility is expressed as:
φ
µφ =
φ
φ
Fig. 5 Schematic for illustrating calculation of = 1+ (3)
rotation φ
Figure 6 shows the rotational time history of the deck. where φ is the ultimate curvature and φ is the yield
It is observed that the normal bridge has hardly had curvature, which can be determined from the idealized
any resultant in-plane residual rotation. During the moment curvature plot. φ is the maximum plastic
strong shaking,(from 8 sec to 25 sec), the normal curvature of the hinge. The curvature ductility
bridge model shows some rotation, which is very demands were found out from the plastic rotations and
small compared with the rotations of other bridge was used as an indication of damage.
models. As expected, all the skew models have
rotational behavior and the rotation increases with an VULNERABILITY ASSESSMENT
increase in the skew angle. For the load case An assessment of seismic vulnerability of structures
LA21_LA22, the maximum rotation was 0.09° and like buildings and bridges are generally carried out by
the residual rotation is 0.07°. It can be stated that the means of fragility curves. A seismic fragility curve
skewness causes excessive in-plane permanent provides a conditional probability that a structure or
rotation, which is undesirable. The excessive rotation component will exceed a certain damage level given a
and longitudinal displacements may cause the bridge ground motion with specific intensity has occurred.
to fail temporarily or collapse completely due to The intensity of a ground motion is generally
unseating of the deck slab. expressed in terms of peak parameters such as PGA or
spectral parameters. A fragility curve is often
expressed in the form of a two-parameter log-normal
distribution with the unknown median and log-
standard deviation. According to Shinozuka et al.

(2001), there are two methods for estimating the Since the girders are placed on the bearing pads, it is
fragility parameters. In this study, the method that logical to assume that the maximum width of the
uses the same log-standard deviation for different bearing pads that can be used is less than or equal to
damage levels (eliminating the possibility of crossing the breadth of the girders placed on the bearing pads.
of fragility curves associated with various damage So for all the models, regardless of the skew angles
levels) is used. the maximum limit of the transverse displacement is
taken as 450mm.
DAMAGE STATES DEFINITIONS
The fragility curves have been developed for 1) Unseating due to In-Plane Rotation
damage due to deck unseating and 2) damage due to For the deck to get fully unseated due to rotation, the
failure of columns. Although several researchers have relative displacement between the two corners should
generated fragility curves for column damage in terms be equal to seat length. Hence, maximum in-plane
of ductility demand, which may be expressed in terms rotation for deck unseating is given by
of displacement ductility, rotational ductility or
curvature ductility, hardly any work is reported in the B
literature for fragility curves involving unseating of θ (degrees) =
L cosβ
bridge decks in skew bridges. Since deck unseating, is 360
a common damage phenomenon occurring in skew × (5)
2π
bridges, fragility curves for deck unseating can be
very useful for risk mitigation in earthquake prone
areas with large number of skew bridges. In this study
three states of damage, 1) Minor damage 2) Moderate where N = the seat width required as per Equation 3.1,
damage and 3) Major damage have been defined B = width of the bridge and β = Skew angle (degrees).
based on the performances of the bridges under
different ground motions.
Fragility Curves for Deck Unseating

The unseating of the deck can happen due to
excessive longitudinal displacement, transverse
displacements or rotation of the deck slab or a
combination of these. For generating the fragility
curves for the damage due to unseating, separate
limits are set for the damage due to these three
individual parameters and the maximum damage state
among these three are taken.
Unseating due to Longitudinal Displacement

For the bridges considered in this study, the girders of
the deck are placed on the bearings at the two Fig. 7 Fragility curve for unseating for 45° model
abutments. Therefore, in case there is an excessive
longitudinal displacement of deck, the deck can In each of these three damage modes, minor,
unseat from the bearing pads. The minimum seat moderate and major damage states are defined by
width to be provided is given in all prominent codes assuming the limiting values to be 25%, 50% and
and in AASHTO-LRFD (2007). According to Clause 75% of the maximum value specified for each mode,
4.7.4.4 of this code, the empirical seat width shall be respectively. Figure 7 shows the fragility curve for the
taken as deck unseating damage for the 45-degree model. The
fragility curves have also been developed for all other
N = {200 + 0.0017L + 0.0067H}{1 models.
Figure 8 shows the variation of median PGA
+ 0.000125 S } (4)
(corresponds to 50% probability of exceedence) value
where N = minimum support length measured normal for different damage states with skew angles. There is
to the centerline of bearing (mm), L = length of the no definite trend seen in the median value for all
bridge deck (mm), H = average heights of columns damage states. However for major damage, it is found
supporting the bridge deck (mm), S = skew of support that the 45° model is the most vulnerable.
measured from the line normal to the span (degrees).
Unseating due to Transverse Displacement

The failure due to deck unseating can also happen by
excessive displacement of the deck in a direction
perpendicular to the longitudinal axis of the bridge.

Fig. 8 Variation of median PGA with skew angles for
different unseating damage states Fig 11. Comparison of median values with varying
skew for different states of column damage.
Damage State Definitions for Column Damage
According to Dutta & Mander (1998), five different CONCLUSIONS
damage states namely Àlmost no', `Minor', Focusing on skew bridges, a systematic investigation
`Moderate', `Major', and `Collapse' can be defined has been conducted in this work to assess seismic
based on column drift, which is defined as the ratio of vulnerability of skew bridges with varying skew
maximum displacement response at the top node to angles when subjected to bidirectional ground
the height of the column. Priestley et al., (1996) motions. For this purpose, a normal bridge and four
recommended that the drift limit corresponding to the skewed bridges of 15°, 30°, 45° and 60° skew but
yield state is 0.005. Therefore, curvature ductility with similar dimensions are designed as per
demand corresponding to Àlmost no damage', which AASHTO-LRFD (2007) guidelines and modeled in a
corresponds to a drift limit of 0.5%, is equal to 1. The widely used commercial finite element software, SAP
curvature ductility demand corresponding to 2000, v14.0.0- Advanced (2009). Sixty ground motion
`Collapse' is computed based on a drift limit of 7.5%. time histories representing hazard levels of 2%, 10%
The ductility demands corresponding to other damage and 50% probability of exceedence in 50 years are
states are obtained in proportion to the ultimate drift used and seismic seismic fragility curves are
limits (Banerjee and Shinozuka, 2007), i.e., by linear generated using PGA as ground motion intensity
interpolation between these two end values. Table 3 parameter and considering different damage states for
shows the curvature ductility demand corresponding deck unseating and column failure. Specific findings
to different damage limit states. In this study, `Major of this study are as follows:
damage' and `Complete damage' is not taken into
account as the bridge models were seismically 1. It has been found that irrespective of skew angle,
designed and no cases surpassed these upper damage rotational mode is the fundamental mode. However,
states. as the skew angle increases, the structure becomes
Figure 9 shows the variation of median of different stiffer, which in turn attracts more forces and shows
column damage states with skew angles. It is higher responses compared to a normal bridge of the
observed that the skew bridges with skew angles 30°, same dimensions.
45° and 60° are more vulnerable compared to 15° and
normal bridge. As the skew angle increases, the force 2. As the skewness of a bridge increases, it has been
demand generally increases and this explicitly found that the amount of transverse displacement
increases the vulnerability, even if the columns are increases during the ground shaking.
designed for that particular skew. Also it can be seen
that for the 45° skew model, the median value is 3. It is observed that the normal bridge suffers
minimum, implying the most vulnerable model. negligible in-plane residual rotation as a result of
ground excitation, whereas, due to torsional coupling,
Table 3: Peak curvature ductility demand for columns all the skew models suffer significant rotation, which
Damage Normal 15° 30° 45° 60° increases with an increase in the skew angle.
states model model model model model
4. As far as damage due to deck unseating is
Almost No 1.00 1.00 1.00 1.00 1.00
considered, it is observed that the 45° model is the
Minor 2.77 2.72 2.71 2.72 2.70
most vulnerable and the vulnerability reduces with an
Moderate 8.09 7.89 7.86 7.86 7.78
increase or a decrease in skew angle from 45°.
Major 16.95 16.5 16.43 16.44 16.26
Collapse 25.81 25.12 25.00 25.01 24.74

5. In the case of column damage, it is found that the
skew bridges with skew angles 30°, 45° and 60° are
more vulnerable compared to 15° or the normal bridge
model.
6. A comparison of fragility curves shows that the

failure probability for deck unseating is much more
than that of the column failure when similar damage
states are considered.
REFERENCES
AASHTO-LRFD (2007). AASHTO LRFD Bridge
Design Specifications. American Association of
State Highway and Transportation Officials.
Banerjee, S. and Shinozuka, M. (2007). Nonlinear
Static Procedure for Seismic Vulnerability
Assessment of Bridges. Computer-Aided Civil and
Infrastructur Engineering, 22, 293-305.
Barker, R. and Puckett, J. (1997). Design of highway
bridges-Based on AASHTO LRFD Bridge design
specifications. John Wiley and Sons, Inc., New
York
Caltrans (2010). Seismic Design Criteria Version 1.6.
California Department of Transportation,
California.
Deepu, S.P. (2011) Seismic performance assessment
of skew bridges, M. Tech. Thesis, IIT Kanpur,
Kanpur, India.
Dutta, A. & Mander, J. B. (1998), Seismic fragility
analysis of highway bridges, in Proceedings of
INCEDE-MCEER Center-to-Center Workshop on
Earthquake Engineering Frontiers in
Transportation Systems, Tokyo, Japan, 311–25.
IBC (2006). International Building Code.
International
Building Code Council.
Pan, Y., Agrawal, A., and Ghosn, M. (2007). Seismic
Fragility of Continuous Steel Highway Bridges in
New York State. Journal of Bridge Engineering,
12(6),689-699.
Pottatheere, P. and Renault, P. (2008). Seismic
Vulnerability Assessment of Skew Bridges.
Proceedings, 14th World Conference on
Earthquake Engineering, Beijing, China.
Priestley, M., Seible, F., and Calvi, G. (1996). Seismic
design and retrofit of bridges. John Wiley and
Sons,
Inc., New York.
SAC Steel Project. Seismic Design Criteria for Steel
Moment Frame Construction.
http:\\www.sacsteel.org.
SAP 2000, v14.0.0- Advanced (2009). Computers and
Structures Inc. Berkeley, California, USA.
Shinozuka, M., Feng, M., Kim, H., Uzawa, T., and
Ueda, T. (2001). Statistical Analysis of Fragility
Curves. Technical Report MCEER, USA.

ISBN 978-81-921121-0-7
GEOCHRONOLOGICAL CONSTRAINTS OF EASTERN DHARWAR

CRATON: A REVIEW
D. Prakash
Assistant Professor, Centre of Advanced Study in Geology, BHU, Varanasi- 22 1005, Email: dprakashbhu@yahoo.com
P. Chandra Singh
Research Scholar, Centre of Advanced Study in Geology, BHU, Varanasi- 22 1005, Email: pchandrageo@gmail.com
S. Tewari
Research Scholar, Centre of Advanced Study in Geology, BHU, Varanasi- 22 1005, Email: suparnatewari67@gmail.com
I.N. Sharma
Research Associate, Centre of Advanced Study in Geology, BHU, Varanasi- 22 1005, Email: indraneel_vns@rediffmail.com
ABSTRACT: The rocks of the Eastern Dharwar Craton span a vast period of time and have been dated by a
number of workers. On the basis of available geochronological data three major tectonothermal events have
been recognized in this part of the Dharwar Craton around 3.0 Ga (MI), 2.5 Ga (MII) and 2.0 Ga (MIII). Dates
from the sedimentary sequences indicate that the lower most part of the sequence may be older than 1.5 Ga and
the upper most part is around 0.5 Ga although no definite dates are available on the uppermost part. On the basis
of the available geochronological data from Eastern Dharwar craton, the emerging concept of different continent
assembly is endorsed.
INTRODUCTION
Geochronological studies of an area are necessary for Table 2. The major differences between EDC and
determining the age of different rocks and correlation WDC are given in Table 3. In the east, Eastern Ghats
of their formation with the events that happened front separates it from the Eastern Ghats Mobile Belt.
during that period. Of the various methods available, In the north of it lies the Deccan Basalts and the
the radiometric dating is the best and most widely Permo-carboniferous sedimentary formations of
accepted as this gives the absolute age. This is Godavari Graben. The later separates the EDC from
particularly true of the Archean rocks due to the the Bastar Craton. In the south the rocks of the EDC
absence of floral or faunal evidences. The most gradually merge into the Southern Granulites.
widely used pairs for the radiometric dating of the The prominent exposures in the EDC are the
Archean rocks are Rb-Sr, U-Pb, Pb-Pb, Sm-Nd and supracrustal enclaves (Sakarsanahalli Group) and the
K-Ar. Out of these Sm-Nd dating technique is the best schist belts (Kolar, Hutti and Ramgiri), gneisses
but unfortunately in the Eastern Dharwar Craton (Peninsular Gneisses), granites (Closepet, Patna,
(EDC) this method has not been used so far. Another Hyderabad and Karimnagar), later dolerite dykes and
new technique SHRIMP (Stimulated High Resolution late Proterozoic sedimentary suites. The metamorphic
Ion Micro Probe) dating has been used in the EDC by rocks of the EDC show an increase in the grade of
Friend and Nutman (1991) and Peucat et al. (1993) in metamorphism southward from greenschist (Sandur,
few places. This method gives an error of around ± 5 Hospet, Ramgiri belts) to amphibolite (Kolar belt) to
Ma which is very good in view of the dates obtained. granulite facies (Krishnagiri-Dharmapuri).
The radioactive mineral dates or whole rock dates are Karimnagar granulite terrain shows granulite facies
meager in the EDC. The evolution of rocks and events rocks further north of the greenschist facies belt in
in this part of the Dharwar Craton are normally EDC.
inferred from those of the Western Dharwar Craton The metamorphic and igneous rocks of the EDC are
(WDC). But there are some specific dates available in covered at places by almost undeformed sedimentary
the EDC of gneisses, granites, charnockites, dykes sequences like Cuddapah and Bhima basins. Bhima
and sedimentary sequences. Chronostratigraphy and basin do not show any deformation, whereas,
the tectonothermal events of the EDC have been given Cuddapah basin shows mild tectonism throughout the
in Table 1. Present paper attempts to club these extent.
scattered radiometric dates in one place for
convenience and discuss the various tectonothermal GEOCHRONOLOGY OF THE ROCKS OF
events in this part of the Dharwar Craton. EASTERN DHARWAR CRATON
The available radiometric dates from the EDC are
REGIONAL GEOLOGY summarized in Table 4. The oldest data available in
The EDC occupies the central portion in the the EDC is 3.3 Ga (U-Pb isochron from zircon) for
peninsular India (Fig. 1). It is separated from the the Peninsular gneiss from Kabbaldurga area (Stahle
WDC by a N-S trending thrust located about 25 km et al., 1987) hence the rocks of the Sakarsanahalli
west of the N-S trending arcuate closepet granite. The Group occurring as enclaves within these gneisses
stratigraphic succession of the Dharwar craton, based must be older. The Peninsular gneisses are dated
on swaminath and Ramkrishna (1981) is given in around 3.0 Ga. At few places younger dates of around

2.5 Ga presumably due to resetting in the greenschist to granulite facies from north to south
concentration of the radioactive minerals during the around 2.6/2.5 Ga. This MII event affected the
formation of charnockites and intrusion of granites Peninsular Gneiss with enclaves of Sakarsanahalli
around 2.5 Ga. Similarly the younger dates for the Group of rocks which were already metamorphosed to
Closepet granites may be due to resetting during the amphibolite facies conditions during the MI event.
dolerite dyke emplacement around 2.0 Ga. MII event is also related with the widespread
Kimberlites from Wajrakarur are highly weathered, charnockitisation and was closely followed by
hence exact age determination is difficult, and widespread granite intrusion such as Closepet, Patna,
younger K-Ar dates of around 1.0 Ga has been Hyderabad and Karimnagar granites. This MII event
reported. These lower was later on followed by a milder MIII event during
early Proterozoic (~ 2.0 Ga) and the rocks formed
during MI and MII events retrograded along the E-W
and N-S trending faults. This MIII event coincides
with the emplacement of dyke swarms.
In the sedimentary sequences 1.58±0.15 Ga date has
been obtained from the amygdaloidal basalts and a
date of 0.98±0.11 Ga has been obtained from the
altered dolerite dyke cutting the Vempalle dolomite,
both from the lower Cuddapah sequence. Thus the
dates available for the sedimentary sequences indicate
that the lower most part of the sequence may be older
than 1.5 Ga and the upper most part be around 0.5 Ga,
although no definitative dates are available on the
uppermost part. The dates of 0.5-0.6 Ga are still
disputed and not well established in the EDC.
Based on the lithology, Kameshwar Rao (1989) and
Rajesham et al. (1993) proposed two different
stratigraphic sequence for the rocks of the Karimnagar
area. Kameshwar Rao’s stratigraphic classification
does not have charnockite as one of the important
litho-unit of the area, where Rajesham et al. (1993)
does not include the peninsular gneissic complex in
their classification. Due to these descripancies a new
stratigraphic classification is being proposed here on
the basis of the lithology and geochrnological data
(Table 5).
TECTONIC SETTING
The primitive Earth consists of a magma ocean, in
which the less dense material floats over the dense
magma. The condensed less dense layer formed into a
number of small plates, which floats over the dense
magma. These plates characterize the early Archaean
Fig. 1(a). Reference map of India, (b) Lithological
crustal evolution (Vansutre and Hari, 2010). The
patterns for Precambrian rocks of the Eastern subduction of Buster craton beneath the Dharwar
Dharwar Craton along with craton boundaries. The
craon took place around 2600Ma (Fig.2a). After this,
abbreviations used are: WDC – Western Dharwar
the TTG (tonalite–trondhjemite– Granodiorite) suite
Craton, SG – Southern Granulite, EGMB – Eastern
formation occurred as a new continental crust with the
Ghats Mobile Belt, B – Bastar Craton, Hy –
partial melting of hydrated basaltic crust (Condie,
Hyderabad, K – Karimnagar, C – Cuddapah, BA –
1988; Martin et al., 2005) between 4 and 2.5 Ga. This
Bangalore, KR – Krishnagiri.
was followed by intense plume activity which is
evident by accretion of juvenile crustal materials, M II
dates may be due to escape of Ar as it is a gaseous event metamorphism is associated with widespread
phase. The major dates available are 3.2 Ga, 3.0 Ga, charnockitization and the peak condition of
2.5 Ga and 2.0 Ga. The earliest recognizable
metamorphism in Dharwar craton found close to 2500
tectonothermal event is MI, which affected the
Ma. The heterogeneity in the density of the plates
Peninsular Gneisses (3.2-3.0 Ga) and Sakarsanahalli
was produced due to the cooling of the crusts, which
Group (> 3.0 Ga) around 3.0 Ga. During this event the
might have caused the development of subduction
rocks were metamorphosed to amphibolite facies
zones, subsequently thickening of the crust (Vansutre
conditions. The major event designated as MII was
and Hari, 2010) (Fig.2b). The presence of granulite
responsible for the prograde metamorphism from

crust in Archaean cratons is recognized on a global The 2.5 Ga old collisional mobile belt at the northern
scale, which was result of magmatic underplating margin of the EDC against the Bastar craton, was
(Lopez et al., 2006). The geochemical signatures from superposed by rift activity during -1.5 Ga creating the
the rocks of Bhopalpatnam granulite belts (BGB), PG basin. This sequence of events closely matches
suggest a provenance for the protolith in close with the accretion and break-up of Columbia. The
proximity to K- rich lithology (Vansutre, 2007) (Fig. ages of 1.7 Ga from the dyke swarms demarcate the
b and c). The clockwise exhumation paths deduced last phase of activity for Columbia supercontinent. No
from BGB and Karimnagar granulite belt (KGB) Grenville ages (ca. 1.0 Ga) corresponding to the
suggest a collisional event between Dharwar craton Rodinia accretion or late Pan-African ages (ca. 550
and Baster craton between 1.9 to 1.6 Ga. Ma) corresponding to the Gondwana amalgamation
There are two dates are available in Karimnagar, one are recorded from KGT or from parts of the PG basin
is from younger granites (2600-2700 Ma by Rb- Sr) located away from the Eastern Ghats Front.
by Rajesham et al., (1993) and another from zircon Therefore, we infer that the KGT and major parts of
and monazite probe dating of the chanockites the PG basin did not witness any of these younger
(between 3.1 and 2.6 Ga) by Santosh et al., (2004). In supercontinent histories. On the other hand, 1.6-1.5
Karimnagar granulite belt, granites and charnockites Ga collision along the amalgamation at the northern
have inclaves of mafics (Prakash and Sharma, 2011), margins of the combined Dharwar-Bastar-Singhbhurn
which are very important to understand the evolution cratons had no rift activity but was superposed by
of these areas. The fluid inclusion and mineral phase second collision during ca 1.0 Ga. These relations
equilibria studies of BGB rocks reveal clock wise would imply that in certain cases, cratonic
exhumation path and hence collision environment convergence was followed by rifting during 1.6-1.5
should be operated at that time (Fig. 2d) between BC Ga, which affected marginal parts of southeast India
and EDC (Santosh et al., 2006). Subsequently due to and possibly coastal East Antarctica.
intracontinental orogeny at around 1.5 Ga, the P-G rift
basin was formed (Fig. 2e).
IMPLICATIONS ON SUPERCONTINENT
HISTORY
The ages of formation and reworking of Precambrian
terrains yield important information for the history of
supercontinents. The Eastern Dharwar Craton
contains juvenile gneissic crust slightly before 3.0 Ga
and having widespread granitic activity at ~2.5 Ga
indicating plume activity beneath the crust at that time
(Jayananda et al., 2000). The West and East Dharwar
cratons were welded by oblique convergence during
late Archaean (Chadwick et al., 2000). The Bastar
craton on the other hand, was multicomponent in
nature with crustal and mantle derived magmatic
additions, and depositional and tectonic reworking
during 2.7-2.1 Ga (Ramachandra et al., 2001).The
3.0-2.5 Ga ages from the KGB rocks show that the
terrain was a part of the Eastern Dharwar Craton. The
present study has established that the KGT represents
a 2.5 Ga old collisional orogen at the southern margin
of the Bastar craton against the Dharwar craton. Thus
the Dharwar, Bastar and Singhbhum cratons of the
Indian subcontinent may not have formed at a 3.0 Ga
old contiguous assembly. Recently Rogers and
Santosh (2002) proposed the configuration of a
Mesoproterozoic supercontinent Columbia that
preceded Rodinia and began to accrete at -1.9-1.8 Ga,
reaching maximum packing at 1.6-1.5 Ga, and began
rifting at 1.5 Ga. Configuration of Columbia was also
proposed by Zhao et al. (2002) which is an alternative
one. Several configurations have also been proposed
for the younger supercontinent Rodinia, which existed
at ca. 1.0 Ga (Moores, 1991; Karlstrom et al., 1999;
Fig. 2 Schematic model showing the evolution of northern
Burrett and Berry, 2000).
part of Eastern Dharwar craton (after Vansutre and Hari,
2010)

CONCLUSIONS Peninsular gneiss, Karnataka, South India. Journal
In the EDC few radiometric dates are now available Geological Society of India, 38, 357-368.
from gneisses, granites, charnockites, dykes and Jayananda, M., Moyen, J.-E, Martin, H., Peucat, J.J.,
sedimentary sequences. The oldest date available in Auvray, B. and Mahabaleswar, B. (2000). Late
the Eastern Dharwar Craton is 3.3 Ga for the Archaean (2550-2520) juvenile magmatism in the
Peninsular gneisses from Kabbaldurga area. Three Eastern Dharwar craton, southern India: constraints
major tectonothermal events around 3.0 Ga (MI), 2.5 from geochronology, Nd-Sr isotopes and whole-
Ga (MII) and 2.0 Ga (MIII) have been recognized in rock geochemistry. Precambrian Research, 99, 225-
this part of the Dharwar Craton. MII event is the 254.
strongest and is responsible for the present nature of Jayaram, S., Venkatsubramaniam, V.S. and
the craton except for the sedimentary basins. This MII Radhakrishna, B.P. (1976). Rb-Sr ages of cordierite
event is also correlated with the widespread gneisses of Southern Karnataka. Journal Geological
charnockitisation and granitic activity. Only scanty Society of India, 17, 557-561.
and patchy dates are available for the sedimentary Kameshwar Rao, T., (1989). Petrological and
sequences but on the basis of the two dates around 1.5 geochemical studies of high grade supracrustal
Ga and 0.98 Ga from the lower part of the Cuddapah rocks and associated granite and migmatitic
sequence, it may be said that these sedimentary rocks gneissic complex in parts of Karimnagar district,
were formed between 1.5 Ga and 0.5 Ga. Andhra Pradesh. Extended Abstract Geological
Survey of India, 122(5), 433-436.
Acknowledgements: We are thankful to Head of the Karlstrom, K.E., Harlan, S.S., Williams, M.L.,
Department of Geology, Banaras Hindu University McLelland, J., Geissman, J.W. and Ahall, A.I.
for providing the working facilities. DST is gratefully (1999). Refining Rodinia: geologic evidence for the
acknowledged for providing financial support to first Australia-western U.S. Connection. GSA Today, 9,
author. 1-7.
Krogstad, E.J., Hanson, G.N. and Rajamani, V.
REFERENCES (1988). U-Pb ages and Sr, Pb and Nd isotope data
Balakrishnan, S., Hanson, G.N. and Rajamani, V. for gneisses near the Kolar Schist Belt: Evidence
(1988). Geochemistry of the amphiboles from the for the juxtaposition of discrete Archaean Terranes.
Kolar Schist Belt. Journal Geological Society of Journal Geological Society of India, 31, 60-62.
India, 31, 9-11. Lopez, S., Fernandez, C.and Custro, A. (2006).
Bhalla, N.S., Gupta, J.M.T. and Vasudeva, S.G. Evolution of the Archaean Continental Crust: in
(1978). Rb-Sr geochronology of rocks from Kolar sight from the
schist belt, South India. In: Archaean experimental study of Archaean granitoids. Current
Geochemistry; Proceedings First International Science 91 (5), 607–621.
Symposium on Archaean Geochemistry, The Origin Martin, H., Smithies, R.H., Rapp, R., Moyen, J.F. and
and Evolution of Archaean Continental Crust Champion, D. (2005). An overview of adakite,
(editors, B.F. Windley and S.M. Naqvi), Elsevier, tonalite–trondhjemite–granodiorite (TTG) and
Amsterdam, 79-84. Sanukitoids: relationship and some implications for
Bhaskar Rao, Y.J., Sivaraman, T.V., Pantulu, G.V.C.; crustal evolution. Lithos 79, 1–24.
Gopalan, K. and Naqvi, S.M. (1992). Rb-Sr ages of Moores, E.M. (1991). Southwest U.S.-East Antarctic
Late Archaean metavolcanics and granites, Dharwar (SWEAT) connection: a hypothesis. Geological
Craton, South India and evidence for early Society of America, 19, 425-428.
Proterozoic thermotectonic event(s). Precambrian Peucat, J.J., Mahabaleswar, B. and Jayananda, M.
Research, 59, 145-170. (1993). Age of younger tonalitic magmatism and
Burrrett, C. and Berry, R. (2000). Proterozoic granulitic metamorphism in the South Indian
Australia-Western United States (AUSWUS) fit transition zone (Krishnagiri area); comparison with
between Luarentia and Australia. Geology, 28, 103- older Peninsular Gneisses from the Gorur-Hassan
106. area. Journal of Metamorphic Geology, 11, 879-
Chadwick, B., Vasudev, V.N., Hegde, G.V., (2000). 888.
The Dharwar craton, southern India, interpreted as Prakash, D. and Sharma, I.N. (2011). Metamorphic
the result of Late Archaean oblique convergence. evolution of the Karimnagar granulite terrane,
Precambrian Research, 99, 91-111. Eastern Dharwar Craton, south India. Geological
Condie, K.C., (1988). Plate-Tectonics and Crustal Magazine, 148(1), 112-132.
Evolution. Pergamon Press, NewYork. p. 292. Rajesham, T., Bhaskara Rao Y.J. and Murti, K.S.
Crawford, A.R. (1969). Reconnaissance Rb-Sr dating (1993). The Karimnagar granulite terrane – a new
of the Precambrian of Southern Peninsular India. sapphirine-bearing granulite province, South India.
Journal Geological Society of India, 10, 117-166. Journal Geological Society of India, 41, 51-59.
Friend, C.R.L. and Nutman, A.P. (1991). SHRIMP U- Ramachandra, H.M. and Roy, A. (2001) Evolution of
Pb geochronology of the Closepet granite and the

Bhandara-Balaghat granulite belt along the Karimnagar and Bhopalpatnam Granulite Belts,
southern margin of the Sausar mobile belt of central Central India, Gondawana Research, 7, 501-518.
India. Proceedings of the Indian Academy of Santosh, M., Tsunogae, T., Iki, T., Vansutre, S. and
Sciences, Earth and Planetary Sciences, 110, 351- Hari, K.R. (2006). Petrology, fluid inclusion and
368. metamorphic history of Bhopalpatnam granulite,
Rogers, J.J.W. and Santosh, M. (2002). Configuration Central India. Journal of Asian Earth Sciences, 28,
of Columbia, a Mesoproterozoic supercontinent. 81–98.
Gondwana Research, 5, 5-22. Sarkar, S.N. (1968). Precambrian stratigraphy and
Santosh, M., Yokoyama, K. and Acharyya, S.K. geochronology of Peninsular India. Report to the
(2004). Geochronology and Tectonic Evolution of
subcommission on Precambrian Stratigraphy, Unpublished Ph.D. Thesis, PanditRavi Shankar
Dhanbad Publications, 1-33. Shukla University, Raipur, India, 134.
Stähle, H.J., Raith, M., Hoerness, S. and Delfs, A. Vansutre, S. and Hari, K. R. (2010). Granulite belts of
(1987). Element mobility during incipient granulite central India with special reference to the
facies formation at kabbaldurga, Southern India. Bhopalpatnam supercontinent Journal of Asian Earth
Journal of Petrology, 28, 803-834. Sciences, 39, 794-803.
Swaminath J. and Ramakrishna M. (1981). Early Venkatasubramanian, V.S. and Narayanaswamy, R.
Precambrian supracrustals of southern Karnataka. (1974). Studies in Rb-Sr geochronology and trace
Geological Survey of India Memoir, 112, 350. element geochemistry in granitoids of Mysore
Taylor, P.N., Chadwick, B. and Friend, C.R.L. (1988). craton, India, Journal Indian Institute of Science
New Age Date on the Geological Evolution of (Bangalore), 56, 19-42.
Southern India. Journal Geological Society of India, Zhao, G., Cawood, P.A., Wilde, S.A. and Sun, M.
31, 155-157. (2002). Review of global 2.1-1.8 Ga orogens:
Vansutre, S., 2007. Geochemistry and implications for a pre-Rodinia supercontinent. Earth
Thermobarometry of the Corundum Bearing Science Reviews, 59, 125-162.
Gneisses and Charnockites in a Part of BGB.
Table 1. Chronostratigraphy of the Igneous and Metamorphic rocks of Eastern Dharwar Craton
Age (Ma.) Stratigraphy Tectonothermal Event
Deformation along E-W and N-S shear
2100-1700 Dolerite zone, retrograde metamorphism (MIII)
and thermal resetting of mineral dates.
Karimnagar Granite
2600.2400 Closepet Granite, Patna
Granite & Karimnagar
Granite-Gneisses
MII, 2600/2500 Ma, Deformation and
metamorphism greenschist to low
pressure
And-Sil type in Eastern Dharwar
Craton.
<3000 Kolar Group
------------------------unconformity----------------------- uplift and erosion
3000 Peninsular Gneisses MI, 3000 Ma, Deformation and granite
intrusive, high grade amphibolite facies
metamorphism throughout the Dharwar
Craton
> 3000 Sakarsanahalli Group
Basement Sialic (not exposed)
Oldest date is 3400 Ma of Peninsular Gneiss from Kabbaldurga area near the junction of the Western and
Eastern Dharwar Cratons.
Table 2. Chronostratigraphic Succession in Dharwar Craton (After Swaminath and Ramakrishna, 1981)
Age (Ma) Western Dharwar Craton Eastern Dharwar Craton
Middle to Late Proterozoic Kaladgi, Badami and Bhima Groups
----------------------------------------------unconformity-------------------------------------------------------------
2100-1700 -----------------------------------------------Dolerite Dykes-----------------------------------------

2600-2400 Chitradurga Granite Closepet and Patna
Granites
2600 to <3000 Dharwar Supergroup:
Chitradurga Group
-----------unconformity-----------
Bababudan Group Kolar Group
------------------------------------------unconformity-----------------------------------------------
3000 Peninsular Gneiss
>3000 Sargur Group Sakarsanahalli Group
3400 Tonalitic gneiss with Basement not exposed

to enclaves of basic and
3580 ultrabasic rocks
Table 3. Major Differences between the Two Blocks of the Dharwar Craton
Western Dharwar Craton Eastern Dharwar Craton
1. Characterised by medium to high pressure 1. Characterized by low to medium pressure
kyanite-sillimanite type of facies series. andalusite-sillimanite type of facies series.
2. The younger supracrustals include continental 2. The younger supracrustals include volcanics,
type shallow water clastic sediments grading immature clastic and chemical sediments of
into greenstone-greywacke-chert association of oceanic character (Keewatin or Kolar type
oceanic affinity (Dharwar type greenstone). greenstone).
3. Tonalitic gneiss basement with enclaves of

basic and ultrabasic rocks. 3. Basement not exposed.
Table 4: Geochronology of the rocks of Eastern Dharwar Craton

S.No. Rock Type Locality Age (Ga) Method Reference
1. Gneiss Kabbaldurga 3.3 U-Pb (Zircon) Stahle et al. (1987)
2. Schist Kolar 3.2 U-Pb (Zircon) Krogstad et al. (1988)
3. Amphibolite Hutti 3.2 K-Ar (whole rock) Sarkar (1968)
4. Cordierite- Western & 3.01±0.9 Rb-Sr, Sr87/Sr86 of Jayaram et al. (1976)
gneisses Eastern side 0.701±0.001
of Closepet
Granite
5. Peninsular Gneiss Kabbaldurga 2.96 ±0.004 U-Pb (Zircon) Friend & Nutman (1991)
87 86
6. Gneiss & Granites Bangalore 2.95±0.11 Rb-Sr, l Sr /Sr of Venkatasubramanian &
0.706±0.004 Narayanaswamy (1974)
7. Amphibolite Kolar 2.9 Pb-Pb (whole rock) Tayler et al. (1988)
8. Champion Gneiss Kolar 2.89±0.96 Rb-Sr, Sr87/Sr86 of Bhalla et al. (1978)
0.702±0.002
9. Amphibolite Kolar 2.73±0.155 Pb-Pb isochron Balakrishnan et al. (1988)
10. Amphibolite Kolar 2.69±0.14 Pb-Pb isochron Balakrishnan et al (1988)
11. Aplite Bangalore 2.625±0.90 Rb-Sr, Sr87/Sr86 of Venkatasubramanian &
0.703±0.004 Narayanaswamy (1974)
12. Charnockite Karimnagar 2.612±0.100 Rb-Sr, Sr87/Sr86 of Bhaskar Rao (1992)
0.7033±0.08
13*. Peninsular Gneiss Bangalore 2.59±0.04 Rb-Sr, Sr87/Sr86 of Crawford (1969)
7016±0.001
14*. Peninsular Gneiss Kolar 2.59±0.04 Pb-Pb (whole rock) Tayler et al. (1988)
15. Closepet Granite Kolar 2.578 Pb-Pb (whole rock) Tayler et al. (1988)
16. Granite-gneiss Krishnagiri 2.557±0.016 U-Pb (Zircon) Peucat et al. (1993)
17. Charnockite Krishnagiri 2.553±0.005 Pb-Pb (Zircon) Peucat et al. (1993)
18. Tonalite-gneiss Krishnagiri 2.552±0.014 Pb-Pb (Zircon) Peucat et al. (1993)
19. Patna Granites Kolar 2.550 U-Pb(Zircon, Sphene) Krogstad et al. (1988)

20. Charnockite Krishnagiri 2.548±0.003 Pb-Pb (Zircon) Peucat et al. (1993)
21. Granite-gneiss Krishnagiri 2.546±0.004 Pb-Pb (Zircon) Peucat et al. (1993)
22. Tonalitic-gneiss Krishnagiri 2.533±0.003 Pb-Pb (Zircon) Peucat et al. (1993)
23. Tonalitic-gneiss Krishnagiri 2.532±0.014 U-Pb (Zircon) Peucat et al. (1993)
*
24 . Peninsular Gneiss Kabbaldurga 2.528±0.005 U-Pb (Zircon) Friend & Nutman (1991)
25. Closepet Granites Kabbaldurga 2.513±0.005 U-Pb (Zircon) Friend & Nutman (1991)
26. Charnockite Krishnagiri 2.517±0.010 Pb-Pb (Monazite) Peucat et al. (1993)
27. Charnockite Krishnagiri 2.506±0.010 Pb-Pb (Monazite) Peucat et al. (1993)
87 86
28. Granites Hyderabad 2.490±0.115 Rb-Sr, Sr /Sr of Crawford (1969)
0.7045±0.0044
29. Granite-Gneisses Karimnagar 2.483±0.204 Rb-Sr, Sr87/Sr86 of Bhaskar Rao (1992)
0.7047±0.026
30. Granite-Gneisses Karimnagar 2.478±0.148 Rb-Sr Rajesham et al. (1993)
31. Closepet Granite Gadag 2.456 Rb-Sr Bhaskar Rao et al. (1992)
32. Closepet Granite Toranagulli 2.452 Rb-Sr Bhaskar Rao et al. (1992)
33. Charnockite Karimnagar 2.446±0.065 Rb-Sr Rajesham et al. (1993)
87 86
34. Closepet Granite 2.380±0.035 Rb-Sr, Sr /Sr of Crawford (1969)
0.7049±0.0014
35. Closepet Granite Sandur 2.377 Rb-Sr Bhaskar Rao et al. (1992)
87 86
36. Charnockite Krishnagiri 2.337±0.010 Rb-Sr,Sr /Sr of Peucat et al. (1993)
0.7102537
37. Granite Karimnagar 2.286±0.100 Rb-Sr, Sr87/Sr86 of Bhaskar Rao (1992)
0.7184±0.011
38. Grey-Gneisses Karimnagar 2.217±0.108 Rb-Sr, Sr87/Sr86 of Bhaskar Rao (1992)
0.7133±0.003
39**. Closepet Granite 2.000 Rb-Sr Bhaskar Rao et al. (1992)
40. Charnockite Karimnagar 3.1 (Core) U-Th-Pb Santosh et al. (2004)
2.6 (Rim) Zircon
41. Charnockite Karimnagar 2.42±0.08 Monazite, Uranitite Santosh et al. (2004)
2.47±0.03 (U-Th-Pb)
* Age is lower than the commonly assumed age of gneisses, may be due to the effect of Closepet Granite.
** Age is lower due to thermal resetting during early Proterozoic, may be due to dolerite dykes.
Table 5. Stratigraphic Sequence of the Karimnagar area

SEDIMENT COVER
Quartz Veins Intrusives
Dolerite Dykes 2.1 to 1.7 Ga
Karimnagar Granites 2.3 to 2.2 Ga
Deformation and metamorphism MII (2.6/2.5 Ga) and
intrusion of granites
Karimnagar Granite-Gneisses comprising of granitic and 2.6 to 2.5 Ga
tonalitic rocks
Charnockites 2.6 to 2.5 Ga
Basic Granulites
Amphibolites
Meta-ultramafics High-Grade Supracrustals*
Quartz-free granulites
Cordierite-bearing gneisses and biotite schists (> 2.5 Ga)
Banded magnetite quartzites interbedded with quartzites
BASEMENT NOT EXPOSED
 These high grade supracrustals are not in stratigraphic order, as they occur together as enclaves within
granite-gneisses and charnockites and must be older than them.

ISBN 978-81-921121-0-7
VELOCITIES IN NON-UNIFORM FLOW ON CLOSELY PACKED

ROUGH BED
Kesheo Prasad
Assistant Professor, Dept. of Civil Engineering, Institute of Technology, BHU, Varanasi, India, e-mail: kesheo_bhu@yhoo.com
C.S.P. Ojha
Professor, Dept. of Civil Engineering, Indian Institute of Technology, Roorkee, India
K. M. Singh
Assistant Professor, Dept. of Mechanical and Industrial Engineering, Indian Institute of Technology, Roorkee, India,
ABSTRACT: Series of experiments are conducted over closely packed hemispheres fixed bed of a channel and the
velocities in 3-D is measured using Acoustic Doppler Velocimiter (ADV). An attempt has been made to study the
variation in the flow velocities in the longitudinal direction of flow in wall region (y/h<0.2) and outer region
(0.2<y/h<1) in open channel, h being the fow depth of flow and y is the vertical distance measured with reference to
bed . In addition to this, an attempt is also made to understand the effect of the bed roughness on dip. Results for one
discharge 50m3/s (Reynolds and Froude number are 1.076x105 and 0.62, respectively indicate that the maximum
velocity is attained at 0.26h from water surface.
INTRODUCTION
Flow of water in hilly rivers above stones, air equivalent roughness height and y0 is the distance of
movement above the cities and forest, and flow of fluid virtual bed level from the bed of the channel.
in turbo machinery are some prominent areas where Y
turbulent flow appears in nature. Very often in Water surface
Um
Flow depth, h
engineering applications, the transport rate of a scalar
contaminant is also fixed by transport mechanism
occurring at the wall. Turbulent flow produced by

rough bed surface is complex in nature and has
remained area of interest of the engineers since long u
time. This has been studied with various flow y Virtual bed level
conditions; non-uniform flow with gradually yo k ks X

accelerating flows on a smooth surface by Tsujimoto et Fig. 1: Definition sketch, virtual bed level, rough
al. (1990), Song and Graf (1994), Song and Chiew surface height and equivalent height.
(2001), Cardoso et al. (1991). Kironoto and Graf (1995)
determined velocity profiles for accelerating and Keulegan (1938) suggested that logarithmic law of
decelerating flows in a rough channel, whereas, velocity distribution for smooth and rough pipes can
Bergstrom et al. (2002) studied turbulent flow in wind also be extended to smooth and rough open channel
tunnel. Nezu et al. (1994) measured turbulence flows and he suggested the following power laws:
fluctuations for accelerating and decelerating flows in for small R / k values
smooth channel. Nikuradse (1932) stated that U / u in
u 1/ 6
the wall region is a function of yu / , where, U is the  9.1 y / k  , (2)
u
average mean velocity, u is the shear velocity, y is the
for large R / k values
local depth of flow and  is the kinematic viscosity.
u 1/ 7
However, for rough pipe flow, an additional parameter  10.2  y / k  , (3)
R  ku / (1) u
[where, R is the hydraulic radius and k is the Eqs. (2) and (3) are simple but range of small and large
roughness size] can also be considered to determine R / k was not defined; hence, for central vertical
velocity distribution a parabolic law was introduced as
U / u in the wall region. Laufer (1954) also supported 2
this theory. Figure 1 shows the conceptual definitions u m  u   6.3 1  y  (4)
of terms used in the Eq (1). In Fig. 2, Um and  where
represent the maximum velocity and the boundary layer um  um / u and u   u / u* (5)
thickness, respectively; u is the local velocity at a
Vanoni (1941) observed that the vertical velocity
distance y, k is the height of roughness element, ks is the
curves of the central region of channel were identical.

He also stated that these identical velocity curves will where, u* is determined using
be in the region of no dip and the phenomenon of dip
occurs in the vicinity of the side walls. Further, there u*  ghS (10)
always exists a central region, in which the logarithmic [where, h is depth of flow and S is slope of channel] for
law extends up to the free surface, when value of aspect smooth and rough surface. Afzalimher and Anctil(2000)
ratio is less than 5. Coles (1956) introduced a purely and Chen and Chiew (2003) stated that Eq. (10) can be
empirical expression to predict velocities in the fully used to non uniform flow in an open channel flow.

turbulent flow ( k s > 70) for inner region as well as for Kirgoz (1989) determined Coles constant as 0.2 and
stated that velocity defect law is valid for y/>0.03.
outer region Swamee (1993) proposed an explicit equation of
u 1 u y   y velocity profile for smooth as well as rough surface
 ln *  A  w   (6)
u*   k   which covers entire inner region as
0.3
Where, = 0.55 with K = 0.4, A = 5.1, and w( y /  ) is  10/ 3  1  9 y  
10/ 3

u   y 
 
  k ln 1   (11)
 
law of the wake given by 2sin 2  y / 2  for zero    1  0.3   
pressure gradient. Clauser (1956) stated that velocity  um
profile for turbulent flow layers depends upon both where    , in which  is the average roughness

Reynolds number and degree of roughness of the wall.
He developed a correlation between velocity u/um with height of the surface. Swamee (1994) developed an
average velocity equation for the uniform flow to a
y/ along the flow direction at a distance y from the
rigid boundary known as resistance equation
wall and concluded that no sample rescaling of u/um and
y/ will reduce the curves to a single curve. But if   0.221 
v  2.547 gRS0 ln    (12)
(um  u ) / u* is used instead of u/um, with y/, than  12 R R gRS 
 0 
turbulent velocity may collapse to a single curve. Vittal where g is gravitational acceleration (varying between
(1972) stated that a single logarithmic law is not 9.78 m/s2), ε is average roughness height of the channel
applicable over the entire depth of flow. Sarma et al. lining (m) and So is channel bed slope. Nikora and
(1983) reported an expression for velocity distribution Smart (1997) stated that log wake law gives the
power law as velocity distribution in longitudinal direction as
n
u / u*  C  yu /  (7) u 1 y
 ln  w( y / h) (13)
which fits best with C=8.3 and n=1/7 in wall region. u*  yo
Further, the equation where u, u* , K, y, yo and w(y/h) are local mean velocity,
2
uc  u  2y  shear velocity; von Karman constant=0.4, distance
 k w 1   (8)
u*  B  from channel bed, roughness parameter and the wake
gives the better results in the outer side region, function, respectively. Afzalimehr and Anctil (2000)
where k w is a constant as 2.4. Sarma et al. (1983) suggested that due to the uncertainty in the
measurement of the distance from the bed y and in the
determined u* using the intercept by plotting semi log estimation of the shear velocity u* , the law of the wall
of u and y/H at y/H=1. This value of u* was used to can be formulated as
determine the value of constant C in Power law and the u 1  y  Ac d50 
value of C was found 8.3. Sarma et al. (1983)  ln   A (14)
u*   ks 
concluded that the aspect ratio in the range of 2.0 to 0.8
had no significant effect in the inner region of the bed, where K is the von Karman constant=0.4, Ac d50 is
side wall and outer region of bed. Kirkgöz (1989) reference level, Ac is the value producing best
studied the mean velocity in fully developed turbulent
correlation coefficient for the regression, A is constant
flow on rough and smooth surfaces using velocity
of integration, ks = d50 is Nikuradse grain roughness.
measuring instrument LDA. He determined u* from They also observed that the maximum velocity appears
measured velocity distributions by shifting the virtual at y/h =0.6. The variation in choosing k s affects the
bed level at a distance yo from the bed. He also integration constant in the law of the wall, but the value
explained that the velocity, in the outer region is mainly of ks does not influence the shear velocity. Sarma et al.
controlled by the turbulent shear and velocity (2000) stated that the phenomenon of velocity dip is
distribution. He derived velocity defect law as caused by the secondary currents in open channel and
um  u 1 y   y  maximum velocity occurs below the water surface. Rao
 ln  2 cos 2   (9)
u*     2  and Kumar (2009) stated that in wall region, the

viscosity and wall shear stress are the important reading represents the vertical drop in the channel. The
parameters which govern the velocity distribution, incoming discharge was measured continuously using a
whereas in the outer region, the wall reduces velocity Portable Transit Time Flow meter and the strength of
below the maximum velocity and it independent of sound was checked and found more than 30DB during
viscosity. Song and Chiew (2001) defined that log law the experiment.Three dimensional velocities with
u 1  y  yo  sampling rate 10 Hz for a length of time series 150s
 ln   B (15) (total 1500 samples) were collected at six locations
u*   k s 
(three sections at central line of channel and three at
[where, yo is the virtual distance from bed] can 10.5cm away from the side wall at distance 4.815m,
represent the non-uniform in the inner region in open 5.275m and 5.505m from inlet) in the longitudinal
channel flow, limited by y/h<0.20. The constant B as direction (Fig. 2). These velocities were measured for
8.5 is commonly accepted value for fully turbulent one discharge at fixed bottom slope, using a
rough bed. They rearranged the Eq. (15) as downlooking intrusive instrument Acoustic Doppler
u  2.5u* ln( y  yo)  Ac (16) Velocimeter (ADV). The temperature and salinity of
where, water during the experiment were measured and were
Ac  B  2.5ln k s (17) found to be 20 degree centigrade and 20ppm,
respectively.
Eq. (17) reveals that the slope of u versus ln(y) line will
give value of B and the value of ks can be evaluated.
Further, the estimation of ks has been attempted by
many researchers. Afzalimehr and Anctil (2000)
considered the equivalent Nikuradse grain roughness ks
as d50, Hey (1979) as 3.5d84, and Ackers & White (1973)
as 1.23d35, Whiting and William (1990) as
k s  Ad84 since the value of ks mainly affects the
velocity profile and measured discharges.
EXPERIMENTAL SETUP
The experiments were conducted in a 10.5m long, Fig. 2: Schematic diagram of vertical sections on rough
0.46m wide, and 0.60m deep horizontal flume. The bed
sides of channel were made up of glass and the bottom
of the channel was steel base with a 5mm fiber on top
of it. The flow in the flume was supplied through a pipe
of 202mm outer diameter through the pump. The water
depth in the flume was regulated using a tailgate at the
end of channel. To create turbulent flow, hemispheres
of concrete in 2.3cm height were prepared and fixed
(Fig. 2) over the bed of channel in 1.5m length in a
reach 1.5m at 4.5m from the upstream end of the flume. Fig. 3: Closed packed hemisphere on bed
At entrance, honey comb wall of small bricks was
constructed to reduce the large scale disturbances in the Speed of sound at 20 degree centigrade was also
approaching flow. A floating wooden baffle of 1m was recorded. The velocity range of flow was kept fixed 100
used to minimize the disturbances at free surface. To cm/s since the variation of actual local velocity was in
measure levels of water, a movable carriage with a the range of 70cm/s to 95cm/s and the local velocity at
pointer gauge (The pointer gauge arrangement had a a point was determined using
least count of 1mm) mounted on the pipe rail at the top 1T 1 N
of the channel was used. The slope of the channel was u  u(t )dt   ui ti (18)
T0 T i 1
measured using two containers, connected at their
bottom with a long plastic tube, placed on the channel Due to technical restriction of ADV, 3-D velocities data
bottom, one each at two predetermined locations along using downlooking ADV within the depth 7cm below
the length of the channel. Care was taken to eliminate from water surface were not available and it was also
air bubbles in the plastic tube. Nearly 18 hours time observed that water at surface was making vortex
was allowed for equalization of the water level in both around the rod of ADV and it was creating an air cavity
the containers and than water levels in both the around the rod of ADV, hence transducer of ADV
containers was noted by means of the pointer gauges within depth 7cm from water surface may produce
mounted on the rails. The difference in the gauge erroneous data. Therefore, to collect error free data, the

downlooking probe was kept 3cm below inside water dy
and a 7cm deep section from water surface was Se  S0 
dx
1  Fr2  (19)
measured using upward looking ADV. Table 1 shows
where, y is the depth of flow. Clauser's (1956)
the hydraulic parameters collected at 5.275 m from inlet described another method to estimate the shear velocity
of channel. from the parameters of the regression of the mean point
velocity u and ln[(y+yo)/ks] as
Table 1: Summery of Flow Parameters and Cross
y  yo
Section Geometry u   ln( ) B (20)
Q (m3/s) Re (x105) Fr Bed Slope w/h Ratio ks
0.05 1.067 0.62 0.000717 2.71 and Eq (17) leads to
u*  k (21)
Table 2: The estimated u* from intercept method and Which confirms the validity of the law of the wall for
Run u  gHS (m/s) u i (m/s) ks (m) the inner region of the flow. The spatially periodic
* *
nature of the roughness bed features makes the
50H M1 0.034508 0.058 0.0057 definition of streamwise average meaningful, e.g. the
50H M4 0.034508 0.047 0.0027 mean energy slope Se. The total shear velocity, derived
50H M6 0.034508 0.043 0.0011 from Eq (10) and Eq (19) as
50H SR1 0.034508 0.072 0.0140
u*  ghSe (22)
50H SR4 0.034508 0.042 0.0011
50H SR6 0.034508 0.066 0.0017 is examined with the u* using Eq (10), as shown in
figure 4 (a &b). It is found that average men energy
The estimated u*i from intercept method and u* using slope Se is not showing much deviation with values
Eq (10) is shown in the Table 2 along with the determined using Eq.(10). Hence, estimated u* using
estimated ks for different runs. Eq (10) is considered in the present study.
0.18
50H at Centre linel
DISCUSSION AND RESULTS 0.16
SWM 50H M1S
0.14
Estimation of velocity distribution over the rough bed SWM 50H M1So
0.12
of hemisphere of arranged closely are attempted in this 0.1
SWM 50H M4 So
y (m)
study. The velocity distribution in a two dimensional 0.08

SWM 50H M4 S
uniform open channel flow usually follows Eq. (15) for 0.06
SWM 50H M6 S
inner region. For the convenience, Eq. (15) is used in 0.04

SWM 50H M6 S0
the present nonuniform flow conditions. The velocity 0.02
profile in two regions (i) Inner region and (ii) Outer 0

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
u (m/s )
region is examined along with the estimated value of ks
using Eq (17). (a)
0.18
50H at 10.5cm from centre
Velocity distribution in inner region 0.16
SWM 50H M1S
From Eq (17), it is evident that the profile mainly 0.14
SWM 50H M1So
depends upon the chosen value of u* , yo and ks.. Hence, 0.12
SWM 50H M4 So
0.1
proper estimation of these parameters plays a vital role
y (m )
SWM 50H M4 S
0.08
to define the velocity profile. Figs 4(a) & (b) are plotted SWM 50H M6 S
between u+ versus y+, which depict that a wall region is 0.06
SWM 50H M6 S0
0.04
seen up to y/h<0.28 and it also follows log law in
0.02
central line region. However, near the side wall of
0
channel y/h<0.42, the reason of this may be due to the 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
u (m/s)
secondary currents.
(b)
Shear velocity estimation Fig. 4: A velocity profile using Eq (10) and Eq (22) at;
Chow (1959) proposed the friction slope application for (a) central line section and (b)10.5cm from centre
uniform flow and the negative bottom slope associated
with this type of flow may prevents the use of the zero Estimation of roughness height
pressure-gradient model, u*  gRS , where S is the The roughness height plays an important role to define
bed slope. Use of conservation of momentum equation the velocity profile in rough open channel. The values
leads to the bed shear velocity expression as of ks for all runs are determined using Eq (17).

28
50H M1
28
50H SR depict that as the intercept is increasing, shear stress is
26
26
decreasing in the flow direction in the central region.
24 24
22 22
Velocity Distribution in outer region
y+
u+
20
20
18 18 The phenomenon of dip in the velocity profiles in open

16
16 channel flows occurs if the aspect ratio of the channel
14
100 1000 10000
14
100 1000 10000
width to water depth is less than 5.0 (Sarma 1990). In
y+ u+
our case w/h = 2.71 and the dip is attained in the central
(a) (b) line region of the channel. In addition to this, the value
Fig. 5: Validation of the logarithmic, Run 50H M1 and of Coles’ constant is determined for closely packed
50H SR1. hemispheres (Fig 8) and are found in the range -
Kironoto and Graf (1995) indicated that the log law is 0.068<  <-0.724 for the central region, similarly in the
valid up to y/h<0.2 and the zone y/h>0.2 will not be range of -0.206<  <-0.472 at 10.5 distance from the
affected by the accelerating flow. Figure 6(a) & (b) glass wall.
shows y/h versus u / um along with the velocity data 15 15
[500VPs and 600 VPS] of Afzalimehr et al. (2010). 500 VPs

50H M1
10 10
1 1
(umax -u)/u*
(umax-u)/u*
500 VPs 500 VPs
0.8 0.8
600 VPs 600 VPs
5 5
0.6 50H M1 0.6 50H SR1
y/h
y/h
50H M4 50H SR4

0.4 0.4
50H M6 50H SR6
0 0
0.2 0.2 0.01 0.1 1 10 0.01 0.1 1 10
y/ y/
0
0 0.3 0.6 0.9 1.2
0
0 0.3 0.6 0.9 1.2 (a)   0.219 (b)   0.724
u/um u/um
15 15
(a) (b) 50H M4 50H M6

Fig. 6: Horizontal mean velocity profile at (a) central 10 10
line of channel, (b) 10.5cm from right side wall of

(umax-u)/u*
(umax -u)/u*
channel. 5 5
These two plots Fig 6(a) & (b) depict that velocity
profile over regular hemispheres is more affected in the 0 0
lower depth of flow comparative to the bed surface with 0.01 0.1 y/ 1 10 0.01 0.1 y/ 1 10
vegetation near to the channel side wall. The maximum (c)   0.564 (d)   0.068
velocity, um for accelerating flow should be 0.6 at y/h as 15
15
stated by Kironoto and Graf (1995) but according to

50H SR1
Afzalimehr and Dey (2009) um occurred at y/h=0.45 500 VPs 6
10
10
when w/h=3 and, it occurred at y/h=0.6 when w/h=4.5

(umax -u)/u*
(umax -u)/u*
at the central axis of flume, while in our case it 5

5
occurred at y/h= 0.74 when w/h= 2.71.
0 0
1 1 0.01 0.1 y/ 1 10 0.01 0.1 1 10
y/
50H M1 50H SR1
50H M4 50H SR4 (e)   0.826 (f)   0.472
0.8 50H M6 0.8 50H SR6
u (m/s)
u (m/s)
15 15
0.6 0.6 50H SR4 50H SR6
10 10
(umax -u)/u*
(umax -u)/u*
0.4 0.4
0.001 0.01 0.1 0.001 0.01 0.1
y+yo (m) y+yo (m)
5 5
(a) (b)
Fig. 7: Velocity distribution and fitted log law at center 0 0
0.01 0.1 1 10 0.01 0.1 1 10
line and at 10.5cm from right side wall. y/ y/
(g)   0.346 (h)   0.206

Fig. 7(a) and (b) are drawn by fitting the u with y+yo for
central line region and at 10.5cm away from the right Fig. 8: Calculation of 
side wall in the flow direction. Figs. 6(a) & (b) depict
that as one moves towards the down stream, the shear The non-uniformity along the depth of flow in the
stress is going to decrease, further Figs. 7(a) & (b) also velocity distribution is due to change in the shear stress.

CONCLUSIONS Bergstrom, D.J., Kotey, N.A., Tachie, M.F., (2002).
The following conclusions are drawn from this study “The effects of surface roughness on the mean
velocity profile in a turbulent boundary layer.” J.
1. In the estimation of the u*, Eq(10) performs well
Fluids Engineering, 124, 664-670.
compared to Eq (22) when the difference between bed
Cardoso, A.H., Graf, W.H., and Gust, G. (1991).
slope and energy line is not significant.
“Steady gradually accelerating flow in a smooth
2. For the present experiments, the maximum velocity
open channel.” J. Hydraul. Res., 29(4), 525–543.
is found at y/h=0.74 when w/h=2.71.
Chen Xingwei and Chiew, Y.M. (2003). “Response of
3. The Coles parameter for the closely packed
Velocity and Turbulence to Sudden Change of Bed
hemisphere were found in the range of -0.068 <  < -
Roughness in Open-Channel Flow.” J. Hydraulic
0.724 for central region.
Engineering, 129(1), 35-43.
4. It is observed that  reduces in the direction of
Coles, D. (1956). “The law of the wake in the turbulent
flow.
boundary layer.” J. Fluid Mechanics, 1, 191 - 226.
Chow, V.T (1959). Open Channel Hydraulics,
NOMENCLATURE
McGraw-Hill College
d50=Mean size of particle Clauser, F. H. (1956) “The turbulent boundary layers.”
d84= 84% finer Adv. Appl. Mech. 4, 1-15.
Fr= Froude number Hey, R.D. (1979). "Flow resistance in gravel-bed
h= Depth of flow in cannel rivers." J. Hydraulic Division, ASCE, 105(4), 365-
 = von Karman constant=0.4 379.
ks= Equivalent roughness height Keulegan, C.H. (1938). “Laws of turbulent flow in open
N = Number of observation channels.” J. Research National Bureau of
R= Hydraulic mean depth Standards, 21, 707-740.
S= Energy slope line Kirkgöz, M.S. (1989). “Turbulent Velocity Profiles for
S0= Bed slope of channel Smooth and Rough open Channel Flow.” J.
R = Radius of pipe Hydraulic Engineering, 115(11), 1543-1561.
Re = Reynolds number Kironoto, B., and Graf, W.H. (1995). “Turbulence
U = Average mean velocity characteristics in rough nonuniform open-channel
u= local mean velocity flow.” Proc. Inst. Civ. Eng., Waters. Maritime
Energ., 112, 316–348.
u = Shear velocity Laufer, J. (1954). “The structure of turbulence in fully
um= Maximum velocity of flow developed pipe flow.” Tech. Report No. 1174,
y = Depth of flow in vertical direction National Advisory Committee for Aeronautics,
yo = Virtual depth from bottom of channel Washington.
 = kinematic viscosity Nezu, I. (1994). ‘‘Compound open channel turbulence
 = Average roughness height and its role in river environment’’ keynote address
 = Constant of 9th APD-IAHR Congress, Delft, The
 = Distance of maximum velocity from channel bed Netherlands, 1-24.
Nikora, V.I., and Smart, G.M. (1997). ‘‘Turbulence
REFERENCES characteristics of New Zealand gravel-bed rivers.’’
Ackers, P. and White, W. R. (1973) Sediment transport: J. Hydraulic Engineering, ASCE, 123(9), 229-246.
new approach and analysis. J. Hydraulic Division. Nikuradse, J. (1932). English Translation: Law of Flow
ASCE , 99(11), 2041-2060. in Rough Pipes. TM 1292, NACA, USA.
Afzalimher H., Najfabadi, E.F., Singh, V.P., (2010), Rao, A.R., Kumar, B. (2009) “Transition of turbulent
“Effect of Vegetation on Banks on Distributions of pipe flow.” J. Hydraulic research, 47(4), 529-533.
Velocity and Reynolds Stress under Accelerating Sarma, K.V.N., Lakshminarayana, P., and Rao, N.S.L.
Flow”, J. Hydrologic Engineering, 15(9), 708-713. (1983). “Velocity distribution in smooth
Afzalimher, H. and Anctil, F. (2000). “Accelerating rectangular open-channels.” J. Hydraulic
shear velocity in gravel-bed channels.” Engineering, ASCE, 109(2), 271-289.
Hydrological Sciences, J. -des Sciences Sarma, V.N. Kandula, Pasad B.V.R. and Sarma A.K.
Hydrologiques, 45(1), 113-124. (2000). “Detailed Study of Binary Law for Open
Afzalimehr, H., and Dey, S. (2009). “Influence of bank Channel.” J. Hydraulic Engineering, 126(3), 210-
vegetation and gravel bed on velocity and 214.
Reynolds stress distributions.” Int. J. Sediment Res., Song, T. and Graf, W.H. (1994). “Non-uniform open
24(2), 236–246. channel flow over a rough bed.” J. Hydroscience
and Hydraulic Engineering, 12(1), 1-25.

Song, T. and Chiew, Y.M. (2001). “Turbulence
Measurement in Nonuniform Open-Channel flow
using Acoustic Doppler Velocimeter (ADV).” J.
Engineering Mechanics, ASCE, 219-232.
Swamee, P.K., (1993). “Generalized inner region
velocity distribution equation.” J. Hydraulic
Engineering, ASCE, 119(5), 651-656.
Swamee, P.K., ISH, M., and Pathak S.K. (1994).
“Velocity Distribution Equations for Laminar and
Turbulent Boundary Layers.” The Indian Society
for Hydraulics J. Hydraulic Engineering, 8(1) 50-
59.
Tsujimoto, T., Saito, A., and Nitta, K. (1990). “Open-
channel flow with spatial acceleration or
deceleration.” KHL Progress Rep., Hydraulics
Laboratory, Kanazawa Univ., Kanazawa, Japan.
Vanoni, V.A. (1941). ‘‘Velocity distribution in open
channels.’’ Civ. Engrg., ASCE, 2(6), 356–357.
Vittal. N. (1972). “Flow over triangular roughness in
open channels.” Ph. D. Thesis, Civil Engineering
Department, IIT Roorkee.
Whiting, P.J., and Dietrich, W.E. (1990). ‘‘Boundary
shear stress and roughness over mobile alluvial
beds.’’ J. Hydraulic Engineering, ASCE, 116(12),
1495-1511.

ISBN 978-81-921121-0-7
ANALYSIS AND CONSTRUCTION OF A LOW COST FOLDABLE

DOME
Rajesh Kumar
Associate Professor, Institute of Technology, Banaras Hindu University,Varanasi
Aniruddh Vashisth
UG Student, Institute of Technology, Banaras Hindu University,Varanasi
Vineet Singla
UG Student, Institute of Technology, Banaras Hindu University,Varanasi
ABSTRACT: The paper investigates the strength of domes which could be folded and the area coverage is
reduced to one-sixth of the original dome structure. Also an attempt has been made to test the durability of the
structure. The major advantage of such housings is listed, such as better dispersion of moments created due to
loads applied on the structure, along the surface of the structure due the curved surface area. It also enumerates
the various problems faced by such types of housings, as it is an unconventional method of housing. Geometry
of one structure was specified and analysis of Finite Elements gave the deflection of the structure, which
suggests that a dome is able to bear much greater loads as compared to conventional shelter techniques.
Keywords: domes, Finite element method, hoop, foldable dome, mesh
INTRODUCTION
Civil engineering does not restrict itself to c) Better Space Utilization
construction of buildings with the stereotypical Bearing walls and columns are not required even in
square or rectangular structures. It also includes extremely large domes.
unconventional structures such as domes, which are d) Lifespan
buildings with hemispherical design, usually having a With the exterior protection of polyurethane
circular, square or octagonal base. insulation and the waterproof roof membrane, the
Domes steel reinforced concrete dome will not weather or
The dome is basically an engineered system of decay. The life span of these buildings will be
triangular space frames that create self-reinforcing measured in centuries.
roofs and wall units. These units are based on e) Strength
mathematically precise divisions of the sphere and The monolithic concrete dome is probably the
each unit may be manufactured to exact specs in a strongest building that can be built with a specific
factory and assembled on site by simply bolting them amount of time and materials. This means it has
together. This structure reduces the quantity of very high resistance to the natural forces of
building materials needed per square foot of usable earthquakes, hurricanes, tornadoes, and floods
area by about 30% over conventional construction f) Fireproof
(deduced by simple mathematical calculation). In another case in California, fire-fighters took
Throughout history, the dome has been the refuge in a dome home to escape a ravenous forest
architectural form of choice wherever efficiency and fire surrounding it. The fire caused only minimal
strength are required of a structure. From the simple damage to the home. In one industrial example in
igloo that shelters the Arctic hunter through the Channelview, Texas, an electrical fire caused 300
ravages of a blinding storm, to the awe-inspiring gallons of transformer oil to ignite and burn three
magnificence of the Sistine Chapel, the dome has wood-framed structures while the Monolithic
been used in every culture, on every continent, as one Dome storage unit remained intact, preserving the
of man's most versatile constructions. materials inside.
g) Low Maintenance
Advantages of Domes A dome is also significantly easier to maintain than
a) High Energy Efficiency a regular building. There is no worry about roof
With a totally seamless layer of polyurethane repairs, wood rot, termite damage or any other sort
insulation sandwiched outside of the concrete and of inconvenient maintenance required in a
under the roof membrane, the dome easily regulates conventional structure. Suspend lighting and other
its internal temperature. A 50% savings of energy can important features from the thin shell.
easily be attained over a conventional building. h) Lower Construction Costs
b) Rapid Construction Domes are often more economical when compared
The dome is constructed inside the enclosed airform, to large conventional free span structures.
enabling construction to progress rapidly regardless
of weather.

loading points and supports, must be carefully
Problems faced by Dome structures analyzed with the use of a sufficiently refined
mesh.
 The shape of a dome house makes it difficult to
conform to code requirements for placement of sewer 2. A BRIEF HISTORY OF THE FEM
vents and chimneys. Although the label finite element method first
 Air stratification and moisture distribution within appeared in 1960, when it was used by Clough
a dome are unusual, and these conditions tend to (Clough; 1990) in a paper on plane elasticity
quickly degrade wooden framing or interior paneling. problems, the ideas of finite element analysis date
 Privacy is difficult to guarantee because a dome back much further. In fact, the questions, who
is difficult to partition satisfactorily. originated the finite element method and when did
 Sounds, smells, and even reflected light tend to it begin, have three different answers depending on
be conveyed through the entire structure. whether one asks an applied mathematician, a
 As with any curved shape, the dome produces physicist, or an engineer. All of these specialists
wall areas that can be difficult to use and leaves some have some justification for claiming the finite
peripheral floor area with restricted use due to lack of element method as their own, because each
headroom. developed the essential ideas independently at
 Circular plan shapes lack the simple modularity different times and for different reasons.
provided by rectangles.
The first efforts to use piecewise continuous
 Furnishers and fitters usually design with flat
functions defined over triangular domains appear
surfaces in mind, and so placing standard beds results
in the applied mathematics literature with the work
in a crescent behind the sofa being wasted.
of Courant (Courant; 1943) in 1943. Courant used
 Wall areas are difficult to use due to the shape.
an assemblage of triangular elements and the
 Banks, insurance companies and building
principle of minimum potential energy to study the
departments may not be familiar with this type of
St. Venant torsion problem. In 1959 Greenstadt
construction
(Greenstadt;1959), motivated by a discussion in the
 Dome design may not be compatible with book by Morse and Feshback (Morse and
surrounding homes Feshback;1953) , outlined a discretization approach
involving “cells” instead of points; that is, he
INTRODUCTION TO FINITE ELEMENT imagined the solution domain to be divided into a
METHOD set of contiguous sub-domains. In his theory he
The finite element method (FEM) is a numerical describes a procedure for representing the
technique, well suited to digital computers, which can unknown function by a series of functions, each
be applied to solve problems in solid mechanics, fluid associated with one cell. After assigning
mechanics, heat transfer and vibrations. The approximating functions and evaluating the
procedure to solve problems in each of these fields is appropriate variational principle to each cell, he
similar; however our area of interest is its application uses continuity requirements to tie together the
to solid mechanics problems. In finite element equations for all the cells. By this means he
models the domain (the solid in solid mechanics reduces a continuous problem to a discrete one.
problems) is divided into a finite number of elements. Greenstadt’s theory allows for irregularly shaped
These elements are connected at points called nodes. cell meshes and contains many of the essential and
Displacements in each element are directly related to fundamental ideas that serve as the mathematical
the nodal displacements. The nodal displacements are basis for the finite element method as we know it
then related to the strains and the stresses in the today.
elements. The finite element method tries to choose .
the nodal displacements so that the stresses are in While the mathematicians were developing and
equilibrium (approximately) with the applied loads. using finite element concepts, the physicists were
The nodal displacements must also be consistent with also busy with similar ideas. The work of Prager
any constraints on the motion of the structure. and Synge (Prager and Synge; 1947) leading to the
development of the hypercircle method is a key
The finite element method converts the conditions of example. As a concept in function space, the
equilibrium into a set of linear algebraic equations for hypercircle method was originally developed in
the nodal displacements. Once the equations are connection with classical elasticity theory to give
solved, one can find the actual strains and stresses in its minimum principles a geometric interpretation.
all the elements. By breaking the structure into a Outgrowths of the hypercircle method (such as the
larger number of smaller elements, the stresses one suggested by Synge (Synge; 1952)) can be
become closer to achieving equilibrium with the applied to the solution of continuum problems in
applied loads. The accuracy of the obtained solution much the same way as finite element techniques
is usually a function of the mesh resolution. Any can be applied.
regions of highly concentrated stress, such as around

In the 1930, when a structural engineer encountered a the finite element method. During the late 1960s
truss problem such as the one shown in Figure a, he and early 1970s (while mathematicians were
immediately knew how to solve for component working on establishing errors, bounds, and
stresses and deflections as well as the overall strength convergence criteria for finite element
of the unit. First, he would recognize that the truss approximations) engineers and other practitioners
was simply an assembly of rods whose force– of the finite element method were also studying
deflection characteristics he knew well. Then he similar concepts for various problems in the area of
would combine these individual characteristics solid mechanics.
according to the laws of equilibrium and solve the
resulting system of equations for the unknown forces ANALYSIS BY A PART BY FEM
and deflections for the overall system.
Fig. 1 Truss structure Fig. 2 Geometry of the Structure

This procedure worked well whenever the structure
in question had a finite number of interconnection There are a number of steps in the solution
points, but then the following question arose: What procedure using finite element methods. ANSYS
can we do when we encounter an elastic continuum has been used to carry out the analysis. It requires
structure such as a plate that has an infinite number the user to go through the following steps for
of interconnection points? For example, in Figure b, analysis of the structure by FEM.
if a plate replaces the truss, the problem becomes
considerably more difficult. Intuitively, Hrenikoff a) Specifying Geometry - First the geometry of
(Hrenikoff; 1941) reasoned that this difficulty could the structure to be analyzed is defined. This
be overcome by assuming the continuum structure to can be done either by entering the geometric
be divided into elements or structural sections information in the ANSYS or by importing the
(beams) interconnected at only a finite number of model from a solid modeller like
node points. Under this assumption the problem Pro/ENGINEER.
reduces to that of a conventional structure, which b) Folding Mechanism- Each hoop is made up of
could be handled by the old methods. Attempts to eight sides. Each side is further divided into
apply Hrenikoff’s “framework method” were three more parts such that each hoop could be
successful, and thus the seed to finite element folded to form a shorter octagon. The smaller
techniques began to germinate in the engineering octagon formed after the fold has a side that is
community. Shortly after various scientists began to less than half the length of the original octagon
publish a series of papers extensively covering linear The specifications of the geometry are done in
structural analysis and efficient solution techniques Fig 3. And Fig 4
well suited to automatic digital computation.
The actual solution of plane stress problems by

means of triangular elements whose properties were
determined from the equations of elasticity theory
was first given in the 1956 paper of Turner, Clough,
Martin, and Topp (Turner et. al; 1956). These
investigators were the first to introduce what is now
known as the direct stiffness method for determining
finite element properties. Their studies, along with
the advent of the digital computer at that time,
opened the way to the solution of complex plane
elasticity problems. After further treatment of the
plane elasticity problem by Clough (Synge; 1952) in
1960, engineers began to recognize the efficacy of Fig 3. Geometric Specifications- Top View

defining the types of elements into which the
structure will be broken, as well as specifying
how the structure will be subdivided into
elements (how it will be meshed). This
subdivision into elements can either be input
by the user or, with some finite element
programs can be chosen automatically by the
computer based on the geometry of the
structure (this is called auto meshing).
Fig 4. Geometric Specifications- Side View
c) .
Fig 5. (a) Figure shows the link between two ends of

the tube welded to a hinge (b) show the section Fig 6. Meshed Structure- Top View
view of the tube.
Fig 6. Figure shows the connection at an inclined

junction where tube is welded to hinge on both
sides. Fig 7. Meshed structure Side View
d) Specify Element Type and Material Properties - f) Apply Boundary Conditions and External
Next, the material properties are defined. In an Loads - Next, the boundary conditions (e.g.
elastic analysis of an isotropic solid these consist location of supports) and the external loads are
of the Young's modulus and the Poisson's ratio specified.
of the material.
Live load of 3kN/m of UDL
Material Properties Dust load of 1kN/m varying with the slope of
Ay=Az= 3.5×10-5 m2 periphery of the dome.
Iy=Iz=6.1017mm4 Wind load of 50m/s is applied( with context to
Depth of the Tube=0.025m local conditions)
Width of the tube=0.025m
Elasticity=205 kN/mm2 g) Generate a Solution - Then the solution is
Density=7833.41kg/m3 generated based on the previously input
Poisson ratio=0.3 parameters.
e) Mesh the Object - Then, the structure is broken

(or meshed) into small elements. This involves

h) Postprocessing - Based on the initial conditions bending moment is negligible throughout the
and applied loads, data is returned after a length of the structure. Therefore much thickness
solution is processed. This data can be viewed in of the outer surface is not required.
a variety of graphs and displays. As per 2001 census report the slum population of
India in cities and towns with a population of
i) Refine the Mesh - Finite element methods are 50,000 and above was 42.6 million. The total urban
approximate methods and, in general, the homeless population is 7, 78,599 people. It could
accuracy of the approximation increases with the be conclued that
number of elements used. The number of 1.) This type of houses can be useful in slum areas
elements needed for an accurate model depends and also for homeless population of our country.
on the problem and the specific results to be 2.) It can prove a great alternative for small houses
extracted from it. Thus, in order to judge the which would be as safe and stable as the normal
accuracy of results from a single finite element houses but would be very cost effective.
run, you need to increase the number of elements Therefore, people with low budgets could have
in the object and see if or how the results change. their own houses. It would surely increase the
standard of living of the whole society.
j) Interpreting Results - This step is perhaps the 3.) Government can also consider it as a good
most critical step in the entire analysis because it remedial option for providing living space and
requires the fundamental knowledge of also to relocate victims of any kind of disasters.
mechanics to interpret and understand the output 4.) It can also prove benefitial for army as they can
of the model. This is critical for applying correct carry it along their campaign.
results to solve real engineering problems and in
identifying when modelling mistakes have been Such structure could be benefical in various
made. spheres of life and prove to be very durable as well
as economical at the same time.
RESULTS AND CONCLUSION
Height of the model increases in the Y axis and is REFERENCES
spread over the X-Z plane. The results have been Hrenikoff, “Solution of Problems in Elasticity by
represented in form of graph for deflection in Figure the Framework Method,” J. Appl. Mech., Vol.
8 and Figure 9 8, 1941, pp. 169–175.
J. Greenstadt, “On the Reduction of Continuous
Problems to Discrete Form,” IBM J. Res. Dev.,
Vol. 3, 1959, pp. 355–363.
J. L. Synge, “Triangulation in the Hypercircle
Method for Plane Problems,” Proc. R. Irish
Acad., Vol. 54A21, 1952.
M. J. Turner, R. W. Clough, H. C. Martin, and L.
C. Topp, “Stiffness and Deflection Analysis of
Complex Structures,” J. Aeronaut. Sci., Vol.
23, No. 9, 1956, pp. 805–823, 854
P. M. Morse and H. Feshback, Methods of
Theoretical Physics, McGraw-Hill, New York,
1953, Section 9.4.
R. Courant, “Variational Methods for the Solutions
Fig 8. Graph of deflection vs. distance from fixed end of Problems of Equilibrium and Vibrations,”
Bull. Am. Math. Soc., Vol. 49, 1943, pp. 1–23.
R. W. Clough, “The Finite Element Method in
Plane Stress Analysis,” Proceedings of 2nd
ASCE Conference on Electronic Computation,
Pittsburgh, PA, Septembern8–9, 1960.
W. Prager and J. L. Synge, “Approximation in
Elasticity Based on the Concept of Function
Space,” Q. Appl. Math., Vol. 5, 1947, pp. 241–
269.
Figure 9. Graph Resulatnt Deflection vs. Distance

from Fixed End
Analysis shows that structure is safe and stable under

given loading conditions. As it is a curved surface,

ISBN 978-81-921121-0-7
GROUNDWATER PROSPECTING OF URBAN AREA OF THE

JABALPUR DISTRICT USING REMOTE SENSING AND GIS
Medha Jha
Assistant Professor, Institute of Technology, Banaras Hindu University, Varanasi-221005U.P.India .jha_medha@rediffmail.com
Sanjay Tignath
Professor,Govt. Autonomous Science College, Jabalpur- 482001, M. P., India. tignathsanjay@rediffmail.com
ABSTRACT: The remote sensing data combined with Geographical Information System (GIS) technique has
proved to be very efficient in assessing the groundwater potential of any area. In the present paper, IRS 1C, LISS III
data and data derived from the toposheet have been used to identify the groundwater potential zones by integrating
various thematic maps generated on 1: 50,000 scale. These maps were integrated using ILWIS 3.4. Depending upon
their characteristics and the groundwater potential zones, the urban areas of the Jabalpur district, Madhya Pradesh
India were demarcated. The area of investigation has been classified into three categories of groundwater
potentiality. The present results show that integration of all attributes provides more accurate results in identification
of groundwater potential zones.
INTRODUCTION
The remote sensing and GIS tools have opened new early October. Winters start in early November and last
paths in water resources studies. Temporal data from until early March.
remote sensing enables identification of groundwater
aquifers whereas geographical information system (GIS)
enables integration of multi-thematic data. The concept
of integrated remote sensing and GIS has proved to be
an efficient tool in groundwater studies (CGWB, 1985 ;
Moore, G., & Waltz, F.A., 1986; K.S.R. Murthy, 2000;
Krishnamurthy et al., 1996; Saraf & Chaudhary, l998;
Singh A. K. et al, 2000; Erhan Sener et al, 2011; Khan
and Mohrana, 2002). Keeping this in view, an attempt
is made to identify groundwater potential using
integrated approach of remote sensing and GIS
techniques in urban areas of Jabalpur, district, M.P.
India .
STUDY AREA
Jabalpur is a city in the State of Madhya Pradesh in
India. It is located at 23°10′N 79°57′E / 23°17′N
79.95′E (Fig 1). It has an average elevation of 411
meters (1348 feet).The city is located in the
Mahakoshal region in the geographic centre of India
and is one of the largest cities of Madhya Pradesh. Fig 1: Map of the Study Area
Jabalpur was the 27th largest urban conglomeration in
India in the year 2001 (2001 Census). On a global scale, MATERIALS AND METHODS
Jabalpur was the 325th largest city or urban area in the The present work aims at studying the groundwater
world in the year 2006 and is estimated to be the 294th potential zones of the urban areas of Jabalpur. The three
largest city in the world by the year 2020. Jabalpur types of data sets have been used for the study:
stands 121st in terms of the fastest growing cities and Remotely sensed data, viz. IRS 1C LISS III of scale
urban areas in the world in 2006. As of 2001 India 1:50,000. The survey of India toposheet no. 55 M / 16
census, Jabalpur and its continuous suburbs had a scale 1:50,000. Field data of geology, geomorphology,
population of 1276853. Jabalpur has a sub-tropical and depth to water level data were used. The study has
climate, typical of North- Central India. Summers start been made in the GIS environment to generate
in late March and last up to early June. May is the important maps and data using ILWIS 3.4 software.
hottest month with an average temperature of 41.9 0C. The base map was prepared using the Survey of India
They are followed by the monsoon season (average (SOI) toposheet (scale, 1:50000) no. 55 M / 16.
precipitation of nearly 1200 mm.) which lasts up to

In order to demarcate the groundwater potential zones using remote sensing and Geographical Information
of study area different thematic maps on 1:50,000 scale System for demarcation of different ground water
were prepared from remote sensing data, topographic potential zones, especially in diverse geological setup.
maps, depth to the water level data. The thematic maps This gives more realistic groundwater potential map of
on geology and geomorphology were prepared using an area, which may be used for any groundwater
IRS 1C LISS-III data by visual interpretation on development and management programme.
1:50,000 scale. Depth to water level map was prepared
through GIS.These thematic layers were integrated in
GIS environment. The integrated layer was grouped
into different groundwater potential zones and the final
groundwater potential zone map was thus generated.
ANALYSIS AND DISCUSSION

The area represents rocks of nearly all geological eras.
The general litho-stratigraphic sequence of the area
is(Fig 2):Alluviums (Newer and Older of Holocene and
late Pleistocene time) ,Traps - Basalts with Inter-
trappean formations (Cretaceous-Tertiary
boundary),Lameta Formations-Green sands, Lower
lime stone, Clays, Mottled nodular lime stone (Jurassic-
cretaceous),Gondwana Formation - Sandstone, Shales Fig. 2: Showing the Geology of the Area
and clays (Triassic),Madan mahal Granites
(Proterozoic),Mahakoshal Metamorphics - Schists,
Slates and Dolomiticmarbles (Paleao-
proterozoic).Geomorphological map was prepared from
remotely sensed data(Fig 3). The geomorphology of the
area shows Granitic torred hills, Granitic gentle
undulations, Gondwana lowland, Gondwana flatland,
alluvial plains, Alluvial rolling lands and Regolith
passage.
Groundwater is developed in the study area through
dugwell, dug-cum borewells, borewells to meet
domestic and irrigation needs. Depth to water level map
has been prepared by using water level data of the
dugwells (Fig 4). The map shows that depth to water
level is quite variable, ranging between 0 to 2.5m,
2.5m to 5m and >5m. Fig. 3: Map Showing the Geomorphology of the Area
Integration of thematic layer through GIS
Each one of the classes in thematic layers was
qualitatively placed into one of the following categories,
viz. (i) excellent(ii) good, (iii) moderate and(iv) poor,
depending on their ground water potential level. After
understanding their behavior with respect to
groundwater control, the different classes were
integrated in GIS environment and the area was
classified into excellent, good, and moderate ground
water prospecting zones. (Table 1).
CONCLUSION
In order to delineate the groundwater potential zones,
in general, different thematic layers viz: geology,
geomorphology, depth to water table have been
integrated through GIS technique. This provided a
broad idea about the groundwater prospect of the area. Fig 4: Map Showing Depth to Water table
The above study has demonstrated the capabilities of

Erhan Sener, Aysen Davraz and Mehmet
Ozcelik(2011),An integration of GIS and remote
sensing in groundwater investigations: A case
study in Burdur, Turkey ,Hydrogeology Jour.Vol.
13, Numbers 5-6, 826-834,
Khan, M.A. and Moharana, P.C. (2002). Use of remote
sensing and GIS in the delineation and
characterization of groundwater prospect zones,
Jour. Indian Soc. Remote Sensing, 30(3):131-141.
Krishnamurthy, J. Venkataesa Kumar, N, Jayraman, V.
and Manivel, M. 1996 : An approach to demarcate
groundwater potential zones through Remote
Sensing and GIS. International Journal of Remote
Sensing, 17, No. 10, 1867-1884.
Moore, G., and Waltz, F.A. 1986, Objective procedure
Fig 5: Groundwater Prospect Map of the Area for lineament enhancement and extraction,
photogrammetric Engineering and Remote Sensing,
Table 1: Table showing Groundwater Prospecting 49, 641-647.
categories Murthy K.S.R., 2000, Groundwater potential in a semi-
Category of arid region of Andhra Pradesh : A geographical
Depth to Ground Index of Information System approach, International
Geomorphology
water table water Map
prospect journal of Remote Sensing, Vol. 21 No. 9, 1867-
0-2.5 Granitic torred hills Excellent GTH 1884.
Granitic gentle Saraf, A.K. and Chaudhary, P.R., 1998: Integrated
0-2.5 Excellent GTU
undulations remote sensing and GIS for groundwater
0-2.5 Gondwana lowland Excellent GL exploration and identification of artificial recharges
0-2.5 Gondwana flatland Excellent GF sites, International Journal of Remote Sensing, Vol.
19, No. 10, 1825-1841.
0-2.5 Alluvial plains Excellent AP
Singh A.K, (2000). Delineation of groundwater
Alluvial rolling
0-2.5
lands
Excellent ARL potential zonesin Bakhar subwatershed Mirzapur &
0-2.5 Regolith passage Excellent RP Sonebhadra districts, U.P., using integrated studies
of Remote Sensing, Geoelectrical and GIS
2.5-5 Granitic torred hills Good GTH
techniques, Proc. Od ISRS National symp. held at
Granitic gentle
2.5-5
undulations
Good GTU Kanpur from Nov. 21-22, 2000,pp. 320-329.
2.5-5 Gondwana lowland Good GL
2.5-5 Gondwana flatland Good GF
2.5-5 Alluvial plains Good AP

Alluvial rolling
2.5-5 Good ARL
lands
2.5-5 Lameta hills Good LH
2.5-5 Low ridge Good LR
>5 Gondwana lowland Moderate GL
>5 Gondwana flatland Moderate GF
>5 Alluvial plains Moderate AP
>5 Lameta hills Moderate LH

Basalt capped
>5 Moderate BCLH
Lameta highland
>5 Saddle Moderate S
REFERENCES
Central Ground Water Board, (CGWB), (1985), Report
on hydrogeology and groundwater potential of
Mirzapur district U.P.

ISBN 978-81-921121-0-7
PARAMETRIC STUDY OF PREDICTION OF SCOUR DEPTH IN

SPILLWAYS - A NEURAL NETWORK APPROACH
Arun Goel
Associate Professor, Civil Engg. Dept., N.I.T. Kurukshetra-.136119. Haryana. E mail drarun_goel @yahoo.co.in.
ABSTRACT: The paper presents prediction of maximum depth of scour downstream of a spillway with linear
regression and back propagation neural network techniques based upon different combinations of input
parameters. The study shows that BP feed forward NN performed better than linear regression when more
number of individual parameters as inputs is considered. Moreover, the performance of model also gets affected
by the type of input parameter to a great extent. The results are also compared with different models with an aim
to study relative importance of parameter on the performance of the BPNN modeling and linear regression
based on root-mean square error and correlation coefficient performance criteria. Furthermore, the BPANN
exhibits inherent advantages leading to a unique optimal and global solution compared to the conventional linear
regression models.
INTRODUCTION based modeling and linear regression simultaneously

Prediction of maximum depth of scour downstream of on the same data set by considering relative
spillways is important for proper planning, design and significance of each individual parameter. The present
construction of any hydraulic structures. Scouring is a study aims to explore utility of the back propagation
complex and dynamic phenomenon affected by many based ANN and linear regression for the maximum
parameters which are often interrelated and difficult to scour hole modeling and comparing its relative
understand because flow in open channel is turbulent; performance with different combinations of input
geometry is irregular and varies with time (Dey & parameters of this data set.
Sarkar 2006, Mason & Arumugam 1985). There are
various factors governing the depth of scour namely ARTIFICIAL NEURAL NETWORKS
discharge intensity q, height of fall H1, bucket radius A neural network is an artificial intelligence technique
R, bucket lip angle, Ø, type of rock, degree of rock that mimics a function of the human brain. Neural
homogeneity, run time and mode of operation of networks are general-purpose computing tools that
spillway etc (Fig.1). can solve complex non-linear problems in the field of
pattern recognition, classification, speech, vision and
control systems. The network comprises a large
number of simple processing elements linked to each
other by weighted connections according to a
specified architecture. A neuron consists of multiple
inputs and a single output. The number of neurons in
the input and output layers are fixed by the problem
being modeled as the number of input variables equals
number of input neurons and number of output
variables equal number of output neurons. The
determination of optimal number of hidden layers and
The literature review indicates that a regression hidden neurons is usually cumbersome, as no general
mathematical model for predicting maximum depth of methodology is available for their determination.
scour is not readily available using different These networks learn from the training data by
combinations of flow, material and fluid parameters. adjusting the connection weights. There is a range of
The researchers have mainly relied on the artificial neural network architectures designed and
conventional experimental approach to study the used in various fields of hydrology and hydraulics. In
scouring by using physical modeling. Recently this study, a feed-forward neural network with back
(Azmathullah et al. 2005, 2006, 2008a) have applied propagation learning algorithm is applied. The basic
soft computing modeling (ANN) for the prediction of element of a back-propagation neural network is
scour parameters downstream of ski jump type processing node and structure of commonly used back
spillway successfully. Many investigators have propagation neural network (Fig.2). A three layer feed
applied soft computing techniques namely Genetic forward ANN has been shown in Fig.2, which
algorithm, ANFIS, SVM and M5 tree (Azmathullah et consists of three layers known as input, hidden an
al. 2008b, Azmathullah et al. 2009, Goyal & Ojha C output layers. Input layer neuron are called as x1, x2,
2011) on prediction of scour downstream of x3; hidden layers neurons are h1, h2, h3 and output
spillways. However, literature review (Goel 2008, layers neurons are O1,O2,O3. A neuron consists of
2009, 2011) indicates that no one has attempted ANN multiple inputs and a single output. The sum of inputs

and their weights lead to a summation function. The spillway. The correlation coefficient (CC) and RMSE
output of a neuron is decided by an activation values are used as shown in equation (1) and (2)
function, which can be step, sigmoid, threshold and mainly for the performance evaluation of models and
linear etc. comparison of the results for prediction of scour using
Each processing node behaves like a biological BPANN and the linear regression. The higher value of
neuron and performs two functions. First, it sums the a correlation coefficient and lower value of RMSE
values of its inputs. This sum is then passed through mean a better performance of the model. To study the
an activation function to generate an output. Any scatter around the line of perfect agreement (i.e. a line
differentiable function can be used as activation at 45 degrees) was plotted for the data set.
function. All the processing nodes are arranged into
layers, each fully interconnected to the following Error Measure
layer. There is no interconnection between the nodes Following error measure criteria has been used in the
of the same layer. In a back propagation neural present study in order to compare the performance of
network, generally, there is an input layer that acts as different modelling techniques.
a distribution structure for the data being presented to
the network. This layer is not used for any type of 1. Correlation coefficient (r)
processing. After this layer, one or more processing
layers follow, called the hidden layers. The final r
 xy
2 2
processing layer is called the output layer in a
network. This process is repeated until the error rate is
x y
minimized or reaches to an acceptable level, or until a (1)
specified number of iterations have been Where x = X - X’, y = Y - Y’ where X = observed
accomplished. All the interconnections between each values; X’ = mean of X, Y = predicted values, Y’ =
node have an associated weight. The values of the mean of Y.
interconnecting weights are not set by the analyst but 2. Root mean square error
0 .5
are determined by the network during the training   X  Y  2 
process, starting with randomly assigned initial RMSE   
weights. There are a number of algorithms that can be  n 
used to adjust the interconnecting weights to achieve
minimal overall training error in multi-layer networks. (2)
The generalized delta rule, or back-propagation is one
of the most commonly used methods (Rumelhart et al MATERIAL AND METHODS USED
1996). Neural networks are now being commonly applied to
the problems related to the hydraulics and hydrologic
modeling (ASCE 2000a, 2000b). The data set
(Azmathullah et al. 2005, 2006, 2008a) used in the
present study were from the laboratory and the field
for an ogee spillway. The data set comprises of a total
of 95 runs on different types of spillway models.
A neural network based modelling approach also
requires setting up several user-defined parameters
like learning rate, momentum, optimal number of
nodes in the hidden layer, number of iterations and the
number of hidden layers, so as to have a less complex
network with a relatively better generalization
capability. Further, training a neural network requires
a number of iterations and a large number of training
iterations may force ANN to over train, which may
affect the predicting capabilities of the model.
The cross-validation is a method of estimating the
accuracy of a classification or regression model. The
Fig. 2 Three layer feed forward neural network input data set is divided into several parts (a number
defined by the user), with each part in turn used to test
PERFORMANCE COMPARISON CRITERIA a model fitted to the remaining parts. In this study, the
data sets of the laboratory and field data were used for
The data sets mentioned in a study (Azmathullah et al. both creating and testing the models by BPANN and
2005, 2006, 2008) are used in the present study for the linear regression techniques. Graphs have been
model building and validation to assess the potential plotted for difference in actual and predicted values of
of MLP based ANN modeling and linear regression in the scour results as shown in Figure 3 to 7 and two
predicting the scour parameters downstream of +10% and -10% error line are also plotted on the

diagrams. The values of correlation coefficients and 0.0534),(r =,0.8865, rmse = 0.0459) and (r = 0.9302,
RMSE obtained for different models are shown in rmse = 0.0369) respectively.
Table 1.
Table 1. Values of important parameters of BP ANN Table 2.Comparison of results
S No Model Type of CC Root mean
details technique (r) squared
S.No Type of error
input BPANN(MLP) (RMSE)
parameter
Model 1 ds, ws, ls, ANN(MLP) 0.6798 0.0788
H1, R, Ø
Hidden Learning without q
Momentum Iterations
nodes rate ds, ws, ls, Linear 0.6574 0.0749
H1, R, Ø regression
1 ds 8 0.2 0.3 500
without q
2 ws 8 0.2 0.3 500 q, R, Ø
without
3 ls 8 0.2 0.3 500 H1
ds, ws, ls, Linear 0.6594 0.0739
4 q 8 0.2 0.3 500 q, R, Ø regression
without
H1
5 H1 8 0.2 0.3 500
q, H1, Ø
6 R 8 0.2 0.3 500 without R
ds, ws, ls, Linear 0.7140 0.0689
7 Ø 8 0.2 0.3 500 q, H1, Ø regression
without R
RESULTS AND DISCUSSIONS q, H1, R,
with out Ø
The first set of analysis was carried out by using
ds, ws, ls, Linear 0.6757 0.0729
BPANN with data for predicting the maximum depth q, H1, R, regression
of scour on downstream of spillway. Seven input with out Ø
parameters namely ds, ws, ls, H1, R, Ø and q were Model 5 ds, ws, ls, ANN(MLP) 0.9302 0.0369
considered from the data sets provided by the studies q, H1, R,
carried out by (Azmathullah et al. 2005, 2006, 2008a) Ø all
to predict the maximum depth of scour (ds).In the ds, ws, ls, Linear 0.8407 0.0528
beginning, a model named a Model 1 consisting of q, H1, R, regression
six parameters namely ds, ws, ls, H1, R, Ø without q Ø all
were considered. A number of trials were carried out
to reach at the various user-defined parameters
required for the back propagation ANN based The same data set was utilized for predicting then
algorithms using WEKA software. The details of user scour by using liner regression modeling technique.
defined parameters are mentioned in Table 1 for this Five models namely Model 1 to 5 were tested in
combination of input parameters keeping the similar manner as in case of BP ANN by using
performance parameters in view. The obtained values WEKA software. The results of r and rmse in models
of correlation coefficients and rmse for the best ANN 1, model 2, model 3, model 4 and model 5 are ( r =
model are for the given data set of Model is also given 0.6574, rmse = 0.0749), (r = 0.6594, rmse =
in Table 2. Measured versus calculated values of the 0.0739), (r = 0. 7140, rmse = 0.0689.), (r =,0.6757,
maximum scour depth by BPANN are shown as a rmse = 0.0729.) and (r = 0.8407., rmse = 0.0528.)
scatter plots (Fig.3). respectively.
Similarly a different combinations of input parameters The perusal of above exercise indicated that the
such as Model 2 ds, ws, ls, q, R, Ø without H1, Model values of correlation coefficient (r) for ANN
3 ds, ws, ls, q, H1, Ø without R, Model 4 ds, ws, ls, q, regression modeling varied form 0.6798 to 0.9302,
H1, R, with out Ø and Model 5 ds, ws, ls, q, H1, R, Ø while for linear regression 0.6574 to 0.8407. It is
were considered and tested as above in Model 1 and maximum in case of back propagation ANN approach
the results obtained are plotted in Fig 4,5,6 and 7 (= 0.9302) and minimum in case of linear regression
respectively. The values of performance parameters (= 0.8407) as shown in the Table 2 for the best model
namely r and rmse in model 1, model 2 model 3, 5 when all parameters are considered as input and
model 4 and model 5 are r = 0.6798, rmse = 0.0788), performance of the BPANN is relatively superior than
(r = 0.7760, rmse = 0.0662), (r = 0.8737, rmse = linear regression. It is evident from Figures 3 to 7 that
more number of points are lying on the 45o line when

all the input parameters are used as input in the model Fig.3 Variation of actual maximum scour depth with
5 (Fig.7) as compared to model 1 (Fig.3) in both the predicted maximum scour depth by BPNN without q
modeling techniques i.e. BPANN and the linear
regression. While comparing the performance of
0.6
model 5 and 6 both are performing the same as BPNN
indicated by the r values. It means that the bucket +10% line
0.5
radius and its angle are not affecting the scour
Best fit line
Predicted max depth of scour without H1

downstream of the spillway. It means that the factor q
which is discharge per unit width on the spillway is 0.4
-10% line
affecting the performance of the model in both the
modeling techniques. The study has revealed the 0.3
relative importance of one parameter over

performance of the BPANN and linear regression 0.2
modeling techniques for predicting maximum scour

depth downstream of the spillway. 0.1
CONCLUSIONS 0
0 0.1 0.2 0.3 0.4 0.5 0.6
This study was carried out to judge the relative Actual max depth of scour w ithout H1
importance of input parameters on the performance of
Back propagation based ANN and the linear
regression modeling methods for the prediction of Fig.4 Variation of actual maximum scour depth with
maximum depth of scour downstream of an ogee predicted maximum scour depth by BPNN without
spillway. Based upon the present study, the following H1
conclusions are drawn
1. A conclusion from this study is that the
performance of the Back propagation based ANN and 0.6
the linear regression techniques are affected when any Regression

+10% line
BPNN
one of the input parameters is not considered as 0.5
Best fit line
Predicted depth of max scour without R
indicated by performance parameters.

2. The discharge per unit width of the spillway is the 0.4 -10% line
most significant parameter affecting the BPANN and
the linear regression technique as compared to 0.3
downstream parameters like bucket angle and its

radius of spillway. 0.2
3. The prediction capability of BPANN is better than

linear regression on the same data set under similar 0.1
conditions. The input parameters have significant
effect on the accuracy of the modeling technique. 0
4. The findings of this study would assist in selecting 0 0.1 0.2 0.3 0.4 0.5 0.6
most significant and a suitable parameter may be Actual depth of max scour w ithout R
chosen for estimation of the maximum scour

downstream of spillway to make it cost effective.
Fig.5 Variation of actual maximum scour depth with
predicted maximum scour depth by BPNN without R
0.6 +10% line
Regression
Best fit line
BPNN
0.5
Predicted max depth of scour without q
-10% line
0.4
0.3
0.2
0.1
0
0 0.1 0.2 0.3 0.4 0.5 0.6
Observed max depth of scour w ithout q

Downstream of a Ski jump Bucket. J. of Hydraulic
0.6
Engg, ASCE, Vol. 131, No. 10, pp. 898-908.
Regression
BPNN
Azamathulla H. Md., Deo, M.C. and Deolalikar,
0.5 P.B. (2006). Estimation of Scour Below Spillways
Predicted depth of max scour without phi
+10% line
Using Neural Network. Journal of Hydr Research,
0.4 Best fit line Intl Association Hydraulic Engg., Vol. 44 No. 1,
-10% line pp. 61–69.
0.3 Azamathulla, H.Md., Deo, M.C. and Deolalikar, P.B.
(2008a). Alternate Neural Networks to Estimate
0.2 the Scour below Spillways. J of Advances in Engg
Software,Vol.39, No.8, pp. 689–698.
0.1 AzmathullahHM, Ghani AAB, Zakaria NA,Hin LS,
Kiat CC, Siang LC, Hasan ZA (2008b) Genetic
0
programming to predict ski-jump bucket spillway
0 0.1 0.2 0.3 0.4 0.5 0.6 scour. J of Hydrodynamics, 20 (4):477–484.
Actual depth of max scour w ithout Phi
Azmathullah HM, Ghani AAB, Zakaria NA (2009)
ANFIS based approach for predicting maximum
scour location of spillway. Water Management ICE
London, 162 (6):399–407.
Fig.6 Variation of actual maximum scour depth with Dey, S. and Sarkar, A.(2006). Scour Downstream of
predicted maximum scour depth by BPNN without an Apron due to Submerged Horizontal Jets. J.
Phi Hydraulic Engineering, Proc.ASCE, Vol. 132, No
3, pp. 246-257.
0.6
Goel A (2008), Prediction of Maximum Scour
Downstream of Ski jump type Spillways. National
Regression +10% line Conference on Hydraulics and Water Resources,
0.5
BPNN Best fit line
MNIT Jaipur, 15-16 Dec 2008.
-10% line Goel A. (2009) Prediction of Maximum Scour
Predicted max depth of scour
0.4
Downstream of Ski jump type Spillways
using M5 model tree. Specialty National
0.3
conference on River Hydraulics, MMU Mullana.
Goel A (2011) Predicting Max Scour Depth.
0.2
International Water Power and Dam Construction,
UK, pp.42-54
0.1 Goyal, M K and Ojha C. S. P. (2011) Estimation of
Scour Downstream of a Ski-Jump Bucket Using
0 Support Vector and M5 Model Tree. Water
0 0.1 0.2 0.3 0.4 0.5 0.6
Actual max depth of scour
Resources Management, 25:2177–2195.
Mason, P. J. and Arumugam, K. (1985). Free Jet
Scour Below Dams and Flip Buckets. Journal of
Fig.7 Variation of actual maximum scour depth with Hydraulics Engineering, Proc. ASCE, Vol. 111, pp.
predicted maximum scour depth by BPNN with all 220-235.
parameters Rumelhart D.E, Hinton G.E., and Williams R.J
(1996) Learning Internal Representation by Error
ACKNOWLEDGMENTS Propagation’, In: Parallel Distributed Processing:
Author wish to heartily and sincerely acknowledge the Explorations, in the Microstructures of Cognition,
authors for data taken from their paper Azamathulla et Cambridge, MIT Press, pp 318-362.
al. (2005). Weka software version 3.4.13
http://www.cs.waikato.ac.nz/~ml/weka/.
REFERENCES
ASCE Task committee on application of ANNs in
Hydrology (2000a), Artificial neural networks in
hydrology, I: preliminary concepts, J. Hydraulic
Engineering, ASCE, 5 (2), pp.115-123.
ASCE Task committee on application of ANNs in
Hydrology (2000b), Artificial neural networks in
hydrology, II: hydrologic applications, J.
Hydraulic Engineering, ASCE, 5 (2), pp.124-137.
Azmathullah, H. Md., Deo, M. C., and Deolalikar, P.
B. (2005). Neural Networks for Estimation of Scour

ISBN 978-81-921121-0-7
URBAN INFRASTRUCTURE PLANNING USING GIS A CASE STUDY

OF SURAT
Shri Chetan R. Patel

Asst. Professor, M.Tech. (Planning), CED, SVNIT-SURAT, Gujarat (India), chetanrpatel@rediffmail.com
Nikunj B. Shah
M.Tech. Scholar (Planning), CED, SVNIT-SURAT, Gujarat (India), nik_shah71@yahoo.com
ABSTRACT: This paper describes how simple analytical GIS techniques can be used to support the planning
of social infrastructure in Surat city, India. The planning context in Surat consists of a micro and a macro level
by planning mechanism of Town Planning Schemes and Development Plan. The population and the population
density play important role in this context so it’s difficult for the urban planner to have decision for provision of
the social infrastructure. This paper illustrates how GIS-based analytical techniques can be usefully applied in
support of more strategic spatial planning of social infrastructure.
INTRODUCTION and rapid informal development in urban periphery

Cities are considered as engines of growth, within often leading to deficient poor planning of urban land
cities infrastructure provision, more specifically social use and infrastructure. In this the GIS based planning
infrastructure provision are important to people. The will assist the urban planner to deal with such
low quality or absence of these kinds of infrastructure complex system. In light of that this paper evaluates
has a direct impact on living conditions, health and the process and provision of infrastructure, especially
potential for economic development in rapidly the planning for social infrastructures in the city of
growing cities in developing countries. Creating good Surat using GIS-based analysis.
quality basic infrastructure therefore is the first and
foremost step to achieve sustainability and create STUDY AREA
healthy living conditions in our cities. Cities in Surat City is one of the rapidly developing cities in
developing countries also have to plan for urban India. The city is located in the southern part of
amenities, like physical infrastructures (road, Gujarat State. It has experienced a very rapid growth
drainage, water supply, electricity, etc) and social during the last 20 year with highest decadal growth
infrastructures (school, health care facility, etc) with rate of 47% (Census, 2001) and second highest
specific circumstances, like lack of financial support population density after Ahmadabad.
Table1: Population and Growth Rate of Major Cities of Gujarat.
Name of municipal Area in sq.km Population in Growth rate DENSITY ppha

corporation lakhs
1991 2001
Ahmadabad 190.94 35,15,361 36.44 154.02 184.20
Surat 324.16 24,33,785 62.38 134.83 216.76
Vadodara 108.26 13,06,035 32.44 952.7 120.64
Rajkot 104.86 9,66,642 53.12 810.7 92.18
Bhavnagar 90.16 5,10,958 - 44.62 56.67
Source: Surat Municipal Corporation
The spatial development of the city is controlled by redistributes the remaining land to the original
two authorities. The Surat Municipal Corporation owners, and provides for physical and social
(SMC) is responsible for the planning in the city core, infrastructure
and the Surat Urban Development Authority (SUDA)
plans for the newly developing and the peri – urban Social infrastructure provided within the TPS are
areas. Urban development is regulated in a usually lower order services, such as health care
hierarchical manner: at the macro scale by a strategic facilities, schools, neighborhood centers, parks, and
Development Plan which is revised every twenty open space. Individual planners decide upon the
years; at the micro scale by means of Town Planning quantity and quality of social in a particular TPS area.
Schemes (TPS). A TPS is a method of land Their decision is generally based upon issues such as,
development which pools non-urban land, rearranges for example, the population projection, and the
the irregular plots of land into a regular pattern, situation in the surrounding area. Therefore, the
reserves part of the land for public space and decision is subjective to the planner who plans the

land allocation for each TPS. Furthermore, multiple This approach can easily lead to a no optimal
agencies (staff of SMC, SUDA or private consultant) distribution of social infrastructure within a TPS area
are involved in developing TPS. Because different and among adjacent TPS areas. Therefore, there is a
agencies work on separate TPS, this results in TPS need to evaluate whether planned social infrastructure
being prepared in isolation with each other, with none in TPS area is sufficient in quantity and if it is
or little coordination with the surrounding TPS areas. adequately distributed in geographic space.
Fig 1 Surat city
DATA COLLECTION AND ANALYSES the health care facility of the Surat city. Following
Before evaluating and proposing the new table 2 shows the existing health care facility of the
infrastructure facility its prime important to identify Surat city.
the problem in existing condition. So at the most the
study is carried out to evaluate the present situation of
TABLE 2: HEALTH CARE FACILITY IN SURAT

Government Private Charitable Total No Of
Sr.No Zone Hospitals Hospitals Trust SMC Hospitals Beds
1 south west zone 2 17 3 3 25 2000
2 south zone 0 23 0 3 26 791
3 south east zone 0 14 0 6 20 434
4 east zone 0 43 5 6 54 1350
5 west zone 0 90 14 3 107 1984
6 north zone 0 60 4 5 69 1303
7 central zone 0 101 18 16 135 5292
TOTAL 2 348 44 42 436 13154
Above table shows the present condition of health in central zone of city and that will create location
care facility in Surat city. Most of the hospital located disparities.

Evaluation of the Social infrastructure facility
Analysis of social infrastructure is based on network
distance. While locating all health facility together
and assume that the demand is absolutely satisfied,
but with respect to the location and travel distance /
time to reach facility scenario will completely
changed. Population in area near the boundary of
SMC needs to travel large distance to reach facility.
For distance analysis in Arc GIS proximity analysis is
done which gives nearest distance to the facility.
Determines the distances from each feature in the
Input Features to the nearest features in the Near
Features, within the Search Radius. The results are
recorded in the output table. For health care facility
this will be done for government hospitals, private
hospitals, charitable hospitals and urban health
centers.
Following Figures 3 to 6 shows the allocation of the
government hospital, intermediated hospital and urban
Figure 2 Distribution of health care facility in Surat health centre. It also gives the idea for the served and
city un served area of the city by the social infrastructure
provision.
Mapping of the existing is done and presented in the
fig 2 after that the GIS based analysis of the served
and un served area is carried out in Arch GIS
software.
¯
!
!
!
!! !
! !!
!
! ! !
! !! !!! !!
!
!
!! !
! !
!
! ! ! ! !!
! !
!!
! ! !!!!! !
!! !!! ! !!
!
! !
!
! !!!! !!!
!
!! ! ! !
!! ! !!! !
! !
! !!!!
!! !! ! !
! !!!
! !
!! !!! !!!
!
! ! !! ! ! ! !
!! !!!!!
!
!
! ! ! !! !
!! !
! ! ! ! !
!! !
! !!
!
!
!! !! !
! ! !
!! !! ! ! !
! ! !!!!!! !
!!!! !!!
!
!!!!
! !!!
!
!
!
! ! ! !
! !!
! !
! ! ! !
!! ! !
!
! !!!
! ! !!
! !!!! !!
!
! ! !! !! !!! !! ! !! !
! !!
! ! !
!! ! !
!! ! !
! ! ! !
!! !
! ! !
! ! !!
! !!!! !
!!
!
! ! !
! !! !!! !
!
! !! !
!! !! !!
!
! !! !
!
!
!!
! ! ! !
! ! !
!
! !!
!
!
Legend
! PRI VATE_HOSPI TAL_SRT Ev ents
! GOVERMEN T_H OSPITAL_SR T Ev ents
! SMC_UH C_SRT Ev ents

! CH ARI TABLE_TR UST_SRT Ev ents
City _R oad_N etwork
Riv er
Hexagon_SR T
Figure 3: Location Of Different Healthcare Facilities Figure 4: Un Served Area For 2500m Distance:
Government Hospital

Figure 5: Un Served Area For 2500m Distance : Figure 6: Un Served Area For 2500m Distance: Urban
Intermediate Hospital Health Centre.
application in planning." International Journal of
For the above analysis the GIS is the best tool to deal Geographical Information Systems 9: 67-80.
with the complex urban problems and it provide Handy, S. L., Niemeier, D.A., (1997). "Measuring
answer to the urban planner in the geographical point Accessibility: An Exploration of Issues and
of view. The ability of GIS to make model and predict Alternatives." Environment and Planning A 29:
the future need is excellent and its best compared to 1175 -1194.
the other database management. Hanson, S. and M. Schwab (1987). Accessibility and
interurban travel. Environment and Planning A.
The central area is on higher side in terms of social 12: 735- 748.
infrastructure provision, but sometimes this close Ritsema van Eck, J.R., de Jong, T., (1999).
cluster which result into under utilization of the "Accessibility Analysis and Spatial Competition
facility. Further it’s not possible for the peripheral Effects in the Context of GIS-Supported Service
area to utilize the same due to far distance. Location Planning". Computers, Environment
and Urban Systems 23: 75-89.
CONCLUSION Spiekerman, K. and M. Wegener (2000). Freedom
The study demonstrates how simple GIS techniques from the tyrrany of zones: towards new GIS-
like accessibility and allocation analysis can be used based spatial models. Spatial models and GIS. A.
to better allocated and plan social infrastructure S. Fotheringham and M. Wegener. London,
facilities. It also demonstrates that such Measures can Taylor and Francis: 45-60.
be applied to improve the existing land use and land Talen E., Anselin L., (1998). “Assessing Spatial
allocation plans. The GIS based model will assist the Equity: An Evaluation of Measures of
urban planner to deal with the complex systems. accessibility to Public Playgrounds”.
Environment and Planning A 30 595-613.
REFERENCES Truelove, (1993). “Measurement of spatial equity”.
Geertman, S. and J. R. Ritsema van Eck (1995). "GIS Environment and Planning C: Government and
and models of accessibility potential: an Policy 11 19-34.

ISBN 978-81-921121-0-7
A STUDY ON LANDFILL LEACHING USING THREE DIMENSIONAL

COLUMN LEACH APPARATUS
P.J .Barman
PhD student, Department of Civil Engineering, IITGuwahati Guwahati – 781039, Email:pjb@iitg.ernet.in
S.A. Kartha
Assistant Professor, Department of Civil Engineering, IITGuwahati, Guwahati -781039, Email: kartha@iitg.ernet.in
B. Pradhan
Assistant Professor, Department of Civil Engineering, IITGuwahati, Guwahati-781039, Email:bulu@iitg.ernet.in
ABSTRACT: The focus of this study is to understand the leaching mechanisms in the landfills by
experimental investigation using column leaching apparatus. The leaching setup comprises of a three
dimensional trapezoidal apparatus (having bottom width of 90 cm, top width of 55 cm and slanted height of 60
cm). A rainfall simulator is fitted at the top. The apparatus is placed on an iron stand of height 70 cm. The
bottom of the apparatus contains a perforated net through which the leachate may percolate and get collected in
a collector device fitted just at the bottom of the net. The leachate is finally collected in a container and placed
in closed bottles. Afterwards by use of Flame photometry the concentrations of Sodium, Calcium and Potassium
are determined.Various experiments are performed by increasing the contaminated height. It has been observed
that with the increase in contaminated height, there is variation in the leached concentrations of different
elements. It was found that Calcium exhibits more fluctuations in leaching behavior as compared to Sodium and
Potassium.
Keywords: Landfill leaching, contaminated soil, Leaching behavior, Flame photometry
INTRODUCTION bedding planes. Chotpantarat et al (2011) carried out

Groundwater contamination due to various an experimental investigation to determine
anthropogenic sources has emerged in recent years as competitive sorption and transport of lead, Nickel,
a high priority topic. One of the major sources of Manganese and Zinc in lateritic soil columns. It was
pollution is leaching from the landfills. Lack of observed that the sorption capacity of lead was higher
scientific knowledge leads to improper design of in lateritic soil than that of other metals. Estevez et al
landfills in many developing countries. Leachate is (2008) carried out a review on mobility and
produced whenever rainwater gets mixed with the degradation of pesticides in soil. French et al (2001)
hazardous chemicals. Leachate may percolate through carried out an experimental investigation to determine
the vadoze zone to the underlying aquifer and pose a the transport and degradation of propyleneglycol and
serious threat to pollution of groundwater resources. potassium acetate in the unsaturated zone. Ozkaya et
Various models were developed to quantify and al (2006) carried out mathematical simulation after
describe leaching. long term monitoring of leachate components from
Kay et al (2005) carried out lysimeter experiment to two different landfill cells. Foo and Hamid (2009)
determine the leaching of veterinary antibiotics presented an overview of landfill leachate treatment
through a clay soil. They observed that the through activated carbon adsorption process. Freijer et
incorporation of the slurry which has been applied on al (1997) presented mathematical description of
soil had little effect on the mobility of the antibiotics. degradation and leaching of pesticides for a column
Yulkeselen and Alpaslan (2001) carried out a column experiment.The mathematical solutions presented by
leach test to determine the leaching of metals from soil their model allow interpretation of the degradation of
contaminated by mining activities. They observed that the compound in terms of half life and sorption
the degree of leaching of heavy metals is highly coefficient of the compound from the initial amount in
dependent on soil pH. Dousset et al(2007) carried the leachate and the amount remaining in the leachate.
column experiment for undisturbed and disturbed By all these reviews, it has been observed that
condition to investigate the transport characteristics of numerous attempts have been made by the scientific
bromide and isoproturon. Their experimental results community to study the effect of leaching behaviour in
showed faster transport of isoproturon in the soil by experimental investigation. However the
undisturbed soil due to higher preferential flow. Deiss challenging task of a model is to effectively
et al (2004) carried out a set of peat column characterize various geometric, hydraulic and soil
experiments to determine the transport potential of parameters which affect the leaching behaviour and
lead and diesel range organics (DRO) in palustrine are responsible for groundwater contamination. An
slope wetlands near Juneau. It was observed that lead attempt is made to study the effect of various field
was surprisingly mobile, both along and across the

parameters like flowrate, depth of contaminated soil changing the flowrate at the same contaminated
etc using column leach experiment. height. The variation in concentration with time for
Sodium, Calcium and Potassium are shown in Fig. 2.
EXPERIMENTAL SET UP
The experimental set up for column leach experiment
comprises of a three dimensional trapezoidal apparatus
having bottom width of 90 cm, top width of 55 cm and
slanted height of 60 cm. A rainfall simulator is fitted at
the top to pour water over the contaminated soil to
simulate a scenario similar to rainfall. The apparatus is
placed on an iron stand of height 70 cm. The
experimental set-up is shown in Fig1. The bottom of
the apparatus contains a perforated net through which
the leachate may percolate and get collected in a
collector device fitted just at the bottom of the net. The
leachate is finally collected in a container and placed
at the bottom and finally collected in closed bottles.
Fig. 2 Concentration vs time plot for Na, Ca and K at

H1/H2=6/10 and flow rate=125 ml/sec
It has been observed from Fig. 2 that concentrations of

Sodium, Calcium and Potassium increases gradually
with time. For the same time intervals Calcium
exhibits highest leached concentration whereas
Potassium shows the least leached concentration. The
patterns of increase in leached concentration obey
approximate logarithmic relationships as shown in the
Figure 2. The effect of increase in contaminated height
at the same flowrate is also studied. Contaminated
height is increased from 6cm to 8 cm keeping the
flowrate same as 125 ml/sec.The experimental
observation is plotted as shown in Fig 3
Fig. 1 Three dimensional Column leach apparatus
PROCEDURES
Contaminated soil samples are collected from the
landfill site of Guwahati city. Uncontaminated soil
samples are also taken from the nearby area.
Uncontaminated soil has been filled as a base soil in
the column over which contaminated soil has been
placed. Water is poured over the contaminated soil
from the rainfall simulator device at a predetermined
inflowrate.The inflowrate is measured using stopclock
and a measuring bottle. Water is sprinkled over
contaminated soil for 100 seconds. After sometime
the leachate comes out from the mesh at the bottom of Fig. 3 Concentration vs time plot for Na, Ca and K at
the device which is collected in a collector H1/H2=8/10 and flowrate=125 ml/sec
arrangement comprising of tin sheet with a slope
fitted below the cylindrical portion of the apparatus. It is observed that for each element (i.e Sodium,
In this way the leachate is finally collected in a Calcium and Potassium) the leached concentration
container at the bottom. increases with time which follows approximate
logarithmic relationships and shown in Fig. 3.
RESULTS AND DISCUSSION
Experiment is conducted by changing the The percent variations in concentrations for Sodium,
contaminated height at the same flowrate and then Calcium and Potassium when contaminated height

increases from 6cm to 8cm at constant flowrate of 125 The effect of increase in flowrate keeping the
ml/sec are shown in Fig.4, Fig.5 and Fig6 respectively. contaminated height constant is studied. Contaminated
height is kept constant at 8cm and the flowrate is
increased from 125ml/sec to 200ml/sec.The
experimental observation is plotted and shown in Fig.
7
Fig. 4 Percent variation in concentration of Sodium

when H1 changes from 6cm to 8 cm at flowrate of
125ml/sec
Fig .7 Concentration vs time plot for Na, Ca and K at

H1/H2=8/10 and flowrate=200 ml/sec
The percent variations in Concentration for Sodium,

Calcium and Potassium when flowrate increases from
125 ml/sec to 200ml/sec are shown in Fig.8, Fig.9 and
Fig10 respectively.
Fig. 5 Percent variation in concentration of Calcium

when H1 changes from 6cm to 8 cm at constant
flowrate of 125ml/sec
Fig. 8 Percent variation in concentration of Sodium

when flowrate changes from 125ml/sec to 200 ml/sec
at H1=8cm
Fig. 6 Percent variation in concentration of Potassium

when H1 changes from 6cm to 8 cm at constant
flowrate of 125ml/sec
It has been observed that the leached concentration of

Sodium, Calcium and Potassium increases with the
increase in contaminated height. However the
variation of the same element is not same at different
time intervals. It can be observed from the bar
diagrams that there is an average 8% increase in
leached concentration for Sodium whereas in case of
Calcium and Potassium the average variation is Fig. 9 Percent variation in concentration of Calcium
observed to be approximately 6% and 7% respectively when flowrate changes from 125ml/sec to 200 ml/sec
when the contaminated height increases from 6cm to at H1=8cm
8cm.

columns” Environmental Modeling & Software
13, 139–149
Dousset S, Thevenot M , Pot V ,Šimunek J , Andreux
F(2007)“Evaluating equilibrium and non-
equilibrium transport of bromide and isoproturon
in disturbed and undisturbed soil columns” Journal
of Contaminant Hydrology 94 , 261–276
Deiss J, Byers C, Clover D, Amore D D,Love
A,Menzies A M,Powell J,Walter T M(2004)
“Transport of lead and diesel fuel through a peat
soil near Juneau, AK: a pilot study “Journal of
Contaminant Hydrology”74 1– 18
Fig. 10 Percent variation in concentration of Chotpantarat S , Ong S.K. Sutthira C, Osathaphan
Potassium when flowrate changes from 125ml/sec to K(2011) “Competitive
200 ml/sec at H1=8cm sorptionandtransportofPb2+, Ni2+, Mn2+,
andZn2+ in lateritic soil columns “Journal of
It has been observed that the leached concentration of Hazardous Materials “190,391-396.
Sodium, Calcium and Potassium increases with the Estevez E A.M , Periago P L E , Carballo M
increase in flowrate. However the variation of the E,Gandara S J,Mejuto J C, Rio G L(2008)” The
same element is not same at different time intervals. It mobility and degradation of pesticides in soils and
can be observed from the bar diagrams that there is an the pollution of groundwater resources” Review
average 25% increase in leached concentration for The mobility and degradation of pesticides in soils
Sodium whereas in case of Calcium and Potassium and the pollution of groundwater resources
the average variation is observed to be approximately “Agriculture, Ecosystems and Environment 123,
12% and 7% respectively when the flowrate increases 247–260
from 125 ml/sec to 200ml/sec. French K H, Van der Zee S.E.A.T.M., Leijnse
A(2001) “Transport and degradation of
propyleneglycol and potassium acetate in the
CONCLUSION unsaturated zone” Journal of Contaminant
1) The leached concentration of Sodium, Calcium and Hydrology 49 , 23–48
Potassium increases with time and maintains an
approximate logarithmic relationship.
2) Calcium exhibits highest leached concentration
whereas Potassium exhibits least leached
concentration.
3) The effect of increase in flowrate is observed to be
more prominent than small increase in contaminated
height.
REFERENCES
Kay P, Blackwell A P, Boxall A B A (2005)”
Column studies to investigate the fate of veterinary
antibiotics in clay soils following
Ozkaya B, Demir A, Bilgili S M (2006)
“Mathematical simulation and long-term
monitoring of leachate components from two
different landfill cells” Journal of Hazardous
Materials A135, 32–39
Foo Y K, Hameed H B (2009)” An overview of
landfill leachate treatment via activated carbon
adsorption process” Journal of Hazardous
Materials 171, 54–60
Yukselen A M, Alpaslan B(2001)” Leaching of
metals from soil contaminated by mining
activities” Journal of Hazardous Materials B87
,289–300
Freijer J I , Veling M.J.E , Hassanizadeh M S(1998)”
Analytical solutions of the convection–dispersion
equation applied to transport of pesticides in soil

ISBN 978-81-921121-0-7
WATER SCARCITY: ISSUES, CONCERNS AND OPTION

(A CASE STUDY OF KOLKATA AND SUB- URBAN AREAS)
Sabita Madhvi Singh
Research Scholar in Civil Engg. Dept.,Institute of Technology,BHU, Varanasi, India.
E-Mail: sabitamadhvi@yahoo.co.in
ABSTRACT: The spatially uneven and temporally skewed rain water in almost every Indian state is an
extremely difficult challenge that often leads to the expansion of the twin calamities of flood and drought with
time, despite ever-increasing investment in water management. The same holds true for the state of West Bengal
too, in spite of being the land of plentiful rainfall (1200 to 2500 mm. annual rainfall) and rich alluvial aquifers
holding 31 billion cubic meters (BCM) of ground water, which is accessible at 5 to 10 meters below the ground
level in 95 percent of villages. West Bengal, located within Ganga–Meghna-Brahmaputra basin, covers 2.7
percent of the national territory and renders home to 8 percent of the Indian population. Significantly, it also
shares 7.5 percent of the water resources of the country. Rainwater harvesting is the process to capture and store
rainwater for its efficient utilization and conservation to control its runoff, evaporation and seepage, basically to
replenish ground water. The paper intends to highlight the different processes of Rainwater harvesting at small
micro-levels in Kolkata and suburban areas with effective net result achieved in water management. The
methods elaborated , if adopted and properly implemented, can bring about a difference though a marginal one.
Keywords: - Rainwater harvesting, Drought prone areas, Water conservation.
INTRODUCTION the harvesting of the plentiful rainwater received by

WEST BENGAL the city of joy weeps for water. the state with better management. Rainwater
West Bengal region with the largest population harvesting is the process to capture and store
density in India and wide spread poverty has also been rainwater for its efficient utilization and conservation
identifying by many as an area of quickly emerging to control its runoff, evaporation and seepage,
water crisis in spite of having some of the India basically to replenish ground water.
largest rivers. The challenges in water management
are rooted in the common objective of all the Inadequate Civic Infrastructure
countries in the region – poverty alleviation and The civic infrastructure of the metropolitan area is not
sustainable development. adequate to cater to the city’s growing population.
The spatially uneven and temporally skewed rain Unplanned development for the better part of the
water in almost every Indian state is an extremely city's three centuries of existence has taken its toll.
difficult challenge that often leads to the expansion of
the twin calamities of flood and drought with time,
despite ever-increasing investment in water
management. The same holds true for the state of
West Bengal too, in spite of being the land of plentiful
rainfall (1200 to 2500 mm. annual rainfall) and rich
alluvial aquifers holding 31 billion cubic meters
(BCM) of ground water, which is accessible at 5 to 10
meters below the ground level in 95 percent of
villages. West Bengal, located within Ganga–
Meghna-Brahmaputra basin, covers 2.7 percent of the

national territory and renders home to 8 percent of the
Indian population. Significantly, it also shares 7.5 Fig. 1 Population growth in Kolkata from 1901 to
percent of the water resources of the country. But, this 2015
is not just enough to cater all the primary needs of the
state during lean months, on account of uncontrolled The shortage of funds and low revenue generation has
population growth, expansion of irrigation networks also affected development work. Housing shortage
and development issues. Elementarily, the dam-canal has resulted in the growth of slums. Roads are few
networks yield 30-40 percent efficiency only due to while the traffic has increased tremendously. Public
evaporation and transit losses. The problem also transportation is extremely crowded. Sanitation, water
accounts for the fact that only 40 percent of the supply, pollution, health, access to education and the
surface water is utilizable, out of 13.29 Mham. state of the economy are major areas of concern. The
Surface water available to the state from 22 river demographic density during 1981 was 22,260 people
basins and 4 large drains covering 7.62 Mham. of per sq km and during 1991 this rose to 23,670 persons
catchment area. This forces no option but to opt for per sq km in Kolkata.

to augment the supply. With the setting up of Garden
Water Sources Reach works in 1982 to supply water to these areas,
the critical water supply distribution system in South
Kolkata is expected to improve. Presently, Garden
Reach supplies 50 million litres per day, which is
expected to rise to 217 million litres per day when the
construction of various units of the water works is
completed.
Fig. 2 Water sources for Kolkata

The residents get their water supply from three main
sources:
 The KMC supplies treated water through an
underground pipeline network
 The roadside public bore wells that KMC has dug
 The innumerable private bore wells that the Fig.3 Water demand and supply senario
residents have dug up.
Though there is no dearth of water, the city does not Present status
have enough resources to treat it or maintain and According to B K Maiti, Deputy Chief Engineer,
manage the distribution system. Water Supply Dept, KMC, the demand as on 2004 is
around 334 million gallons per day (MGD) and
History of Water Supply in Kolkata supply from surface and groundwater source is about
Potable water supply began in Kolkata in 1869 with 320 MGD with a gap of 10 MGD.
the installation of a slow sand filtration unit having a In 2002, KMC supplied 1062.18 MLD of water,
capacity of 6 million gallons per day (mgd) at Palta. falling short of the demand by almost 189 MLD.
The plant’s capacity was raised, in stages, to 32 mgd However, in 2003, the average supply rose to 1262.52
by 1911. The city expanded rapidly between 1948 and MLD, meeting nearly 90 percent of the requirement.
1966 and had to cater to a huge influx of people for Keeping the needs of the growing population in mind,
various reasons, including the Second World War, the KMC has undertaken several capacity enhancement
infamous Bengal famine, Independence and partition and up gradation programmes and hopes to exceed the
of Bengal. Large settlements grew on both side of demand by 2005.
River Hooghly in the Gangetic delta. The Kolkata The New Palta Water Works at Indira Gandhi Water
Municipal Authority (KMA) looked after the civic Treatment Plant is a notable step in this process. The
needs of a population of more than 14.6 million, 378 MLD water generation project, which was
settled over 178,500 hectares. The first water works at commissioned in July 2004, further augments Palta’s
Palta for supply of 27.28 million litres of filtered capacity by 226.8 MLD. This, along with the
water every day to a population of 0.4 million was increased supply from Garden Reach Water Works,
constructed in 1870 for a community water supply will increase the per capita water supply to 234 litres
scheme with a supply of 68 litres per capita per day. per day from the present 202 litres per day.
Today, the Palta Water Works, which is situated on
195 hectares along the riverfront, supplies about 780 Kolkata’s Water Supply at a Glance
million litres of water every day on an average. The Table.1 Kolkata water supply
water from Palta flows through four pipes (of 1 to 1.8 Total daily potable water supply (in 1350
m in diameter) to the reservoirs and pumping station million litres)
at Tallah. From Tallah the water is distributed to the Per capita availability of water per day 202
city by four zonal mains, which feed the vast (in litres)
distribution network measuring about 3,800 No. of tube wells
kilometres in length. Big diameter 455
However, the Tallah-Palta water supply system is not Small diameter 7875
enough to serve the consumers in the southern parts of No. of connections
the city, particularly in areas that have been recently Domestic 2,12,000
added to the Kolkata Municipal Corporation like Industrial and commercial 25,000
Tollygunj, Garden Reach, South Suburban and Public Access
Jadavpur. Considering the low pressure and Stand posts (in numbers) 17,019
inadequate supply in these areas, supplementary Street hydrants that use unfiltered 2,000
arrangement from local groundwater sources is made water (in numbers)

No. of reservoirs litres of water per day. An unaccounted number of
Present 7 deep tubewells by private industries and business
Under construction 14 establishments, housing estates and highrise
Combined capacity (in million 96 apartment blocks have also been functioning. There
gallon) are also several thousand small diameter tubewells
After the commissioning of the New Palta Water owned by private households.
Works, the per capita water availability in
Kolkata will be 234 litres per day Groundwater Resource
Source: KMC Due to large-scale withdrawal of groundwater from
the confined aquifers, a depression of piezometric
Problems that Palta Water Works faced surface in Central and South Central Kolkata has
Rapid growth of population in the city placed a huge developed. The magnitude of the depression is 6-8
burden on the plant. Increased salinity in River metres, and has developed over a period of about 40
Hooghly, excessive leakage through the old years (1958-98) in the core sector covering
distribution network and loss of pressure head as Narkeldanga- Park Circus – Bhowanipur. As a result,
water had to be conveyed over long distances were the general southerly flow in the confined aquifers has
among the chief problems that the plant faced. But, become radial in a much larger area surrounding the
with the construction of Garakka Barrage and cone of depression.
diversion of sweet water to River Hooghly, there has The central part of Kolkata Municipal
been marked reduction in the salinity of river water. Corporation area is drawing water from all directions
With the establishment of Kolkata Metropolitan resulting in its radial flow. The total quantity entering
Development Authority (KMDA), development work form the immediate vicinity of KMC area into the
for improvement of surface water to KMC and other central depression zone comes to 204 MLD in the
settlements in the Kolkata Metropolitan Area have absence of precise Census data of the withdrawal
been undertaken. structures it is difficult to evaluate precise ground
The plant, which had an initial capacity of 22.68 water draft in KMC area.
million litres per day (MLD), generated filtered water Groundwater withdrawal status
through sedimentation in pre and final settling tanks Table 2 Ground water status
and slow sand filtration in the Old Series of 12 filter
beds. Another 24 beds, each with a capacity of 3.78 Source Numbers Average Assumed Total per
discharge hours of day
MLD, were added between 1888-1893. In 1905, filter
pumping withdrawal
beds with 7.56 MLD capacity, called New Series, was Larger 325 68 cubic 7 hours 153 MLD
started. Between 1920 and 1936 when 17 beds of diameter metres per
11.34 MLD capacity and one bed of 7.56 MLD tubewells hour
capacity, called Extension Series, were introduced. Small private 4834 27 cubic 7 hours 914 MLD
With an output of 491.4 MLD the plant seems to have tubewells metres per
(4.5 inch hour
perennial capacity to serve. In 1952, a 12-bed rapid diameter)
gravity unit with a capacity of 75.6 MLD was added. Shallow 12,000 0.68 cubic 7 hours 56.7 MLD
This process helped in reducing the sedimentation hand pumps metres
time. In 1968, another 226.8 MLD rapid gravity Total 1123.70
filtration unit was added. MLD
Source: Central Ground Water Board (CGWB)
GROUNDWATER SCENARIO IN KOLKATA According to the Central Ground Water Board
History of Groundwater Extraction (CGWB) the withdrawal is 1123.70 MLD but the
The first move to tap sub-surface water in Kolkata inflow is only 204 MLD, which results in the fall in
was made as early as 1804. Between 1918 and 1940, the piezometric head. In 2003-04 according to KMC
nearly 200 medium diameter tube wells with an the number of larger diameter tubewells has increased
average yield of 27,275 litres per hour were set up in to 455, and smaller diameter borewells to 9875 and
what is known today as Kokata Metorpolitan District. the number of shallow hand pumps is 17,019. Hence
Many more tubewells were drilled during the World the current groundwater based on the withdrawal
War II to ensure water supply during emergency. figure by CGWB will be about 2120.5 MLD. Hence it
The population growth after partition made it is clear that between 1998 and 2003 the groundwater
necessary to install heavy-duty tubewells to extraction has increased by 1.9 times in five years
supplement Tallah-Palta water supply, particularly in period
southern Kolkata. Though initially conceived as an
emergency and stopgap measure, the system has now Rainwater Harvesting possibility in Kolkata
come to stay for good. Rainwater harvesting will be an ideal option to
Presently, Kolkata Municipal Corporation runs 455 prevent the alarming fall in the water level in the city.
large diameter deep tubewells and 7825 small The study of the rainfall pattern and sub surface
diameter tubewells contributing about 136 million

geology shows that both storage and recharge Total 1765.1 84.0
Table no.3 40 years average rainfall (1964-2005)
Water Harvesting Potential of the City

Kolkata Municipal Area = 187 sq km
Annual average rainfall = 1500 mm
Total rainwater falling over the city = 187 x
1,000,000 x 1500
= 280,500,000,000 litres = 768 MLD
Present water supply = 1209.6 MLD
Even if we assume 70 per cent of the area to be
developed, 50 per cent of it to be roofed and we
collect 70 per cent of the water falling over it, the
quantity of rainwater that can be harvested works out
methods can be successful applied in the city.
to 188 MLD. This is a sizeable quantity compared to
Fig. 4 Rainfall pattern in Kolkata
current water supply.
Rainfall Pattern in Kolkata
How much water can be harvested from individual
The city lies in a monsoon region, with most of its
houses ?
average annual rainfall of 1500 mm falling from June
If a house in Kolkata has a rooftop area of 100 sq m,
through September. Though winters are mild, with an
one can harvest about 1,20,000 litres of water. The
average January temperature of 19° C (67° F), the
formula for calculating the amount of rainwater that
temperature sometimes dips to 10° C (50° F). From
can be harvested is as follows.
March through September, Kolkata is hot and humid,
Runoff =A x R x C
with an average July temperature of 29° C (85° F); in
A=Area
the months of May and June the temperature may rise
R=Rainfall in millimetres
as high as 38° C (100° F).
C=Runoff coefficient (0.8)
Here’s how:
Area (A) =100 m2
Rainfall = 1500 millimetres
Runoff coefficient = 0.8
Runoff (100 x 1500 x 0.8) = 1,20,000 litres
Storing rainwater
The design parameter for storage depends on the
number of dry days and requirement. The details for
calculation an optimum storage tank is given below.
Rainy days in the City
Fig. 5 40 years average rainfall vs rainy days(1964-

2005)
Av. Annual Rainfall
Months Rainy Days
(mm)
Jan 12.1 1.0
Feb 24.5 1.8
Mar 44.8 2.4
Apr 55.7 3.7
May 153.6 6.8
Fig. 5 Rain Pattern
Jun 311.9 12.8 The above graph shows that the number of dry spell
Jul 332.5 17.2 or non-rainy days in a year is 200. So, we have to
design the size of our storage tank in such a way that
Aug 349.5 17.0 during the dry periods, water is available for drinking
Sep 287.3 13.3 purpose. The below mentioned calculation can be
applied to determine the size of the storage tank.
Oct 147.0 6.3
Nov 36.9 1.2 No. of persons in a household :10
No. of dry days in a year :200
Dec 9.4 0.5 Per capita consumption (lpcd) :6

Average annual rainfall (mm) :1500
Rooftop area (sqm) :20
Runoff coefficient :0.80
Size of the storage tank:10 x 200 x 6=12,000 litres
Check :
Water available from roof : 1500 x 20 x 0.8
= 24,000
Hence, sufficient water is available to meet the
demand of 12,000 litres
The storage tank can be Ferro cement tank, plastic
tank or brickwork tank. The tanks can be placed of Fig.6 Collecting the rain water from factoryshed
surface or below the ground level also. RAINWATER TRANSPORTATION
Relevant Case Study: Hindalco Industries Limited,

Belur Works, West Bengal
Unit Profile
The Belur Unit of Hindalco Industries Limited is
located in the State of West Bengal and is just 5 kM
from the city Kolkata. The plant was established in
the year 1938 and is the oldest aluminium sheet
rolling plant in the country. It has gradually grown
over the years and currently has a capacity of 45,000
tonnes per annum. The plant had a turnover of Rs 519
crores during 2004-05. It is one of the most
diversified aluminium sheet factories in the world
capable of manufacturing 32 different alloys and Fig.7 Rainwater Transportation
supplying to the Defence, Packaging, Bottle Closure,
Pressure Cooker, Automobile and Building industry RAINGAUGE AND LEVEL INDICATOR
throughout India. The unit is ISO 9001, ISO 14001
and OHSAS 18001 certified. The current employee
strength is 800.
Rain water Harvesting

The plant usually depends upon ground water to meet
all its water requirement. The input water is treated in
a water treatment plant far making it suitable for
industrial use. This process starting from pumping of
ground water, its treatment and then distribution
consumes lot of energy. The plant utilized the captive
pond to hold the rain water that is collected from the
overhead roof. Thus during rainy season it has able to
eliminate lifting of ground water and its treatment by
using the rain water directly in the process.
Fig. 8 Raingauge and level indicator
CAPTIVE POND
RAINWATER VS. GROUNDWATER
Captive pond volume: 5785 KL.
 Average Savings : 75 KL/Day
 Energy saved : 80,000 units per annum = Rs. 3.6
Lakhs
 Investment : Rs. 2 Lakhs
 Simple payback : 7 Months
RAINWATER COLLECTION
Fig.9 Groundwater and rainwater used in industry

disastrous results, such as aresentization. The current
RAINWATER QUALITY management regime of state determination of ’what is
good for people’ can go sour and may have a little
hope of ameliorating poor water quality. Accepting
the challenge would involve recognition of
multifaceted nature of water as a resource, accepting
that the local communities have the right over its use
and control, and that water can’t be reduced
categorized and classified and manipulated and
treated other objects. A number of NGOs have
Table. 3 Rainwater and groundwater quality demonstrated the effectiveness of alternative
paradigms but there up scaling replication and
REUSE OF REJECT WATER mainstreaming in to government programmes still
remain difficult tasks. This rain water harvesting
methods elaborated here, if adopted and properly
implemented, can bring about a difference though a
marginal one.
REFERENCES
Water Harvesting Manual, Centre for Science and
Environment
http://www.cgwaindia.com/suo/home.htm
A Water Harvesting Manual for Urban Areas: Case
Studies from Delhi. 2003. New Delhi: Centre for
Science and Environment.
Centre for Science and Environment. 2003. Site
Fig.10 Reuse of rejected water dedicated to Rainwater Harvesting. Accessed on
Use of Reject water of WTP for non-critical usage various dates at http://www.rainwaterharvesting.org/
(Road washing/ Toilet flushing) - 50 KL/Day Government of India. 2003. Ground Water in Delhi:
Improving the sustainability through Rainwater
REUSE OF ETP TREATED EFFLUENT Harvesting, Central Ground Water Board, Ministry
of Water Resources.
Government of India. 2003. Rainwater Harvesting: A
necessity in South and Southwest Districts of NCT,
Delhi. State Unit Office, Delhi, Central ground
Water Board, Ministry of Water
Resources.Government of India. 2003. Details on
Water Harvesting. Accessed on various dates at
http://www.cgwaindia.com/
Key Issues. September 2000. Water harvesting: urgent
need to reap rich rewards [Article] Indian Energy
Sector, TERI 2000. Accessed on 15 May 2003,
http://www.teriin.org/energy/waterhar.htm
Kumar, M. Dinesh. 2003. Paper: Roof Water Harvesting
Fig. 11 Treated water from ETP to Urinal for Domestic Water Security: Who gains and who
loses?
RESULT
The wastage of water is reduced which implies Re-
scheduling of the timing for supplying water to the
Community.
Previous: @ 43 KL/Day (avg.)
Previous Timing: 24 hours
Present: @ 8 KL/Day (avg.)
Present Timing: 5 am - 2 pm & 5 pm - 9 pm
CONCLUSION
Water is at the heart of politics between the nations as
they increasingly face the challenge of meeting the
growing demands of their populations. The issue of
inadequate knowledge base within which modern
management practices are adopted often leads to

ISBN 978-81-921121-0-7
SITE SELECTION FOR GROUND WATER RECHARGE USING

TREATED MUNICIPAL WASTEWATERS- A CASE STUDY OF
VARANASI (INDIA)
P.K.Singh
Professor, Dept. of Civil Engg, IT-BHU, Varanasi
Anurag Ohri
Corresponding author, Assistant Professor, Dept. of Civil Engg, IT-BHU, Varanasi, Email aohri.civ@itbhu.ac.in
Abhishek Kumar Bhardwaj
Department of Environmental Science, Veer Bahadur Singh Purvanchal University, Jaunpur
ABSTRACT: Artificial ground water recharge has been practiced in many parts of the world, including India to
supplement the under ground water reserve. Application of reclaimed waters from municipal sewage treatment
plants for such purpose is relatively a new development. GIS and expert knowledge are used as a decision
support tool to determine potential ground water recharge (GWR) sites. These sites are identified using a single-
objective multi-criteria analysis. This paper intends to report a Geographical Information System (GIS) based
methodology for the GWR site selection when the source is surplus reclaimed water from Sewage Treatment
Plants (STPs). A set of six criteria (land use/land cover, altitude, distance from STP, distance from drinking
water pumping station, type of soil, and depth of ground water table) have been used for GWR site selection.
The selection of criteria and acceptable ranges is based on expert opinion and published guidelines. Analytic
hierarchy process (AHP) has been used to give weightage to different factors based upon expert opinion. The
criteria were aggregated and Attribute Value (A) is generated using weighted linear combination (WLC)
technique in GIS environment. The suitability index (0-10) values are classified into five categories: Not
Suitable, Least Suitable, Moderately Suitable, Suitable, and Most Suitable, to select the best site for the purpose.
The results indicate that in Varanasi city whereas adequate land under “most suitable” category appears
available near Bhagwanpur and DLW STPs, the lands available for the purpose near Dinapur STP fall under
“moderately suitable” category. It is found that GIS based multi-criteria decision analysis can be a powerful tool
for such applications.
Keywords: Ground Water Recharge using reclaimed water, Geographical Information System (GIS), Multi-
Criteria Decision Analysis (MCDA), Analytical Hierarchy Process (AHP), Weighted Linear Combination
(WLC) technique.
INTRODUCTION and environmental pollution, provides a realistic
The artificial recharge to ground water aims at framework for considering reclaimed wastewater as a
augmentation of ground water reservoir by modifying water resource rather than a liability. The inclusion of
the natural movement of surface water utilizing groundwater recharge in a wastewater reuse project
suitable civil construction techniques. Artificial may provide psychological and aesthetic benefits as a
recharge techniques normally address to following result of the transition between reclaimed municipal
issues: wastewater and groundwater. However, the lack of
(i) To enhance the sustainable yield in areas specific criteria and guidelines governing the artificial
where over-development has depleted the aquifer. recharge of groundwater with recycled municipal
(ii) Conservation and storage of excess surface wastewater is currently hampering the implementation
water for future requirements, since these of large-scale groundwater recharge operations.
requirements often changes within a season or a
National Ganga River Basin Authority (NGRBA) has
period.
been tasked with the preparation of a Ganga River
(iii) To improve the quality of existing ground
Basin Management Plan (GRBMP), implementation
water through dilution.
of which is expected to result in the restoration of the
The basic purpose of artificial recharge of ground
‘wholesomeness’ of all the rivers in the Ganga basin.
water is to restore supplies from aquifers depleted due
to excessive ground water development (CGWB, Two important objectives of GRBMP are the
2000). restoration of 1) ‘Nirmal’ dhara and 2) ‘Aviral’ dhara
As noted by Asano and Cotruvo (2004), for more than in the rivers of the Ganga basin. The restoration of
a quarter century, a recurring thesis in environmental ‘Nirmal’ dhara in all rivers of the Ganga basin will
and water resources engineering has been that require the following steps concerning sewage
improved wastewater treatment provides a treated collection, treatment and disposal processes:
effluent of such quality that it should be put to 1. Complete stoppage of the discharge of sewage,
beneficial use. This conviction in responsible either treated or un‐treated, from towns into all rivers
engineering, coupled with increasing water shortages of the Ganga basin.

2. All effluents from urban domestic sources must be The Varanasi city is located in the North Eastern part
collected and treated up to tertiary level. of the Nation and is one of the biggest cities of Uttar
3. The treated water should be recycled or reused for Pradesh and is the district headquarters. The
various purposes, i.e., industrial, irrigation, geographical coordinates of the city are 25º 0’ to 25º
horticultural, non‐contact/non‐potable domestic uses, 16’ North latitude and 82º 5’ to 83º 1’ East longitude.
groundwater recharge, etc. It is bounded on north by Jaunpur and Ghazipur, on
south east by Mughal Sarai, on east by Bihar State, on
For this purpose, map showing areas where treated west by Bhadohi and on south by Mirzapur. The area
sewage can be used for groundwater recharge needs to of the city is 78.5 km2. The Varanasi Nagar Nigam
be prepared (GRBMP Report, 2010). has divided the area of the city in 90 wards for
Recharge is generally considered to occur in administrative and management purposes. Apart from
topographic highs and discharge in topographic lows being a place of pilgrimage and tourist centre,
in humid regions, whereas in arid alluvial-valley Varanasi is a fast growing commercial, industrial and
regions recharge is usually focused in topographic trading centre of Uttar Pradesh and Central India. The
lows, such as channels of ephemeral streams. general features of the city are presented in Table 1.
Delineation of hydrogeomorphic settings on the basis Table 1: General features of Varanasi City
of topographic attributes – including slope classes, S. Information Data
breaks in slope, curvature, and elevation – is greatly No.
facilitated by the use of geographic information 1. Population ~ 14 lacs as per 2001
systems (GIS) and digital elevation models and is Census
used widely in Australia (Salama et al. 1994). 2. Area 78.5 sq.kms
3. Altitude 80 meters above msl
Ghayoumiana & Mohseni (2006) reported application
of GIS techniques to determine areas most suitable for 4. Temperature 22.6 to 41.5 °C
artificial groundwater recharge in a coastal aquifer in Summer
southern Iran. Parameters considered in the selection 5. Temperature Winter 9.5 to 23.4 °C
of groundwater artificial recharge locations are 6. Rainfall 1113.4 mm
diverse and complex. In this study factors such as There are two sources of water available at Varanasi,
slope, infiltration rate, depth to groundwater, quality viz., ground water and river water (surface). Both the
of alluvial sediments and land use were considered to sources are utilized for drinking water requirement of
determine the areas most suitable for groundwater the city. These water sources are also used for
recharge to determine the relationships between industrial purposes, irrigation purposes and other
geomorphological units and the appropriate sites for domestic usage of the city. In the unconsolidated
groundwater artificial recharge, land-use and alluvial sediments ground water occurs in the pore
geomorphological maps were developed from satellite spaces in the zone of saturation. The ground water in
images. The results of the study indicated that about the unconsolidated deposits of the older alluvium in
12% of the study area is considered as appropriate and the area comprises of two bodies, (i) a shallow ground
8% moderately appropriate sites for artificial water body which occurs principally in clay and
groundwater recharge. Rao et al. (2009) used GIS kankar beds of back swamp deposits and (ii) deep or
analysis for identification of groundwater potential main ground water body which occurs in thick sands
zones in and around Madhurawada (Andhra Pradesh, of the meander belt deposits. These two water bodies
India). are hydraulically distinct. The back swamp deposits
are thick but are likely to be inter connected when
Various studies have been done to locate the site for they are comparatively thin. The shallow ground
different purposes using GIS and AHP. Negi and Jain water in back swap deposits is generally unconfined
(2008) have developed a methodology for site and its static water level is only few meters bgl. It
selection of municipal solid waste landfill using GIS supplies water to the dug wells. The deep ground
and AHP, which is one of the methods of MCDA. water body in meander belt deposits is considered to
Siddiqui et al. (1996) have developed a methodology be hydraulically continuous and is confined in nature.
to find the best location for landfill by integrating GIS It supplies water to the tube wells for irrigation,
and AHP. Thus there is appreciable amount of industrial, and domestic uses. The back swamps
information available in literature for site selection deposits and the meander belt deposits differ
using GIS and MCDA. Varanasi city has three considerably in their capacity to absorb and transmit
functional sewage treatment plants at present. The water to the zone of saturation. The back swamp clays
objective of this study is to find the GIS based are massive and contain Kankar and in thin lenses of
methodology for selection of suitable sites for ground fine grained sand. Because this material is generally
water recharge in Varanasi by treated municipal too fine grained to transmit water readily, it prevents
wastewaters from STPs. the downward percolation of water from the land
surface and also confines water in the underlying
THE STUDY AREA meander belt sand deposits. The meander belt deposits
consist of medium to coarse grained sand and

underlying impervious back swamp clays, are Varanasi city has following three existing STPs
moderately to highly permeable and constitute a 1. STP at Bhagwanpur - 8 MLD
potential ground water reservoir. 2. STP at Dinapur - 80 MLD
3. STP at DLW - 12 MLD
The population of Varanasi city and in particular of Total = 100 MLD
Trans Varuna area is mostly dependent on the ground
water for its domestic use. About 115 MLD ground METHODOLOGY
water is extracted to fulfill the requirement. There are Based on literature review regarding the site selection
about 111 tube wells in different localities of the city for different purposes and keeping in mind the
and the Jal Sansthan has provided about 1954 hand sensitivity of different environmental parameters for
pumps to meet the requirement of public, who are the purpose, overall six criteria were used for site
unable to get the water connection in their premises. selection of ground water recharge.
Besides these private boring is also very common. i) Land use/land cover area: Land use provides
Ground water resource evaluation conducted reveals important indicators of the extent of groundwater
the information related to ground water recharge, requirements and groundwater utilization, as well as
ground water depth, level of development, and being an important indicator in the selection of sites
category of the city on the basis of the development. for artificial recharge of groundwater. Satellite data,
The exploitable dynamic ground water resource is topographic maps and field information were used as
estimated to be 1172.46 mcm and the ground water reference data. Different classes of land use were
draft is estimated to be 671.87 mcm. The balance of identified (e.g. residential, lontic water, agriculture,
500.59 mcm of ground water is available for future sand, barren land, lentic water) by visual image
development both for domestic and irrigation interpretation of remote sensing image.
purposes. Varanasi Jal Sansthan also draws water ii) Altitude: This criterion reflects the necessary
from Ganga River for supply to residential colonies wastewater pumping devices and related costs. The
after complete treatment. Static Water Table is around economic feasible elevation difference must not
16 m below ground level and a seasonal variation of 6 exceed 15 m (Kallali et al., 2007).
m is observed. Fig 1 shows the geographical map of iii) Distance from STP: The economic criteria
Varanasi and Fig 2 shows the drainage districts and considered include water transfer costs from STP to
the location of sewage treatment plants in the city. GWR sites. This reflects necessary pipe length in
meters and afferent costs to convey treated
wastewater from source to destination. The economic
feasible distance must not exceed 8 km (Kallali et al.,
2007).
iv) Distance from drinking water pumping station:
A safeguard distance of at least 1000 m from drinking
water pumping station to avoid their possible
contamination by wastewater infiltration.
v) Type of soil: Suitable soils should have a very low
clay fraction in order to firstly be permeable enough
to allow treated wastewater infiltration and secondly
perform good complementary wastewater
Fig.1(a) Study area in India map purification. In this respect, soils with greater than
10% clay fraction would exclude GWR site.
vi) Depth of ground water table: Suitable site must
have static groundwater level at least 5 m depth in
order to have a sufficient vadose zone for wastewater
purification. With this minimum aquifer depth, we
expect overcoming aquifer mounding regardless of its
transfer rate. (Kallali et al., 2007)
Accordingly, six input map layers for each of land
use/land cover area, altitude, distance from STP,
distance from drinking water pumping station, type of
soil, and ground water table are prepared to be used in
the analysis in GIS environment. All the data layers
are derived and prepared from related maps by
scanning, geocoding and digitizing the relevant
information. The information compiled from literature
from various sources about the suitability of site for
ground water recharge is used to determine the buffer
Fig. 1(b): Drainage Districts and location of STPs in zones for each layer. After creating the classes for
Varanasi (JICA Report, 2005) each layer by using buffer zones, each layer is

converted into individual raster maps. After raster assigns to those areas which are not suitable for
conversion, each input map is reclassified to give ground water recharge and 10 are assigns to those
them weights based upon the suitability of ant site for areas which are most suitable for locating ground
ground water recharge. Because of different scales water recharge sites. Other numbers are assigns to
upon which criteria are measured, it is necessary that different areas having different suitability foe site
factors be standardized before combination. All the selection. The layers, buffer zones and their attribute
criteria are standardized in the scale of 0 to 10. Zero is weights are summarized in Table 2.
Table 2: Sites Selection Criteria for Ground Water Recharge Using Treated Municipal Wastewaters
S. N. Name of criteria Constrain/ Zones Scale Value Suitability

Factor (xij)
1 Land use/ land Constrain and Residential 0 Not suitable
cover Factor Lontic Water 0 Not suitable
Agriculture 6 Moderately suitable
Sand 6 Moderately suitable
Barren 10 Most suitable
Lentic Water 10 Most suitable
2 Altitude Constrain and Bhagwanpur STP

Factor elevation =76.5
76.5-78.5m 4 Least suitable
>78.5m 0 Not suitable
76.5-72.5m 10 Most suitable
72.5-68.5m 8 Suitable
<68.5m 6 Moderately suitable
Dinapur STP elevation
=78.5)
78-74m 10 Most suitable
74-70m 8 Suitable
70-66m 6 Moderately suitable
DLW STP elevation
=82.5)
82.5-78.5m 10 Most suitable
78.5-74.5m 8 Suitable
74.5-70.5m 6 Moderately suitable
3 Distance from Constrain and 0-1000m 10 Most Suitable
STP Factor 1000-2000m 8 Moderately suitable
2000-3000m 6 Suitable
3000-4000m 4 Least Suitable
>4000m 2 Least Suitable
4 Distance from Constrain and 0-1000 m 0 Not suitable
existing drinking Factor 1000-2000m 4 Less Suitable
water pumping 2000-3000m 6 Moderately suitable
station 3000-4000m 8 Most Suitable
>4000 10 Most Suitable
5 Ground Water Constrain and 0-2 m 0 Not suitable

Table Factor 2-5m 2 Least Suitable
5-8m 5 Moderately Suitable
8-12m 8 Suitable
>12m 10 Most Suitable
6 Type of Soil Factor Alluvial soil 04 Not Suitable
Sandy soil 08 suitable

RESULTS AND DISCUSSION water reservoir. Under the circumstances, assuming
Estimation of Area Required for Ground Water an annual hydraulic loading rate of 200 m/year (range
Recharge using Treated Waste Water 30-300 m/year) and following a 2 days wetting with 4
For methodology development purpose, infiltration days drying schedule of waste water spreading and
rate and hydraulic loading rate as adopted by Bouwer recharge operation, the number of cycles per year is
and Rice (1984) is used for Varanasi city also. The approx. 61. This amounts to a hydraulic loading rate
range of infiltration rates (i.e. the rate water drains of about 1.5 m/d in two days wetting period. Based
into the ground when basin is flooded ) for the in and on this hydraulic loading rate, the infiltration basin
off channel system is about 0.3 to 3 m/d including any area required for spreading the treated waste waters
effect caused by clogging. Systems with year round from STPs of Varanasi city is obtained as given in
recharge and periodic drying and cleaning of the Table 3.
bottom typically have hydraulic loading rates of 30 to Table 3: Land area required for ground water
300 m/year. If the basin bottoms are not covered by recharge at different STP sites of Varanasi
sediment or other clogging materials, and ground S. Location of Design Approx. Approx.
water level are sufficiently low to not effect N. STP Discharge Area size (ha)
infiltration, infiltration rates are about the same as (MLD) Required
vertical hydraulic conductivity of the soil, which may (m2)
be about 0.3 m/d for sandy loam, 1 m/d for coarser 1. Bhagvanpur 8 5333 0.5
sand and 20-50 m/d for fine and clean gravel. Sand 2. DLW 12 8000 0.8
and gravel mixtures gave lower hydraulic 3. Dinapur 80 53330 5.3
conductivities than the sand alone ((Bouwer and Rice,
1984). Generation of suitability maps
Suitability maps have been generated using criteria
In Varanasi, the meander belt deposits consist of given in Table 2. A suitability map for GWR site on
medium to coarse grained sand and underlying the basis of different criteria set for the purpose is
impervious back swamp clays are moderately to shown in Fig. 2 for Bhagwanpur STP.
highly permeable and constitute a potential ground
a b
c d

e f
Fig.2: Suitability map based on (a) Land use/Land cover area, (b) Altitude, (c) distance from STP, (d)
existing drinking water pumping station, (e) soil type, and (f) ground water table for Bhagwanpur STP
The criteria adopted for ground water recharge using them are considered at a time which provides easier
treated waste water has been assigned weightage ranking. The comparison matrix developed for six
based on Analytical Hierarchy Process (AHP). In the included criteria are shown in Table 4. The composite
AHP, the first step is that a complex decision problem weights are produced by means of a sequence of
is decomposed into simpler decision problems to form multiplication. In the process, first, the decision
a decision hierarchy. When developing a hierarchy, matrix is squared; the row sums are calculated and
the top level is the ultimate goal which in this case is then normalized. This procedure is continued till the
ground water recharge site selection. After the differences between normalized weights of the
decomposition stage, cardinal rankings for criteria are iterations reach to a very small value. Finally the
determined which is done by pairwise comparisons. weight has been calculated for different criteria.
Two alternatives and the importance relation between
Table 4: Weights and Comparison Matrix for AHP Calculation

Land use\Land cover
Ground water table

Pumping station
Dist. From STP
Altitude area
Weights (wi)
Type of soil
Land use\Land cover 1 3 4 4 6 7 0.4391

Altitude area 1/3 1 2 2 4 5 0.2090
Dist. From STP 1/4 1/2 1 1 3 4 0.1289
Pumping station 1/4 1/2 1 1 3 4 0.1289
Type of soil 1/6 1/4 1/3 1/3 1 2 0.0559
Ground water table 1/7 1/5 1/4 1/4 1/2 1 0.0382
The consistency ratio is 0.02.
Before calculation of final results, the areas restricted unsuitable areas is done by mask operation. The gray
by rules and physical constraints are excluded from areas in the mask are not suitable areas shown in Fig 3
the study area which are assigned 0 during the data (a). All data layers converted to raster are multiplied
preparation stage. The exclusion of certainly by mask to become ready for ranking.

a b
Fig.3: (a) Mask map, (b) Final Site Suitability Map for GWR at Bhagvanpur STP.
Preparation of Final Maps (0.5 ha) is available even in the most suitable category
of land, the ground water recharge possibilities using
Following geoformula is used to generate the overall
the treated municipal waste waters may be explored
score in terms of Attribute Value (A) in the GIS
further at this site.
environment.
A  ( wi * xij ) * constrain Suitable area obtain in DLW
Where wi is the weight of ith criteria and xij is the scale As per the final site suitability map (Fig. 4 b), lands
value of jth class in ith criteria. under various categories for the GWR purpose near
Table 5: Classification of Area based on Attribute DLW STP are found as follows:
Value Most suitable: 3.74 ha (37400 m2)
Attribute Value (A) Recommendation Suitable: 121 ha (121,0000 m2)
0 Not Suitable Moderately suitable: 945 ha (945,0000 m2)
2-4 Least Suitable Here “most suitable” area lies at 25˚18'0.61"N,
82˚58'4.36"E, near Lahartara and in Narwal. As the
4-6 Moderately Suitable
area required for the purpose (0.8 ha) is much less
6-8 Suitable than the available area in “most suitable” category,
8-10 Most Suitable the GWR option using treated waste waters may be
A constrain value of 0 is assigned for not suitable explored further at this STP site also.
areas and 1 for remaining areas. The output map is Suitability Maps for Dinapur
produced by using the map calculator. The output As per the final map, lands under various category for
values obtained are divided into five classes, one of the GWR purpose near Bhagwanpur STP are found as
which are the masked areas with value of 0 and follows:
defined as restricted areas for GWR. The other classes Most suitable: Nil
in terms of increasing suitability are Least Suitable, Suitable: Nil
Moderately Suitable, Suitable and Most Suitable. Moderately suitable: 138 ha or (138, 0000 m2)
Normalized weight distribution of the factors is An area of about 5. 3 ha have been estimated (Table
shown according to their ranking. Classification of 3) to be required for GWR from treated municipal
area based on attribute value is done, as given in waste waters. There is no land available under “most
Table 5. suitable” and “suitable” category primarily due to
high ground water table (within 3-5m bgl) in the
Potential Sites for Ground Water Recharge region. Area under “moderately suitable” category is
(GWR) using treated wastewaters around STPs available near 25˚20'9.87"N, 83˚33'2.73''E in
As per the final site suitability map (Fig. 3 b), lands Khalispur and Kotawa villages. Hence GWR facility
under various category for the GWR purpose near using treated waste waters from Dinapur STP may be
Bhagwanpur STP are found as follows: developed near this area subject to due precautions for
Most suitable: 0.8 ha (8700 m2) sensitive parameters. Other land requirements include
Suitable: 150 ha (150,0000 m2) area for buffer zones, access roads, berms, and storage
Moderately suitable: 260 ha (260,0000 m2) (if necessary). Buffer zones can be used to screen sites
Most suitable area is situated on 25˚17'8.81"N, from public view. Access roads and ramps, typically 3
83˚01'5.32"E, and suitable area on 25°15'20.73"N, to 3.6 m wide, are needed so that maintenance
83˚01'7.38"E. As the required area for the purpose equipment for surface scratching can enter each basin.

b
a
Fig-4 (a) Mask map (b) Final Site Suitability Map for GWR at DLW STP
a b
Fig.5: (a) Mask map (b) Final Site Suitability Map for GWR at Dinapur STP
CONCLUSION AND RECOMMENDATIONS (AHP) can conveniently be used for giving weightage
Remote Sensing and GIS are important tools which to different criteria for this purpose. In Varanasi,
can efficiently be applied in sites selection for ground assuming a hydraulic loading rate of 1.5 m/d, the
water recharge. Land use/land cover, distance from basic area required for infiltration basins to recharge
drinking water pumping station, distance from STP, the ground water using treated waste waters from
surface water body, altitude and type of soil are Bhagwanpur, DLW and Dinapur STPs at designed
considered important parameters which govern discharge is about 0.5, 0.8 and 5.3 ha respectively.
ground water recharge site selection using treated Whereas adequate land under “most suitable”
waste water. Among Multi Criteria Decision Analysis category appears available near Bhagwanpur and
(MCDA) methods, Analytical Hierarchy Process DLW STPs, the lands available for the purpose near

Dinapur STP fall under “moderately suitable” Ghayoumiana, J. and Mohseni S. M. (2007).
category primarily due to high ground water table. Application of GIS techniques to determine areas
The approx. distances of sites under “most suitable” most suitable for artificial groundwater recharge
category near Bhagwanpur and DLW STPs is 1.75 in a coastal aquifer in southern Iran. Journal of
km. The sites under “suitable” category are around 0.7 Asian Earth Sciences, 30, 364–374.
km and 2.75 km away from Bhagwanpur and DLW GRBMP Report (2010). Plan for construction,
STPs respectively. The distance of “moderately operation and maintenance Sewage Collection:
suitable” lands available for the purpose near Dinapur Treatment and Disposal Infrastructure Proposed
STP is approx. 0.6 km. The relative merits of treated under GRBMP, Ministry of Environment and
waste waters to be conveyed to a distant “most Forest, Government of India.
suitable” site with respect to developing an area under Negi, P. and Jain, K. (2008). Spatial multicriteria
“suitable” category nearer to the STPs need to be Analysis for Siting Groundwater Polluting
examined. Actual field measurement of infiltration Industries. J Enviormental Informatics, 12(1),
rate near STP sites and applicable hydraulic load 54-63.
estimation based on the quality of treated effluent may Rao P. J., Harikrishna, P., Srivastav, S. K.,
be undertaken to plan the ground water recharge Satyanarayana, P. V. V., and Rao, B. V. D.
systems for underground storage, treatment, disposal (2009). Selection of groundwater potential zones
and reuse purposes. in and around Madhurawada Dome,
Visakhapatnam District - A GIS approach. J. Ind.
Geophys. Union ., 13(4), 191-200.
REFERENCES Salma, R. B. (1994). The use of GIS in ground water
Asanoa, T. and Cotruvob, J. A., (2004). Groundwater studied and modeling and dry land salinity risk
recharge with reclaimed municipal wastewater: maps in proceedings of Resource technology.
health and regulatory considerations, Water New apportunities- Best practice, Univerrsity of
Research, 2004, Melbourne, Australia.
doi:10.1016/j.watres.2004.01.023 Siddiqui, M. Z., Everett. And Vieux, B. E. (1996).
CGWB (2000). Guide on Artificial Recharge to Landfill siting using geographic information
Ground Water, Ministry of Water Resources, systems: a demonstration. Journal of
Government of India, New Delhi. Environmental Engineering, 122-6, 515-523.

ISBN 978-81-921121-0-7
ANALYSIS OF GAS DYNAMICS IN A FUNNEL DURING AIR

SUCTION
K.K. Pandey
Associate Professor, Deptt., of Civil Engg., IT-BHU, Varanasi, Email: kkp.civ@itbhu.ac.in
Aniruddh Vashisth
B. Tech Student, Deptt., of Civil Engg., IT-BHU, Varanasi
ABSTRACT: A study of gas dynamics is carried in a funnel pumped with air through a nozzle. The present
investigation predicts the suction flow rate from the atmosphere through the louvers of a funnel (used in naval
and merchant ships) when the high velocity exhaust comes out from a nozzle or a set of nozzles placed inside
the funnel. Conservation equations of mass and momentum have been solved for the funnel with a surrounding
computational domain so that the suction can take place at the louvers entry. The resulting equations have been
solved numerically using finite volume technique in an unstructured grid employing eddy viscosity based two
equation k- turbulence model.
INTRODUCTION
Flow of air in pipes is an example of compressible cause atmospheric air to be sucked into the funnel
fluid flow. Compressible fluid flow implies and mix with the hot gas thus making it to cool
appreciable variations in density throughout the flow down.
field (Zucker and Biblatz 2002), (Anderson 1990), In real life the fluid is a hot gas. In case of
(Thompson 1972). Compressibility becomes an isothermal fluid the density is constant where as in
important factor at high flow speeds or for large case of hot fluid the density is a function of
temperature changes (Acker 1952), (Baham and temperature which takes into account the buoyancy
McCallum 1977). Large changes in velocity require effect. In case of buoyancy driven flow it is
large pressure changes; for gas flow these pressure expected that the mass ingress of air into the funnel
changes are accomplished by significant variations in would be higher due to the extra buoyancy term
both density and temperature. A familiar case of that is included in the momentum equation as the
application of vertical pipes where gas dynamics plays body force. For constant nozzle mass flow rate the
a very important role is an exhaust pipe. A basic velocity at the nozzle exit would be higher
attempt has been made to investigate the phenomenon (compared to a cold jet) due to low density of air
when air suction takes place inside a vertical pipe. The (because of high temperature) for which the mass
analysis was further studied for vertical pipes having suction into the funnel is expected to be more.
louvers which allowed air to come into the pipe freely. An attempt is made to analyse the following cases:-
Normally the combustion products are very hot when  Analysis of Structural Stability.
they reach the outlet of the nozzles protruding into the  Analysis of Air Suction through a funnel and its
funnel. The combustion products come out of the various effects:-
nozzle at a very high speed and there are lots of a. Effect of number of cells on air suction rate
louvers around the periphery of the funnel through b. Effect of louvers opening area on suction
which the atmospheric air from the surroundings rate
rushes into the funnel due to the jet action created by c. Effect of height of the funnel on air suction
the high speed jet coming out of the nozzle placed rate
inside the funnel (Isyumov and Tanaka, 1979) (Snyder d. Effect of funnel diameter on air suction rate
1981). The fresh air that enters the funnel thus mixes e. Effect of wind velocity on suction rate
with the hot gas and the mixture travels upward along
the axis of the funnel and gets discharged into the METHODOLOGY
atmosphere at the height of the. By the time the The funnel is analyzed for structural stability and
mixture reaches the top of the funnel the temperature for the various effects by air suction in a funnel.
of the gas drops considerably due to the mixing with
the cold air from the ambient (Jha et al. 2003). Making Structure Stability
the combustion product to attain a low temperature is The process of analyzing a structure according to
only possible (in a very practical way) by sucking the given algorithm
fresh air from the atmosphere through the louvers of A brief methodology of the analysis could be
the funnel. No extra power through the use of a pump stipulated as follows:-
is needed to do this. The jet action created by the high a) Specifying Geometry - First the geometry of
speed jet coming out from the nozzle is utilized to
the structure to be analyzed is defined.
create low pressure zones in the funnel which will

Model
(Model Generation) funnel) is placed at the centre to supply high
Definition
Define: 1) Elements velocity air jet. Another cylindrical computational
2) Material domain of diameter, Dcd and height, Hcd (same as
3) Boundary funnel height) is placed around the funnel so that
boundary condition can be applied on the boundary
of the computational domain for suction to take
Solution place near the louvers cut on the funnel. A number
Frontal Solution
1) Phase
Formulate Element Matrix of rectangular louvers (simply the opening), of
2) Triangularize Global Matrix height, Lh and width, Lw has been cut around the
3) Back Substitute periphery of the funnel. The vertical distance
between the two consecutive rows of louvers, Lr
has been maintained to be 0.5 m from the view
Results Calculated point of structural rigidity of the funnel. The flow
1) Displacement field in the domain has been computed by using
2) Stress/Strain three-dimensional, incompressible Navier–Stokes
3) Moments/Forces
equations with a two equation k-e turbulence
model. The Ren (Reynolds number based on the
Fig.1. Flow diagram of the analysis procedure
nozzle diameter) vary from 5.3 × 106 to 8.8 ×106.
followed by ANSYS (ANSYS User’s Manual)
For such a high Re flow it is believed that the two
b) Specify Element Type and Material Properties - equation k- turbulence model (high Re k-
Next, the material properties are defined. In an model with wall functions) would produce the right
elastic analysis of an isotropic solid these consist kind of result.
of the Young's modulus and the Poisson's ratio
of the material.
c) Mesh the Object - Then, the structure is broken
(or meshed) into small elements. This involves
defining the types of elements into which the
structure will be broken, as well as specifying
how the structure will be subdivided into
elements.
d) Apply Boundary Conditions and External
Loads - Next, the boundary conditions (e.g.
location of supports, wind mass per second etc.)
and the external loads are specified.
e) Generate a Solution - Then the solution is Fig. 2 Specifications of the funnel under
generated based on the previously input examination
parameters by the general equations by
formulating Element Matrices and MATHEMATICAL FORMULATIONS
Triangularized Global Matrices are solved by
back substitution. Governing equations
f) Refine the Mesh - Finite element methods are The governing equations for the above analysis can
be written as
approximate methods and, in general, the
Continuity
accuracy of the approximation increases with the
. (ρ v )=0
number of elements used.
Momentum
.(ρvv)= - p + .( ) + ρg
Air suction through funnel
Where p is the static pressure, is the stress
The computational investigation is out for a tensor, ρ is the fluid density and ρg is the
cylindrical funnel, of diameter, Df and height, Hf as gravitational body force. It is to be noted here that
shown in the given figure. The funnel is closed at the the pressure, p is a modified pressure because the
bottom and opened at the top to the surrounding isotropic turbulence model used for the simulation
atmosphere. At the bottom of the funnel a nozzle of will scramble the pressure to be (p+2kρ/3) which is
diameter, Dn having a protrusion length of Lp (into the computed iteratively from the momentum and

continuity balance though a standard pressure At the nozzle outlet
correction scheme. vx = 0; vy = 0; vz = vinlet (common for both
The stress tensor is given by analysis)
=µeff ( v+ vT), The turbulent quantities, k and , on the first near

wall cell have been set from the equilibrium log-
Effective viscosity law wall function (Jha and Dash 2002) ,( Jha and
Dash 2004), (Jha et al.2003). At the first near wall
µeff = ( µ + µt ) cell, the values of kp and p are set from the log-
law wall function according to:
Turbulence kinetic energy – k
pupkp1/2cµ1/4 1
--- ln(Ez+)
---------------- =
.(ρvk)= (µt/σk) k +Gk - ρ w k
Where w is the wall shear stress
Rate of dissipation of k
Zpkp1/2cµ1/4
+
.(ρv ) = .(µt/σ ) +( /k)(C1 Gk - Z = ---------------
C2 ρ ) v
The turbulent viscosity µt is computed from Where E= 8.6 and k= 0.41
µt = ρCµ (k2/ ) cµ3/4 kp3/2

p = ---------------
The term Gk representing the production of turbulent kzp
kinetic energy is computed from Numerical Solution Process
Three-dimensional equations of mass, momentum
Gk= µt ( v+( v)T): v and turbulence have been integrated over the
control volume and the subsequent equations have
σk and σ are the Prandtl numbers for k and been discretized over the control volume using the
finite volume technique to yield a set of algebraic
equations which could be solved by an algebraic
multi grid solver of Fluent by imposing proper
boundary conditions as specified above. Second
order upwind scheme was considered for
momentum as well as for the turbulence discretized
equations in order to get better accuracy. The k-
model has been used as a closure model for
turbulent flow. Fine grids were used at the nozzle
entrance as well as at the louvers opening area to
have better accuracy.
Fig.3 Boundary condition for computational domain
RESULTS AND DISCUSSIONS
Boundary Conditions
The boundary conditions can be seen from Fig 3 The Results for Analysis for Structural stability
funnel wall, nozzle wall and bottom wall are solid and Analysis by ANSYS reports that as the height of
have been given a no-slip and no-penetration the tower increases, the deflection due the resultant
boundary condition. Pressure outlet boundary of the force generated by internal air suction and
conditions have been imposed at the outer periphery the external wind pressures, has more prominent
of the computational domain and at the top surface of impact on the structural stability of the funnel.
the computational domain, while velocity inlet ANSYS divided the circumference in to 120
boundary condition has been employed at the nozzle divisions of an angle of 3o(ANSYS User’s
outlet which ejects high velocity air jet into the funnel. Manual). The displacement of the structure
The boundary conditions can be summarized as increases gradually in the beginning but starts to
At the solid walls climb very quickly after a certain point (Girault and
vx = vy = vz = 0 (common for both analysis) Raviart, 1979),( Reddy, 1990). It is at this point
At the outer periphery and at the top surface of the that the impact of the air turbulence inside the
computational domain funnel along with external wind speed begins to
p = pa (atmospheric pressure assumed to be overpower the structural stability. ANSYS
zero);(pressure outlet boundary) (common for both optimisation also suggests that a certain ration of
analysis) the height and the funnel diameter could be

maintained (for a given building material property), The air suction rate is enhanced with increase in
for producing a stable structure. louvers opening area. The Computational Flow
Table 1 Dynamics analysis suggests the louvers should be
located at the bottom half of the funnel and nearer
Parameters Value to the nozzle.
Height of the vertical 20 m
pipe Effect of louvers opening area on suction rate
Wind Speeds at height of 50 m/s The effect of increase in louvers opening area on
20m the air suction rate from the surrounding
Structure material M65 concrete atmosphere has been shown in Fig. 5. In the present
properties study, simulations are carried out for louvers
Thickness of walls 2m opening area of 5.76 m2 (AL/AF = 0.051), 11.52 m2
(AL/AF = 0.1018), 17.28 m2 (AL/AF = 0.1527),
Results for the Air suction in a funnel due to high 23.04 m2 (AL/AF = 0.2037) and 28.8 m2 (AL/AF =
velocity air-fluid jet. 0.2546) with nozzle flow rate of 162 kg/s, 121.5
kg/s and 97.24 kg/s. The opening area is increased
Effect of number of cells on air suction rate by adding additional louver rows placed
In order to start numerical investigation a cylindrical symmetrically around the periphery. It is clear from
computational domain of diameter (Dcd) and height the graph that the air suction rate increases with
(Hcd) has been considered around the funnel. The increase in louvers opening area. It is also evident
circumference of the funnel has been cut with 12 from Fig.9 that, when opening area of the louvers
numbers of louvers per row of size (0.6 × 0.8) m2 and are increased more than 23.04 m2 the air suction
the total louvers opening area was 5.76 m2 per row. rate does not vary appreciably. This can be
Such type of flow normally occurs from the Gas explained from the fact that when the louvers are
turbine of a powerful ship at peak load. The analysis is located towards the top end of the funnel, it does
initiated with a grid independence test for the louvers not create enough low pressure in the funnel so that
opening area (AL) of 5.76 m2 while other parameters atmospheric air can be sucked in.
were kept constant (Mishra and Dash, 2010). Table 3
Table 2 Parameters Value/Range
Parameters Value/Range nozzle diameter 1.32 m
Diameter (Df) 4m protruding length of 0.7 m
Height of the funnel (Hf) 9m nozzle
Nozzle diameter (Dn) 1.32 m funnel diameter 4m
Protruding length (Lp) of 0.7 m funnel height 9m
nozzle computational domain 24 m
The height of the louvers 0.8 m
Curve length of the width 0.6 m
Vertical distance between 0.5 m
the adjacent rows (Lr)
An air flow rate 137.76 kg/s
Fig.5 Variation of air suction rate with louvers

opening area.
The air suction rate is more effective when the

louvers are located at the bottom half of the funnel
and it should be located near to the nozzle.
The air suction rate is independent of number of
Fig. 4. Variation of air suction rate with number of
louvers per row but strongly depends on louvers
cells.
opening area.

Effect of height of the funnel on air suction rate Fig. nozzle diameter. It can be seen from the plot that
6 shows the change of air suction rate with the height (Fig.7) the air suction rate initially increases with
of the funnel for three different cases of nozzle flow the diameter of funnel and then decreases after
rates of 97.2 kg/s, 121.5 kg/s and 162 kg/s. The height attaining a peak, suggesting the existence of an
of the domain has been changed in line with the funnel optimum funnel diameter. If the funnel diameter
height. increases then the low pressure created at the
centreline due to the high velocity jet is not felt
well at the periphery where the louvers are located,
so the mass ingress into the funnel falls after a
certain diameter. The centre line pressure drop is
insignificant when the funnel diameter is 7 m, so
there is less mass ingress into the funnel for this
case. But for funnel diameter of 4 m and 5 m the
centre line pressure drop becomes significant and
the mass ingress into the funnel rises (Isyumov and
Tanaka, 1979), (Jifei et al. 2005).
Effect of wind velocity on suction rate

Fig 8 shows the effect of ambient wind velocity on
mass suction rate (other operating parameters can
be seen from the plot). It can be seen from the plot
Fig. 6. Variation of air suction rate with height of the the mass ingress into the funnel decreases as the
funnel. ambient air velocity increases. Ambient wind
creates a low pressure field around the funnel. Due
All the three curves of mass suction have been to this low pressure field, the difference in pressure
reduced to a single curve by normalizing the mass across the funnel (inside and outside the funnel)
suction with the inlet nozzle flow rate. decreases causing less suction of air into the funnel.
The CFD analysis shows that increasing the height of So it can be concluded the entrainment of air into
the funnel, the air suction rate is increased. But the funnel decreases with increase in ambient wind
increase in height of the funnel is limited due to velocity.
design of other parameters as well as from economical
point of view. But it is suggested that the maximum
possible funnel height should be used.
Effect of funnel diameter on air suction rate

For a particular nozzle flow rate and louvers opening
area, it is predicted through the CFD analysis that, the
diameter of the funnel should be three times the
diameter of the nozzle to get maximum suction effect.
Fig. 8 Effect of ambient wind velocity on air

suction rate.
CONCLUSIONS
The procedure examined is a complete numerical
investigation of parametric study of air suction rate
through the louvers of a cylindrical funnel. The
conservation of mass and momentum equations
have been solved numerically with k- turbulence
Fig. 7 Variation of air suction rate with funnel model using unstructured grid to predict the air
diameter suction rate. Throughout the entire analysis the
Fig. 7 shows the variation of air suction rate with the funnel has been considered to be cylindrical in
diameter of the funnel. The nozzle diameter, louvers shape.
It has been found from the computation
opening area, protruding length of nozzle and height
i) Air suction rate into the funnel increases with
of the funnel can be read from Fig.7. The present
the increase in louvers opening area,
simulation is carried out with three nozzle flow rates.
ii) Air suction rate into the funnel increases with
The suction rate has been normalized with inlet nozzle
the increase nozzle flow rate
flow rate and the funnel diameter is normalized with

iii) Air suction rate into the funnel increases with the Equations, (NASA, Washington, D. C.,), 320-
increase height of the funnel. 345.
iv) There also exists optimum louver opening area Hamming R. W.,(1973): Numerical Methods for
on the funnel where the suction rate becomes the Scientists and Engineers, 2nd edition,
maximum as a function of the funnel diameter. (McGraw-Hill, New York).
Havelka P., Linek V., Sinkule J., Zahradnik J.,
NOTATION Fialova M.,(1997): Effect of the ejector
Lr- vertical distance between rows configuration on the gas suction rate and gas
minlet mass flow rate through the nozzle hold-up in ejector loop reactors. Chem Eng
msuction - air suction rate through the nozzle Sci; 52(11):1701–13.
Isyumov N., Tanaka H.,(1979): Wind-tunnel
Greek modelling of stack gas dispersion difficulties
ρ-density and approximations. Proceedings of the fifth
µ- laminar viscosity international conference on wind engineering,
µt-turbulent viscosity Fort Collins, Colorado, USA.
µeff - effective viscosity Jha P.K., Dash S.K.,(2002): Effect of outlet
- stress tensor positions and various turbulence models on
- rate of dissipation of k mixing in a single and multi strand tundish.
σk- turbulent Prandtle number for k Numer Int J Method Heat Fluid Flow;
σ - turbulent Prandtle number for 12(5):560–84.
Jha P.K., Dash S.K.,(2004): Employment of
Subscripts different turbulence models to the design of
r- radial direction optimum steel flows in a tundish. Int J Numer
- azimuthal direction Methods Heat Fluid Flow; 14(8):953–79.
Z – axial direction Jha P.K., Ranjan Rajeev, Mondal S.S., Dash
Lw- width of louver S.K.,(2003): Mixing in a tundish and a choice
of turbulence model for its prediction. Int J
REFERENCE Numer Methods Heat Fluid Flow; 13(8):964–
Acker H.G., (1952): Stack design to avoid smoke 96.
nuisance. Trans SNAME; 60:566–93. Jifei Gao, Guowei Gu, Zilong Zhao, Yalei
Anderson, J. D., (1990): Modern Compressible Flow, Zhang(2005): CFD numerical simulation
2nd ed., McGraw-Hill, New York,. applied in the design of the jet Aerator.
ANSYS User’s Manual for Revision 5.0, vol. I to IV. Environ Inform Arch;3:226–31.
Developed by Swanson Analysis Systems, Inc., Keller ,E. G. and Doherty, R. E.,(1936):
Houston, PA. Mathematics of Modern Engineering, Volume
Baham G.J., Mc Callum D.,( 1977): Stack design I, (Wiley, New York,), 163-188.
technology for naval and merchant ships. Trans Mishra D.P., Dash S.K.,(2010): Numerical
SNAME; 85:324–49. investigation of air suction through the louvers
Cambel, A. B., and Jennings, B. H.,( 1958): Gas of a funnel due to high velocity air jet ,
Dynamics, McGraw-Hill, New York,. International Journal for Computations in
Cramers P.H.M.R., Dierendonck van L.L., Beenackers Fluids, Elsevier; 1597-1608.
A.A.C.M.,( 1992): Influence of the gas density Nolan R.W.,(1946): Design of stacks to minimize
on the gas entrainment rate and gas hold-up in smoke nuisance. Trans SNAME; 54:42–84.
loop-venturi reactors. Chem Eng Sci; 47(9– Reddy , J. N.(1990): in Eshbach’s Handbook of
11):2251–6. Engineering Fundamentals, 4th ed. edited by
Davis H. T., Introduction to Nonlinear Differential B. D. Tapley, (Wiley, New York,), 2.145-
and Integral Equations, (Dover, New York, 2.168,2.191.
1962), 467-488. Research Laboratory, Office of Research and
Deeg, E. W., “Mechanics of AMPLIFLEX Connector Development, US Environmental Protection
Elements,” AMP J. of Technol. 4 (1994), pp 24 Agency, Research Triangle Park, NC 27711;
to 40. April 1981.
Fox W.R., McDonald A.T.,( 2004): Introduction to Snyder W.H., Guideline for fluid modelling of
Fluid Mechanics, Wiley India Edition, India. atmospheric diffusion Environmental Sciences
Frank White M.( 1991): Viscous fluid flow. New Thompson, P. A.,(1972): Compressible Fluid
York: McGraw-Hill International Co. Dynamics, McGraw-Hill, New York.
Girault ,V. and Raviart, P.A.,(1979): Finite Element Zucker,R.D. and Biblatz,O.,(2002): Fundamentals
Approximation of the Navier-Stokes Equations, of Gas Dynamics John Wiley & Sons, Inc. New
(Springer, Berlin,), 58-86. Jersey.
Goldstein M. E. and Braun W. H.,(1973): Advanced
Methods for the Solution of Differential

ISBN 978-81-921121-0-7
EVOLUTION OF SOME RESISTANCE LAW FOR NON-UNIFORM

ACCELERATED FLOW OVER CLOSELY PACKED ROUGH BED
Kesheo Prasad
Assistant Professor, Department of Civil Engineering, IT-BHU, Varanasi, email: kesheo_bhu@yahoo.com
C.S.P. Ojha
Professor, Department of Civil Engineering, IIT Roorkee, Roorkee, email: cspojha@rediffmail.com
K. M. Singh
Assistant Professor, Department of Mechanical and Industrial Engineering, IIT Roorkee, Roorkee
ABSTRACT: In the present study, experiments were conducted on a closely packed array of hemispheres fixed on
the bed of a channel. Three-dimensional velocities were measured using Acoustic Doppler Velocimeter (ADV).
Effect of the turbulence on discharge measurement is also attempted using resistance laws of Colebrook, Bathurst,
Swamee and Manning. It is seen that in case of fully rough turbulent flow (ks+ >70), Swamee’s resistance law
estimates discharge more precisely. Further, a modified form of Griffith’s equation is also proposed.
INTRODUCTION
Traditional hydrodynamic models assume that flows in mechanism occurring at the wall. Turbulent flow
natural hilly regions are uniform and unidirectional produced by rough bed surface is complex in nature.
(Papanicolaou, 2002) as in mountain streams. Further, it Hence, it has remained area of interest of the engineers
is assumed that the water flows through the gravel layer since long time. It has been studied for non-uniform
in turbulent flow conditions, because mountain-river flow conditions with gradually accelerating flows on
has a bed with steep slope and relatively large gravels. rough surface by Cardoso et al. (1991), Kironoto and
The factors which affect the velocity profile along the Graf (1995). Keulegan (1938) suggested that the
depth of water are cross section, aspect ratio and logarithmic law can be used for estimation of rough
roughness type. Swamee et al. (2002) have described open channel flow velocity. In general, a relation
the design procedure for the optimal cross section. The between the dimensionless velocities and the physical
discharge from any cross-section can be estimated parameters can be represented by the power law:
using law of conservation of mass for large R / k values
u 1/ 7
Q  av (1)  10.2  y / k  (3)
u
where a is cross section area, and v is the average flow
velocity in the direction of flow. The right hand side of where, R is the diameter of pipe. Equation (3) is
Eq. (1) mainly depends on the accuracy of the estimated simple, but range of R / k was not defined. Hence for
value of v. The average flow velocity v is itself a
central vertical velocity distribution, a parabolic law
function of various hydraulic parameters. The flow over
was introduced as
the gravel layer follows the law of flow resistance with 2
the roughness of bed material. The Manning’s equation u m  u   6.3 1  y  (4)
1 where u is shear velocity and
v  R 2 / 3 S 1/ 2 (2)
n
is most commonly used for uniform flow due to its um  um / u and u   u / u* (5)
simplicity and acceptable degree of accuracy in most Y
Water surface
discharge measuring techniques. Ganguillet and Um
Flow depth, h
Kutter’s, and Bazin’s formulas have largely gone out of

use (Swamee et al. 2002). Equation (2) depends on
physical parameters, R and S, of the channel along with 
a suitable value of roughness coefficient, n. Manning u
has classified the different types of roughness as shown y Virtual bed level
in Table 2 (Rangaraju 1993). yo k ks X

Fig. 1: Definition sketch, virtual bed level, rough
Flow in hilly region rivers remains fully in turbulent surface height and equivalent height.
condition. Very often in such cases, the transport rate of
a scalar contaminant is also fixed by transport

In Fig. 1, Um and  defines the maximum velocity and that velocity defect law is valid for y/>0.03. Swamee
the boundary layer thickness, respectively, u is the local (1993) proposed an explicit equation for velocity
velocity at a distance y, k is the height of roughness profile for smooth as well as rough surface which
element, ks is the equivalent roughens height and y0 is covers entire inner region as
the distance of virtual bed level from bed of the channel.  10/ 3

0.3
10/ 3   9 y   (12)
u    y     k 1 ln 1    
Vanoni (1941) also observed identical vertical velocity    1  0.3   
curves of the central region of a channel. In addition to  um
this, he also stated that there always exists a central where    , in which  is the average roughness

region, in which the logarithmic law extends up to the
height of the surface. Further, Swamee et al. (1994)
free surface, when w/h is less than 5. Colebrook (1939)
developed an average velocity equation for the uniform
proposed an empirical relationship between hydraulic
flow over a rigid boundary known as resistance
parameters and depth of flow as
equation given by
 14.8 y 
v  31.6 gRS log10   (6)   0.221 
 ks  v  2.547 gRS0 ln    (13)
 12 R R gRS 
Coles (1956) introduced a purely empirical expression  0 

to predict velocities in the fully turbulent flow ( k s > where g is gravitational acceleration and So is channel
bed slope. Note that the Manning equation (2) is
70) for inner region as well as for outer region applicable to fully rough turbulent flow for the limited
u 1 u y   y bandwidth of relative roughness (Swamee et al. 1994)
 ln *  A  w   (7)
u*   k   given by
where = 0.55, K = 0.4, A = 5.1, and w( y /  ) is law of 30
 0.4
(14)
the wake given by 2sin 2  y / 2  for zero pressure 
gS o Q 
gradient. Clauser (1956) correlated (um  u ) / u* with 
0.004   0.04 (15)
y/, and concluded that the turbulent velocity will R
collapse to a single curve. Vittal (1972) and Swamee On the other hand, Eq. (13) which involves physically
(1994) stated that a single logarithmic law is not conceivable parameters  and v, is applicable for
applicable over the entire depth of flow. Griffiths (1980) hydraulically rough or smooth or transitional flows.
proposed an empirical resistance equation as Also, Eq. (13) holds good for open channel sections in
 9.68 R  VR
v  31.36 log10   (8) the ranges of 103   108 (Hager, 1989). Afzalimehr
 ks  v
Bathurst (1985) stated a resistance equation for channel and Anctil (2000) suggested that due to uncertainty in
with boulders on bed having the ratio of depth of flow the measurement of the distance from the bed y and in
to d84 <5 the estimation of the shear velocity, u* , the law of the
 15.44 y  wall can be rewritten as
v  31.6 gRS log10   (9)
 ks  u 1  y  Ac d50 
Kirkgoz (1989) determined mean velocity using  ln   A (16)
u*   ks 
um  u 1 y   y 
 ln  2 cos 2   (10) where K is the von Karman constant(=0.4), Ac d50 is
u*     2 
reference level, Ac is the value producing best
in fully developed turbulent flow in open channels
using LDV and determined v from these data by correlation coefficient for the regression, A is constant
shifting the virtual bed level at a distance yo from the of integration, ks = d50 is Nikuradse grain roughness. In
bed. In addition to this, he also explained that the addition to this, they also observed that maximum
velocity in the outer region is mainly controlled by the velocity occurs at y/h =0.6. Sarma et al. (2000) stated
turbulent shear and velocity distribution. He estimated that the velocity distribution in the inner and outer
region of a boundary layer can be described by the log
u* using law and the parabolic law, respectively.
u*  ghS (11)
for steady uniform flow in a wide channel. Equation (10) Equations (3), (4), (5), (6), (10) and (16) clearly show
is described by velocity defect law and Coles parameter that u* fundamentally affects the velocity profile.
was estimated as  =0.2 for rough surface. He stated Hence, precise estimation of the shear velocity is
required (Nezu and Nakagawa, 1993). The estimation

of shear velocity has been proposed by several At entrance, honey comb wall of small bricks was
researchers. Chow (1959) proposed the friction slope constructed to reduce the large scale disturbances in the
application for uniform flow, and shear velocity approaching flow. A floating wooden baffle of 1m was
u*  gRS (17) used to minimize the disturbances at the free surface. A
movable carriage with pointer gauges (the pointer
where S is the bed slope. Use of conservation of
gauge arrangement had a least count of 1mm) was
momentum equation leads to the following expression
mounted on a pipe rail at the top of the channel to
for energy slope
measure levels of water. The slope of the channel was
dy
Se  S0  1  Fr2  (18) measured using two containers, connected at their
dx bottom by a long plastic tube, placed on the channel
Equation (17) may be rearranged for non-uniform flow bottom one each at two predetermined locations along
as the length of the channel. Care was taken to eliminate
u*  gRS e (19) air bubbles in the plastic tube. Nearly 18 hours time
was allowed for equalization of the water level in both
Further, Clauser (1956) described another method to
the containers and then water levels in both the
estimate the shear velocity from the parameters of the
containers was noted by means of the pointer gauges
regression of the mean point velocity u and ln[(y+yo)/ks]
mounted on the rails. The difference in the gauge
as
reading represents the vertical drop in the channel. The
u 1  y  yo  incoming discharge was measured continuously using a
 ln   B (20)
u*   k s  Portable Transit Time Flow meter and the strength of
sound was checked and found more than 30DB during
y  yo
u   ln( )B (21) the experiment.
ks
This leads to In this experiment, six velocity profiles were measured
u*  k (22) for one discharge. Three dimensional velocities at a
and confirms the validity of the law of the wall for the point were measured using downlooking intrusive
inner region of the flow, where the regression is fitted instrument Acoustic Doppler Velocimeter with
only to the lower fifth of the velocity profile. sampling rate 10 Hz for a length of time series 150s.
Thus, a total of 1500 samples were collected at six
EXPERIMENTAL SETUP locations (three sections at mid channel and three at
The experiments were carried out in a 10.5m long, 10.5cm away from the side wall) at 4.815m, 5.275m
0.46m wide, and 0.60m deep using a horizontal and 5.505m from inlet in the longitudinal direction as
rectangular laboratory flume at the Department of Civil shown in Fig. 3. The temperature and salinity of water
Engineering, IIT Roorkee. Two sides of channel were during the experiment was measured and found to be 20
made up of glass and the bottom of the channel was of degree centigrade and 20 ppt respectively.
steel base with 5mm fiber on top of it. The flow in the
flume was supplied through a pipe of 202 mm outer
diameter through the pump. The water depth in the
flume was regulated using a tailgate at the end of
channel. To create rough turbulent flow, hemispheres of
cement mortar in 2.3cm height were prepared and fixed
on bed of channel in 1.5m length in a reach of 1.5m at
4.5m from the upstream end of the flume.
Fig. 3: Schematic diagram of vertical sections on rough

bed.
Speed of sound at 20 degree centigrade was recorded.

The velocity range of flow was kept fixed at 100 cm/s
since the variation of actual local velocity was between
70cm/s to 95cm/s.
Fig. 2: Arrangement of hemispheres on channel bed
Due to technical restriction of ADV, 3-D velocities data
using downlooking ADV were restricted beyond 7cm

below from the top of water surface. It was found that In addition to this, the dependency of ks on yo is also
water formed a vortex around the rod of ADV at water examined it is found that ks follows a power law as
surface which created an air cavity around the rod of shown in Fig. 5. Further, variation of yo with  and ks
ADV. The transducer of ADV in 7cm from water with best R2 is shown also in Table 3.
surface produced erroneous data. Hence, to collect error
free data the down looking probe was kept 3cm below 0.07
50H M1
water surface and 7cm deep section water surface was 0.06
measured using upward looking ADV probe. Table 1 0.05

y = 153.32x 2 - 6.1756x + 0.0619
shows the hydraulic parameters at 5.275 m from inlet of 0.04
channel. 0.03
ks
Table 1: Summery of Flow Parameters and Cross 0.02
Section Geometry 0.01
Q (m3/s) Re (x105) Fr Bed Slope w/h Ratio 0.00
-0.01
0.05 1.067 0.62 0.000717 2.71 0.0000 0.0050 0.0100 0.0150 0.0200 0.0250
yo
Table 2: Manning’s coefficient for different types of

roughness Fig. 5: Value of ks versus yo for rum, 50H M1
Surface Recommended value
Table 3 shows the variation of ks with yo affected by
Glass, plastic Brass 0.010
parameter α along with correlation coefficient and the
Timber 0.011-0.014
most correlated value yo=0.0144 and ks (Table 3 column
Cement plaster 0.011 4) is adopted.
Cast iron 0.013
Concrete 0.012-0.017 Table 3: Estimation of u* , yo and ks
Drainage Tiles 0.013 yo  ks R2
Rubble Masonry 0.017-0.025 (i) (ii) (iii) (iv)
Rock cuts 0.035-0.040
0.0000 0.000 0.0623 0.9862
0.0029 0.125 0.0450 0.9881
The velocity distribution data over the rough bed of 0.0058 0.250 0.0310 0.9895
closely packed hemispheres were collected and 0.0086 0.375 0.0194 0.9916
analyzed in the present study. The velocity distribution 0.0115 0.500 0.0109 0.9931
in a two dimensional uniform open channel flow 0.0144 0.625 0.0057 0.9937
usually follows Eq. (20) in the inner region. Hence, the 0.0173 0.750 0.0019 0.9916
velocity profiles in two regions: (i) Inner region and (ii)
0.0201 0.875 0.0004 0.9791
Outer region are examined using estimated value of ks
from Eq. (18) as shown in Fig. 6. Along with the 0.0230 1.000 0.0000 0.8188
virtual bed level shifting yo, the dependency of u* on
 is also studied for all set of observed data and it is The estimated u* from slope of best fit along with u*
found that in most of the cases   0.625 gives the obtained using Eq. (11) is shown in the Table 4 with the
satisfactory results with correlation coefficient of 99% estimated ks. To examine the velocity profile the shear
as shown in Fig. 4. From a linear relation as shown in velocity obtained from Eq. (11) is considered in the
Fig. 4, ks is determined. present study
Table 4: Estimated u* from slope of best fit and Eq. (11)
0.9
0.8 u*  ghS u* using slope of
0.7 Run ks (m)
0.6
y = 0.2068Ln(x) + 1.4226 (m/s) best fit (m/s)
R2 = 0.9897
u (m /s)
0.5 50H M1 0.034508 0.058 0.0057

0.4
0.3
50H M4 0.034508 0.047 0.0027
0.2 50H M6 0.034508 0.043 0.0011
0.1 50H SR1 0.034508 0.072 0.0140
0
0.01 0.1 1
50H SR4 0.034508 0.042 0.0011
(y+yo) (m)
50H SR6 0.034508 0.066 0.0017
Fig. 4: Variation of u with y+yo for run 50H M1

Fig. 6 shows relation between u+ versus y+ for the test underestimates the discharge. The Swamees equation
run 50H M1 and 50H SR1 to check the validity of the (12) performs well, even though it is deviates in the
log law. outer region due to dip formation.
While, Fig. 7(a & b) shows y/h versus u / um for run
0.18
50H M1, 500VPs and 600Vps data. 0.16 50H M1
0.14
28
50H M1 0.12
26 0.1
y (m)
0.08
24
0.06 OB data
22 Swamee
0.04
u+
Bathurst
20 0.02 colebrook
Manning's
18 0
0 0.2 0.4 0.6 0.8 1 1.2
u (m/s )
16
14 (a)
100 1000 1 0000
y+ 0.18
50H M4
(a) 0.16
0.14
28
50H SR 0.12
26 0.1
y (m)
24 0.08
0.06 OB data
22
Swamee
y+
0.04
Bathurst
20 coolbrook
0.02
M annings
18 0
0 0.5 1 1.5 2 2.5
u (m/s)
16
14
(b)
100 1000 10000 0.16
u+
0.14 50H M6
(b) 0.12
Fig. 6: Validation of the logarithmic; (a) Run 50H M1 0.1
and (b) Run 50H SR1.

y (m)
0.08
0.06
A wall region in central line region depicts that inner 0.04 OB data
Swamee
layer lies below y/h<0.28. However, near the side wall, 0.02
Bathurst
Colebroo k
maximum velocity is available at 0.42y from water 0

0 1 2 3 4
Mannings
surface; the reason of this dip may be the secondary u (m/s )
wave movement. (c)

0.18
0.16 50H SR1

1 0.14
5 00 VP s
0.12
6 00 VP s
0.8
5 0H M1 0.1
y (m)
0.08
0.6
0.06
y/h
OB data
0.04
Swamee
0.4 Bathurst
0.02 Colebrook
M annings
0
0.2 0 0.2 0.4 0.6 0.8 1 1.2
u (m/s )
0
0 0.25 0.5 0.75 1
(d)
u/um
0.18
50H SR4
0.16
0.14
Fig 7: velocity profile for 50H M1 along with the data 0.12
[500VPs and 600 VPS] of Afzalimehr et al. (2010). 0.1

y (m)
0.08
0.06
From Fig. 8 (a,b,c) at central line of channel and from 0.04
OB data
Swamee
Fig. 8(d,e,f) at 10.5cm away from the side wall of 0.02

Bathurst
Colebrook
Mannings
channel, it is clear that under the turbulent conditions, 0
0 1 2 3 4 5 6
Manning’s equation (2) either over- or under-predicts u (m/s )
the discharge, while Bathrust and Colebrook equation (e)

0.18
50H SR6
The velocity profile 50H M1 is also plotted (Fig. 9) to
0.16
understand the variation of velocities by various
0.14
0.12
researchers. The measured discharge using modified
y (m)
0.1
Griffiths equation (23) is shown in Table 5 column 7,
0.08 which gives the results in the acceptable limit.
0.06
OB data
0.04 Swamee
Bathurst
CONCLUSIONS
0.02 Colebrook
Mannings The following conclusions are drawn from this study
0
0 0.2 0.4 0.6 0.8 1 1.2
u (m/s) 1. Swamee’s equation provides good results as
(f) compared to other available resistance equations.
Fig. 8: Velocity profile at; (a,b,c) central line, and (d,e,f) 2. Bathurst and Colebrook equation provides under
10.5cm from right side wall. estimated discharge in most of the cases.
3. Accurate estimation of u* , yo and ks is needed to
0.18
0.16 50H M1
use these parameters in velocity measurements.
0.14
4. Even modified Griffiths equation (23) does not
0.12 provide as precise estimate of discharge as
0.1 Swamee’s equation. But in absence of proper u* , it
y (m)
0.08
0.06 OB data
provides comparatively good results.
Swamee 93
0.04 Bathurst
colebrook
0.02
Manning's
swmee 94
NOMENCLATURE
Mod Griffths
0
0 0.2 0.4 0.6 0.8 1 1.2
U = Average mean velocity
u (m/s) y = Depth of flow in vertical direction
 = kinematic viscosity
Fig. 9: Velocities at run 50H M1 by various researchers.  = von Karman constant=0.4
R = Radius of pipe
The above discussed resistance equations are examined
for nonuniform open channel flow and the results are u = Shear velocity
shown in the Table 5. um= Maximum velocity of flow
 = Distance at maximum velocity from bed
Table 5: Estimated discharge using resistance
 = Average roughness height
Equations
Discharge (m3/s)
N = Number of observation
Run Eq. Eq. Eq. R= Hydraulic mean depth
Actual Eq. (2) Eq. (23) S0= Bed slope of channel
(13) (9) (6)
50H
0.050 0.044 0.026 0.026 0.067 0.079 yo = Virtual depth from bottom of channel
M1 h= Depth of flow in cannel
50H
M4
0.050 0.042 0.026 0.026 0.127 0.082 u= local mean velocity
50H d50=Mean size of particle
0.050 0.045 0.028 0.028 0.287 0.090
M6 d84= 84% finer
ks= Equivalent roughness height
Fig. 8 (a,b,c,d,e,f) depicts that the behavior of the S = energy slope line
Manning’s equation, Eq. 2 mainly depends on the Fr= Froude number
estimated value of the ks and Swamees equation, Eq. 13 Re = Reynolds number
may give measured discharge under turbulent flow  = Constant
condition.
REFERENCES
An attempt is also made to use the Griffith equation Afzalimher H. and Anctil, F. (2000). “Accelerating
with the modification in constant value on the basis of shear velocity in gravel-bed channels.”
arithmetic average value. The velocity profile suggested Hydrological Sciences, J. -des Sciences
for determining the velocity profile in turbulent flow Hydrologiques, 45(1), 113-124.
conditions is proposed as Afzalimher H., Najfabadi, EF, Singh, VP, (2010),
 9.68 R  “Effect of Vegetation on Banks on Distributions of
v  0.26 log10   (23) Velocity and Reynolds Stress under Accelerating
 ks  Flow”, J. Hydrologic Engineering, 15(9), 708-713.

Afzalimehr, H., and Dey, S. (2009). “Influence of bank Turbulent Boundary Layers.” The Indian Society
vegetation and gravel bed on velocity and for Hydraulics J. Hydraulic Engineering, 8(1) 50-
Reynolds stress distributions.” Int. J. Sediment Res., 59.
24(2), 236–246. Swamee, PK, Mishra, GC, Chahar, BR, (2002),
Bathurst J.C. (1985) “Flow Resistance Estimation in “Optimal Design of Transmission Canal”, J.
Mountain Rivers”, J. Hydraulic Engineering, Irrigation and Drainage Engineering, 128 (4),
111(4), 625-643 234-243.
Cardoso, AH., Graf, WH., and Gust, G. (1991). “Steady Vanoni, V. A. (1941). ‘‘Velocity distribution in open
gradually accelerating flow in a smooth open channels.’’ Civ. Engrg., ASCE, 2(6), 356–357.
channel.” J. Hydraul. Res., 29(4), 525–543. Vittal. N (1972). “Flow over triangular roughness in
Chow V.T (1959). Open Channel Hydraulics, McGraw- open channels.” Ph. D. Thesis, Civil Engineering
Hill College. Department, IIT Roorkee.
Clauser, FH. (1956) “The turbulent boundary layers.”
Adv. Appl. Mech. 4, 1-15.
Colebrook, CF (1993), “Turbulent flow in pipes, with
particular reference to the transmition region
between the smooth and rough pipes laws, J. Instn
Civ Engrs, 11, 133.
Coles D. (1956). “The law of the wake in the turbulent
boundary layer.” J. Fluid Mechanics, 1, 191 - 226.
Hager, W. H. (1989). “Discussion of Noncircular sewer
design.” J. Environ. Eng., 115(1), 274–276.
Hey, R. D. (1979). "Flow resistance in gravel-bed
rivers." J. Hvdraulic Division, ASCE, 105(4), 365-
379.
Griffiths G.A. (1981) “Flow Resistance in coarse gravel
bed rivers” Proceedings ASCE, Journal of
Hydraulic Division, 107(7), 899-918
Keulegan C.H. (1938). “Laws of turbulent flow in open
channels.” J. Research National Bureau of
Standards, 21, 707-740.
Kirkgöz, M.S. (1989). “Turbulent Velocity Profiles for
Smooth and Rough open Channel Flow.” J.
Hydraulic Engineering, 115(11), 1543-1561.
Kironoto, B., and Graf, W. H. (1995). “Turbulence
characteristics in rough nonuniform open-channel
flow.” Proc. Inst. Civ. Eng., Waters. Maritime
Energ., 112, 316–348.
Nezu, I., and Nakagawa, H. (1993). “Turbulence in
open-channel flows”. A. A. Balkema, Rotterdam,
The Netherlands.
Ranga Raju K.G. (1993) Flow through open channels,
Tata McGraw-Hill.
Papanicollaou, A.N. and Hilldale, R. (2002).
“Turbulence Characteristics in Gradual Channel
Transition.” J. Engineering Mechanics, 128(9),
948-960.
Sarma V. N. Kandula, Pasad B.V.R. and Sarma A K
(2000). “Detailed Study of Binary Law for Open
Channel.” J. Hydraulic Engineering, 126(3), 210-
214.
Swamee, P.K., (1993). “Generalized inner region
velocity distribution equation.” J. Hydraulic
Swamee, P.K., ISH, M. and Pathak S. K. (1994).
“Velocity Distribution Equations for Laminar and

ISBN 978-81-921121-0-7
FACTORS AFFECTING BEHAVIOUR OF PILED RAFT

FOUNDATION
Rajendra Singh Bisht
PG Student, Department of Civil Engineering, Indian Institute of Technology, Guwahati – 781039,
E-mail: r.bisht@iitg.ernet.in
Baleshwar Singh
Associate Professor, Department of Civil Engineering, IIT Guwahati, Guwahati – 781039, E-mail: baleshwar@iitg.ernet.in
ABSTRACT: Piled raft foundation is considered as suitable alternative for foundation in tall buildings instead
of conventional pile foundation. When raft foundation alone cannot satisfy both bearing capacity and settlement
criteria then piles are added to enhance the performance. The primary function of piles in piled raft foundation is
to reduce or control overall and differential settlement. The load sharing between raft and piles as well as total
and differential settlement in a piled raft foundation depends on number of piles, pile diameter, pile length, pile
spacing, location of piles under raft and raft thickness. So in order to control overall and differential settlement
within an acceptable limit and also make piled raft foundation more cost effective as compare to conventional
pile foundation the pile group system geometry and raft geometry have to be designed optimally. In the present
paper, a review of various studies related to effects of pile group system and raft thickness on behaviour of piled
raft is presented.
INTRODUCTION
A piled raft foundation is combination of a raft and a Randolph (1994) proposed three different design
deep foundation. When a raft has adequate bearing approaches for piled raft foundation. Depending on
capacity but the settlement is not within an allowable primary function of piles in piled raft foundation the
limit in that case instead of a fully pile foundation the differential settlement control design approach
piled raft foundation has been proved more efficient suggests that the piles are designed in such a way as
and economical. The piles not only control or reduce to minimize differential settlement at the outset
overall and differential settlement but also increase without necessarily reducing the overall settlement.
the bearing capacity of foundation. Overall settlement The effects of pile group system and raft geometry on
has less effect on the performance of a structure but it piles load carrying capacity and settlements have been
is differential settlement which is main concern in any studied by several researchers. A review of various
foundation. studies is presented in this paper.
There are complex soil structure interactions that take EFFECT OF PILE SPACING
place between raft piles and soil in a piled raft The pile spacing is one of the factors that affect the
foundation. The load is shared between piles and raft behaviour of piled raft. Hain and Lee (1978) studied
and then transferred to the soil. The proportion of load the effect of piles spacing on percentage of total load
shared between piles and raft and the level of carried by the piles for a piled raft system containing
reduction of overall and differential settlement depend flexible raft and compressible piles. Figure 1 shows a
on piles group system geometry (i.e. pile length, pile plot between spacing to diameter ratioand percentage
spacing, pile diameter, number of piles and location of of load taken by the piles for different pile and raft
piles) and raft geometry (i.e. length, breadth and stiffness.
thickness of raft). Here a group of 82 piles were used with length
todiameter ratio, L/d= 25 and Poisson’s ratio of soil
In conventional method of design the piled raft 0.5. The percentage of total load carried by the piles
foundation is designed in such a way that the entire decreases with increase in pile spacing, and increases
load is carried by piles and neglecting the contribution as the pile stiffness increases. However, when the raft
of raft, this approach results in installation of more stiffness is increased the load taken by the piles
piles than that are necessary which will completely decreases.
eliminate settlements rather than control it to an Prakoso and Kulhawy (2001) used a nonlinear plane
acceptable level and it is also uneconomical. The strain finite element model to find the effects of
conventional approach ignore primary objective of system geometry elements on the performance of
addition of piles to piled raft foundation and treats piled raft. They examined the effect of pile spacing on
settlement as secondary issue.Therefore a major total and differential settlement. The piles and raft
design question is how to design piles optimally to parameters used for analysis are listed in Table 1
control the settlement.

The displacement ratio decreases with the decrease in
pile spacing but also the displacement ratio is more
for low pile group to raft width ratio (Bg/Br). So to
minimize the total settlement of piled raft both pile
spacing and pile group to raft width ratio should be
considered.
The effect of pile spacing on center to edge

normalized differential settlement, ΔWC-E is shown in
Figure 3. The differential settlement decreases with
decrease in pile spacing and for certain Bg/Br ratio at
less spacing the differential settlement, ΔWC-E even
become negative i.e. the edge of raft settles more than
the center.
Fig. 1 Percentage of load taken by pile group versus

spacing to diameter ratio (after Hain and Lee 1978)
Table 1 Pile and raft dimensions

Parameters Magnitude (m)
Raft thickness, tr 2
Raft breadth, Br 42
Raft length, Lr 42
Pile depth, D 40
Pile spacing, S 4
Pile diameter, B 1
Height of soil layer, H 80
Fig. 3 Differential settlement versus pile spacing
The pile group to raft width ratio used were in the (after Prakoso and Kulhawy 2001)
range Bg/Br= 0.2 to 1. Here the settlement is
introduced as reference displacement, Wref which
represents a nominal average type of displacement EFFECT OF NUMBER OF PILE
along the raft by considering the displacements at the Poulos (2001) used a hypothetical case of a raft
raft centre, ¼ point, and edge. The differential supported by nine piles and carrying a total load of 12
settlement considered is the center edge differential MN to find out the effect of number of piles in the
displacement, ΔWC-E. Figure 2 shows a plot between behaviour of piled raft. A computer program was used
normalized reference displacement ratio, Rref (Wref for for the analysis. In this case the raft was divided into
piled raft/Wref for raft) and inverse pile spacing at 273 elements and the piles were assumed to exhibit
different pile group to raft width ratios. elastic-plastic load-settlement behaviour and its
stiffness and interaction characteristic were computed
from another computer program.
The various dimensions of piles, raft and soil

properties are given in Table 2.
Table 2 Pile and raft dimensions
Parameters Magnitude (m)
Raft thickness, tr 0.5
Raft length 10
Raft breadth 5
Pile length, l 10
Pile diameter, d 0.5
Pile spacing, S 2
Height of soil layer, H 20
Fig. 2 Reference displacement ratio versus inverse
pile spacing for different Bg/Br (after Prakoso and
Kulhawy 2001)

The effects of number of piles on maximum corner piles shows higher value as compared to the
settlement and maximum moment for both case of uniform loading.
concentrated and uniform load are shown in Figure 4.
The number of piles also affects the percentage load
The overall settlement decreases with the increase in carried on piles. The percentage load taken by piles
number of piles but after a limit the there is no further increases with the increase in number of piles and
decrease. The decrease is similar for both achieved maximum value for certain number of piles.
concentrated and uniform loading. Howeverthe Once the maximum value is reached after this the
increase in number of piles increases the maximum percentage load on piles shows a constant value. So
bending moment for both concentrated and uniform there is a certain higher limit of number of piles
loading however maximum bending moment is more beyond this limit further increase in number of piles
for concentrated loading. For more than 9 piles there does not have any effect on percentage load taken by
is not much variation in maximum moment with piles. The type of loading does not have any effect on
increase in number of piles and smallest moment the percentage load on piles. The percentage load
occurs when there are only three piles under raft. taken by piles is similar for concentrated and uniform
loading.
Fig. 4 Effect of number of piles on maximum

settlement and maximum moment (after Poulos 2001) Fig. 5 Effect of number of piles on differential
settlement and percentage load on piles (after Poulos
Figure 5 shows the effect of number of piles on 2001)
differential settlement between centre and corner piles
and percentage load on piles. Unlike overall Bajad and Sahu (2008) conducted some experiments
settlement the effect of increasing in the number of to find out the effect of number of piles on the
piles on differential settlement between centre and settlement of piled raft on soft clay and used the
corner piles is non-uniform.The differential settlement concept of equivalent pier to analyse piled raft. A
does not decrease with increase in number of piles. small scale model tests were conducted on piled raft
The minimum differential settlement is achieved using a soil bin made of mild steel and solid steel
when only three piles are present under the raft. So in piles. They used a cohesive soil having liquid limit of
order to minimize differential settlement providing 57% and plastic limit 27% with undrained shear
less number of piles below a central portion of raft strength of 10±1kPa. To examine the effect of number
can be more efficient. In the case of concentrated of piles on the settlement and load carried by piles the
loading the differential settlement between centre and equivalent modulus, Eeq, given in eq. (2) settlement

reduction ratio, Sr, given in eq. (1) and load sharing Fig. 6 Effect of number of piles on settlement of piled
ratio, αpr given in eq. (3) were used. raft (after Bajad and Sahu 2008)
Settlement reduction ratio, Sr=δr-δpr/δr …… EFFECT OF PILE LENGTH

(1) Horikoshi and Randolph (1998) used a concept of
Where δr and δpr are the settlement of raft and piled equivalent pier to find the effect of pile length on
raft respectively for a given load. piled raft behaviour. A square piled raft with group of
nine piles was modelled beneath a flexible raft and
Equivalent pier modulus, Eeq=ES+ [EP-ES]*Atp/Ag elastic analyses were performed with different pile
... (2) length and the piles were set at s/b=0.4, where s is
Where EP is the young’s modulus of pile and ES is the spacing between piles and b is half the length of raft.
young’s modulus of soil, Atp is total cross sectional The piled raft stiffness, Kpr was kept constant at 1.02.
area of the piles in the group and Ag is the gross cross Figure 7 shows a plot between differential settlement,
sectional of the pile group. ΔW* and relative pile length, LP/aeq . where LP is the
Equivalent pier diameter, deq=2/√π*√Ag ….. length of pile and aeq is the equivqlent circular raft
(3) radius for rectangular raft of length L and breadth B
where Ag is the gross cross sectional area of the pile and aeq= (BL/π). There is not much variation in
group. differential settlement with the change in pile length
for a constant stiffness of piled raft, Kpr the
The various dimensions of soil bin, pile and raft are differential settlement does not change with change in
given in Table 3. pile length. So long piles are ineffective in reducing
differential settlement.
Table 3 Dimensions of soil bin, pile and raft
Dimensions Magnitude (mm)
Soil bin
Thickness 3
Diameter 570
Height 500
Pile
Diameter 10
Length 100 and 200
Raft
Length 100
Breadth 100
Thickness 10
Figure 6 shows a plot between settlement reduction Fig. 7 Effect of pile length on differential settlement
ratio and equivalent pier modulus at a working load of (after Horikoshi and Randolph 1998)
400N and 250N. The settlement reduction ratio, Sr
increases as the equivalent pier modulus, Eeq increases The effect of pile length on total settlement was
i.e. as the number of piles increases but the reduction studied by Prakoso and Kulhawy (2001). A plane
in settlement is there for certain number of piles and strain finite element model was analysed using
further increase in number of piles does not affect the geotechnical finite element code PLAXIS and a six-
settlement reduction ratio. node triangular element was used.The various
raft,piles and soil properties are given in Table 1. The
effect of pile depth, D on normalized
referencesettlement ratio, Rref(Wref of piled raft/Wref
for raft ) was examined for different pile group to
raftwidth ratio,Bg/Br. Figure 8 shows that the
reference

Fig. 8 Effect of pile depth on reference displacement Fig 10 Pile diameter versus differential displacement
ratio (after Prakoso and Kulhawy 2001) (after Prakoso and Kulhawy 2001)
displacement decreases as the pile depth increases and Oh et. al. (2009) analysed a piled raft foundation on
the decrease is more when a large width pile group is sand using the PLAXIS. A six-layer soil model was
used. the minimum value of refrence settlement is adopted in which there are five different soil layers
obtained with increase in pile depth when width of and layer six is assumed to be rock layer below 30 m.
pile group is equal to the width of raft. The effect of pile diameter on differential settlement
and pile load was examined for three different load
EFFECT OF PILE DIAMETER intensities using a piled raft with thickness of 0.6m
Prakoso and Kulhawy (2001) also studied the effect of and the pile group size is 32. The pile diameters are
pile diameter on refrence displacement ratio and changed as 0.6, 0.8and 1m. the spacing between piles
differential displacement.the various raft,pile and soil are kept as 4d butthe pile group area and raft area
properties are given in Table 1. Figure 9 shows increases due to the increase in pile diameter.
thatthe refrence displacement ratio decreases with
decrease in pile diameter but the decrease in the value Figure 11 shows that the normalised differential
of refrence settlement is very less amd even the settlement increase with the increase in pile diameter.
change in Bg/Br ratio does not show much variation so For a less load intensity the increase in differential
the smaller diameter piles are preferred for reducing settlement is not much significant but for higher load
refrence settlement. intensity the increase in normalized differential
settlement with increase in pile diameter is very high.
Fig. 9 Pile diameter versus reference displacement

ratio (after Prakoso and Kulhawy 2001)
Figure 10 shows the effect of pile diameter on Fig. 11 Normalized differential settlement versus pile
normalized differential displacement. The differential diameter (after Oh et al. 2009)
displacement decreases slightly as pile diameter
increases but for less value of Bg/Br and bigger pile The effect of pile diameter on normalized total pile
diameter the differential settlement even become load,Rg/qB2 where Rg is the pile load, q is the load
negative . so smaller diameter piles are preferred to intensity and B is the breadth of raft is shown in
minimize settlement effeciently and economically. Figure 12 for different load intensities.

diameter of 0.8 and then decreases with further
increase in pile diameter. The reason for this
particular behaviour is that as pile diameter increases
the pile spacing also increases and the pile-pile
interaction become less effective and piles in piled
raft behaves as single pile.
EFFECT OF RAFT THICKNESS

The effect of raft thickness on piled raft behaviour is
studied by poulos (2001). To analyse a piled raft
foundation a computer program was used. The various
parameters of piled raft foundation used in analysis is
already given in Table 2. The effect of thickness of
raft on maximum moment and differential settlement
between centre and corner column is shown in Figure
Fig. 12 Pile diameter versus normalized total pile load 13.
(after Oh et al. 2009)
The total pile load increases with increase in pile

diameter and reaches at maximum value at the pile
The maximum moment increases with increase in raft
thickness but when there is no piles under raft the
bending moment decrease as raft thickness increases.
the larger piles group showing more increase in
maximum moment as compared to small pile group.
The differential settlement showing a positive
response with increase in raft thickness there is a
decrease in differential settlement between centrer and
corner column. Fig. 13 Effect of raft thivckeness on differential
settlement and maximum moment (after Poulos 2001)
CONCLUSIONS
The pile spacing, number of piles in a group, pile
diameter, pile length and raft thickness affect the
behaviour of piled raft foundation. From above study
the following points are observed.
1) The decrease in pile spacing decreases the total
and differential settlement and increases the pile
load carrying capacity. But if the piles are spaced
very close to each other then the differential
settlement can even become negative i.e. the
settlement at the edges will be more compare to
center.
2) The increase in number of piles beyond a certain
limit is not very beneficial. However a large
group of piles placed uniformly under the raft is
not useful instead of this placing a small group of
piles below the central portion of raft can
minimize the differential settlement
3) Long piles are effective in reducing total
settlement but the differential settlement is not
much affected by change in length of pile.
4) The increase in diameter of pile does not show
any significant change in reduction of differential
and total settlement but the percentage piles load
increases with increase in pile diameter. So
generally a small diameter piles are preferred for
reducing differential and total settlement.
5) The increase in raft thickness reduces the
differential settlement but increases the maximum

bending moment. So a thick raft is generally not
preferred and it is also uneconomical.
REFERENCES
Bajad, S. P. and Sahu, R. B. (2008). An experimental
study on the behaviour of vertically loaded piled
raft on soft clay. Proceedings of the 12th
International Conference of International
Association for Computer Methods and Advances
in Geomechanics, 1-6 October, 2008, Goa, India.
Hain, S. J. and Lee, I. K. (1978). The analysis of
flexible raft-pile systems. Geotechnique, 28(1), 65-
83.
Horikoshi, K. and Randolph, M. F. (1998). A
contribution to optimum design of piled rafts.
Geotechnique, 48(3), 301-317.
Oh, E. Y. N., Bui, Q. M., Surarak, C. and
Balasurbamaniam, A. S. (2009). Investigation of
the behaviour of piled raft foundation in sand by
numerical modelling. Proceedings of the 19th
International Offshore and Polar Engineering
Conference, June 21-26, Osaka, Japan.
Poulos, H. G. (2001). Piled raft foundation- design
and application. Geotechnique, 51(2), 95-113.
Prakoso, W. A. and Kulhawy, F. H. (2001).
Contribution to piled raft foundation and design.
Journal of Geotechnical and Geoenvironmental
Randolph, M. F. (1994). Design method for pile
groups and piled rafts. Proceedings of the 13th
International Conference on Soil Mechanics and
Foundation Engineering, New Delhi, 5, 61–82.

ISBN 978-81-921121-0-7
OPTIMISATION OF LOW RISE MULTI-STOREY BUILDING USING

STEEL-CONCRETE COMPOSITE SECTION & SEMI-RIGID BEAM TO
COLUMN CONNECTIONS
P. S. Buradkar
Assistant Professor, Civil Engineering Department, Shri Ramdeobaba Kamla Nehru Engineering College, Nagpur – 440 013,
Email: prachi.buradkar@gmail.com
ABSTRACT: Scarcity of land obstructs the horizontal expansion of accommodations resulting in their vertical
advancement. Steel-concrete composite systems have benefits of combining the two construction materials. Reinforced
concrete is inexpensive, massive, and stiff, while steel members are strong, lightweight, and easy to assemble. Steel–
concrete composite structures, owing to their high capacity for prefabrication and rational use of materials can provide
high levels of performance in terms of ductility and dissipation of energy.
Basic assumption of conventional structural analysis is that beam-to-column joints are either perfectly rigid or hinged.
Therefore, when analyzing a structure, joints are idealized as either fixed or hinged. These assumptions allow a great
simplification in structural analysis and design but they neglect the true behaviour of joints. However, in actual
structures, typical connections do not behave in either a perfectly rigid or perfectly hinged manner. These non-ideal
connections are often referred as semi-rigid connections. The consideration of semi-rigid behaviour results into reduced
midspan moment and increased end moments in the beams. Thus a reduced steel beam section caters the need. This
paper deals with the analysis and design of a low rise, multi-storey building using the benefits of steel-concrete
composite construction and semi-rigid beam – to – column connection.
INTRODUCTION
Multi-storey buildings are being proposed in different steel sections, have high stiffness and can carry heavy
cities of India to resolve the problem of overpopulation loads on long spans. Keeping the span and loading
caused by migration of people searching better living unaltered; a more economical steel section (in terms of
opportunities there. Scarcity of land limits the horizontal depth and weight) is adequate in composite construction
expansion of accommodations meant for various compared with conventional non-composite construction.
purposes. With this requirement, it has become a As the depth of beam reduces, the construction depth
necessity of time in our country to reduce the construction reduces, resulting in enhanced headroom. In decks,
time by adopting fast track construction methodologies as composite systems eliminate the need for formwork. One
well as allowing parallel construction activities. important advantage of composite systems is that
construction is accelerated through separation of trades.
Beam Initially, a bare steel frame is erected to carry the gravity,
The use of composite action has certain advantages. In construction, and lateral loads during construction. As
particular, a composite beam has greater stiffness and erection of the building progresses, concrete is cast in to
usually a higher load resistance than its non-composite form the composite system that will resist the total gravity
counterpart. Considering the increase in stiffness of the and lateral loads.
composite beam section in case of full shear connection This paper focuses on the extraction of the benefits of
between concrete slab and steel beam, a full shear steel-concrete composite construction for the design of
connection has been used to secure bond between the in – low rise (upto eight storeys of height) multistorey
situ concrete slab and the steel beam designed for the buildings. It not only ends here but the technique to
building frame in this paper. The composite action optimize design solution of such buildings is also
between steel beam and concrete slab is achieved through considered. However this optimization technique is
the provision of mechanical shear connectors. The shear applicable to all the non-composite steel frames also. The
connectors required at the steel-concrete interface are composite beams for the frame in question are designed
designed to (a) Transmit longitudinal shear along the as per IS 11384- 1998 based on limit state of collapse.
interface, and (b) Prevent separation of steel beam and
concrete slab at the interface. Concrete is stronger in Column
compression than in tension, and steel is susceptible to The column could have also been designed as a steel-
buckling in compression. By the composite action concrete composite section but the building in question
between the two, we can utilise their respective being a low rise multistorey building having the total
advantages to the fullest extent. For single span beams, number of storeys less than eight, the column could be
sagging bending moments, due to applied vertical loads, designed as a single steel I section using IS 800-2007.
cause tensile forces in the steel section and compression
in the concrete deck thereby making optimum use of each Connections
material. Therefore, composite beams, even with small One of the basic assumptions of conventional structural

analysis is that beam-to-column joints are either perfectly for the specific design. These test results are then made
rigid or perfectly hinged. Therefore, when analyzing a available as data bases. Semi-rigid joints can be modeled
structure, joints are idealized as either fixed or hinged. using a rotational spring which separates a member from
These assumptions allow a great simplification in its surrounding environment. The rotational spring is
structural analysis and design-but they neglect the true characterized by its moment-rotation curve. Simple
behavior of the joints. However, in actual structures, models are proposed in the form of equations with
typical connections do not behave in either a perfectly empirical constants derived based on test results.
rigid or a perfectly hinged manner. The different types of Depending on the degree of accuracy required, the
connections that are commonly used fill the entire moment-rotation characteristics may be idealized as
flexibility spectrum from flexible "hinge like" linear, bilinear or nonlinear curves. The major advantage
connections to semi-rigid connections to rigid of semi-rigid connections is that they are cheaper than
connections. These non-ideal connections are often rigid connections and allow the optimum utilization of the
referred to as flexible connections, semi-rigid beam member. By using a semi-rigid connection we can
connections, spring hinged joints, or sticky hinges. Semi- control the mid span and support moments to the desired
rigid connections have highly nonlinear behaviour, but value.
they present a more realistic structural analysis. For obtaining the moment rotation relationship the Frye-
Morris polynomial model is recommended by IS 800 -
Simple Connections 2007. The model has the form shown in the following
Simple connections are assumed to transfer shear only at equation
some nominal eccentricity. In simple connections no θr = C1 (KM)1 + C2 (KM)3 + C3 (KM)5 (1)
moment transfer is assumed between the connected parts Where, K = a standardization parameter dependent upon
and hence assumed as hinged. Therefore such connections the connection type and geometry and C1, C2, C3 = curve
can be used only in non-sway frames where the lateral fitting constants.
loads are resisted by some alternative arrangement such IS 800–2007 also gives classification proposed by
as bracings or shear walls. Simple connections are Bjorhovde which can be conveniently used with Frye–
typically used in frames up to about five storeys in height, Morris model to model semi – rigid connections.
where strength rather than stiffness govern the design.
Table. 1 Bjorhovde’s Connection Classification
Rigid Connections Nature of the In terms of In terms of
Rigid connections transfer significant moments to the connection strength stiffness
columns and are assumed to undergo negligible Rigid connection m’ ≥ 0.7 m’ ≥ 2.5’
deformations. Rigid connections are necessary in sway Semi – rigid 0.7 > m’ > 0.2 2.5’ > m’ >
frames for stability and also contribute in resisting lateral connection 0.5’
loads. In high-rise and slender structures, stiffness Flexible connection m’ ≤ 0.2 m’ ≤ 0.5’
requirements may warrant the use of rigid connections.
Where, m’ = Mu/Mp and ’ = θr/ θp
Semi-Rigid Connections CONNECTION STIFFNESS, JOINT STIFFNESS
Semi-rigid connections are those fall between simple and AND FIXITY FACTOR
rigid connections. Semi-rigid connections may not have Fig. 1 shows typical M vs r relation observed in semi –
the sufficient capacity to hold the original angles between rigid connections for which the connecton is modeled as
the members and develop less than the full moment a spring. Initially the connection behaves nearly
capacity. Most simple connections do have some degree elastically and the curve is nearly a straightline with a
of rotational rigidity. Similarly rigid connections do slope J=M/r, which represents the rotational spring
experience some degree of joint deformation and this can constant of the connection and is called the “joint
be utilized to reduce the joint design moments. These modulus” or also termed as the “connection spring
connections are used in conjunction with other lateral stiffness”.
load resisting systems for increased safety and
performance. Use of semi-rigid connections makes the
analysis somewhat difficult but leads to economy in J1 J2
member designs. The analysis of semi-rigid connections EI
L
is usually done by assuming linear rotational springs at Strength of spring Mc
the supports or by advanced analysis methods, which
account for non-linear moment-rotation characteristics.
TRUE BEHAVIOR OF CONNECTIONS IN STEEL Initial spring

stiffness J = M/r
FRAMES
Moment – Rotation Relationship of Connections Relative rotation r
The moment-rotation characteristics of the connections Fig. 1 Moment – Rotation Relationship and Connection
have to be determined based on experiments conducted Characteristics

and reinforcing steel used is Fe415 and Fe250. Soil type
On further loading, the connection begins to deform is medium soil and Zone III.
inelastically and the angle of rotation increases rapidly. # 12 #6
The connection stiffness decreases as the load increases
FOURTH FLOOR
and it is characterized by the M-r curve becoming flatter # 11
and flatter as it asymptotically approaches the plastic #5
THIRD FLOOR
moment capacity Mc of the connection. # 10
The magnitude of the stiffness of the spring J (kN-m/rad) #4
does not determine if a joint will behave in a rigid or 20m SECOND FLOOR
flexible manner. Rather it is the ratio of the spring #9
#3
stiffness J to the flexural stiffness of the member (beam) FIRST FLOOR
EI/L that determines what behavior will be exhibited. #8
This ratio is often called the "joint stiffness" and is #2
7m 7m GROUND FLOOR
defined as JL/EI. When the joint stiffness is less than 1.0
#7 #1
the beam behaves as if it is hinged, while if the joint
stiffness is greater than 100 the beam behaves as if it is Fig. 2 Building Frame Under Consideration
fixed. Therefore it may not be worth the effort to model a
connection with a rotational spring if the resulting joint The frame is analyzed for load combinations as per IS
stiffness being used is less than 1 or greater than 100. 1893-2002 viz. (i) 1.5 (DL + IL), (ii) 1.5 (DL ± EL), (iii)
Furthermore, if semi-rigid behavior is desired a joint 1.2 (DL + IL ± EL), (iv) 0.9 DL ± 1.5 EL and IS 800-
stiffness between 1 and 100 should be used. 2007 viz. (v) 1.2 DL + 0.5 IL ± 2.5 EL (vi) DL ± 2.5 EL.
To incorporate the connection flexibility into a computer
based program for frame analysis a term “Fixity factor” is Analysis Results: Beams & Columns
used. Monforton and Wu (1963) defined the fixity factor Table 2 Frame Analysis Results
of a semi – rigid connection as follows Floor Midspan Moment End Moment
Level kN-m kN-m
Fixity factor ri = 1/[ 1 + (3 EI/ Ji L)] (2) Fourth 76.6 120
Third 146 251
Where, i = 1, 2 Second 145 222
The fixity factor ri defines the rotational stiffness of each First 144 235
end connection relative to that of attached of member. For Ground 145 260
a pinned connection the value of end fixity factor is zero Floor Exterior Column Interior
(ri = 0), for a fixed connection the end fixity factor is Level Mt Mb P(kN) Column P(kN)
unity (ri = 1). Therefore, a semi- rigid connection has end Fourth 124 121 112 200
fixity factor value between zero and unity (0 < ri < 1). Third 127 125 340 620
Upon the introduction of this fixity factor, different Second 121 118 566 1053
member end restraint conditions are then readily modeled, First 122 117 788 1492
such as Rigid – Pinned, Rigid – Semi-rigid, Pinned - Ground 91 87 1004 1945
Semi-rigid, etc. simply by setting the end fixity factor at
the two ends of the members to appropriate values. Design Results
Table 3 Design Results: Composite Beam
ANALYSIS & DESIGN Floor Sectio- No. of Minim- +ve M
Analysis of the frame is done using STAAD.pro 2006. Level n shear um slab / Mu
Being a frame analysis, slab is not modeled. Instead the connectors r/f mm2
load is transferred to the beams. The beam element is Fourth ISMB 22 362.5 0.2289
assigned with property of a composite section. Single I 300
steel section property is assigned to the column element. Third ISMB 26 426.5 0.3443
In case of beam element, the partial release of moment (in 350
the major direction only) is assigned through the Second ISMB 26 426.5 0.3419
‘specifications’ option. The frame is reanalyzed and the 350
beam column elements are designed accordingly. First ISMB 26 426.5 0.3395
350
Problem Statement Groun ISMB 26 426.5 0.3419
The two bays – five storey frame considered for the d 350
design is as shown in Fig. 2. Spacing of frames is 3m and
height of columns is 4m. Thickness of slab and floor Table 4 Design Results: Interior Column
finish is 150mm and 50mm respectively. Construction
Floor Level Section P/Pd
load is 0.25 kN/m2 and imposed load is assumed to be
Fourth ISWB 200 0.7044
4kN/m2. Concrete used is M30, structural steel is Fe250
Third ISWB 250 0.9888
Second ISWB 450 0.8513

First ISWB 500 0.8251 Fig. 3 M- Curve for End Plate Connection
Ground ISWB 550 0.8870 Connection classification:
Table 5 Design Results: Exterior Column Table 7 Classification of End Plate Connection as per
Floor Secti- N/Ndz M/Mdz N/Ndz P/Pdz + Frye – Morris Model
Level on + (kz Cmz Pos Stiffness Joint Fixity Type of
M/Mdz Mz/Mdz) iti- of stiffness factor connection
Grou- ISMB 0.479 0.253 0.732 0.751 on connection >0&<1
nd 450
kN-m/rad semi-rigid
First ISMB 0.376 0.344 0.720 0.851
450 #2 2.84 E +6 729.99 0.9959 Semi-rigid
Second ISMB 0.317 0.455 0.772 0.991 #8 2.84 E +6 729.99 0.9959 Semi-rigid
400 #3 3.82 E +6 980.86 0.9970 Semi-rigid
Third ISMB 0.191 0.481 0.672 0.897 #9 3.82 E +6 980.86 0.9970 Semi-rigid
400
Fourth ISMB 0.074 0.622 0.696 0.970 #4 3.30 E +6 848.30 0.9965 Semi-rigid
350 #10 3.82 E +6 980.86 0.9970 Semi-rigid
#5 3.30 E +6 848.30 0.9965 Semi-rigid
Design of Connection #11 3.30 E +6 848.30 0.9965 Semi-rigid
After designing beams and columns as per first analysis #6 2.09 E+6 850.35 0.9965 Semi-rigid
results, three types of beam to column connections are
#12 1.74 E +6 447.39 0.9933 Semi-rigid
designed viz. (i) End plate connection with stiffners, (ii)
T-stub connection, (iii) Top & seat angle connection
consecutively. The Moment – Rotation relationship of Table 8 Classification of End Plate Connection as per
these three connections is derived and depending on it Bjorhovde
their fixity factor in the frame is decided. The frame is Posi- m’ = ’ = r/p Type of connection
then reanalyzed using this fixity factor. Then the frame is tion M/Mp in terms of
redesigned for the new results. Strength Stiffness
#2 1.2884 0.2580 Rigid Rigid
(i)End plate connection with stiffners #8 1.0307 0.1286 Rigid Rigid
Table 6 Design Results of End Plate Connection #3 1.1645 0.2007 Rigid Rigid
Positio- C/c of the Thickness Diameter of
#9 1.1645 0.2007 Rigid Rigid
n outermost of the the bolt, db
#4 1.1001 0.1702 Rigid Rigid
bolt, dg end plate, (mm)
#10 1.2190 0.2783 Rigid Rigid
(mm) tp (mm)
#5 1.0654 0.1574 Rigid Rigid
#2 470 16 30
#11 1.2438 0.2410 Rigid Rigid
#8 470 16 30
#6 0.4932 0.0052 Semi-rigid Rigid
#3 470 25 30
#12 0.8075 0.0188 Rigid Rigid
#9 470 25 30
Similarly connection classification of T-stub connection
#4 470 20 30 and Top & seat angle connection are as follows:
#10 470 25 27
#5 470 20 30 ii) T-stub connection
#11 470 20 30 Table 9 Classification of T-stub Connection as per Frye –
Morris Model
#6 400 20 22
Pos Stiffness Joint Fixity Type of
#12 400 25 27
iti- of stiffness factor connection
on connection
Moment – Rotation relationship of connection
#2 1.05 E +7 2693 0.9989 Semi-rigid
M -  Curve #8 1.05 E +7 2693 0.9989 Semi-rigid
300 #3 8.22 E +6 2109 0.9986 Semi-rigid
Moment (kN)
200 #9 8.22 E +6 2109 0.9986 Semi-rigid

100 #4 8.22 E +6 2109 0.9986 Semi-rigid
0 #10 1.05 E +7 2693 0.9989 Semi-rigid
0.00E+00 1.00E-03 2.00E-03 3.00E-03 4.00E-03 5.00E-03 #5 8.22 E +6 2109 0.9986 Semi-rigid
#11 1.05 E +7 2693 0.9989 Semi-rigid
r (Radians)
#6 4.31 E +6 1753 0.9983 Semi-rigid

#12 5.67 E +6 2307 0.9987 Semi-rigid
Table 10 Classification of T-stub Connection as per
M -  Curve
Bjorhovde
Posi- m’ = ’ = Type of connection 300
Moment (kN)
tion M/Mp r/p in terms of
Strength Stiffness 200
100
#2 1.0307 0.3862 Rigid Rigid
#8 1.2884 0.5341 Rigid Semi-rigid 0
#3 1.1645 0.4565 Rigid Rigid 0.00E+00 1.00E+00 2.00E+00 3.00E+00 4.00E+00
#9 1.1645 0.4565 Rigid Rigid r (Radians)
#4 1.1001 0.3996 Rigid Rigid
#10 1.2190 0.4971 Rigid Semi-rigid
#5 1.0654 0.3821 Rigid Rigid Fig. 5 M- Curve for Top & seat angle Connection
#11 1.2438 0.5179 Rigid Semi-rigid
Table 12 Classification of Top & Seat Angle Connection
#6 0.4932 0.0558 Semi-rigid Rigid
as per Bjorhovde
#12 0.8075 0.1213 Rigid Rigid Posi- m’ = ’ Type of connection
=
tion M/Mp r/p in terms of
The above results are based on the M-q relationship for Strength Stiffness
T-stub connection as follows
#2 0.9118 0.0554 Rigid Rigid
M -  Curve #8 1.0109 0.0896 Rigid Rigid
300 #3 0.9613 0.0690 Rigid Rigid
Moment (kN)
#9 0.9811 0.0757 Rigid Rigid

200 #4 0.9019 0.0513 Rigid Rigid
100 #10 1.0307 0.0958 Rigid Rigid
#5 0.8919 0.0485 Rigid Rigid
0 #11 1.0307 0.0958 Rigid Rigid
0.00E+00 2.00E-05 4.00E-05 6.00E-05 #6 0.4253 0.0007 Semi-rigid Rigid
r (Radians) #12 0.6796 0.0033 Semi-rigid Rigid
 The frame was then re analysed using the values of fixity

Fig. 4 M- Curve for T-stub Connection factor at the corresponding nodes and redesigned for the
three connections in consecution.
iii) Top & seat angle connection  It was observed that beams in the frame with end plate
Table 11 Classification of Top & Seat Angle Connection connection and T – stub connection showed almost the
as per Frye – Morris Model same value of negative moment as in the first analysis
Pos Stiffness Joint Fixity Type of results. Whereas, the negative beam moment in case of
iti- of stiffness factor connection frame with top and seat angle connections was reduced
almost by an average of 16.16%.
on connection
 More the negative moment more the number of lateral
#2 2.91 E +4 7.47 0.7135 Semi-rigid restraints required to resist the lateral torsional buckling
#8 2.91 E +4 7.47 0.7135 Semi-rigid of the composite beam at the construction stage. In this
#3 2.91 E +4 7.47 0.7135 Semi-rigid respect, the beam in top and seat angle connection frame
#9 2.91 E +4 7.47 0.7135 Semi-rigid becomes a better option.
 An increased positive beam moment value in case of
#4 2.91 E +4 7.47 0.7135 Semi-rigid
frame with top and seat angle connection was observed as
#10 2.91 E +4 7.47 0.7135 Semi-rigid compared to that with end plate connection with stiffners
#5 2.91 E +4 7.47 0.7135 Semi-rigid and T – stub connection.
#11 2.91 E +4 7.47 0.7135 Semi-rigid  The positive moment capacity of composite beam even
#6 2.03 E +4 8.24 0.7332 Semi-rigid with smaller steel section is very high and hence, a better
#12 2.03 E +4 8.24 0.7332 Semi-rigid utilization of this capacity of the composite beam section
in case top and seat angle connectioned frame was noticed
as summarized below, though the beam section could not
be reduced because of limited deflection criteria.

Positive moment
Drift (mm)
Moment (kN-m) 200 Top & seat angle connection with double
150 End plate web angle
connection with T - stub connection
100 stiffners
50 End plate connection with stiffners
T - stub
0 15.018
connection 12.167
12.199
Fourth floors
Ground floor
First floor
Second floor
Third floor Top & seat angle
connection with Fig. 7 Comparison of Horizontal Deflection in case of
double web angle
different connections
CONCLUSIONS
The objective of this paper was to analyse and design a
Fig. 6 Comparison of Positive Moments in case of low rise multi-storey building using steel-concrete
different connections composite section. The use of a steel-concrete composite
beam in the building frame results in a lower beam
No major change in the axial loads of the exterior and section than that would have been resulted by the use of
interior columns both of the frames with the three any non-composite section whether RCC or plain steel.
different connections was observed. This enabled greater headroom or vertical clearance. This
First analysis results of moments were retained in case of building frame was then tested for three connections viz.,
frame with end plate connection with stiffners and T – end plate connection with stiffners, T – stub connection
stub connection, whereas, the results found to be and top & seat angle connection with double web angle,
attenuated in case of frame provided with top & seat accounting the true behaviour of the connections. The
angle connection with double web angle. use of semi-rigid connection results in better utilization of
No change of column section was observed in the first the steel-concrete composite beam’s positive moment
two cases of connections while a change of section in capacity. The reduction in the beam’s negative moment
frame with top & seat angle connection was observed enables lesser number of lateral restraints to be used at the
from ISMB 400 to ISMB 350 at the third floor exterior construction stage. The building frame with semi-rigid
column. connections showed an increase in the lateral drift than
The major contribution of the exterior columns’ combined that was observed in case rigid connectioned frames. But
axial load and moment resisting capacity was towards this drift was below 1/250h, as limited by the Indian
resisting the moments. As a result, even though no seismic code which proved the reliability of the semi-
remarkable reduction in axial loads occurred, an average rigid connectioned frame.
reduction of 16.22% of the original moments led to the
reduction in column section at third floor. REFERENCES
Joint stiffness of all the three connection showed great S.A.L. de Andrade, P.C.G. da S. Vellasco, L.T.S.
difference. Whereas the fixity factor for end plate Ferreira, L.R.O. de Lima, (April 2006), Semi-rigid
connection and T – stub connection was approximately composite frames with perfobond and T-rib
same. Still T – stub connection had the greatest value of connectors Part 2: Design models assessment,
fixity and thus it can be differentiated almost as a rigid Journal of constructional steel research
connection. End plate connection can also be said to be a Scott A. Burns, “Recent Advances in Optimal Structural
rigid connection. But top and seat angle connection had Design” (Equation 2)
the least value of fixity factor and can be categorized as a Tazeen Fatema & Md. Toihidul Islam, (June 2006),
semi – rigid connection. Study on connection between precast concrete beam
Though as good as almost 40% of the moment could be and cast-in-situ column in prefabricated building
released in case of top & seat angle connection, it was frames, ARPN Journal of Engineering and Applied
restricted to a maximum value of 29% by providing Sciences
higher sections to secure the connection. - IS: 456 – 2002, “Plain and Reinforced Concrete – Code
The drift values of node number six of the rigid of Practice”
connections were found to be nearly same. In semi – rigid - IS: 800 – 2007, “General Construction in Steel – Code
connectioned frame, the drift value was observed to be of Practice” (Table 1, Equation 1)
higher than that with rigid connections. - IS: 1893 – 2002, “Criteria for Earthquake Resistant
The drift in case of frame with end plate connection was Design of Structures: Part – 1 – “General Provisions
1/1639h, for T-stub connection the value was 1/1644h and Buildings”- IS: 11384 – 1998, “Code of
and it was the largest in case of top & seat angle Practice for Composite Construction in Structural
connection equal to 1/1249h. But all the drift values were Steel and Concrete”
less than 1/250h. So none of the connections violated the
drift laws of IS: 1893 – 2002 (I).

ISBN 978-81-921121-0-7
ARTIFICIAL NEURAL NETWORKS FOR REAL TIME RESERVOIR

INFLOW MODELING AND PREDICTION
A.K. Pardeshi
PG student, Department of Civil Engg. Government College of Engineering, Aurangabad – 431005,
Email:alka.pardeshi@gmail.com
D.G. Regulwar
Associate Professor Department of Civil Engg. Government College of Engineering, Aurangabad – 431005,
Email: regulwar@gmail.com
ABSTRACT: A neural network modeling approach is used to construct a real time catchment flow
prediction model for river basin. The present study aims towards introducing the use of ANN technique to
model and predict the monthly inflow of Jayakwadi Reservoir. The data pertaining to the years 1927-1971
have been explored to develop the predictive model with Tanh Axon non linearity and Levenberg Marquardt
as a training algorithm. The Jayakwadi reservoir, one of the largest irrigation projects in Maharashtra with
catchment area of 21,750 km2, is selected as the study area. ANN model was developed based on the
historical inflow data of the Jayakwadi Reservoir and successfully predicted inflow with less error. In this
study six forecasting model approach are used, which are multilayer Perceptron (MLP), feed forward (FF),
modular neural network (MNN), Jordan/Elman network, time-lag recurrent (TLRNN), CANFIS network
method for the natural inflow at Jayakwadi utilizing the inflow data of the monitoring stations. Inflow data
collected over the last 45 years is used to develop the model and examined its performance. 70% of data is
used to develop the model and rest of data is utilized to test the models. Except CANFIS method in all five
methods i.e. MLP, GFFNN, MNN, J/ENN, TLRNN results shows that coefficient of correlations are above
90% and in CANFIS method coefficient of correlation is 89%. The result of this study showed that ANN
technique is capable with small computational effort and high accuracy of predicting the monthly inflow.
KEY WORDS: Inflow prediction, artificial neural network, tanhaxon, Llevenberg marquardt.
INTRODUCTION and long term inflow of Jayakwadi reservoir are

important for the dam safety.
The management of the water resources is of crucial
importance since it directly affects the design and Neural network technology have provided many
operation of many water resources structures. promising results in modeling complex nonlinear
Conventional procedures for design and analysis are systems, and successful applications of this artificial
basically trial-and-error procedures. In these intelligence in the field of hydrology and water
methods the aim is to predict the monthly inflow of resources modeling have been widely reported, such as
reservoir. At the stage of the planning of the for stream flow and river flow prediction (Sharifi et al.
reservoir size this is a very important subject. Here 2005; Aquil et al. 2007, Sankarasubramanian et al.
the monthly total inflow of the reservoir is the main 2007; Kim 2004; Haddad et al. 2005; Ahmed et al.
data series. It is better to have a data record length 2007; Raju et al. 2004; Gopakumar et al. 2007; Jian et
as long as possible. Regarding water engineering al. 2006; Lin et al.), and inflow prediction (kote 2010;
field, several researchers have incorporated ANN Coulibaly et al. 2000; Wang et al. 2009; Ismail et al.
technique in hydrology, groundwater and hydraulics 2005; Kolen et al. 2000; Shafie et al. 2009),
and reservoir operation to simulate their problems. groundwater modeling ( Coulibaly et al. 2001;Sohail
Due to rapid growth of population and economy in et al. 2007; Agrawal et al.2004; Nauraniet al. 2009; ),
the world, water is becoming a precious and scares rainfall–runoff simulation (Jain et al. 2004; Sminia et
resource as a result of growing demands from al. 2003).
various purposes, such as irrigation, hydropower,
municipal and industrial water supply, navigation, ARTIFICIAL NEURAL NETWORKS
etc. (Jain et al [1]). A neural network is a method that is inspired by the
Artificial neural networks (ANN) are black box studies of the brain and nerve systems in biological
models that are used for forecasting and estimating organisms. Neural networks have the capability of
purposes in so many different areas of the science self-learning and automatic abstracting. Artificial
and engineering. The accurate predictions of short neural networks are important alternatives to the
traditional methods of data analysis and modeling.

Multilayer Perceptron more intermediate layers or subgroups, which are
Multilayer Perceptron (MLP) are the most used to act as a supplier of results to the next layer.
commonly used ANN in hydrological predictions. The network-layered structure consists of a set of
The multilayer perception neural network is built up nodes (neurons) connected by links from one layer to
of several layers by the addition of more hidden its next layer. Following Fig. 2 shows the structure
layers. In these method neurons within the same of FFNN. The main advantage of the feed forward
layer connected to the neurons in the next layers. architecture is that it requires relatively less of
computing time during training.
Modular Neural Network

In a modular neural network there are combinations
of two or more neural networks in one network by
providing appropriate connections between layers of
one subnetwork to layers of others.
Fig. 1 Typical Multilayer Perceptron Network

The connections are forward and the layers are
cascaded together to form the MLP network. Each
MLP is consists of a minimum of three layers these
are: an input layer, one or more hidden layers and
an output layer. A typical three layer network is
shown in Fig. 1. Only three layer MLPs will be
considered in this work.
Feed Forward Neural Network

FFNN has parallel layers or subgroups made up of
processing elements. Each layer of processing
elements makes independent calculations on data
that it receives and forwards the results to next
layer. The next layer again passes the results to yet Fig. 3 Modular Neural Network Architecture
further layer.
Each independent neural network acts as a module
and operates on separate inputs to complete some
subtask to achieve the task of network as shown in
Fig. 3. The mediator takes the outputs from each
module and processes them to generate the output of
the network. The mediator only accepts the modules
outputs it does not respond to it. The modules do not
interact with each other. One of the major benefits of
a modular neural network is the ability to reduce a
large, unwieldy neural network to smaller, more
manageable components.
Jordan / Elman Neural Network

In Elman neural network the input layer is divided
Fig. 2 Generalized Feed Forward Neural Network into two parts: the true input units and the context
Finally a last layer determines the output from this units. The network generated its output signals from
network system. Basically output of neural network the hidden layer to train the network. The
is a weighted sum of its input but threshold function connections are all feed forward, and the weights
is sometimes used to qualify the output of a neuron from the context units can be trained exactly as the
in the output layer. In general, it is composed of back propagation methods. In Jordan neural network
three or more layers: an input layer, which is used to the network determines its output signals from the
present data to the network; an output layer, which is output layer to train the network. Its output signals
used to generate an appropriate output; and one or are fed back to the input layer via context units as
shown in Fig. 4. Again, the weights from the context

units can be trained in exactly the same manner as
those of ordinary type ERN. Levenberg Marquardt Backpropagation
Training
In this study the LMBP (Levenberg-Marquardt
BackPropagation) algorithm method for ANN
training is used. The LMBP, one of the second-order
nonlinear optimization techniques, is usually faster
and reliable. The LMBP uses the approximate
Hessian matrix that can be approximated as follows
H=JTJ (2.1)
Where J is the Jacobian matrix with first derivatives
of the ANN errors with respect to weights and
biases.
MODEL DEVELOPMENTS
Fig. 4 Elman Recurrent Network Description of Study Area

Jayakwadi project is one of the largest irrigation
Time Lag Recurrent Network projects at Paithan in Aurangabad District of
Time lagged recurrent networks (TLRNs) are MLPs Maharashtra State. It is a multipurpose project.
comprehensive with short term memory structures Jayakwadi is one of the largest earthen dams in Asia.
and local recurring connections. The input layer Its catchment area is 21,750 km². Total submergence
used the inputs delayed by multiple time points area due to the reservoir is approximate 35,000 ha.
before presented to the network. The network Its height is approx 41.30 m and length of 9998 m
general architecture has three layers and the (10 km approximate) Nath Sagar is the name of the
feedback connection connected from the hidden reservoir formed due to Jayakwadi Dam. Total area
layer to the input layer as shown in Fig. 5. of reservoir is approx 350 km2. Its total storage
capacity is approx 2.909 km³ and effective live
storage capacity is 2.17 km³. The length of left bank
canal is 208 km and the length of right bank canal is
132 km. It irrigates cultivable area of 237,452 ha in
the Districts
of Aurangabad, Jalna, Beed and Parbhani.
The reservoir with latitude 190 29’ 8.7” N and
longitude 750 22’ 12” E is located on Godavari river.
The hydrological data used in this study included
monthly inflow data.
Model Structure
Fig. 5 Time Lag Recurrent Network
Accurate prediction of reservoir inflow has been
predictable as an important measure for effective
CANFIS Network (Fuzzy Logic)
flood-risk management strategy. A neural network
modeling approach is used to construct a real time
reservoir inflow prediction model for a river basin.
Six types of neural network architecture i.e. MLP,
GFFNN, MNN, J/EN, TLRNN, CANFIS Network
and Levenberg-Marquardt training algorithm, with
adaptive learning examined in this study.
Artificial Neural Network models have developed by
using Neuro Solution Software. Six models are
developed depending one lag of previous one month
using data from 1927-1971. It is very important to
make sure that the validation data should not have
Fig. 6: CANFIS Architecture been used as part of the training process.
CANFIS is a generalized form of ANFIS has the
benefit of non-linear rule formations. CANFIS Model 1 of MLP
method can obtain more than one outputs. The Graph 7 shows the graph of desired output and the
CANFIS model integrates fuzzy inputs with a graph of actual network output result for 2-20-1
modular neural network to quickly solve poorly model developed in Neuro solutions. With two
defined problems. The output of a fuzzy axon is
calculated by the formula as shown in Fig. 6.

input nodes U(t-1), U(t-2), one hidden layers with
20 processing elements and one output node U(t). Desired Output and Actual Network Output
4000
UT
3500 UT Output
Desired Output and Actual Network
Output 3000
4000 UT
UT Output 2500
3500
Output
2000
3000
2500 1500
Output
2000 1000
1500 500
1000 0
500
-500 1 16 31 46 61 76 91 106 121 136 151
Exemplar
0
-500 1 16 31 46 61 76 91 106 121 136 151

Exemplar Graph 8 Generalized Feed Forward Testing
Graph
Graph 7 Multilayer perceptron Testing Graph
Table 2: Result Summary of GFFNN Model
Table 1: Result Summary of MLP Model Performance Ut
MSE 41954.75529
Performance Ut
NMSE 0.141965656
MSE 38373.71651
MAE 115.2845047
NMSE 0.129848209
Min Abs Error 0.059851748
MAE 128.4363943
Max Abs Error 1199.446658
r 0.929627666
r 0.940817632 Model 3 of MNN
Graph 9 shows the result for 2-6-4-1 model
The monthly inflow data of 45 years was taken in developed in Neuro solutions. With two input nodes
model, out of which 70% was used for training and U(t-1), U(t-2), two hidden layers with upper and
30% for testing. The network build is Multilayer lower PEs 6 and 4. In both hidden layer upper and
Perceptron with learning algorithm, Levenberg lower transfer function is TanhAxon and in output
Marquardt. The TanhAxon transfer function has layer sigmoid Axon transfer function has been used
used. The correlation factor r is 0.9408 and with one output node U(t).
Normalized Mean Square Error (NMSE) is 0.1298
as shown in Table 1.
Desired Output and Actual Network
4000 Output
Model 2 of GFFNN
Graph 8 show the graph of desired output and the 3500 UT
graph of actual network output result for 2-25-1 3000
model developed in Neuro solutions. With two
2500
Output
25 processing elements and one output node U(t). 2000

The monthly inflow data of 45 years was taken in 1500
model, 70% was used for training and 30% for 1000
testing. The TanhAxon transfer function has used.
The correlation factor r is 0.9296 and Normalized 500
Mean Square Error (NMSE) is 0.1419 as shown in 0
Table 2. -500 1 16 31 46 61 76 91 106 121 136 151
Exemplar
Graph 9 Modular Neural Network Testing Graph

Table 3: Result Summary of MNN Model correlation factor r is 0.9388 and Normalized Mean
Square Error (NMSE) is 0.1381as shown in Table 4
Performance UT
MSE 37710.10661 Model 5 of TLRNN
NMSE 0.127602699 Graph 11 shows the result for 2-15-1 model
MAE 90.39697836 developed in Neuro solutions. With two input nodes
U(t-1), U(t-2), one hidden layers with 15 processing
elements and one output node U(t). The monthly
Max Abs Error 1201.326289 inflow data of 45 years was taken in model, out of
r 0.936429223 which 70% was used for training and 30% for
testing.
The monthly inflow data of 45 years was taken in
model. The correlation factor r is 0.9364 and
Normalized Mean Square Error (NMSE) is 0.1276 Desired Output and Actual Network
as shown in Table 3. 4000 Output
UT
3500
Model 4 of J/ENN UT Output
Graph 10 shows the graph of desired output and the 3000
graph of actual network output result for 2-12-1 2500
Output
model developed in Neuro solutions. With two
2000
12 processing elements and one output node U(t). 1500
1000
Desired Output and Actual Network 500

4000 Output
UT 0
3500
1 16 31 46 61 76 91 106 121 136 151
3000 Exemplar
2500
Graph 11 Time Lag Recurrent Network Testing
Output
2000 Graph
1500
Table 5: Result Summary of TLRNN Model
1000
500
Performance UT
MSE 51472.08936
0
NMSE 0.174170219
-500 1 16 31 46 61 76 91 106 121 136 151 MAE 149.1692248
Exemplar
Graph 10 Jordan / Elman Network Testing Graph
r 0.929158665
Table 4: Result Summary of J/ENN Model
The network build is Time Lag Recurrent Network
Performance UT with learning algorithm, Levenberg Marquardt
Focused on Gamma Axon. The TanhAxon transfer
MSE 40834.65052
function has used. The correlation factor r is 0.9291
NMSE 0.138175468
and Normalized Mean Square Error (NMSE) is
MAE 115.0383927 0.1741 as shown in Table 5.
Max Abs Error 1192.850927 Model 6 of CANFIS Network
r 0.9388224
Graph 12 the result for 2-0-1 model developed in
The monthly inflow data of 45 years was taken in Neuro solutions. With two input nodes U(t-1), U(t-
model, out of which 70% was used for training and 2), one hidden layers with zero processing elements
30% for testing. The network build is Jordan/Elman and one output node U(t). The network build is
Network with learning algorithm, Levenberg CANFIS Network with learning rule, Momentum.
Marquardt. The TanhAxon transfer function has been The correlation factor r is 0.8998 and Normalized
used in hidden layer and SigmoidAxon transfer Mean Square Error (NMSE) is 0.2048 as shown in
function has been used in output layer. The Table 6.

predicted well, researchers are seeking for better
model to predict the future inflows with better
Desired Output and Actual Network accuracy in terms of high flows and low flows.
4000 Output The results shows that MLP, GFFNN, MNN,
3500 UT J/ENN and TLRNN trained with Levenberg-
3000 Marquardt are able to forecast the reservoir inflow
2500 up to one month in advance with reasonable
prediction accuracy, it means one lag is sufficient
Output
2000
for predicting better inflows.
1500 In Model 1 of MLP ANN shown best result with
1000 maximum correlation coefficient (r) as 0.94 and
500 minimum value of Normalized Mean Square Error
0
(NMSE). In all ANNs model accuracy is so
significant that all error measures were improved
-500 1 16 31 46 61 76 91 106 121 136 151
by using correlation. The result of a given case
Exemplar study shown that ANN has some significant
advantages for short and long term prediction of
Graph 12 CANFIS Network Testing Graph inflow in hydrology.
The ANN methodology has been reported to
Table 6: Result Summary of CANFIS Model provide reasonably good solution for circumstances
where there are complex systems that may be
Performance UT poorly defined or understood using mathematical
MSE 60540.27553 equations.
NMSE 0.204854965 Artificial Neural Networks are considered as a
promising tool for reservoir inflow prediction when
MAE 136.2111536
detailed time series of hydrological metrological
Min Abs Error 1.805163592 data are available. When trying to decide the input
Max Abs Error 1595.529907 parameters of the neural network a correlation
r 0.899828108 coefficient analysis of the parameters can give
hints of which parameters are important and should
Justification for Error be included.
The methods were used to predict inflow in
Jayakwadi river basin at paithan, the model REFERENCES
results indicate that reasonable prediction accuracy Ahmed El-Shafie, Alaa E. Abdin, Aboelmagd
was achieved for most of models for one month Noureldin and Mohd R. Taha. (2009). Enhancing
ahead forecast with correlation > 0.91. However, Inflow Forecasting Model at Aswan High Dam
the model accuracy deteriorates as the lead time Utilizing Radial Basis Neural Network and
increases. When compare, a 2-20-1 multilayer Upstream Monitoring Stations Measurements.
Perceptron and 2-25-1 generalized feed forward Springer Science Environmental Management,
network, has produced better performances on 2289–2315.
indicators related to average goodness of prediction Alka Sunil Kote and V. Jothiprakash. (2008).
for the one month ahead river inflow compared to Reservoir Inflow Prediction Using Time
CANFIS model based on fuzzy logic. Thus the Lagged Recurrent Neural Networks. Emerging
result of the study show that MLP, GFFNN, MNN, Trends in Engineering & Technology,
J/ENN and TLRNN trained with Levenberg- International Conference, IEEE Computer
Marquardt are able to forecast the reservoir inflow Society, 618-623.
up to one month in advance with reasonable Coulibaly P, Anctil F, Boobe B. (2001). Multivariate
prediction accuracy. It means one lag is sufficient Reservoir Inflow Forecasting Using Temporal
for predicting better inflows. ANN predicts much Neural Networks. Journal of Hydrology Engg.6,
better that is magnitude of each and every error 367-376.
term leads to higher accuracy than CANFIS method Dae-Il Jeong and Young-Oh Kim. (2005). Rainfall-
based on fuzzy logic. Table 7 shows result Runoff Models Using Artificial Neural Networks
summary of ANN models. for Ensemble Streamflow Prediction.
Hydrological Processes. 3819–3835.
CONCLUSIONS Ismail Kilinc and Kerem Cigizoglu. Reservoir
Artificial neural networks, one of the artificial Management Using Artificial Neural Networks.
intelligence tools which capture the pattern 14th. Reg. Directorate of DSI (State Hydraulic
between input and output is found to be suitable in Works) Istanbul Turkey.
predicting inflows. Even though the conventional Jain S.K., Das, and Shrivastava D.K. (1999).
models show that the future value could be Application of ANN for Reservoir Inflow

Prediction and Operation. J. Water Resources Wensheng Wang ,Juliang Jin and Yueqing Li.
Planning and Management, ASCE, Vol.125, (2009). Prediction of Inflow At Three Gorges
No.5, 263-271. Dam In Yangtze River With Wavelet Network
John F. Kolen and Rattikoen Hewett. (2000). Model. Springer Science Water Resour Manage,
Predication of Lake Inflows with Neural 2791–2803.
Networks. IEEE, 572-576. Yu-Min Wang and Seydou Traore.(2009). Time-
Majid Heydari and Parisa Hosseinzadeh lagged Recurrent Network for Forecasting
Talaee.(2011). Episodic Event Suspended Sediment Load in
Prediction of flow through rockfill dams using a Typhoon Prone Area. International Journal of
neuro-fuzzy computing technique. The Journal of Physical Sciences, Vol. 4 No.9, 519-528.
Mathematics and Computer Science, Vol .2 No.3,
515-528. Acknowledgement
Muhammad Aqil , Ichiro Kita , Akira Yano and The authors are thankful to command area
Soichi Nishiyama. (2007). Neural Networks for development authority, Aurangabad, Maharashtra
Real Time Catchment Flow Modeling and state, India. For providing necessary data for the
Prediction. Springer Science Water Resour analysis.
Manage, 1781-1796.
Table 7 Result Summary of ANN Models
Model % Training Network & Learning Epochs Testing result

& Testing Transfer algorithm
function NMSE MAE r
ANN (2-20-1) 70:30 MLP (TA-TA) LM 25 0.129 128.436 0.94
ANN (2-25-1) 70:30 GFFNN (TA-TA) LM 27 0.141 115.284 0.92
ANN (2-6-4-1) 70:30 MNN (TA-SA) LM 37 0.127 90.3969 0.93
ANN (2-12-1) 70:30 J/ENN (TA-SA) LM 18 0.138 115.038 0.93
ANN (2-15-1) 70:30 TLRN (TA-TA) LM 27 0.174 149.169 0.92
ANN (2-0-1) 70:30 CANFIS LM 481 0.204 136.211 0.89

ISBN 978-81-921121-0-7
FLOOD FORECASTING SYSTEM IN KOSI BASIN

L. B. Roy
Associate Professor & Head, Civil Engineering Department, NIT Patna, Ashok Rajpath, Patna – 800 005 (India), Email:
lalbahadurroy@yahoo.co.in,
Sunil Kumar
Lecturer, Department of Civil Engineering, MIT Muzaffarpur, Bihar, email: sunil_mit1@yahoo.in
Jay Prakash Kumar
Research Scholar,MIT Muzaffarpur, B. R. Ambedkar Bihar University, Muzaffarpur, Bihar
ABSTRACT: Flood Forecasting System, which is reducing the flood damages by giving timely warning, is
now fully realized, although it can in no way prevent the recurrence of flood. The real time flood forecasting is
one of the most effective non-structural measures for flood management. For formulation the flood forecast in
the real time, observed meteorological and flow data are transmitted to the forecasting stations through different
means of data communication which include telephone, wireless and network of telemetry stations etc. The
collected meteorological and flow data in real time are then used into the calibrated & validation real time flood
forecasting model to forecast the flood flow and corresponding water level for different lead periods varying
from few hours to few days depending on the size of catchment and purpose of forecast. Structure of the model
should be simple and it should not have excessive input requirements, but the same time the forecast flood must
be as accurate as possible. The Central Forecasting Organization of CWC has been conducting the flood
forecasting work in Kosi river system since 1970. At present there are three forecasting sites on the river Kosi,
,namely Basua, Baltara and Kursela with two base station at Barahkshetra and Birpur barrage. All these sites are
equipped with wireless system for efficient communication purpose. The method of forecasting is practically on
the basis of gauge to gauge co-relation among these stations in the network.
INTRODUCTION to the stages of river above as specified water level

Flood and drainage are the two vital interrelated called “Warring level”. This warning level is
problems of this poverty stricken state of Bihar, generally 1 meter below the danger line fixed in
particularly in North Bihar. The north Bihar plains consultation with the beneficiary i.e. the concerning
are drained by some Himalayan rivers and the state authority.
Ganges being the trunk drain. These rivers are
Utility of Forecast
perennial as these are rain fed as well as snow fed.
Utility of forecast is dependent on both accuracy and
Important rivers of North Bihar include the Kosi, the
time of prediction. At the time of issue of the
Gandak, the Baghmati, the Burhi Gandak, the
forecast, the value and utility of the forecast can be
Kamala, the Kamala Balan, the Mahananda and few
considered to be zero. The entire operation of the
others. All these rivers have relatively youthful
Flood Forecasting Service has to he planned around a
topography and are engaged in channel deepening.
time-factor keeping in view the following factors:
They move parallel to the Ganges in the south-cast
direction and then drain into it. The rainfall in North (a) Availability of Operational data
The operational data being generally poor and
Bihar is monsoonal in character and its distribution
incomplete, a compromise between a theoretical
shows wide temporal and spatial variation. Rainfall
desirable and operationa1ly available is required.
is confined to few months and is higher in north-
(b) Adoption of Appropriate Technique
eastern part. This variation in rainfall distribution
The use of appropriate techniques or methods to
causes frequent floods in North Bihar, huge losses to
meet the requirements of accuracy and period of
crops as large areas are inundated every year.
warning to different locations are required
Obviously the principal cause of recurrent flood is
heavy and erratic rainfall and inadequate drainage.
Hence, flood forecasting system in such rivers e. g. Table 1: Flood Problem of India
Kosi may prove to be an effective means to
SI No Items Areas/states/bas
overcome the flood damages in the basin. Flood in
problem of India and the Flood damages of India are
1 Geographical Area 329 m ha
given in Table 1 and Table 2 respectively.
Flood forecasting may be defined as “the process of 2 Total Flood Prone Area
estimating the future stages or flows and its time As Assessed by RBA 40 m ha
sequence at selected points along the river during
floods”. Flood forecast refers to prediction of the As reported by the State to the
working group on flood control
crest and its time of occurrence and logical extension

programme for 10th five year plan 45.65 m ha mathematical relationship, are developed with the
3 Major Flood Prone States Assam, Bihar, help of historical data, using the statistical analysis.
West Bengal, These include simple gauge to gauge relationships,
UP, Orissa, A.P. with some additional parameters and rainfall-peak
4 Major Flood Prone Basin Ganga, stage relationship etc. These relationships can be
easily developed and are most commonly used in
Brahmputra,Mah
anda River
India as well as other countries of the world.
But the most systematic approach for development of
forecasting model will be one which is based on the
Table 2: Flood Damages (India) During 1953-2005 concept regarding the mechanism of formulation and
SI Items Average Maximum propagation of flood i.e. transformation of rainfall
No (Year) falling over the basin into run off, its time
1. Area Affected 7.55 17.50 (1978) distribution and further propagation down below.
(Million Hectare) Based on the data used for formulation of forecast,
2. Crop Affected 3.54 10.15 (1988) the various methods of flood forecasting can be
(Million Hectare) classified in three major groups
3. Population 3.286 7.045 (1978)
1. Forecast on the basis of stage-discharge data,
Affected (Crores)
4. Human Lives 1,589 11,316 2. Rainfall-runoff methods and
Lost (Nos) (1977)
3. Meteorological methods.
5. Cattle Lost (Nos) 94,839 6,18,248
(1979) FLOOD FORECASTING SYSTEMS EXISTING
6. House Damages 12,17,918 35,07,542 IN THE KOSI RIVER
(Nos) (1978) The Central Water Commission has a good network
7. Value of 710.63 4246.62 for the forecast of flood in river Kosi and its
Damages to (2000) tributaries under the Middle Ganga Division IV,
Crops (Crores) Patna. Information relating Rain gauges, river
8. Value of 270.59 1307.89 gauges, communication network and methods
Damages to (1995) adopted for forecasting is being described below.
House (Crores)
9. Value of 820.75 5604.46 Rain Gauge Network
Damages to (2001) A network of 10 Nos. of raingauges in the plains
Public Utility serves for the real-time data input in the flood
(Crores) forecasting operations during flood season. These
10. Values of 1805.18 8864.54 raingauges are ordinary as well as self-recording
Damages to (2000) type. Most of raingauges are installed by Flood
Crops and Public Meteorological Organization (FMD) of the IMD, and
Houses (Crores) upkeep of the instruments are their responsibility.
However the observations are made by CWC
(c) Dissemination of Forecast officials. The list of various rain gauge stations being
maintained for forecasting is given in Table- 3
Dissemination of forecast is equally important as
accuracy. There should be an efficient arrangement River Gauge Network
for dissemination of the forecast. Attempts should be The 15 nos. of hydrological stations on Kosi and its
made to convey the forecast/warning to the authority tributaries are functioning under Middle Ganga
concerned so that they get adequate time to organize Division IV of CWC Distributions of these stations
necessary measures, if required in the light of the are given in Table- 4. In addition to these stations
forecast issued. at least 67 gauge and discharge sites are reported to
be in operation under Irrigation Department of Govt.
Methods of Flood Forecasting of Bihar
On the basis of the analytical approach for Communication Network
development of the forecasting model, the method of
flood forecasting can be classified as: The wireless communications, a vital operational
component in flood forecasting activities, are used
1. Methods based on statistical approach and for real time transmission of data & flood
2. Methods based on mechanism of formation and information etc. Means of communication being used
propagation of flood. to communicate information of flood are given
In the first category, the forecasting methods which below.
are generally graphical or in the form of

1. CWC wireless network developed and used. This method helps to tide over
errors due to aggradation or degradation in the
2. Custom and Central Excise wireless network
forecasting station.
3. Bihar Police wireless network
Also, correlation between the change in stage of the
4. Inter State Police Wireless network forecasting station and change in the stage of base
stations during T’ hours (T = Time of travel of flood
5. Telephone wave between the base station and forecasting site).
6. Telegrams is developed and used This method helps to tide over
errors due to aggradation or degradation in the river
7. Special Messenger section depending upon flows.
Description of wireless sets at various stations in Further, correlation between the N hour and (N+T)
plain of Kosi, maintained by Middle Gang.: Division hour stages of the forecasting station with change in
IV CWC is given in Table- 5. stages at the base station during past ‘T’ hours as
METHODS OF FLOOD FORECASTING USED variable is also used and mathematical equations are
BY CWC IN THE BASIN developed corresponding to rising and falling
conditions of the river.
Method based on statistical approach has been used
to analyze the historical data with an objective to In tributary like Kamala Balan, where wide
develop methods for the formulation of flood fluctuations in the stages of the upstream station
forecasts. The methods thus developed have been causes relatively identical/less fluctuations at the
presented in the form of graphical downstream station due to valley storage in between
relations/mathematical equations. a large number of the base and forecasting stations, relationship has
data covering a wide range of conditions have been been established between Nth and hour (N+T)th hour
analyzed to derive the relationship between gauge to stage of the forecasting site with change in the stage
gauge relationship with or without parameters and of the forecasting site in the past ‘T’ hours and
rainfall-peak relationship. average gauge of the base station. .
Methodology The mathematical equation developed is based on the
fact that it should have fairly constant travel time &
The variables which affect the stage at forecasting having considerable discharge & no major tributary
site are identified and mathematical relationship thus between the reach.
established between the stage at forecasting site and
the identified variables (Chander S and Prasad T., Table 3: Rain Gauge Stations in the Kosi Basin for
1981). The probable variables which used to affect at Flood Forecasting (CWC, 1989 & 1992)
forecasting points are given below. Sl. Name Location Freq. Type
(a) Stage and Discharge of the base station, N. of
Longitude Latitude
Station
(b) Change in stage and discharge of the base
station,
(c) Travel time at various stages,
d) The rainfall (amount, intensity and duration)., in 1. Birpur 87°00’E 26°25’N Hourly Ord/
the intercepting catchment. SR
(e) Topography, nature of vegetation etc. 2. Basua 86°36’E 26°07’N Hourly Ord/
SR
(f) The climatic and atmospheric conditions,
3. Baltara 86°50’E 25°30’N Hourly Ord/
(g) Stage and discharge of any tributary joining the
SR
main stream between base and forecasting stations.
4. Kursela 87°16’E 25°30N Hourly Ord/
Factors (a) to (c) being the basic parameters have
been used in the development of the correlation SR
curves and factor (g), being very important, has been Kamla Balan River
considered where required.
5. Jainagar 86°14’ 26°.34’ Daily Ord/
Direct Correlation between Gauges at U/S and E N SR
D/S
6. Jhanjhar 86°40’ 26°16’N Hourly Ord/
Correlation between Nth hour stage of base station pur E SR
and (N+T)th hour stage of the forecasting station,
where ‘T’ hours (T=Time of travels of flood wave 7. Balan 86°20’ 26°37’ Hourly Qrd/
between the base station and forecasting site) is E N SR

Sl. Name Location Freq. Type S. Name of the Danger Highest Flood
N. of N. Site Level Level Level
Longitude Latitude Year
Station
15. Barahachatra 120.30 132.18 1968
16. Birpur 74.70 75.53 1990
Adhwara Group Rivers

Therefore, precisely the travel time characteristics of
the North Bihar rivers can be put in three distinct
- categories:
8. Sonbarsa 85°30’ 26°25’ Hourly Ord/ (a) For rivers in the upper reaches, the travels time
E N S.R. increases uniformly with the increase in flood
9. Saulighat 85°50’ 26°25’ Hourly Ord/ stage as the rivers are very shallow and has less
E N S.R. water carrying channel capacity.
10. Kamtaul 85°48’ 26°10’ Hourly Ord/ (b) The rivers in the middle reaches have got a
E N S.R constant travel time as the river water remains
confined to its banks.
(c) As soon as the flood water starts spilling over
However, in North Bihar rivers, the travel time is not banks in flat plains, there is large increase in
constant and varies with water level and found to travel time for even very small increase in flood
give considerable errors under different conditions. stage due to the inundation on both sides of the
Therefore, while issuing forecasts, the following river getting large.
aspects are taken into account:
(i) The variation in travel time,
Table 5: Wireless Network in the Kosi Basin (CWC,
(ii) Varying conditions during rising and falling 1989 & 1992)
stages of the flood.
No.
(iii) Characteristics of flood wave, Sl. Type
River Station of
No. of set
Table 4: Gauge sites on Kosi and its Tributaries for sets
Flood Forecasting CWC, 1989 & 1992) 1. Bagmati Runisaidpur - -
S. Name of the Danger Highest Flood 2. Bagmati Benibad 1 15
N. Site Level Level Level
Year watt
Bagmati River 3. Bagmati Hayaghat - -
1. Benibad 48.68 49.66 1987 4. Adhwara Sonebarsa - -
2. Hayaghat 45.72 48.96 1987 Group
3. Runisaidpur 53.73 54.46 1987 5. Adhwara Kamtaul 1 15
4. Dheng bridge 70.10 72.00 1975
Group watt
Kamla Balan River 6. Adhwara Saulighat - -

Group
5. Jhanjharpur 50.00 52.73 1987
6. Jainagar 67.75 71.35 1965
7. Kamla Jainagar 1 15
Balan watt
7. Balan H/W 75.34 79.56 1987
8. Kamla Jhanjharpur 1 20
Adhawara group River
Balan watt
8. Kamtaul 50.00 52.99 1987
9. Kamla Balan - -
9. Ekmighat 46.94 49.27 1987 Balan
10. Sonbarsa 80.66 82.60 1987
10. Kosi Birpur 1 15
11. Solighat 52.37 54.90 1987 watt
Kosi River 11. Kosi Basua 1 20
12. Basua 47.75 48.65 1987 watt
14. Karsela 30.00 31.85 1978 12. Kosi Baltara 1 15

No. Travel time from (i) Basua to Baltara is 18 hours, &
Sl. Type
River Station of
No. of set (ii) Hayaghat to Baltara is 28 hours.
sets
Case I- Basua & Hayaghat both Rising
watt
BALN+T = BALN + 0.875876 ( * BAS)1.123508
13. Kosi Kursela 1 15
watt + 0.9376 (* HAY)1.302
Case II- Basua Rising & Hayaghat Falling:
Mathematical Relationship BALN+T = BALN + 0.875876 ( * BAS)1.123508
With the synopsis as generalized above, method and + 0.5 (* HAY)
mathematical equations used for individual Case III- Basua Falling & Hayaghat Rising:
forecasting stations under Middle Ganga Division
IV, CWC, Patna, are given below BALN+T = BALN + 0.7 ( * BAS) + 0.9376 (*
HAY)1.30
Equation for Basua
Basua is the first forecasting site on the river Kosi
where co-relation between change in discharge with Case IV- B.asua & Hayaghat both Failing:
change in gauge at Barahkshetra base station and BALN+T = BALN - 0.7 ( * BAS) +-0.5 (* HAY)
corresponding gauge observations at Basua is
developed having time of travel i.e warning time is Where,
24 hours. However the warning time at Basua
BALN & BALN+T = Gauge at BALTARA at Nth &
forecasting site is reduced to 16 hours when co-
(N+T)th hour
relationship of discharge released and gauge
respectively
observation at Birpur barrage and corresponding
stages at Basua established and the mathematical (*BAS) = Change in Gauge at BASUA during
equation developed for this forecasting site as follow. past
a. For Rising Stage ‘T’ hours.
0.86851
BASN+T = BASN + 0.64607(BIRN - BIRN.-T) (*HAY) = Change in Gauge at Hayaghat during
past
b. For Falling Stage ‘T’ hours.
0.86851
BASN+T = BASN — 0.64607(BIRN-T - BIRN)
Equation for Kursela (Kosi)
Where, BASN & BASN+T are Gauge at Basua at
Kursela is the third and the last forecasting site on
Nth &
river Kosi. The gauges at Kursela are influenced by
hour (N+T)th respectively the back water effect of Ganga. Accounting for this
aspect, a two quadrant graph is being used. The base
and BIRN & BIR N-T are Gauge at Birpur at Nth
stations are, Hathidah on Ganga & Basuà (Kosi). The
& ((N-T)th hour respectively. travel time for the forecast reach is as given below:
T = Travel time, 16 hours. (i) Basua to Kursela is 26 hours &
Equation for Baltara (Kosi) (ii) Hathidah to Kursela varies from 20 to 45 hours.
Baltara is the second station on the river Kosi having The mathematical equation for the forecasting site is
co-relations of gauge observation between Basua and as follow
Baltara. The river Bagmati along with river Kamla-
Case I- Hathidah & Basua Both Rising:
Balan meets the river Kosi at a point which is
slightly upstream of Baltara so the gauge at Hayaghat KUR N+T = KURN + 0.7143 (*HTD) + 0.5 (*BAS)
site on the river Bagmati is also co-related with the
gauge at Baltara which gives another parameters for
forecasting at this site. Case II- Hathidah & Basua both Falling:
Forecasting for this site is done with due change in KUR N+T = KURN + 0.75 (*HTD) + 0.5 (*BA5)
gauge at base station Basua during the past ‘T’ hours
Case III- Hathidah Rising & Basua Falling:
in the 1st quadrant and incorporating the change in
stage at the tributary base station Hayaghat during KUR N+T = KURN + 0.7143 (*HTD) + 0.5 (*BAS)
past {(N-10)-(N-28)} hours in the second quadrant.

Case IV- Hathidah Falling & Basua Rising: imparted to the concerned officials on a regular basis
so that they are well versed in the interpretation of
KUR N+T = KURN + 0.75 (*HTD) + 0.5 (*BAS)
the forecast and taking precautionary measures in the
Where. event of an imminent threat to the life and property.
This training programme should become regular
KUR N+T & KURN = Gauge at Kursela at features before the flood season every year as per
(N+T )th & Nth hours respectively Second Bihar State Irrigation Commission Report,
(1994).
(*HTD = Change in Hathidah gauge during
REFERENCES
Past ‘T’ hours
Central Water Commission, (1989), Manual on
(*BAS) = Change in Basua gauge during Flood Forecasting, River Management Wing,
Past ‘T’ hours New Delhi.
Central Water Commission, (1992), Manual on
Flood Forecasting, New Delhi.
CONCLUSIONS Chander S and Prasad T., (1981), Forecasting and
Prediction of Floods, Indo- US Workshop on
Flood forecasting has proved to be of great help in Flood Mitigation, New Delhi.
issuing warning to the people in flood prone areas IWRS, (2001), Theme Paper on Management of
,organizing flood fighting and safety measures for Floods and Droughts Indian Water Resources
the engineering works, timely evacuation of people Society, Roorkee.
from affected area, salvation of movable properties Second Bihar State Irrigation Commission Report,
besides mobilizing relief operations as per IWRS (1994), Flood and Drainage Problem of Bihar
(2001),. and Their Remedial Measures, Volume V,
It has come to the notice that the overall performance Patna.
of the forecasting made at all these three stations
have been found to be quite satisfactory when
compared with the actual gauge level. However, the
forecasting can be improved by increasing the
warning time if more gauge discharge sites being
established by CWC in the upper reach of the river in
Nepal portion start functioning.
Although there is wide appreciation of Flood
Forecasting System and Warning issued by CWC,
there is very little feedback on the procedures
specified or evolved by the civil administration and
the engineering organization for undertaking
relief/rescue/ precautionary action on the basis of the
forecast. It is also not known as to how effectively
the necessary advice is being given to the people.
On receipt of the forecast, its dissemination to the
local population in terms of likely depth of
inundation and its duration in the areas by the
administrative authorities is very important so that
affected population, cattle, movable properties etc.
are evacuated before the area gets submerged by the
flood waters which cause damage. For this, a
network of wireless stations and telephone system
are necessary in the basin near critical/vulnerable
reaches of embankment and towns etc. especially
where other means of communications are not
dependable or adequate. Flood Warning to smaller
areas in villages may be conveyed through public
address system or in its absence by beat of drum.
Specific advice should be given to the people
regarding evacuation the areas likely to be affected
and also about the locations which could be
considered safe for the level indicated in the flood
forecast .Necessary training in this regard should be

ISBN 978-81-921121-0-7
PUSHOVER ANALYSIS OF RCC BUILDING WITH AND WITHOUT

INFILL WALL
R. E. Dalvi
Post Graduate student, Structural Engineering Department, V.J.T.I.,Mumbai-400019,
Email: ranjud123@gmail.com , ranjana_dalvi@yahoo.co.in
ABSTRACT: A large number of buildings in India are constructed with masonry infills for functional and
architectural reasons. Masonry infills are normally considered as non-structural elements and their stiffness
contributions are generally ignored in practice. However, infill walls tend to interact with the frame when the
structure is subjected to lateral loads. In the present study, effect of masonry infill wall on frame structure
building is studied along with mass irregularity at ground and 5th story level. Pushover analysis is carried out on
G+7 story rectangular 3D frame RCC building using nonlinear analyses option of commercial software
SAP2000(V.14.2).The building elements first design according to the IS456:2000. Both geometric and material
nonlinearity is considered in the structure. For nonlinear analysis, hinges assign to beam and column. The
diagonal strut approach is adopted for modeling of masonry infill walls as compression only member with pin
joint at the both ends. Presence of masonry infill wall alters displacements and base shear of the frame. Masonry
infill walls increase global stiffness and strength of the structure. But discontinuous and irregular arrangement of
infill wall drastically changes the failure mechanisms of building frame.
INTRODUCTION walls develop cracks once their ability to carry

A large number of buildings in India are constructed horizontal load is exceeded. Thus, infill walls
with masonry infills for functional and architectural develop cracks under severe ground shaking but help
reasons. In buildings, the ordinarily occurring vertical share the load of the beams and columns until
loads, dead or live do not pose much of a problem, but cracking. Although the infill panels significantly
the lateral loads due to earthquake tremors area matter enhance both the stiffness and strength of the frame,
of great concern and need special consideration in the their contribution is often not taken into account
design of buildings. These lateral forces can produce because of the lack of knowledge of the composite
the critical stress in a structure, set up undesirable behavior of the frame and the infill.
vibrations and in addition, cause lateral sway of the
structure which can reach a stage of discomfort to the Masonry infill wall panels increase strength, stiffness,
occupants. In many countries situated in seismic overall ductility and energy dissipation of the
regions, reinforced concrete frames are infilled fully building. More importantly, they help in drastically
or partially by brick masonry panels with or without reducing the deformation and ductility demand on RC
openings. frame members. This explains the excellent
performance of many such buildings in moderate
Masonry infills are normally considered as non- earthquakes even when the buildings were not
structural elements and their stiffness contributions designed or detailed for earthquake forces. The
are generally ignored in practice. However, infill reinforcement in the infills does not contribute
walls tend to interact with the frame when the significantly towards stiffness and strength; in fact, it
structure is subjected to lateral loads, and also exhibit may lead to reduction in stiffness and strength due to
energy-dissipation characteristics under seismic increased mortar thickness in the layers containing the
loading. Masonry walls contribute to the stiffness of reinforcement. However, the reinforcement helps in
the infill under the action of lateral load. The term improving the post-cracking behaviour of the
‘infilled frame’ is used to denote a composite masonry and in preventing out-of-plane collapse.
structure formed by the combination of a moment
resisting plane frame and infill walls. The composite Conversely there is a common misconception for infill
behaviour of an infilled frame imparts lateral stiffness wall that it increase the overall lateral load carrying
and strength to the building. capacity. This would lead to undesirable performance
Normally, infill walls are not connected to of moment resisting frames because the infill which
surrounding RC columns and beams. When columns was not considered during design stage would modify
receive horizontal forces at floor levels, they try to the inherent of RC frame members. As a consequence,
move in the horizontal direction, but masonry walls failure in different forms would be the result due to
tend to resist this movement. Due to their heavy additional loads on the stiffened members
weight and thickness, these walls attract rather large The present study is investigates the contribution of
horizontal forces. Infill walls are behaves like infill walls to the building’s seismic response.
compression strut in moment resisting RC frame Hypothetical eight-storeyed RCC frame buildings
However, since masonry is a brittle material, these with brick masonry infills are considered for the

present study. Different type of infill arrangements hi = Height of floor i measured from base and
considered in plan and elevation of building. Pushover n = Number of storey in the building.
analysis for bare framed and infilled framed RCC
building is carried out with mass irregularities in Calculation of Width of Diagonal Strut
structure. Formula used for width calculation for the equivalent
diagonal compression strut member is as given in the
STRUCTURAL MODEL DETAILS FEMA 356 (7.5.2.1)
A 8 story RCC frame building is analyzed using a=0.175(λ1hcol) –0.4 rinf (3)
sap2000 (nonlinear version 14) Building dimension is Where
16mX12m. Beam size is 230X450 mm and column
size is 600X400 mm for ground, 1, 2 story and
450X400 is used for 3rd to 5th story. Slab thickness is
120mm. Typical story height is 3m. Bay width is 4m
in X direction and 3m in Y direction. hcol = Column height between centerlines of beams, m
hinf = Height of infill panel, m.
Material Properties Efe = Expected modulus of elasticity of frame
The grade of concrete used M20 for beam and M25 material, kN/m2
for column having Young modulus 22360 N/mm2 and Eme = Expected modulus of elasticity of infill
25000 N/mm2 respectively with Poissons Ratio 0.2. material, kN/m2
Grade of Steel used Fe 415 having, Youngs Modulus Icol = Moment of inertia of column, m4
2.1 x 105N/mm2 and Poissons Ratio 0.3. Unit weight Linf = Length of infill panel, m
Bricks masonry is 20 KN/m3 and Youngs Modulus rinf = Diagonal length of infill panel, m.
2530 N/mm2 with Poissons Ratio 0.15 .The modulus tinf = Thickness of infill panel and equivalent strut, m
of brick masonry is obtained using FEMA 356(2000)
recommendations i.e Em = 550fm where fm is Both material and geometric nonlinearity is
compressive strength of brick masonry. considered in analyses which are available in
SAP2000. Plastic hinges assign to beam (M3) and
Loading data column (PM2M3) and tension limit taken as zero in
The building frame is subjected to dead load of 1 equivalent diagonal strut member. P-delta effects are
KN/m2 on floor 1.5 KN/m2 on roof and live load of 4 considered in analysis.
KN/m2 on floor and 2 KN/m2 on roof. The structure is
considered to located in the Indian seismic zone IV PARAMETRIC STUDY
with hard soil strata. Response Reduction Factor (R) Typical pushover analysis is achieved using
of 5 is considered for design the structural system (IS displacement control strategy, where in the whole
1893:2000-Part I) consistent with the code. structure is pushed to evaluate the seismic
Importance Factor for building is taken as 1. performance of the buildings using preselected lateral
load pattern until the roof displacement reaches the
Equivalent Lateral load pattern: IS 1893 (part I), target value. In addition to these lateral loadings the
2002 structures are subjected to dead loads and live loads.
This method uses the equivalent lateral forces due to The results of various models were discussed
fundamental period of vibrations. The code lateral separately to have proper comparison with that of the
load shape represents the forces obtained from the bare frame model.
predominant mode of the vibration and uses the
parabolic distribution of lateral forces along the height Following different models are analyzed using
of the building. The following expression has been sap2000
used to calculate the load pattern as per IS 1893 (Part- Model I : Bare frame
I): 2002. Model II : Masonry infill are arranged in outer
VB = AhW (1) periphery
Model III: Masonry infill are arranged in symmetric
manner
(2) Model IV: Masonry infill are arranged in
unsymmetrical manner
These models are analyzed with following cases.
Where, 1. Without soft story
VB = Design Base Shear as per IS 1893(Part-I): 2002 2. Soft story at ground floor
Ah = Design horizontal acceleration spectrum value 3. Soft story at 5th floor
as per IS 1893(Part-I): 2002
W= Seismic weight of the building as per 7.4.2.
Qi = Lateral Force at Floor i ,
Wi = Seismic weight of floor i ,

These models are as follows:
Figure 1: Elevation- CASE 1 Figure 2 : Elevation- CASE 2 Figure 3 :Elevation- CASE 3
Figure 4: Plan –Model 1 Figure 5: Plan –Model 2 Figure 6: Plan –Model 3 Figure 7: Plan –Model 4
OBSERVATION TABLES
Case 1 : Without soft story
Table 1: Linear analysis for earthquake loading , CASE1
Model No. 1 2 3 4
Fundamental time
1.19 0.8404 0.8482 0.8890
period(sec)
Roof
0.0201 0.0082 0.00892 0.00898
displacement(m)
Table 2: Pushover analysis, CASE1

Model No. 1 2 3 4
Base Reaction(KN) 1312.12 3197.12 2825.04 2782.88
IO
Roof displacement(m) 0.0350 0.03434 0.03256 0.03234
Base Reaction(KN) 1976.35 6099.68 5804.26 6085.48
LS
Base Reaction(KN) 2210.58 6704.4 6242.97 6153.38
CP
CASE 2 : With soft story at ground level

Table 3: Linear analysis for earthquake loading, CASE 2
Model No. 2 3 4
Fundamental time period(sec) 0.8799 0.8816 0.9233
Roof displacement(m) 0.008628 0.009291 0.009314
Table 4: Pushover analysis, CASE 2
Model No. Model No 2 3 4

Base Reaction(KN) 2579.98 2428.6 2412.185
IO
Base Reaction(KN) 5895.87 5529.42 5401.6
LS
Base Reaction(KN) 5918.113 5592.12 5632.0
CP

CASE 3 : With soft story at 5th story level
Table 5: Linear analysis for earthquake loading , CASE 3
Model No. 2 3 4
Fundamental time
0.8352 0.8358 0.8739
period(sec)
Table 6: Pushover analysis, CASE 3

Model No. Model No 2 3 4
Base Reaction(KN) 2098.95 1783.70 1779.03
IO
Base Reaction(KN) 4951.16 4509.54 4482.60
LS
Base Reaction(KN) 5203.76 5169.67 4839.81
CP
CONCLUSION Masonry Infill Walls” , International Journal of

The based on above results, the following conclusion Science & Technology Volume 2, No 2, 155-164
are drawn. Kashif Mahmud , Md. Rashadul Islam and Md. Al-
 Fundamental time period reduces with Amin (2007) “Study The Reinforced Concrete
introduction of infill wall in RC frame panel. Frame With Brick Masonry Infill Due To Lateral
 Presence of masonry infill wall also alters Loads” , International Journal of Civil &
displacements and base shear of the frame. Environmental Engineering IJCEE-IJENS Vol: 10
 Masonry infill walls increase global stiffness No: 04
and strength of the structure. MatjaˇZ DolˇSek_, Peter Fajfar (2008),“The Effect Of
 Presence of masonry infill wall reduces the Masonry Infills On The Seismic Response Of A
ductility of the structure. Four-Storey Reinforced Concrete Frame—A
If infill wall arrange unsymmetrically, Deterministic Assessment” Engineering Structures
 Reduction in fundamental time period 30
decreases Sujatha , Jiji Anna Varughese, Bindhu K.R.(2009),
 Increase in base shear reduces. “The Influence Of Masonry Infill In Rc Multi-
Storey Buildings” , 10th National Conference on
 Reduction in displacement also decreases.
Technological Trends
When there is a mass irregularity in the structure,
S.M. Wilkinson , R.A. Hiley (2006) “A Non-Linear
 Consideration of infill wall drastically
Response History Model For The Seismic Analysis
changes the failure mechanisms of building
Of High-Rise Framed Buildings” Computers and
frame.
Structures 84
 Failure of frame elements takes place at level
Combescure, D. et al. (1995). Modelling of the in-
where change in infill pattern is occur.
plane behaviour of masonry infilled frames.
 Increase in base shear decreases as compared European Seismic Design Practice, Rotterdam,
to symmetric and continuous infill wall Balkema
arrangement. Haddad, M. H. E. (1991). "Finite element analysis of
infilled frames considerng cracking and separation
phenomena." Computers & Structures 41(3): 439-
REFERENCES 447.
V R Murty and Sudhir K Jain (2000), “Beneficial Polyakov, S. (1960). "On the Intereaction between
Influence of Masonry Infill Walls On Seismic Masonry Filler Walls and Enclosing Frame when
Loaded in the Plane of the wall."
Performance of Rc Frame Buildings” 12th World Holmes, M. (1961). Steel Frame with Brickwork and
Conference on Earthquake Engineering, Paper No. Concrete Infilling
1790 Stafford Smith, B. (1962). "Lateral stiffness of Infilled
Robin Davis, Praseetha Krishnan1, Devdas Menon, . Frames." 182-197.
Meher Prasad, (2004) “ Effect Of Infill Stiffness On Stafford Smith, B. (1966). "Behaviour of Square
Seismic Performance Of Multi-Storey Rc Framed Infilled Frame
Buildings In India ’’ 13th World Conference on Mainstone, R. J. (1971). On the Stiffness and Strength
Earthquake Engineering, Paper No. 1198 of infilled frames: 57-90
Kasım Armagan Korkmaz, Fuat and Mustafa (2007) Mallick, D. V. & Garg, R. P. (1971). Effects of
“Earthquake Assessment Of R/C Structures With openings on the lateral stiffness of the infilled
frames. Institute of Civil; engineering

Asteris, P. (2003). "Lateral stiffness of brick masonry
infilled plane frames." Structural Engineering 129:
1071-1080
IS code 1893:2000 ( Part1)
Applied Technology Council, ATC-40. Seismic
evaluation and retrofit of concrete buildings, vols. 1
& 2. California; 1996
FEMA-356, (2000). “Prestandards and commentary
for Seismic rehabilitation of buildings
Pankaj Agarwal And Manish Shrikhande , “A Book
Of Earthquake Resistance Design Of Structures”
Anil K. Chopra, “A Book of Dynamic Of Structure”

ISBN 978-81-921121-0-7
DEFORMATION AND STRENGTH CHARACTERISTICS OF FLY ASH

ADDED SOILS
Tapas Das
Post Graduate Student, Department of Civil Engineering, Indian Institute of Technology Guwahati, Guwahati - 781039
Email: d.tapas@iitg.ernet.in
Baleshwar Singh
Associate Professor, Department of Civil Engineering, Indian Institute of Technology Guwahati, Guwahati – 781039
Email: baleshwar@iitg.ernet.in
ABSTRACT: At present, for various types of construction, engineers are forced to lay foundation on soft
ground. The characteristics of soft ground are low strength, instability, high total settlement and differential
settlement. Ground improvement by treating soft soil with various types of industrial by-products and binders is
an attractive alternative and often economical compared to other ground improvement methods. Addition of
industrial by-products and binder alters the engineering properties of the existing soil to create a new site
material which is capable of enhancing strength. In India there are many industrial by-products, which are being
dumped as waste material, creating disposal problem and pollution hazard. Fly ash is a by-product of the
combustion of pulverized coal in thermal power plants. Geotechnical use of fly ash with soil provides viable
solutions for its large-scale disposal and utilization. Efforts have been made by different researchers in the
laboratory to understand the effect of fly ash addition on soils. This paper presents a review of the studies on
deformation and strength characteristics of fly ash added soils.
INTRODUCTION engineering applications such as construction of

Coal-based electrical power production has an embankments, dams, back-fill material behind
important role over the world. The combustion of coal retaining walls, base or sub-base material and land fill
is used throughout the world to raise steam for power reclamation (Joshi & Nagaraj 1987). The advantages
generation. About 80-90% of the ash formed from the of using fly ash in geotechnical applications are its
burnt coal is carried out of the furnace, extracted from low unit weight, high shear strength (Sridharan et al.
the flue gas and is known as fly ash. The remaining 1998), low compressibility, low specific gravity
coarser fraction falls to the bottom of the furnace (Pandian et al. 1998), insensitivity to moisture
where it sinters together to become the bottom ash. variation and pozzolanic nature (Sivapullaiah et al.
The quantity of fly ash produced depends on different 1998), which could result in significant engineering
factors such as the quality of coal, degree of benefits in addition to facilitating mass disposal of fly
pulverization, furnace design, changes in coal supply, ash. The effectiveness and economy of using fly ash
changes in boiler load, and firing conditions. The has been explored by researchers.
structure, composition and properties of the fly ash
particles depend upon the structure and composition Several research works and studies have already been
of the coal and the combustion processes by which the carried out to improve soil strength with the addition
fly ash is formed. of fly ash. In this paper an attempt has been made to
review the effects of fly ash addition on deformation
In addition to the fossilised organic matter, coal and strength characteristics of soils.
contains an appreciable amount of inorganic material
in the form of quartz, clay and fine-grained minerals. FLY ASH PROPERTIES
Fly ashes are mainly classified in two groups, “C” and Since rock detritus or mineral matter in coal is likely
“F” classes. When fly ash consists of more than 10% to vary from one coal to another, variations are to be
CaO, the fly ash is categorized as class C. Class C expected between ashes. Further variations are
ashes which are produced from lignite coals have introduced by the actual firing process, the degree of
more pozzolanic and cementing activity than class F pulverization the coal receives, the type of furnace,
ashes which are produced from bituminous coals. and the firing conditions. At thermal power plants, the
Both types are available in India. amount of producing fly ash trapped from chimney
gasses mainly depend on the level of the technology
The increase in the demand for power has led to the of the plant, which is about 15% of the coal burnt.
commissioning of large coal-fired power stations in
India. The generation of fly ash at present is about 100 Mineralogical and Chemical Properties
million tons per year, but its use is less than 10%. The The majority of mineral matter consists of clays,
management and disposal of the remaining huge pyrite and calcite. More than 85% of most fly ashes
quantity of fly ash has created tremendous consist of glassy and crystalline compounds formed
environmental problems. Bulk use of fly ash or fly by the thermal treatment of minerals. In considering
ash/soil mixtures is possible in geotechnical the composition of fly ashes, a convenient division

based on their chemical nature can be made, notably and aluminous materials (pozzolans) in the fly ash. A
between those derived from sub-bituminous and similar reaction can occur in class F fly ash, but lime
lignite coal sources and those derived from must be added because the lime content of the ash is
bituminous coal. Sub-bituminous and lignite coal too low. Lime stabilization of soils occurs in a similar
ashes are characterised by a relatively high proportion manner, where the pozzolanic reactions depend on the
of CaO and MgO to Fe2O3, as distinct from siliceous and aluminous materials provided by the
bituminous coal ash, which is usually low in CaO and soil.
MgO and relatively rich in Fe2O3. Some of the coal
particles remain unburned and are collected with the EFFECT OF FLY ASH ON SOIL PROPERTIES
fly ash as carbon particles. The amount of unburned
coal is determined by such factors as the rate of Compaction Behaviour
combustion, the air/fuel ratio and the degree of In many studies, fly ash-soil ratio by weight have
pulverization of the coal. Typical chemical preferred between 5% and 30%. Since the specific
components of fly ashes which exhibit varying degree gravity of soils is greater and the optimum water
of self hardening characteristics are summarized in content is smaller than fly ashes, the increase in the
Table 1 (Joshi 2010). fly ash content decreases the maximum dry density of
the mixture, on the contrary the fly ash content
Table 1 Typical chemical composition of fly ash increases the optimum water content. Compaction
Constituents % by weight characteristics of clay-fly ash mixtures vary between
SiO2 29 - 58 100% of clay and 100% of fly ash depending upon the
Al2O3 14 - 25 fly ash/clay ratio as expected.
Fe2O3 3-8
CaO 10 - 30 Tan & Iyisan (2001) have investigated the effects of
MgO 1-8 the admixture of 10% and 20% fly ash (class C
SO3 0-4 contains 12% CaO) on the mechanical behaviour of
Loss on Ignition, LOI 0-1 the clay. Compaction curves under the Standard
Proctor energy have been given in Figure 1.
Physical and Geotechnical Properties
Physical and index properties of fly ash mainly
depend on the quality and grinding size, burning
technology of the plant, ash collecting and stocking
method. Fly ash is generally consist of grey, some
black portion when unburned, depending upon
combustion efficiency, 1 to 100 µm rounded particles,
commonly in silt size, non-plastic material. The
specific gravity of fly ash paticles varies from 2.05 to
2.66 (Joshi 2010). Fly ash containing high amounts of
unburnt carbon and hollow spherical particles, called
cenospheres, has low specific gravity. Larger
proportions of iron-rich particles and other
compounds impart a high specific gravity to the ash.
Pozzolanic Activity
Formation of cementitious material by the reaction of
lime with the pozzolans (Al2O3, SiO2, Fe2O3) in the
presence of water is known as hydration of fly ash.
The hydrated calcium silicate gel or calcium
aluminate gel (cementitious material) can bind inert
material together. The pozzolanic reactions for soil
stabilization are as follows (Senol et al. 2002): Fig. 1 Compaction curves of clay and mixtures
(after Tan & Iyisan 2001)
CaO + H2O => Ca(OH)2
Ca(OH)2 => Ca++ + 2[OH]-
++ -
Ca + 2[OH] + SiO2 => CSH Deformation and Strength Behaviour
(silica) (gel) Pandian et al. (2001) have studied the effect of
Ca++ + 2[OH]- + Al2O3 => CSH Raichur fly ash (class F contains 1.7% CaO) on the
(alumina) (gel) CBR characteristics of the black cotton (BC) soil.
Figures 2 and 3 show the stress versus penetration
For class C fly ash (containing more than 15% CaO), plot and variation of CBR values with the addition of
the lime present in the fly ash reacts with the siliceous fly ash content respectively for soaked condition.

The results clearly bring out that the fly ash could be initial tangent modulus increases by adding fly ash to
an effective additive (about 20%) to improve the CBR the soil. As can be seen from the test results (Table 2),
of BC soil (about 200%) significantly. On the other cohesion of the fly ash admixtures soil increases about
hand, the CBR of fly ash could be improved (about 20% by adding fly ash. An obvious increment is
50%) with the addition of 30% BC soil. The mixing observed in angle of shear strength by both time and
of fly ash and BC soil in suitable proportions amount of added fly ash (Table 2). Fly ash content has
improves the compacted strength as well as CBR major effect than the time most probably due to
values considerably. rounded particles of the fly ash. By the time
pozzolanic or cementing properties increases the shear
strength.
The deformation modulus increases by fly ash content

and by the time as can be seen in Figure 4, which
should signify increase in strength. As in the shear
strength, in tangent modulus increment of fly ash
content is more effective than the time. Tangent
modulus increases very sharply from from 10% to
about 75% but the increment by the time is around
25~50% in 30 days.
Another parameter of the stress-deformation

characteristics is the deformation to reach the failure.
While the tangent modulus increases by fly ash
content but no significant changes in failure
deformation (Fig. 5).
Table 2 Experimental results of shear strength

Fig. 2 Stress versus penetration plots of CBR test for parameters
BC soil treated with class F fly ash (after Pandian et
al. 2001) (after Tan & Iyisan 2001)
Fly ash/clay Curing time Cohesion, Angle of
ratio (%) (days) c (kPa) shear
0 -- 125 6
10 1 150 11
10 7 150 12
10 33 150 13
20 1 150 15
20 7 150 16
20 33 150 18
Fig. 3 Variation of CBR values with the addition of

class F fly ash to BC soil (after Pandian &
Krishna 2001)
Tan & Iyisan (2001) have investigated the effects of

the admixture of 10% and 20% fly ash (class C
contains 12% CaO) on the mechanical behaviour of
the clay and the time effect. Experimental results
represent that even a day after sample preparation, the Fig. 4 Variation of the tangent modulus with the time
strength increases significantly. Both cohesion and (after Tan & Iyisan 2001)
angle of friction terms of Mohr-Coulomb criteria,

The results show that upon addition of fly ash to BC
soil, the CBR of the mix increases. The increase in
CBR with the addition of fly ash can be attributed to
pozzolanic reaction forming a cementitious
compound, which results in good binding between BC
soil and fly ash particles. The increase in CBR under
soaked conditions (i.e. curing) is due mainly to the
presence of calcium present in the “high lime content
fly ash” that results in a pozzolanic reaction leading to
increased strength. For fly ashes with low calcium, the
soaked CBR will be less than the unsoaked CBR.
Fig. 5 Variation of deformation at failure of mixture

with the amount of added fly ash (after Tan & Iyisan
2001)
Senol et al. (2002) have found that the class C fly ash
can be used without any other activator for improving
the engineering properties of soft soil. Like
unconfined compressive strength and CBR, resilient
modulus increases substantially with increase in fly
ash content (Fig. 6). The rate of increase of CBR
diminishes as the fly ash content increases. The
strength of fly ash added soil can be maximized by
stabilizing at specified water content and minimizing
compaction delay.
Fig. 7 Stress versus penetration plots of CBR test for

BC soil treated with class C fly ash (after
Pandian & Krishna 2003)
Sezer et al. (2006) have investigated the stabilization

of a soft clay subgrade of a military zone in Izmir,
Turkey with a very high lime fly ash obtained from
the Soma Power Plant. It can be clearly seen from
plots of unconfined compressive strength (qu)-fly ash
content (Fig. 9) and cohesion (c)-fly ash content (Fig.
10) that addition of fly ash up to 15% increases both
qu and c values of the resultant mixtures. Increasing
the fly ash substitution level to 20% has a negligible
effect on the above-mentioned properties as compared
to that of the 15% FA bearing mixture.
Fig. 6 Resilient modulus of fly ash added soil at The effects of fly ash addition on the internal friction
different fly ash contents (after Senol et al. angle ( ) of the mixtures are shown in Figure 11. The
2002) increases with time irrespective of the fly ash
content of the mixture. Besides, internal friction angle
Pandian & Krishna (2003) have studied the effect of increases considerably with increasing fly ash
Neyveli fly ash (class C contains 15.5% CaO) on the inclusion level.
CBR characteristics of the black cotton (BC) soil.
Figures 7 and 8 show the stress versus penetration
plot and variation of CBR values with the addition of
fly ash content respectively for soaked condition.

Fig. 11 Variation of  with fly ash content (after
Sezer et al. 2006)
CONCLUSIONS
The study reveals that the deformation and strength
characteristics of soft soils can be improved by use of
Fig. 8 Variation of CBR values with the addition of “high lime content fly ash” without any other
class C fly ash to BC soil (after Pandian & Krishna activator.
2003)
The addition of fly ash to soil leads to significant
improvement on geotechnical behaviour of clay. The
unconfined compressive strength (qu), cohesion (c)
and angle of internal friction increase ( ) with
addition of fly ash and the stabilizing effect of fly ash
increases with time.
The deformation modulus viz. tangent modulus and

resilient modulus increase sharply with fly ash
content. Deformation at failure has no significance on
fly ash content but the stress at failure increases with
fly ash content.
REFERENCES
Joshi, R.C. (2010). Fly ash - production, variability
and possible complete utilization. Indian
Fig. 9 Variation of qu with fly ash content (after Geotechnical Conference – 2010, GEOtrendz, IIT
Sezer et al. 2006) Bombay.
Joshi, R.C. and Nagaraj, T.S. (1987). Fly ash
utilization for soil improvement. Environmental
Geotechnics and Problematic Soils and Rocks,
Balasubramaniam et al. (eds), Balkema,
Rotterdam, 15-24.
Pandian, N.S. and Krishna, K.C. (2003). The
pozzolanic effect of fly ash on the California
Bearing Ratio behaviour of black cotton soil.
Journal of Testing and Evaluation, ASTM, Vol.
31, No. 6.
Pandian, N.S., Krishna, K.C. and Sridharan, A.
(2001). California Bearing Ratio behaviour of
soil/fly ash mixtures. Journal of Testing and
Evaluation, ASTM, Vol. 29, No. 2, 220-226.
Pandian, N.S., Rajasekhar, C. and Sridharan, A.
Fig. 10 Variation of c with fly ash content (after (1998). Studies on the specific gravity of some
Sezer et al. 2006) Indian coal ashes. Journal of Testing and
Evaluation, ASTM, Vol. 26, No. 3, 177-186.

Senol, A., Bin-Shafique, Md. S, Edil, T.B. and
Benson, C.H. (2002). Use of class C fly ash for
stabilization of soft subgrade. Fifth International
Congress on Advances in Civil Engineering,
Istanbul Technical University, Istanbul, Turkey.
Sezer, A., Inan, G., Yilmaz, H.R. and Ramyar, K.
(2006). Utilization of a very high lime fly ash for
improvement of Izmir clay. Building and
Environment, 41, 150-155.
Sivapullaiah, P.V., Prashanth, J.P., Sridharan, A. and
Narayana, B.V. (1998). Reactive silica and
strength of fly ashes. Geotechnical and Geological
Engineering, No. 16, 239-250.
Sridharan, A., Pandian, N.S. and Srinivasa Rao, P.
(1998). Shear strength characteristics of some
Indian fly ashes. Ground Improvement, London,
Vol. 2, No. 3, 141-146.
Tan, O. and Iyisan, R. (2001). Strength and
deformation characteristics of fly ash added clay.
Proc. of the Fifteenth International Conference on
Soil Mechanics and Geotechnical Engineering,
Istanbul, Turkey, Vols. -3, 1859-1862.

ISBN 978-81-921121-0-7
GIS: AS A NECESSARY TOOL FOR URBAN PLANNING

Arati S. Petkar
Assistant Professor, Civil Engineering Department, College of Engineering, Pune – 411005, Email: asp.civil@coep.ac.in
Eshwar M. Bahirat
M. Tech Student, Civil Engineering Department, College of Engineering, Pune – 411005, Email:eshwar.jgd@gmail.com
ABSTRACT: The conventional or traditional method of creating map on a piece of paper for the planning
purposes are time consuming and with less accuracy. Further, the information which is collected for survey has
both spatial attributes as well as its relative non-spatial attributes. This information should be properly stored,
managed and analyzed for getting the desired results with a speedy process and less time. Geographic
Information Systems (GIS) helps to greater extent in storing, managing, and analyzing the data as well to add or
to change the spatial information in existing data, with the facility of linking the attribute data with map. This
paper focuses on how GIS is useful for the data storage, handling, analyzing as well as spatial, graphical,
statistical representation of data. Why it is necessary tool for urban planning, how it will help decision makers
and planners in urban planning process. In order to understand the theory and the explanation given in the paper,
an area from Pune city is selected as an example for the application for GIS.
KEY WORDS: Urban planning, spatial analysis, attribute data, Geo-referencing
INTRODUCTION Geographic Information System, designed in, the mid-

The urban planning process includes multiple aspects 1960s as a computerized map measuring system. GIS
like surveys, data collection, data storage, data is a computer based information system which is used
analysis, spatial data representation, map creation and to represent and analyze the information of the
predictions for the future development. The data geographic features which are present on the surface
which is used for planning purpose is spatial data, of Earth. Like, other information systems, the GIS not
which must be geographically referenced. While the only store, display and modify the data, but also it can
maps which are created for the reference must be in analyze, query, and retrieve. The basic function of
the scale of the geographically referenced spatial data. GIS is to work with data referenced by geographical
For the purpose of efficient planning process these coordinates. It can also deal with the information from
planning aspects needs to be addressed very skilfully different databases for putting them on map. GIS has
and carefully for the predictions in the future two basic features firstly it can generate the map with
development. For this purpose of urban planning geographically referenced objects and secondly it can
process the information system which is used for the link the database associated with the geographically
data collection, storage, analysis, representation, map referenced objects with the set of operations for
creation and retrieval of data must be liable to handle working with data. So with the specific set of
these planning aspects efficiently. So the necessity of operations GIS visualize the backend data on the map
the efficient information system for the purpose of as well as make the analysis from the data for the
efficient urban planning process is real need in this desired outcome for planning issues. This is the basic
technological era. difference between the GIS and other information
system.
THE NECESSITY OF INFORMATION SYSTEM
The purpose of any information system is to store, APPLICATION AREAS OF GIS
analyse and display the data so that whenever required Geographic Information System encompasses both
it can be used and recollected for the specific purpose the digital and geographical techniques involved in
as well as it can be modified whenever necessary. So the systems used for the processing and dissemination
any information system which is dealing with the of geographic information as well it, facilitates
planning aspect for the efficient planning process has combining of data from a variety of sources with
got very much importance for the future development. different map scales and produces the desired results
The information system should be such efficient that as per the user need. GIS, therefore, as a tool, can
it can handle all these aspects effectively. enhance and improve the quality of the planning
process with better and accurate maps creation and
Geographic Information Systems (GIS) is such kind analysis to solve spatial problems with speed. Like
of information system which is used for to store and other information systems GIS has specific functions
analyse the data as well as to display data with to deal with the data. The use of GIS for urban
function of modification of data. Moreover the GIS is planning has following functions which are beneficial
an information system which is for capturing, storing, as compared to other database systems.
querying, analyzing, and displaying geographically
referenced data. The first GIS was the Canada

Data Types it can be projected on the globe or real world. In GIS
The data collected for the planning purpose mostly Geographic Coordinate System (GCS): spherical
comes in three basic forms. coordinates, Cartesian coordinate system (CCS):
planar coordinates, are used for the data projection.
Vector Data: It is the data which contains the location
and shape of geographic objects. This data has three The Geo-Referencing and Projection System
basic shapes to present real-world features: points, Georeferencing is the important tool for GIS, which is
lines, and polygons. not effectively, used in the other information systems.
This tool provides the function to locate the
coordinates of the spatial objects with respect to the
Attribute Data: Attribute data is the descriptive data earth’s coordinate system that is GCS. Once the
that GIS used to link with map features. Attribute data
coordinates of the spatial objects are located with
is collected and compiled for specific areas like names
respect to the earth’s coordinates the georeferencing is
of the locations, population data, temperature and
done and the objects are referenced on their exact
rainfall variation data , and so on and comes with
geographic locations on the globe. So in this way any
reference to vector data. spatial data can be located to their exact locations with
the help of geo referencing tool. The georeferencing
Raster Data: This is a data which is collected from provides the locating the spatial data to its
satellite images, aerial photographs, and scanned geographical locations with GCS. Which will gives
images. the coordinates in latitude and longitude which are
spherical coordinates. This coordinate system is not
Data Models handy for the planning purpose. So it must be
For to processing the data any information system projected into the CCS or planar coordinate system.
requires the data model. GIS has two data models, When GIS obtains the data, it often needs to be
Raster data model and vector data model. referenced and projected. Since the data received is
not processed and it should be process and need to
Vector Data Model: Discrete features, such as place the coordinates to raster image. The earth shape
locations and data summarized by area, are usually is spherical and we must transform its three
represented using the vector model. dimensional surface to create a flat map sheet. And
this process of transformation is termed as a map
projection in GIS. So using the GIS the data can be
Raster Data Model: The raster data model is the data
georeferenced to the geographical coordinates at the
model which is the image data directly used in GIS
same it can be projected to the Cartesian coordinate
for the data analysis. Continuous numeric values, such
system. So the dual purpose is fulfilled by the
as elevation, and continuous categories, such as
georeferencing and projection tool in GIS.
vegetation types, are represented using the raster
model.
Join and Relationship Class
The use of data models: The join provides the function to join the database
The either data model is used in GIS as per the users table from different databases like SQL server,
need and the data collected. Discrete features, such as Microsoft access and other compatible formats in to
customer locations, pole locations or others; and data the GIS map on the basis of common field related to
summarized by area, such as postal code areas or the map attributes. This process of joining the
lakes; are usually represented using the vector model. descriptive data makes the map analysis easy and fast
Continuous categories, such as soil type, rainfall, or as well as representation of the table data with
elevation, are usually represented using the raster reference to the spatial locations. This makes the
model. planners work easy and fast for analyzing the data on
the basis of these attributes. This data can be queried
Coordinate System by using the select by attribute query. For e.g. the
The data which is collected has to be georeferenced so recreational plots in the particular area of Pune city
that it can be projected on the globe or real world. A say kasaba peth need to find out for the area
coordinate system is a reference system which is used calculations and finding the deficiency for the
to represent the locations of geographic features population residing in the nearby areas. So in this
within a common geographic framework. For any case, by using the select by attribute query the
spatial mapping system the coordinate system is recreational areas can be found out in kasaba peth. So
important. GIS provides the coordinate system which by using this query only recreational areas from
is spatially referenced and can be projected in the kasaba peth can be selected and its area can be found
metric system. GIS has both spatially referenced out. So by using this technique any particular data can
system geographical coordinate system (GCS) and be selected and analysed for decision making. So this
Universal Transverse Mercator System (UTM). The is the important tool for planner’s point of view for
data which is collected has to be georeferenced so that

the analysis of the spatial data and making the correct city Development Plan (DP) can be clipped and it can
proposals. be used as different layer with its associated attributes.
Relationship class is important when one to many For e.g. if Sadashiv peth area is to be clipped from the
relations are there for e.g. for one residential land use sector-1 of Pune DP then by using the clip feature it
there are many buildings and when these buildings are will be very easy to clip and make separate layer of
to be related then it is not easy to relate or join in such sadashiv peth, and do particular analysis of the
cases the relationship class is used to relate the data Sadashiv peth. Refer Fig. 1 and Fig. 2.
on the basis of common field in both the databases. In
the case of study area the common field the GIS Buffer and overlay
identity (GISid) is the common field, on the basis of The buffers are the influence areas around the vector
land use data, the one to many fields of buildings the data sets. Buffers are created around point, line and
GISid from sub-land use table the relationship class polygon. The application of buffer in GIS has got the
between the land use table and sub-land use table is highest important as it gives the output very precise
made. and accurate. For e.g. square on laxmi road say
shagun chowk is to be widened for the turning of the
Symbology heavy traffic movement, so in this case the point
For any mapping system the annotation system is buffer will be used at the centre of the square and the
most important for the graphical representation of the buffer value will be given for the square to be
database. Maps are an interface to GIS. They are used widened, so the buffer value will be given as per the
for variety of requirement and the buffer will be created. Now the
method of communicating geospatial data. For the buffer overlaps on the adjoining properties, so by
effective communication the symbology and using the overlay technique, how much part of each
annotation techniques are very important. GIS adjoining plots is to be acquired can be calculated.
provides large scope for the symbology This method can be applicable for the whole Pune city
customization. GIS has different types squares at a time in a single command. By using the
combined and can be displayed as the map. So this same buffer and overlay technique, the road widening
provides the large scope for to represent the data on can be done with the line buffer technique. Also
the purposes. As a visual tool, maps are most effective common areas between the two adjoining areas can be
of symbolization system they are Features, found out by using the polygon buffer. Fig. 3
Categories, technique, the different land uses in indicates use of point buffer on laxmi road and Fig. 4
particular locations say peth area in Pune city can be indicates use of intersect overlay on laxmi road.
represented. So by using this technique the data can
be displayed with the combined coding system for the Join
peth area and the land The join provides us the joining the database table
uses. So for every land use in every peth there will be from
single and combined code. Like these three fields can different databases like SQL server, Microsoft access
be Quantities, Charts, and Multiple attributes using and other compatible formats in to the GIS map on the
these symbolization techniques the data can be basis of common field related to the map attributes,
analysed very which makes the map analysis easy and fast as well as
easily. For e.g. in Pune Development Plan the GIS id representation of the table data on the spatial
is the common field which is used for the data locations, Refer Fig. 5. This process makes the
linkage. So without the data linkage we cannot sort planners work easy and fast for the decision making
the land use data for analyzing the land uses, road network, water
according to land use codes and represent on the map. bodies location, areas of land uses etc. this is the
So after the data linkage the data can be represented important tool for planners point of view for the
by using the land use codes and create the map. As analysis of the spatial data and making the correct
well using the proposals.
combination of the fields used in the data base the
data can be graphically represented on map. So using Relationship class
this basis Relationship class is important when one to many
Of the different data fields which are attached by join relations are there for e.g. for one residential land use
and relationship class. there are many buildings and when these buildings are
to be spatially related then it is not easy to relate or
Clipping join in such cases the relationship class is used to
Clipping is the method of cutting required set of relate the data on the basis of common field in both
feature or polygons from the total map and creating the databases, Refer Fig. 6. In case of study area the
the new map with new layer along with all the common field the GISID is the common field, on the
attributes with the clipped map. This method is very basis of land use data, the one to many fields of
useful to create the map of required feature from the buildings the GISID from sub-land use table the
original map. By this method any area from the Pune

relationship class between the land use table and sub-
land use table is made.
CONCLUSION
In developing country like India the urban planning
process is continuous process, and for this purpose a
vast data is required to be handled repetitively for
different purposes. The GIS is the tool which can
handle this data for the effective urban planning
process with precise, accurate and speedy data
processing as well as useful for analysis and graphical
data representation. For efficient and integrated
planning process all the urban planning authorities of
India should adapt the GIS technology. The data
collection, manipulation and the data processing
should be done according to GIS technology from the
beginning of the planning stage for the effective usage
of the GIS technology.
ACKNOWLEDGEMENTS
We would like to thank Dr. Pratap Raval and Dr.
Nikhil Shejawalkar for their valuable support
REFERENCES
Anshu Gupta, Dr. Jagdish Singh, An Investigation of
Reliability on Remote Sensing and GIS Data as an
Aid to Urban Development Plan: A Case Study on
Bhopal, Institute of Town planners, India Journal 7
- 4, 90 - 100, October - December 2010
Desai C.G., Patil M.B,, Mahale V.D. and Umrikar B.,
Application of remote sensing and geographic
information system to study land use / land cover
changes: a case study of Pune Metropolis,
Advances in Computational Research, ISSN: 0975–
3273, Volume 1, Issue 2, 2009, pp-10-13
Dr.M.Krishnaveni, J.Prakashvel, Dr.M.Kaarmegam,
(2003), Capabilities for Interlinking of Indian
Rivers, Map Asia conference 2003- Water
Resources.
K. Ram Mohan Rao, Ismat Bakshihov and B.S.
Sokhi, A Customized GIS application for
Dehradun Tourist Information System, Institute of
Town Planners, India Journal 6 - 2, 86-96, April -
June 2009
M. Anji Reddy (2003), remote sensing and
geographical information systems, second Edition,
BS Publication, Hyderabad.
Vivek N Patkar (2003), Directions for GIS in Urban
Planning, Map Asia conference 2003- Geospatial
Application Papers.

Fig. 1 Pune city, Sector-1
Fig. 2 Clipping of sadashiv peth, Pune city
Fig. 3 Use of point buffer on laxmi road, Pune

Fig. 4 Use of intersect overlay on laxmi road, Pune
Fig. 5 Joining the table to the map
Fig. 6 Relationship class on the basis of common field- GISID

ISBN 978-81-921121-0-7
CANAL REGULATION AND SCADA

Kinnari R. Mishra
Water Resource Management Department, L.D.College of Engineering, Ahmedabad, Gujarat.
E-mail: kinnari.mishra@gmail.com
N.K.Sherasia
Water Resource Management Department, L.D.College of Engineering, Ahmedabad, Gujarat.
INTRODUCTION collection stations, automated gates and valves,

The water which enters into the main canal from the river has to be Remote Terminal Units (RTU), communication
divided into different branches and distributaries, in accordance networks, and SCADA control systems that will save
with the relative urgency of demand on different channels. This time (labour and support costs), water (reduce or
process of distribution is called ‘Regulation’. To distribute water eliminate spillage) and money (increase overall
effectively, the discharge has to be adjusted to any desired value. efficiency).
This aim is achieved by means of regulators. The ability to know the exact status of all aspects of
The works which are constructed in order to control and regulate your water delivery system at all times and
discharges, depths, velocities etc, in canals, are known as canal efficiently control operation is now possible by
regulation works. These structures ensures the efficient functioning innovative products and turn-key SCADA control
of a canal irrigation system. Canal Regulators such as distributary systems.
head regulator and cross regulator Flumes are the structures which Technology can assess needs and provide a system
help in regulating the canal. that continuously collects station or system-wide data
in real time, efficiently operates automated
Necessity of the regulators: equipment in SCADA systems, and distributes real
 To effectively control the entire Canal Irrigation System. time data to those who need it. Having data at your
 To regulate or control the supplies entering the off-take fingertips allows for sound decision making, but is
channel. also necessary for accurate and timely billing, record
 To control silt entry into the off-take channel. keeping, and reporting to state agencies.
 To serve as meter for measuring discharge.
 To maintain water level. IMPORTANCE OF SCADA OVER OTHER
 To absorb fluctuations in various sections of the canal system. CONVENTIONAL METHOD
 To prevent the possibilities of breaches in the tail reaches Today, competition for scarce water resources and
competition in the marketplace for quality and price
are placing increased pressures on irrigated
Water distribution infrastructure has been targeted agriculture, and consequently on the operation and
for upgrades and modernised to promote efficiency performance of irrigation water delivery systems.
and conservation. Systems in place for 50 years or While knowledge of irrigation and methods for
more are common. These old systems depend on distributing water over large areas slowly advanced
occasional manual observation and control of over the millennia. Fixed structures Many early
conditions within a system and assume spillage. schemes for the diversion, conveyance, and
Labour and fuel were cheap and conservation not distribution of water relied on fixed structures. This
critical or mandated. is an advancement over use of a shovel or digging
In the last years, digital controllers became a very tool to open earth berms. Fixed structures, when
interesting alternative (low costs and higher properly constructed, are efficient at "water
accuracy) to the analogue or to the hydrodynamic spreading." Such systems are supply oriented and
traditional controllers in water supply canal resemble water disposal systems. Water diverted at
automation, in order to match water supply to water the head finds the path of least resistance as it winds
demands. This kind of hydraulic systems needs its way through the canal network to the fields. Such
particular research for control applications because systems do not provide the flexibility and service
they are big scale systems, open and characterized by required for efficient utilization of irrigation water
big delays and great inertia. Low-cost communication supplies. The need to constantly adjust to changing
technology, automation, easily deployable sensors, demands, due to weather conditions and the variety
and computer software provide the tools to of crops and farm irrigation systems, is increasingly
modernize how a system is operated to reduce costs, important. Irrigation water delivery systems are
optimize conservation, and report operational data in being forced to become more and more demand
real time as required. Technologies has the oriented, which requires better water control.
demonstrated ability to design, manufacture, install, Pressurized systems would satisfy these
and maintain data requirements, but they are prohibitively expensive
for large structure.

SCADA' Impact On Regulation:
Remote monitoring with SCADA systems has had a
major impact on the operation of irrigation canal
systems over the last several decades. SCADA
systems allow operators to monitor and control water
distribution from a central location, rather than from
riding or driving to the canal where they could see
and control only one gate structure at a time.
SCADA systems are now economical for even the
smallest water districts. Good water measurement is
also essential for improving water control. In general,
check structures do not give accurate water
measurement. Over the last several decades, modern
flumes and weirs have provided simple, accurate, and
inexpensive means for measuring water flow at all
Fig. 1 SCADA software
bifurcation or division points. In places where such
measurement is difficult, acoustic flow meters are
now becoming common. The ability to tie remote Working of SCADA
water measurement information into SCADA SCADA refers to a system that collects data from
software allows continuous monitoring of flows various sensors at a factory, plant or in other remote
within the system. This has also proven useful for locations and then sends this data to a central
water accounting and documentation of annual water computer which then manages and controls the data.
uses. There are many parts of a working SCADA system.
A SCADA system usually includes signal hardware
SYSTEM CONTROL AND DATA ACQUISITION (input and output), controllers, networks, user
interface (HMI), communications equipment and
About Software software. All together, the term SCADA refers to the
SCADA software is used to provide a graphical entire central system.
interface between an operator and a network of
sensors and control devices deployed in the field. It A key tool of integrated water resources management
can include the entire program for operating the is automation of the water distribution system based on
SCADA system, including communications, or be an introducing the state-of-the-art system of supervisory
adjunct to a Master RTU (Remote Terminal Unit) control and data acquisition (SCADA). This system
that communicates to a deployed set of slave RTUs, allows improving the quality, flexibility and reliability
but in all cases it is used to display data, alarms, and of water distribution management and reducing
equipment status while providing for control of unproductive losses of water resources.
deployed devices, SCADA software was developed
for process control applications typically associated Supervisory control consists of bringing system-wide
with large-scale industrial operations. Now SCADA information from remote sites to a single master
is developed by different agencies suited for station. Supervisory monitoring can give a water
Maritime and Water Management markets that master the power to view the whole project without
brings this powerful technology to tasks of more leaving the office. Supervisory control changes the
modest size than the typical industrial setting. This target points of local controllers and empowers a
saves programming time and money along the way. water master to make rapid coordinated changes at
Software is uniquely configured for each SCADA key structures. Supervisory control was implemented
project. The objectives of the HMI operation are to in the 1970s on several irrigation systems, such as
provide clear images, efficiently generate commands the Salt River Project and the Coachella Irrigation
and reports and communicate with RTUs in the field District. With further advances in equipment
and remote operators. How efficiently software can Supervisory Control and Data Acquisition (SCADA)
be deployed and operated depends on the complexity has now spread to a number of irrigation districts in
of the task. the all over the world. Adoption of SCADA has
considerably improved the performance of irrigation
districts.
Software can be managed and modified by the
system operator. This feature makes a vast difference
in the deployed cost of a program and certainly its
maintenance since a field engineer may not be
required to make typical changes or additions to a
project.

Few SCADA projects have been implemented in
developing countries and many of them have failed
for various reasons. The most critical phase in an
automation project is implementation: Selection of
the equipment, the integration of hardware and
software components, installation and testing.
Shortcomings in electronic/communications-based
automation are possible at all stages: Design,
implementation and operation, control algorithm
limitation, poor integration of components,
malfunctioning of equipment, lack of training of
operation staff, lack of spare parts and poor
maintenance.
Fig. 2 Supervised control
REFERENCES
Irrigation and Hydraulic Structure By S.K.Garg
http://program-plc.blogspot.com/2011/01/canal-
automation-history.htm
http://en.wikipedia.org/wiki/SCADA
http://www.ncs.gov/library/tech_bulletins/2004/tib_0
4-1.pdf
Fig. 3 Centralised control
Fig. 4 Distributary control

ISBN 978-81-921121-0-7
DELINEATION OF GROUNDWATER SATURATED FRACTURE

ZONES IN HARD ROCK TERRAINS USING INTEGRATED
RESISTIVITY SURVEY APPROACH
N.P. Singh
Associate Professor, Department of Geophysics, Banaras Hindu University, Varanasi- 221 005, Email:
singhnpbhu@yahoo.co.in
ABSTRACT: Groundwater in hard rocks owes its occurrences to secondary porosity, generally developed due
to weathering, fracturing, jointing, faulting and fissures etc. In present study, an attempt has been made to
delineate the fractured zones saturated with groundwater using resistivity survey techniques in Birsa Agriculture
University campus, Kanke, Ranchi District, Jharkhand. Gradient profiling, Wenner profiling and geoelectrical
soundings were carried out in the area during the December 27, 2008 to Jan 10, 2009. Gradient profiling survey
was conducted along two parallel profiles with fixed current electrodes separation of 600m and potential
electrodes separation of 10m for the points lying between central regions of the profile (i.e. about 1/3rd of the
total spread length of geoelectrical sounding survey) at station interval of 5m and covering a total length of
500m . Wenner profiling with electrode separation ( a  30m ) was conducted along a specific profile line of
gradient profiling for comparative study. In Wenner profiling, the magnitude of observed apparent resistivity is
lower in comparison to the gradient profiling due to the predominant contribution of the surface layer as
compared to the underlying hard rock, which happened because the depth of penetration in Wenner profiling
was low since the current electrode separation was taken as 3 0 m only. However, in case of gradient profiling
the magnitude of the apparent resistivity is higher due to the predominant contribution of hard rocks compared
to the overlying surface layer of clay because the current electrode separation was 600m which resulted in a
relatively larger depth of penetration. The resolution of gradient profiling is more compared to Wenner
profiling. In case of gradient profiling, apparent resistivity varies between 7 3 ohm-m to 905 ohm-m, whereas in
case of Wenner profiling the apparent resistivity varies between 28 ohm-m and 128 ohm-m. Based on the
analyses of responses of gradient survey, four ‘low’ resistive and one ‘high’ resistive points were selected, and
thereafter at these selected locations geoelectrical sounding survey were carried out for detailed analyses with
depth. The interpreted sounding results show presence of moderately to highly fractured zones. Results of
soundings conducted at ‘low’ resistive points indicate that the resistivity of highly to moderately fractured zone
ranging between 38 ohm.m to 181 ohm.m, with varying thickness of 45m to 108m, a possible indication of
water saturated fractures. However, the interpreted results of the sounding conducted at ‘high’ resistive point
indicate high resistivity rocks below 15m from the ground surface. The study, clearly demonstrates that an
integrated resistivity survey approach (combination of gradient profiling and geoelectrical sounding techniques)
can be used successfully to investigate the presence of water saturated fractures in hard rock terrains.
INTRODUCTION for detection of cavities in limestones of Kurnool

In hard rock area, water saturated fractures are district, A.P., India. Yadav and Singh (2007 and
randomly distributed. Therefore, in such areas, if 2008) have applied Gradient profiling survey
geoelectrical sounding is conducted randomly then it followed by geoelectrical sounding for delineation of
may not coincide exactly with the fractured zone and water saturated fracture zones in Gurudev Nagar,
consequently fails to locate the presence of fractures Mirzapur, and Robertsganj, Sonebhadra district, U.P.,
in the area. Due to this reason, many unnecessary India, respectively in Vindhyan terrains. Their study
geoelectrical soundings would have to conduct in the show promising preliminary results in these areas. As
area which requires relatively more time and costs a confirmatory test for the preliminary success of
associated with the survey. Hence, there is need to test Yadav and Singh (2007 and 2008) and an
the various approaches of conducting geoelectrical experimental study for assessing the suitable
surveys and establish the particular one, which is combination of geoelectrical techniques in hard rock,
suitable for hard rock areas. Among the various an integrated survey approach consisting of gradient
approaches, the gradient profiling is possibly the one profiling followed by geoelectrical sounding was
way which might be used for mapping the fracture carried out in hard rock area of Ranchi, Jharkhand in
zones and their inter connectivity for groundwater the granitic terrain and the results have been discussed
exploration in hard rock areas. with supporting data.
Application of various techniques in hard rock areas
had been carried out by many researchers [Chandra et LOCATION MAP OF STUDY AREA
al. (1987), Yadav and Singh (2007 & 2008)]. Chandra The study has been carried out in the granitic terrain
et al. (1987) have applied an integrated approach of of Ranchi urban area, the capital city of Jharkhand
geoelectrical techniques in an experimental field study state. Jumar river flows in approximately east-west

direction with flow direction from east to west, and surface of a homogeneous earth is shown in Fig. (2).
was in flowing condition during the survey work In upper portion of the figure (Fig. 2a), the fixed
(December 27, 2008 to Jan 10, 2009). The survey area separation between current electrodes A & B is taken
falls under Chotanagpur plateau with granitic-gneiss as 600m. The central 1/3rd region between the current
rock type. The location map of the study area is electrodes (-100 m to 100m on both side of the centre)
presented in Fig. 1. is scanned at a regular station interval of 5m for the
smallest possible spacing of potential electrodes. The
smallest spacing of potential electrode can be taken as
JHARKHAND STATE
PA
KI
ST
AN
NE
PAL
10m which satisfies the condition L  x   5b to
B ANG
measure the potential gradient (Kunetz, 1966;

LAD
ESH
JHARKHAND
RANCHI Summer, 1976; Bertin and Loeb, 1976; Kearey and
Brooks, 1984; Sharma, 1997; and Telford et al.,
Study Area 1998). The current electrode separation can be
LAYOUT OF THE GEOPHYSICAL SURVEY AT BIRSA increased or decreased depending on the availability
AGRIICULTURE UNIVERSITY CAMPUS RANCHI
of the space and accordingly the observations within
the central region may be increased or decreased to
nn er
50 m
& We
fulfill the condition for the measurement of potential

GP- 1
C23
C13
gradient. For gradient profiling array, one can easily

C12
C22
JU M
AR
NA
DI
derive the formula for the field apparent resistivity as
following.
500 m
VES 5
VES 3 V (1)
C21 a  K
I
C11 VES 2
GP-2
VES 4
VES 1 BIRSA AGRICULTURE UNIVERSITY CAMPUS
RANCHI KANKE ROAD
KRISHAK
BHAW AN where K is the geometrical factor, which can be
ROAD
ROAD
written using Fig.(1a) as
 1 1 1 1  (2)
K  2     
 MA MB NA NB 
Fig. 1: Study area and locations of survey points.
Dark circles show the centre of the sounding (VES) With
along the profile line. MA (L x  b)2  d2 ; MB (L x  b)2  d2 ;
THEORETICAL BACKGROUND AND FIELD NA (L x  b)2  d2 ; andNB (L x  b)2  d2
PROCEEDURES
There are number of techniques for conducting In the above equation, 2 L & 2b are distances
geoelectrical surveys for groundwater investigations. between current electrodes and potential electrodes;
The gradient profiling array comprising of two fixed x is the distance between the centre of the current
current electrodes placed at large separation and one electrodes and centre of potential electrodes, d is the
moving potential dipole with considerably small distance between main profile and parallel sub-profile,
electrode separation is one of the technique used for  V is the observed potential difference, and I is the
groundwater exploration. The resistivity input current. The lower portion of Fig. 2b, shows
measurements are made using potential dipole in the theoretical responses of potential mV  and potential
central region, i.e. between the central 33% region of
the current electrodes separation along the line joining gradient mV / m over a homogeneous earth for a
the current electrodes or parallel to it. The potential fixed current electrode separation AB  600m . The
field in this region is constant, and as a consequence potential gradient is computed using I / 2  5000
of this a uniform horizontal electric field is generated I  1000mA,   31.4ohm  m  .
over a homogeneous earth. Therefore, in principle,
any inhomogeneity encountered within this region The potential gradient curve shows almost constant
produces variation in the response and thereby can be value in the central 1 / 3rd region and a significant
easily identified using gradient profiling technique. change in potential gradient beyond this region. Thus,
This forms the logic behind the application of the array can be used to measure the potential gradient
integrated gradient profiling and geoelectrical within the limit specified by L  3x  0
sounding techniques for exploration of ground water
in hard rock areas. and L  3d  0 . It is worthwhile mentioning here that
the field strength and depth of investigation will be
The general layout of gradient profiling array and nearly constant in the central region bounded by
variations of potential and potential gradient along a x  L / 3 and d  L / 3 in case of homogeneous
straight line passing through the point source at the
earth. Under the similar earth condition, if the

potential electrodes MN are kept out-side this region, towards the electrode end A or B . This problem can
the depth of investigation decreases and the horizontal be minimized provided the observations are confined
electric field deviates due to the influence of either to the central one-third region of AB . However, in
current electrode A or B depending upon its position. this study, the objective of applying gradient profiling
These changes in electric field and depth of is to use the technique only for qualitative
investigation, in real field condition, would produce interpretation of the data to identify the possible
responses leading to false interpretation of the locations of fracture zones, where geoelectrical
presence of inhomogeneities, which are actually not sounding can be conducted for further details using
there. the same array as a Schlumberger array (while
L  5b , and d  0 x  0 ).
y y
M N
A 2b B
Profile Lines
-x
O d (x,d)
x Gradient Profiling (GP)
Initially, two Gradient profiles (GP) were carried out
-y
2L
-y for a fixed current electrodes separation of 600m
(a) using the Resistivity meter- Syscal-R2. The potential
electrodes separation was taken as 10 m with
500 25 observation interval of 5m along a profile to cover
rd
200 m (  100m to 100 m ) span in the central 1/3
Potential region of current electrodes spacing. Computation of
apparent resistivity has been carried out for each
V/2b (mV/m)
Potential Gradient
gradient profile using equation (1). The data (apparent

Potential V (mV)
resistivity and distance) are plotted on linear scales.
V E S- 1
1000
Potential Gradient
0 0
-300 -200 -100 0 100 200 300 Gradient Profile -1

Distance (m)
A p p a re n t R e s is tiv ity ( o h m .m )
Gradient Profile
800
Separation AB = 600 m
V E S -2
Separation MN= 10 m
VE S - 4
(b) 600
-500 -25 V E S- 3
Wenner Profile
Fig. 2: (a) Lay-out of gradient measurement of 400
Separation AB (3a) =30 m
potential field, where AB = current electrodes and Separation MN (a) =10 m
VES- 5
MN = potential electrodes, and (b) Distribution of

200
potential and gradient of potential along a line at the
surface of a homogeneous earth for d  0 ,
I  1000 mA , and   31 .4ohm  m . 0
-100 -50 0 50 100 150 200 250 300 350 400

The prime aim of gradient profiling is to locate the Distance in m (Station interval=5m)
lateral inhomogeneities in the ground. Therefore, the
Fig. 3: Comparison of Gradient profiling response
above array can be used as Gradient profiling for
along the profile GP  1 for fixed current and potential
taking observations along the main profile AB when
electrode separations  AB  600m, MN  10m with
d  0 , and along the parallel sub-profile for a fixed
value of d for the same setup of current electrodes. Wenner profiling response for fixed current and
The measurements along parallel sub-profiles are potential electrode separations  AB  30, MN  10m .
normally made to trace the lateral extents and The selected prominent lows are marked as VES 2 ,
orientation of anomalous feature. In case, if the zone VES 3 , VES 4 & VES  5 and a prominent high is
of interest exceeds the 1/ 3rd central region or one marked as VES  1 on the profile.
wants to increase the coverage, the entire profile is Figure 3 shows Gradient profiling response (upper
shifted along the same line having same current
curve) and Wenner profiling response (lower curve)
electrodes spacing in such a manner that the shifted
for a total length of 500 m . The Gradient profiling
1/ 3rd central region overlaps the former one and cover curve is obtained for the potential electrode separation
the exceeded zone of interest.
of MN  10m for the fixed current electrode
Apart from the proven field advantages, the gradient
separation of 600m . The initial 200 m (  100m to
profiling suffers from certain serious limitation, such
as variation in depth of investigation within the 100 m ) of the response presents the result of first setup
current electrode interval. Depth of investigation is of the Gradient profile where as the remaining 300 m
deeper towards the centre of AB and shallower of response presents the result for next setup of the

gradient profile. The lower portion of the curve
presents the response of Wenner profiling for current Vertical Electrical Sounding (VES)
electrode separation of AB  30m . The comparison of
1000 9
both the response curves for Gradient profiling and 8
7
Wenner profiling is quite interesting. 6
5
4
The nature of the curve for Gradient profile shows 3
similar trend (considering moving average) as that for 2
VES-3 VES-4
the Wenner profile curve, except for the magnitude of

VES-2
apparent resistivity and the amplitude fluctuations. In 100 9
Apparent Resistivity (Ohm.m)

8
case of Wenner profiling the magnitude of apparent 7
6
resistivity is lower compared to the Gradient profiling 5

4
due to the predominant contribution of the surface 3
layer compared to the underlying hard rock – which is 2

VES-1
VES-5
due to the fact that the depth of penetration is low

10 9
(since the current electrode separation is taken as 8
7
30 m only). However, in case of Gradient profiling 6
5
the magnitude of the apparent resistivity is of higher 4

3
range due to the predominant contribution of hard 2
rock compared to the overlying surface layer as the
current electrode separation is 600m giving a 1
relatively larger depth of penetration. Variation in the 1
2 3 4 5 6 7 89
10
2 3 4 5 6 7 89
100
2 3 4 5 6 7 89
1000
Half- Current Electrode Spacing (AB/2) in m
apparent resistivity from left to right in Wenner
profile (lower curve) reflects the presence of different Fig. 5: Field apparent resistivity data and curves
nature of surface layer which is very well obtained through geoelectrical soundings at VES  1 ,
corroborated with the existing geomorphological VES  2 , VES  3 , VES  4 & VES  5 locations.
setup along the profile. In case of gradient profiling,
apparent resistivity varies between 73 ohm  m to Five vertical electrical sounding (VES) using
Schlumberger configuration were carried out at one
905 ohm  m , whereas in case of Wenner profiling the
‘high’ and four ‘low’ points selected from the
apparent resistivity varies between 28 ohm  m and
Gradient profiling along the gradient profile GP1 for
128ohm  m . Further, on the basis of analyses of the
the maximum current electrode spacing of 500m
responses of gradient survey, four ‘low’ resistive
using the same Syscal-R2 Resistivity-meter unit.
points and one ‘high’ resistive point were selected for
the vertical electrical sounding. The data obtained in the field at each sounding
1000
Gradient Profile -1 & 2 location contains the records for the value of current
and potential electrode spacing. Each data set was
A p p aren t R e sistiv ity (o hm .m )
Gradient Profile -1
800
Separation AB = 600 m processed and plotted for AB / 2 versus apparent
Separation MN= 10 m resistivity on a bi-logarithmic scale and interpreted in
600
Gradient Profile -2
terms of layer parameters i.e. thickness and resistivity
Separation AB = 600 m of individual layers. The field data and curves are
Separation MN= 10 m
400
presented in Fig. 5. During interpretation, the layer
parameters are initially obtained using the partial
curve matching technique with the help of three layer
200
master curves (Rijkswaterstaat, 1969) and auxiliary
point charts (Ebert, 1943). These parameters are
0 further used as initial models for computer assisted
-100 -50 0 50 100 150 200 250 300 350 400 software, IPI 2WIN . The final layer parameters are
Distance in m (Station interval=5m)
presented in Table-1.
Fig. 4: Gradient profiling curves along the profile
GP  1 and GP  2 (parallel profiles, 50 m apart) for
potential electrode separation MN  10m and current
electrode separation  AB  600m  .
Figure 4 shows Gradient profiling (( GP  1
& GP  2 ), parallel to each other and 50 m apart)
response over a total length of 500m. The first
200m (  100m to 100m ) of the response presents
the result of first setup of the Gradient profile where
as remaining 300m of response presents the results of
next setup of the Gradient profiling.

investigation of lateral variation of apparent resistivity
Table 1: Final interpretation in terms of lithological and selection of interesting point (‘low’) where further
units based on the results of geoelectrical vertical electrical surveys should be conducted for
interpretation in terms of layer parameters (Thickness variation of resistivity with depth. Thus, the gradient
and Resistivity) profiling survey are very much helpful in deciphering
the prospective locations for conducting detailed VES
Geoelectrical
Interpretation survey for exploration of ground water resources and
VES Lithological avoiding unfruitful attempts of conducting VES
Layer Resistivi
No. Thicknes Interpretation
s (m)
ty survey at several randomly selected locations, and
(ohm.m) thereby minimizing the cost and time of geoelectrical
1 1.0 41.6 Surface layer
Highly
surveys to a great extent.
2 1.7 10.7 Weathered The overall study demonstrates that the integrated
Formation resistivity survey (Gradient profiling and Vertical
VES-1
Highly Fractured electrical sounding) approach can be successfully
3 14.8 110.0
Formation applied for exploration of ground water resources in
Compact
4 1172.0
Formation hard rock terrains.
2 2.9 272.9
Weathered ACKNOWLEDGMENTS
Formation Author is grateful to Dr. P. Chandra, Regional
Highly
Director and Dr. S.K. Singh, Scientist, CGWB, MER,
VES-2 3 6.4 35.7 Weathered
Formation Patna for providing logistic and technical support for
Fractured conducting the survey. The author also acknowledges
4 108.0 181.4
Formation the support of M. Sc. (Tech) Geophysics- II year
5 251.5 Fractured Formation (Session 2008-09) students, who were actively
Weathered
involved in the survey work.
2 4.2 168.2
Formation
Highly REFERENCES
3 2.3 46.3 Weathered Bertin, J. and Loeb, J. (1976). Experimental and
VES-3
Formation
Theoretical Aspects of IP: Presentation and
Highly Fractured
4 45.5 92.0
Formation Application of the IP Method — Case Histories,
Compact Vol. 1, Gebruder Borntraeger, Berlin, 250pp.
5 67453
Formation Chandra, P.C., Tata, S.N. and Raju, K.C.B. (1987).
1 1.6 39.5 Surface layer Geoelectrical response of cavities in limestones: an
Weathered
2 4.4 71.1 experimental field study from Kurnool district,
Formation
Highly A.P., India. Journal of Applied Geophysics, 24,
VES-4
3 5.4 32.6 Weathered 483-502.
Formation Kearey, P. and Brooks, M. (1984). An Introduction to
Semi-Fractured Geophysical Exploration. Blackwell Scientific
4 429
Formation
Publications, Oxford, London, 296 pp.
Highly Kunetz, G. (1966). Principles of Direct Current
2 1.4 49.5 Weathered Resistivity Prospecting. Gebruder Borntraeger,
Formation Berlin, 250pp.
Weathered Sharma, P.V. (1997). Environmental and Engineering
3 9.7 138.0
Formation
VES-5 Highly Geophysics. Cambridge University Press, New
4 13.9 7.2 Weathered York. 475pp.
Formation Summer, J.S. (1976). Principles of IP for Geophysical
Highly Fractured Exploration. Elsevier, New York, 278pp.
5 94.2 38.9
Formation
Fractured
Telford, W.M., Geldart, L.P., Sheriff, R.E. and Keys,
6 155.0
Formation D. A. (1988). Applied Geophysics, 2nd Ed.,
Cambridge University Press.
CONCLUSIONS Yadav, G. S. and Singh, S.K. (2007). Integrated
This paper presents the efficacy of combined gradient resistivity surveys for delineation of fractures for
profiling and vertical electrical sounding groundwater exploration in hard rock areas. Journal
(Schlumberger sounding) surveys for delineating the of Applied Geophysics, 62, 301-312.
weathered and fractured zones in hard rock area. The Yadav, G. S. and Singh, S.K. (2008). Gradient
results presented for Wenner and gradient profiling profiling for the investigation of groundwater
surveys clearly indicate that gradient profiling survey saturated fractures in hard rock of Uttar Pradesh,
is more efficient and yields improved results with India. Hydrogeology Journal, 16, 363-372.
better resolution as compared to the Wenner profiling
survey. The gradient profiling surveys are useful in

ISBN 978-81-921121-0-7
VISCO-ELASTO-PLASTIC ANALYSIS OF LAYERED PLATE: A

UNIFIED OVERLAY APPROACH
P. Srimani
Post-Graduate Student, Department of Civil Engineering, Jadavpur University, Kolkata – 700032 Email:
psrimani@gmail.com
S. Roy Chowdhury
Assistant Professor, Department of Civil Engineering, Jadavpur University, Kolkata – 700032 Email: srceju@yahoo.com
K. K. Ghosh
Professor, Department of Civil Engineering, Jadavpur University, Kolkata – 700032 Email:g_kallol@vsnl.net
ABSTRACT: An overlay model is employed to provide a unified approach for visco-elasto-plastic analysis
of layered finite element for thick plate bending problems under monotonically increasing transverse static
loads. The plate element based on Reissner-Mindlin theory, is discretized into different layers across its
thickness to incorporate gradual propagation of plastic zone from outer most surface to mid plane as well as to
consider the possibility of variation of material properties of the plate in that direction. Furthermore each layer is
again divided, for mathematical convenience only, into different number of overlays. Each overlay can posses
different material properties or time dependent characteristics but undergoes same deformation. The total stress
field for a particular layer is obtained by summing the weighted contributions of each overlay. A basic
component is added which allows an instantaneous elastic response followed by viscous deformation, the onset
of which is governed by the yield limit of a parallel friction slider. By combining several of these components in
the form of overlays it is shown that a single layered finite element formulation is capable of modelling time-
dependent behaviour of materials in elastic as well as in plastic stage. Further numerical experiments are carried
out to prove its applicability.
INTRODUCTION Some attempts have been taken by Khan & Zhang

The deformation of the structures in inelastic range (2001) and Lin & Schomburg (2003) in recent year to
always has time rate effects to some extent. The combine viscoplasticity theories with viscoelasticity
prediction of structural response is significantly theories to characterize the mechanical behavior of
influenced by this effect and its degree depends on polymers. But the same approach has not been
several factors. The structures subjected to both reported for the analysis of structural components like
mechanical load and thermal load simultaneously slab, plate etc.
during a long period of time, exhibit simultaneously Alternatively ‘overlay approach’ has been followed
both the phenomena of creep and viscoplasticity. The by some researchers in the past in the similar
first one is essentially a redistribution of stress/strain applications. Zienkiewicz et al. (1972) introduced the
with time under elastic material response and that’s concept of a finite element ‘overlay’ model for elasto-
why can be termed as visco-elasticity. The second one plastic deformation of solids. Owen et al. (1974)
i.e. viscoplasticity is a time-dependent plastic firstly employed the same in the solution of several
deformation i.e. after initial yielding of the material numerical examples and compared with alternative
the plastic flow occurs and the resulting stresses and approaches. Pande et al. (1977) indicates how overlay
strains become time dependent. models can be employed in time-dependent material
Viscoelastic & viscoplastic characterizations of studies.
materials have always been an active research field in The objective of the present work is to develop a
continuum mechanics and material science since three unique constitutive model based on ‘overlay
or four decades. In the available literatures approach’ capable of both viscoelastic and
(Zienkiewicz et al. 1968, Dinis & Owen 1978), it is viscoplastic analysis of layered plate under transverse
found that visco-elastic and visco-plastic analysis of mechanical load.
different materials are performed separately by using
different formulations and different sets of material THEORY
parameters. But experimental observations cannot Elastic-viscoplastic analysis of layered plate
distinguish between these two phenomena. For the The first phase of this work is to develop an elastic-
theoretical interest, they are treated separately during viscoplastic finite element model for the analysis of
the time of analysis of such structures. layered plate under transverse load. This finite
This is why there is a need to develop a unique element plate model is developed in a layered
numerical method for predicting both types of time approach to analyze the plate made of isotropic,
dependent response of structural members that are orthotropic or any other composite materials and to
subjected to thermal loads in addition to mechanical simulate stress variation across the thickness of the
loads. plate.

The main reason to adopt the layered approach are- (i) F  F0
to incorporate variation of material properties of and  (F )  ( )
F0
composite plate or reinforced concrete slab etc. across
its thickness, (ii) to get the variation of stresses or
strains across the thickness of the plate, (iii) to locate Overlay approach – a unified model
the propagation of zone of plastification of the “Overlay concept” is applied in the same viscoplastic
materials in both inplane & transverse direction of the model to convert that into a unified approach for
plate. visco-elasto-plastic analysis of plate.
Each layer is again divided into different number of
For nonlinear continua problem    e   vp (1)
overlays. Here, in place of a layer, it is considered that
So,   e   vp (2) different overlay can posses different material
properties or time dependent characteristics.
Total stress rate   De Where, D = Elasticity Physically there is no existence of overlay. By setting
matrix. different suitable values of different material
properties like modulus of elasticity, yield stress,
fluidity parameter of different overlays, viscoelastic,
viscoplastic and combined visco-elasto-plastic
analysis have been done in this work.
Fig. 2 Overlay model
Fig. 1 Division of layered plate element and basic one

dimensional elasto-viscoplastic model
The onset of viscoplastic behavior is governed by a

scalar yield condition: F ( ,  vp )  F0  0 (3)
where F0  Uniaxial yield stress which may be a
function of hardening parameter. It is assumed that
viscoplastic flow occurs for the value of F>F0 only.
So, F ( ,  vp ,  )  F ( ,  vp )  F0  0 (4)
According to viscoplastic rule:  vp     ( F )  Q (5)

where, γ = Fluidity parameter. Fig. 3 Overlays in a single layer and the unified model
Q  Q( ,  vp ,  ) = Plastic potential.
All overlays present in a single layer undergo same
  (F )  Positive monotonic increasing function. deformation. This results in a different stress field
  ( F )   ( F ) for F>0 { i } in each overlay which contribute to the total
  ( F )  0 for F<0 stress field { } of the layer according to the overlay
For associated plasticity situation F  Q
thickness t i ,
F
So, vp     ( F )    a (6) k
 In which t i 1 (7)
F i 1
where,  a = Flow vector


    .t
k
i i (8)
i 1
Distribution of modulus of elasticity
k
E  E .t
i 1
i i
In the unified model viscoplastic strain rate at n’th

time step for r’th overlay of a single layer becomes: Table 1 Equivalent overlay model:
 Frn r n Number of layers considered = 10
 [ F0 ]  3
{ vp }r   r   { r }n (9) Number of overlay in each layer = 2
r n
[ F0 ]  1/ 2 Overlay Thk σy / Y γve ν
  2( j 2 ) r E
where, (t/cm2) (t/cm2) (min/t/
cm2)
[ Fr ] n  yield function & = zero [or arbitrary small
value] for viscoelastic overlay. So, Viscoelastic strain 1 0.5 ∞ 159.0 ----- 0.3
rate at n’th time step for r’th overlay of a single layer 2 0.5 0 8.525 3.01 0.3

becomes: {ve } r   r Frn  [F0 r ] n  3
 r n (10)
Load-deflection (vertical displacement at central point
2( j 2  ) 1r / 2
of the plate) curve has been shown in Figure 5 for
NUMERICAL EXAMPLES simply supported condition. As it is a time dependent
elastic deformation the load-deflection curve is linear.
A study of viscoelastic response with the unified To get the nature of time dependent nature of
overlay approach: deflection properly the load has been applied in six
A square plate [50cm X 50cm] of 1.25cm thick steps as: 1.40x10-4 t/cm2, 2.80x10-4 t/cm2, 4.20x10-4
(Fig.4) has been considered and analysis has been t/cm2, 5.20x10-4 t/cm2, 6.20x10-4 t/cm2, 7.00x10-4
done under increasing uniformly distributed t/cm2. The time dependent viscoelastic deformations
transverse loading. The material properties are: are shown in Figure 6.
E=83.7625 t/cm2, fluidity parameter of visco-elastic
response γve= 3.01x10-5 min/t/cm2, poisson’s ratio ν =
0.3.
Fig. 5 Load-displacement curves of viscoelastic

deformation
It is clear that viscoelastic deformation consists of two

part, one is instantaneous and other is time dependent.
For this problem the time depndent deformation is
Fig. 4 Finite element discretization of the plate quite small.
For viscoelastic analysis of the above problem two

overlay has been considered in each layer. Out of
these, one is maxwell overlay representing the time
dependent elastic deformations and other is fully
elastic overlay representing time indipendent elastic
behaviours. In this present model, maxwell overlay
has been developed by considering = 0 in classical
viscoplastic model and fully elastic overlay is the Fig. 6 w/h vs. time curves of viscoelastic deformation
viscoplastic overlay with .
The variation of stresses along the thickness of the
plate at the central point with the load increments has
been shown in the Figure 7.

Fig. 7 Stress distribution across the thickness at
central point of the plate in viscoelastic situation
Parametric study 1: Distribution of ‘E’ between two

overlays
The material considered here has the following
properties: E= 85 t/cm2, fluidity parameter of
viscoelastic response γ= 3.01x10-5 min/t/cm2. A study
has been performed for viscoelastic responses with Fig. 10 Effect on stress diagram across thickness:
different distributions of E in two overlays according [element 16, gauss point 4] (simply supported
to: E = [E1 x t1+E2 x t2] condition)
The combinations are considered as: (1) E1 = 100
t/cm2, E2 = 70 t/cm2 (2) E1 = 120 t/cm2, E2 = 50 t/cm2 Again Figure 10 shows that although the time
(3) E1=140 t/cm2, E2=30 t/cm2 (4) E1=160 t/cm2, dependent deformation increses with the increment of
E2=10 t/cm2. E2, as the total E = [E1 x t1+E2 x t2] remain constant in
From the load-deflection curves presented in Figure 8, different combinations, the stress distribution across
it is found that for [E1=160 t/cm2, E2=10 t/cm2], the the thickness remain unchanged for all type of
curve is stiffer than others and it is very close to the distributions of E.
fully elastic load-deflection curve with E = 85 t/cm2. This parametric study is very effective for the
It is also clear from Figure 8 that with the increse of E selection of distribution of E in different overlays for
of maxwell overlay (i.e. E2), the viscoelastic load- viscoelastic analysis of materials in overlay approach,
displacement curve becomes more softer. This is according to the experimental load-displacement
because with the increment of E2, the time dependent curve.
deformatin becomes more prominent. From Figure 9,
it is also clear that the time dependent deformation Parametric study 2: Variation of ‘γ’ in Maxwell
increses with the increment of E2. overlay
Another parametric study has been performed here
with different values of viscoelastic fluidity parameter
‘γ’. It is found that γ is a very important parameter for
the time dependent part of the viscoelastic
deformation. Figure 11 shows quicker time step
convergence with the increase of γ. That means
viscoelastic strain rate increases with the increment of
fluidity parameter. Another important observation is
that the γ has no effect on load-deflection curve as
Fig. 8 Effect on load-displacement curve shown in Figure 12.
Fig. 9 Effect on displacement-time curve

Fig. 11 Effect on load-displacement curve: [E1=100
t/cm2, E2=70 t/cm2]

Fig. 12 Effect on displacement-time curve:[E1=100
t/cm2, E2=70 t/cm2] Fig. 13 Load-displacement curves of visco-elasto-
plastic deformation.
A study of visco-elasto-plastic response with the
unified overlay approach
Same square plate [50cm X 50cm] of 1.25cm thick

under increasing uniformly distributed transverse
loading is considered. The material properties are:
E=83.7625 t/cm2, uniaxial yield stress = 0.057 t/cm2,
fluidity parameter of viscoelastic response γve= 3.01x
10-5 min/t/cm2, fluidity parameter of viscoplastic
response γvp = 2.15x10-5 min/t/cm2, poisson’s ratio ν =
0.3 Fig. 14 w/h vs. time curves of visco-elasto-plastic
Like previous case study, by employing at least one deformation.
“Maxwell type” overlay and one visco-plastic overlay, The liner portion of the curve represents the visco-
the complete behaviour in the visco-elastic range as elastic load-deflection response and after yielding it is
well as irrecoverable creep deformations can be gradually converted to visco-plastic response. The
generated. The maxwell overlay represents the time whole load-deflection path is time dependent.
dependent part of viscoelastic deformations and other To get the time dependent nature of deflection
viscoplastic overlay with a definite value of uniaxial properly, as shown in Figure 14, the load has been
yield stress (represented by the friction slider) gives a applied in five steps as follows:
fully elastic behaviour untill the yield stress has been a) 6.00x10-5 t/cm2, b) 1.00x10-4 t/cm2, c) 1.40x10-4
crossed. Thus the model behaves as a “standard linear t/cm2, d) 1.80x10-4 t/cm2 and e) 2.00x10-4 t/cm2.
solid” until the failure of the friction slider in the
visco-plastic overlay after which it behaves as a four
parameter fluid i.e. viscoplastic behaviour. In fact a
fifth parameter , the yield limit of the slider, must
also be defined. These parameters are material
characteristics and their values must be
experimentally determined.
Fig. 15 Stress distribution across thickness: [element

16, gauss point 4]
Figure 15 shows the stress diagram across the
thickness of the plate near the central point i.e at gauss
point 4 of element 16 at different stage of loading.
Table 2 Equivalent overlay model:
Number of layers considered = 10 Parametric study 1: Distribution of ‘E’ between two
Number of overlay in each layer = 2 overlays
Overl Thk σy / Y E γ (min/t/ ν Similar to the previous case study the distribution of E
ay (t/cm2) (t/cm2) cm2 ) in different overlays has been studied for visco-elasto-
1 0.5 0.114 159 2.15 0.3 plastic analysis also. The same material properties and
2 0.5 0 8.525 3.01 0.3 combinations of E distributions has been considered
here and similar type of results has been obtained.
Load-deflection (vertical displacement at central point
of the plate) curve has been shown in Figure 13 for
simply supported conditions.

Fig. 16 Effect on load-displacement curve Fig. 20 Effect of increase of γvp with a constant value
of γve i.e. 3x 10-5 min/t/cm2
Note: Although all results presented here are for
simply supported plate, but both the numerical
problems have been analyzed for clamped plate also
and similar type of results have been obtained.
CONCLUSIONS
The overlay approach is one of the better alternative
for the development of unified finite element
Fig. 17 Effect on displacement-time curve formulation that can be used for both visco-elastic and
viscoplastic analysis of layered plate structure. This
Parametric study 2: Variation of ‘γ’ in both approach is very effective for proper modelling of
Maxwell and viscoplastic overlay material behaviour without considering a large
Here also the similar type of results has been obtained amount of experimental informations needed for it.
as case study for viscoelastic analysis. It is found (Fig. The distribution of material properties like modulus of
18) that both γve and γvp has no considerable effects on elasticity (E), fluidity parameter (γ), thicknesses (t)
load displacement curve. But they have major effects etc. are to be properly selected to model the time
on the rate of time dependent viscoelastic and dependent behaviour of the material of the layered
viscoplastic deformations respectively. plate in a more realistic manner. Slight change in
some of the parameters of different overlays may
affect the overall structural response to a great extent.
Application of this overlay model can be suitably
extended for the visco-elasto-plastic analysis of the
plates made of layered composites, reinforced
concrete etc. This requires further attention for
incorporation of complicated material response and
failure criteria of those materials.
Fig. 18 Effect on load-displacement curve: [E1=100
t/cm2, E2=70 t/cm2] REFERENCES
Dinis, L.M.S. and Owen, D.R.J. (1978). Elastic-viscoplastic
In Figure 19 γve varies with a constant value of γvp i.e. analysis of plates by the finite element method.
Computer & Structures, 8, 207-215.
2x10-5 min/t/cm2. In which it is clear that variation of
Khan, A. and Zhang, H. (2001). Finite deformation of a
γve affects only on the viscoelastic time dependent polymer: experiments and modeling. Int .J. Plasticity,
deformation. With the increment of γve the rate of time 17, 1167-1188.
dependent viscoelastic displacements also increases. Lin, R.C. and Schomburg, U. (2003). A finite elastic-
On the other hand if γvp increases with a constant viscoelastic-elastoplastic material law with damage:
value of γve, only the viscoplastic strain rate increases theoretical and numerical aspects. Comput. Methods
while the viscoelastic time dependent deformation Appl. Mech. Eng.,192, 1591–1627.
remains unaffected Figure 20. Owen D.R.J., Prakash, A. and Zienkiewicz, O.C. (1974).
Finite Element Analysis of Composite Materials by use
of Overlay Systems. Computer & Structures, 4, 1251-
1267.
Pande, G.N., Owen, D.R.J. and Zienkiewicz, O.C. (1977).
Overlay models in Time-Dependent Non-Linear
Material Analysis. Computer & Structures, 7, 435-443.
Zienkiewicz, O.C., Watson, M. and King, I.P. (1968). A
numerical method of viscoelastic stress analysis. Int.
J. Mech. Scencei, 10, 807–827.
Zienkiewicz, O.C., Nayak, G.C. and Owen, D.R.J. (1972).
Composite and ‘overlay’ models in numerical analysis
Fig. 19 Effect of increase of γve with a constant value of elasto-plastic continua. Int. Symp. on Foundations of
Plasticity, September, 1972, Warsaw.
of γvp i.e. 2 x 10-5 min/t/cm2

ISBN 978-81-921121-0-7
AUTOMATED HIGHWAY SYSTEMS
S. Lakshminarasimhan
Student, Final year B.E civil engineering, Sri sairam engineering college Chennai-600044, Email – ar14jun@gmail.com
K.S. Yuvaraaj
Student, Final year B.E civil engineering, Sri sairam engineering college Chennai-600044, Email - yuvaraajkd@yahoo.com
ABSTRACT: The ever growing population and the increasing standard of living signifies the rapid and
uncontrolled growth in the traffic volume and since this growth is not properly met by the present road
infrastructure there is a possibility of reduction in the functional efficiency and therefore leads to the ultimate
situation of congestion and accidents. Not only this lead to some 50 billion collective hours of delay annually
but also an estimated productivity loss of $500 billion worldwide. The traditional solution to this problem is to
increase the road network but that is no longer a viable option owing to the high financial, social and
environmental impacts of such giant projects. Somehow we need to meet the unsatisfied demand for the
freedom and mobility of cars and other vehicles and find ways to operate the existing system more efficiently
and effectively. The idea is to have a magnetic reference sensor system for lateral control, an electronic throttle
actuation system, a communication protocol for vehicle-to-vehicle communication.
Keywords: Automated Highways, intelligent cars, lateral control, environmental impacts.
intelligence to support decision-making at the vehicle

AHS-THE EMERGING CONCEPT: level.
Definition
The Automated Highway System (AHS) concept Maximally Adaptable Concept:
defines a new relationship between vehicles and the This concept of maximum adaptability envisions a set
highway infrastructure. AHS refers to a set of of standards by which vehicles could operate under a
designated lanes on a limited access roadway where variety of conditions, depending on local design
specially equipped vehicles are operated under decisions.
completely automatic control. AHS uses vehicle and
highway control technologies that shift driving AHS - IMPLEMENTATION PROCESS
functions from the driver to the vehicle. Throttle,
steering, and braking are automatically controlled to The following describes the various steps in the
provide safer and more convenient travel. planning process and identifies which issues will be
most relevant for each step.
LEVELS OF AHS
A driver electing to use such an automated highway The Site Selection:
might first pass through a validation lane, similar to This step involves identifying the various potential
today's high-occupancy-vehicle (HOV).These lanes sites and conducting Field Tests (FT) to determine the
are also known as carpool lanes, commuter lanes, suitability.
diamond lanes, express lanes, and transit lanes.
Basically the level of Intelligence between the Site Characteristics:
vehicles and the infrastructure can be of the following  The site characteristics include descriptions
four types (Randolph W. Hall et al, 1997). of existing roadways,
 Surrounding land-uses and surrounding
Independent Vehicle Concept: communities, along with
This is a very futuristic model where vehicles will  Planned Non-AHS roadway changes and
have a high degree of autonomous operation,  Climatic conditions.
allowing them to operate with or without roadside Physical Characteristics
systems.  Number of Lanes, Width and length of
Existing Roadway
Cooperative:  Number of tight radius curves (and their
Vehicles have the capability to broadcast messages radius)
and to receive messages broadcasted by other  Number of Signal Controlled Intersections
vehicles or the roadside, in addition to possessing (if any)
sensing capabilities.  Connections to other highways
 Grades (maximum ascent, descent)
Infrastructure Supported:
In this concept, the infrastructure plays an active role
to enhance performance. Automated vehicles would Performance Characteristics
travel on dedicated lanes, which use infrastructure  Vehicle hours of delay per day

 Accidents/fatalities per 1,000,000 vehicle The California Partners for Advanced Transit and
miles Highways (PATH) Program has developed the
 Average daily traffic level technology whereby magnets buried at intervals in the
 Average peak hour traffic level roadbed would be sensed by magnetometers in
vehicles, providing a way to monitor their location
Characteristics of Proposed Right-of-Way and velocity. This approach could cost around
 Width of ROW (min/max/variation) $10,000 per mile to install, far less than the millions
 Number/Value of Existing Structures of dollars to build each mile of new highway.
 Property Value of ROW
PLANNING AND PROGRAMMING

This step involves the selection of the project,
identifying the proper site, preparing the field review
form which consists of the physical and performance
characteristics of the site selected, and conducting the
field review.
Based upon the site characteristics data obtained the

viability of a site is determined. For this first we have
to have a basic idea of how to select. The selection is
essentially based on two things, the need for
development in the site and the potential available for
the development. There is more than just planning of Fig (1-3) Drilling done to insert magnets at regular
the design of the highway involved, various other intervals.
parameters need to be developed such as
 Advanced Transport Management Systems Advanced Traveler Information System (ATIS):
(ATMS) Congestion is building on the nation’s streets and
 Advanced Traveler Information Systems highways and is predicted to get worse. ATIS is about
(ATIS) providing citizens, businesses and commercial
 Automated Vehicle Control Systems carriers with the right information at the right time to
(AVCS) improve the quality and convenience of their trip and
the overall performance of the transportation system.
Advanced Transport Management Systems (ATMS): From a technical standpoint, ATIS has three steps:
The need to have a new and improved system of  Collecting data (e.g., using loop and other
managing the traffic has led to the making of detection systems as well as cameras),
specialized software’s which are intelligent and  Consolidating that information, then
merge different facets of technology to bring about  Communicating the “unified” information to
the desired change in traffic governing. This area has various agencies and/or the public.
huge implications when it comes to managing
automated traffic and one advantage would be that It is important to remember that the exact mix of
the system could directly interact with vehicle and technologies for collecting, consolidating and
exchange information. The traffic system has to be in communicating information depends on your region’s
pristine condition to provide a safe and faster journey. unique needs and, ultimately, on the broader trends in
Traffic Detection is the first step in making an the information services industry. With only the
advanced transport management system work. For development of the desired road structure and the
traffic detection we can make use of the following vehicle controls the AHS cannot be implemented. An
techniques: integrated information providing system must be
 Inductive Loops Detection System centrally present.
 Close Circuit Television (CCTV)
 Wide Area Video Detection System Any ATIS will incorporate the following four
 Infrared Laser Detection System fundamental features for providing support.
Other ways to manage traffic efficiently is to properly  Geographical Information System (GIS)
regulate signals, emergency management, having an  Global Positioning System (GPS)
incident command centre to manage traffic flow.  Electronic Data Interchange (EDI)
Basic to the automated-highway schemes are  Mobile Communications (MC)
computer simulations to indicate that such systems
may be successful. Automation could double or even Automated Vehicle Control System (AVCS)
triple highway capacity. To many people the subject of self-guided
"automatic" automobiles has a "science fiction"

flavor, typical of projects that are either far beyond experience lower property values, traffic, noise, and
the state of the art or impractical from a cost/benefit pollution.
standpoint. Actually, recent advances in computers,
sensors and other related technology have made such Environmental issues will obviously be encountered
a system feasible in the relatively near term and throughout the environmental analysis phase, but may
enormous benefits can justify major development and first arise when the project is initially selected. The
deployment costs. An automated system can have most significant impact these issues will have on the
major advantages over the current system in the areas project is in additional costs required for
of highway space utilization and safety. environmental remediation, as well as time delays in
Listed below are the five layers in which an conducting the EIS. Public acceptance will also be a
automated control system will be able to co-ordinate significant concern once the construction phase of the
the movement between vehicles. process begins and creates inconveniences such as
Table 1: Five layers of automated control traffic congestion, noise and dust.
S.No Layer Function Legal and liability concerns should primarily be

1 Network Controls Entering\Exiting addressed in the engineering and design phase of the
Traffic and regulates flow project. Organizational capacity and resource issues
2 Link Computes and broadcasts will predominate throughout the process when it
routes, maneuvers and comes time to obligate funds to conduct studies,
speed for each vehicle in acquire right-of-way, and begin construction.
each section
3 Co-ordination Communicate and co- Lastly competition with other projects will be
ordinate with peers and encountered early in the planning process in trying to
selects one maneuver to be get approval for the project. The urban commuter
executed sites present the greatest deployment challenges, both
4 Regulation Executes the maneuver, in terms of infrastructure cost and garnering support.
maintains speed, and In general, the attractive feature of the urban sites is
performs lane change etc. that it would be relatively simple to attract
5 Physical Decouples lateral and participants from any vehicle classification and there
longitudinal control. are few climatic or terrain challenges. The drawback
is that the AHS would likely require construction of a
grade-separated roadway, which would expose the
project to lots of environmental reviews and huge
cost. Provision of dedicated on and off ramps could
prove to be especially expensive. With average daily
traffic in the range of 20,000 - 35,000 vehicles per
lane per day, it would be impossible to develop AHS
within existing lanes. The task of completing the
project on time would be involving a huge risk factor
and delays would be unacceptable in case of urban
sites.
ENGINEERING, DEVELOPMENT AND

FIG 4: Example of an automated car DESIGN
In conjunction with the environmental assessment or
ENVIRONMENTAL ANALYSIS EIS, this step develops engineering studies for
This step includes the preparation of a preliminary alternative analysis. Once the assessment or EIS is
environmental study, obligating funds for preparing approved, the final design is started and completed
the environmental document, completing the and preparation of construction documents begins.
environmental assessment or Environmental Impact Engineering deals with the specific needs of the
Statement (EIS), approving the assessment or EIS, selected site. The development and simulation of
and approving the location of the project. models is required in case of engineering and these
must be adequately equipped with parameters for the
Environmental Infrastructure Issues deal with the present as well as the likely trends to emerge in the
actual infrastructure changes required for AHS future due to presence of AHS. The development of
deployment. These infrastructure issues will further required services along with the automated vehicle is
intensify concerns about the urban form and sprawl essential. The amount of participation brings about
implications of freeways. Communities and the success of the AHS and participation is attracted
neighborhoods adjacent to freeways typically by giving incentives to the users.

DEVELOPMENT OF AHS each other, reducing drag by half. This arrangement
Road-Side System could result in a 20-percent boost in fuel economy.
The roadside control system’s primary objective is to
optimize the capacity and traffic flow of the overall While this expanded capacity seems like the
AHS. The models used in the link layer involve commuters’ dream, one must be careful of its future
aggregated vehicle densities and traffic flows but not effects. Recall how when the Highway System was
individual vehicles. Thus, vehicle safety cannot be implemented, growth developed around it to utilize
monitored, much less enforced. The roadside control the new infrastructure. This growth ultimately
system can control the network and link layers in continued until the highway system could no longer
ways that tend to increase vehicle safety, such as support it. This growth has to be expected to happen
maintaining sufficiently low aggregated vehicle around the AHS also. This is where the danger comes
densities and decreasing the inlet traffic flow into in. Operation of the AHS must be controlled so as to
links where aggregated traffic density is very large. allow for gradual growth over time instead of having
At the link layer a large number of vehicles are large population centers developing overnight.
controlled in a decentralized but coordinated manner.
Finally there the issue of what operational method the
On-Board System AHS should operate under, user equilibrium or
The vehicle in order to communicate with the road system optimization. With the present highway
side system has to have an information gathering system this was not an issue, since drivers naturally
system inside and apart from that it has to have an conform to user equilibrium while there is no way to
intelligent decision making system within itself to alter their travel patterns to conform to system
process and bring out the necessary effect on the optimization. But with vehicles under automated
vehicle. An important perceptual task during driving control, system optimization now becomes a very real
is the ability to detect and avoid collisions. Failure to option. By implementing a system optimal approach,
accurately perform this task can have serious one would be able to keep more local traffic on the
consequences for the driver and passengers. Many freeway by regulating vehicle flow. And by
recent researches have developed 3D sensing models implementing a system optimal operational approach
(e.g., range or distance) these models are based on the the total travel time experienced through the system
visual information available to the driver. It uses will be lower, thereby increasing the efficiency of the
visual angle and change in visual angle to regulate system simply by not allowing travelers to choose
speed during car following. routes according to user equilibrium. Thus one must
conclude by saying AHS can be seen as a viable
CONCLUSION option in controlling the congestion and various other
While it has been agreed upon that traffic congestion problems. But projects like these can be only
is a major problem in the world, the solutions to this implemented in steps. So people must have the
problem have ranged from the simple to the exotic. patience to wait to reap the benefits of such
On paper, an automated highway system looks like a improvements in taking the process of driving to a
solution to future traffic problems, but technical whole new level.
feasibility has not been proven beyond a few lab and
controlled test track demonstrations. These REFERENCES
technologies have to be pretty much proved S.L. Dhingra, Integrated Transit Systems for
otherwise, there's no way to avoid the legal liabilities sustainable development, IIT Bombay, 2008.
involved. Farokh H. Eskafi, Modelling and simulation of
While this seems like a lot, converting one lane of automated highway systems, California PATH,
existing freeway into an AHS lane amounts to the 1996.
construction of two new lanes to the existing Rillings, James H., Automated Highways. Scientific
infrastructure. American, 1997
Another is the restriction of maximum operational R. Horowitz and P. Varaiya, Control design of
speed by compromising geometrics. The need to Automated Highway syatems, Berkeley, 2000.
provide safe and comfortable conditions while Rodney Lay Vehicle–Highway Automation
negotiating a curve restricts the ability of vehicles to Directions, Challenges, and Contributing Factors
travel at unlimited maximum speeds, and therefore JUAN Zhicai, WU Jianping , Mike Mcdonald, Socio-
places a ceiling on the maximum achievable capacity. Economic Impact Assessment Of Intelligent
While the capacity would be controlled more by Transport Systems, Southampton, 2006.
weaving and merging sections, these geometrics Varaiya, Pravin, and Steven E. Shladover, Sketch of
create the need to transition vehicles between higher IVHS Systems Architecture. California PATH,
speed operations and lower speed operations. A new 1997.
concept which applies exclusively to the Automated
Highway System is that of platooning. With platoon
spacing set at a half a car length, cars can draft off

ISBN 978-81-921121-0-7
A CASE STUDY OF MULA RESERVOIR SEDIMENTATION

ASSESSMENT USING BATHYMETRY ALONG WITH SATELLITE
REMOTE SENSING TECHNOLOGY
D.D.Bhide
Director Genera, Maharashtra Engineering Research Institute, Nashik -422004, Maharashtra,
P.R.Bhamare
Chief Engineer, Maharashtra EngineeringTraining Academy, Nashik -422004, Maharashtra
V.B.Pandhare
S.E. & Jt. Director, Maharashtra Engineering Research Institute, Nashik -422004, Maharashtra
S.N.Kulkarni
Research Officer, Maharashtra Engineering Research Institute, Nashik -422004, Maharashtra
S.K.Kalvit
Asst. Res. Officer,Maharashtra Engineering Research Institute, Nashik -422004, Maharashtra
E mail – rssd_meri@hotmail.com
ABSTRACT: Differential Global Positioning System based bathymetric survey and Satellite Remote Sensing
technique are two widely accepted methods for evaluation of storage capacity and assessment of sedimentation
in reservoirs. The Hi Tech DGPS bathymetric survey is more accurate, reliable and covers almost entire
reservoir portion if the survey is conducted at F.R.L. condition. However it is expensive and takes more time
and manpower as compared to remote sensing method. DGPS bathymetric survey does not cover reservoir
portion of shallow depth of 1.5 - 2 m since the survey vessel requires minimum depth of 1 to 1.5 m for its free
movement and accurate echo sounding. Satellite remote sensing method is fast, less laborious, fairly reliable. It
is cost effective one and hence preferred over DGPS based bathymetry for repetitive surveys. However,
limitation of this method is that its application is generally restricted to live storage zone of the reservoir. Since
both SRS technique and DGPS bathymetric survey have inherent advantages and limitations as mentioned
above, a hybrid technique of using both the methods in conjunction provides better solution to make the survey
cost effective, fast, accurate and complete one. In this paper a case study of Mula reservoir sedimentation survey
using Remote sensing technique and bathymetry is briefly discussed.
INTRODUCTION reliable, a cost effective one and hence preferred over

Reservoirs are important fresh water resources. DGPS based bathymetric survey for monitoring of
Today many of them are under the threat of sedimentation in reservoirs at regular interval of
excessive sedimentation. The excessive soil erosion time. However, limitation of this method is that its
in the catchment area due to various reasons leads to application is generally restricted to live storage zone
heavy sedimentation in the reservoirs. The eroded of the reservoir. Since both SRS technique and
soil settles down in the reservoirs at different DGPS bathymetric survey have inherent advantages
elevations reducing their storage potential. For and limitations as explained above, a hybrid
realistic and effective planning of reservoir water it is technique of using both the methods in conjunction
important to know how much storage is lost due to provides better solution to make the survey full
sedimentation over a period of years so that the net fledged, cost effective, fast, accurate and reliable.
storage available for planning can be ascertained. It Maharashtra Engineering Research Institute
can be known by conducting sedimentation surveys (M.E.R.I.) Nashik, a research organization of Water
of reservoirs. Today Differential Global Positioning Resources Department Govt. of Maharashtra, has
System based bathymetric survey and Satellite conducted number of reservoir sedimentation
Remote Sensing technique are widely accepted assessment surveys so far, by both DGPS based
methods for assessment of sedimentation in bathymetric survey and by Satellite remote sensing
reservoirs and evaluation of storage capacity. Hi method either independently or in conjunction. The
Tech DGPS bathymetric survey is more accurate, survey of Mula irrigation reservoir in Ahmadnagar
reliable and covers almost entire reservoir portion if district of Maharashtra was conducted using both the
the survey is conducted at F.R.L. condition. However techniques in combination.
it is expensive and takes more time and manpower as
compared to remote sensing method. DGPS STUDY AREA
bathymetric survey does not cover reservoir portion Mula reservoir lies between following geographical
of shallow depth of 1.5 - 2 m since the survey co-ordinates.
vessel requires minimum depth of 1 to 1.5 m for its Latitude from 190 15’ 15” to 19 0 23’
free movement and accurate echo sounding. Satellite 29”
remote sensing method is fast, less laborious, fairly

Longitude from 740 24’ 55” to 740 37’ 5. Project chart and range-lines
16”
The reservoir is constructed on river Mula, a
tributary of river Godavari, in Ahmadnagar district of
Maharashtra. It is a major irrigation project. Total
catchment area of the reservoir is 2275.86 sq. km.
The designed gross storage capacity of the reservoir
at FRL is 738.02 Mm3 and live storage capacity
between FRL 552.30 m to MDDL 534.016 m is 609
Mm3. The dead storage capacity below MDDL is 129
Mm3. The submergence area at FRL is 5360 ha. The
construction of this reservoir was completed in the
year 1972. As per the present practice, the useful life
period of major reservoir is considered as 100 years.
The reservoir is nearly 38 years old. Thus nearly 38
Fig: 1 Schematic sketch of DGPS bathymetric survey
% of its useful life span is already over. Hence it was
felt necessary to assess the loss in its storage capacity
The survey vessel with twin mounted 60 HP engines
on account of sedimentation. The sedimentation
was transported to dam site & launched with the help
assessment survey of this reservoir was conducted by
of the crane from a suitable jetty place at the
Maharashtra Engineering Research Institute, Nashik
reservoir. The boat mounted equipment the GPS
in the year 2009-10 using DGPS based bathymetry in
mobile station, GPS antenna, UHF data link with
conjunction with satellite remote sensing.
UHF antenna, batteries of boat engine, and batteries
of equipment were installed. The survey boat was
APPROACH & METHODOLOGY
also fitted with the small navigation echo sounder for
The DGPS Bathymetric survey of this reservoir was
the boat driver. Echo sounder was installed in the
planned to be conducted at FRL condition so as to
boat. The Echo sounder was calibrated with the help
cover maximum reservoir portion under bathymetric
of calibrator. The transducer was fitted to the
survey. But due to some reasons the survey started
transducer well provided in the boat properly and
late and by the time survey started the water level in
connected to the echo-sounder. A Pentium laptop PC
reservoir had depleted almost 7 meters below FRL.
installed with Navitronic software was used for the
So bathymetric survey alone would not have covered
hydrographic survey work. Mobile GPS was
entire reservoir portion. It was therefore planned to
interfaced with the echo sounder and the echo
conduct satellite remote sensing survey for the
sounder was interfaced with the Laptop. The laptop
reservoir portion not covered under bathymetric
used, had inbuilt VGA port which is interfaced with
survey. Thus Mula reservoir sedimentation survey
the additional LCD monitor (Helmsman display) for
comprised of two parts viz. (I) Bathymetry and (II)
the boat driver to view the range lines to be surveyed.
remote sensing.
The range lines with spacing 100m were generated to
conduct the survey at this interval. The boat driver
Part (I) DGPS Based Integrated Bathymetric
was guided to position the boat on the selected track
Survey
with the help of helmsman’s display. After
In DGPS based hydrographic survey two or more
positioning the boat on the line, the data logging was
satellite based GPS receivers are used in differential
started. The time base (2 sec) for data logging for
mode for positioning and echo sounder is used for
depth & positioning was done. The boat speed of 2.5
depth measurement. GPS reference receiver is
to 3.5 knots was maintained. The software
located at a known latitude and longitude position
automatically recorded the position and depth data at
and GPS mobile receiver is located on the survey
the time interval set. During the reservoir survey,
boat. Both the GPS operate in differential mode
daily water level was collected at the start of the
through a communication link. The location accuracy
survey. The water level was required for
increases with the adoption of differential global
transforming the depth data into reservoir level. At
positioning system involving the reference station
the reservoir upstream end area, the survey was
and the mobile station
conducted at range line interval of 200 m, and 300 m.
Mula reservoir survey preparation was done in
The survey area of about 25 sq. km inclusive of 270
following stages.
range lines was completed within 10 days period.
1. Installation of DGPS reference station.
The shallow regions of the reservoir approximately
2. Launching of boat and installation of mobile
less than 1.5 m depth could not be surveyed, as it was
station.
not possible to maneuver the boat in these regions.
3. Installation of Echo sounder.
4. Installation of data acquisition PC and helmsman
display.

Part (II) Remote Sensing in Reservoir analysis NDVI method and classification technique
Sedimentation Study were adopted to measure the water spread area in
Reservoir sedimentation surveys are essentially each image
based on mapping of water-spread areas at the time NDVI is one index, which enhances vegetation and
of satellite over pass. It uses the fact that water- water. NDVI is generated using the formula given
spread area of the reservoir reduces with the below.
sedimentation at different levels. The water-spread NDVI = (NIR –R)/ (NIR+R)
area and the elevation information are used to Where ‘NIR’ is digital number in near infrared band
calculate the volume of water stored between and ‘R’ is digital number in red band. The ratioed
different levels. These capacity values are then image is then density sliced. Water pixels generally
compared with the previously calculated capacity occupy lower range of histogram in ratioed image.
values to find out change in capacity due to Since Classification technique was found more
sedimentation at different levels. appropriate in most of the cases its results were
In this case, the Remote Sensing survey covered a finally retained
reservoir portion between R.L. 541 m and FRL
552.30 m. Water level information on the dates of Table showing IRS P6 LISS III images of different
satellite passes was collected from the field office. water levels used in this study and respective
The Indian Remote Sensing Satellite IRS P6 estimated water spread area is given below
(RESOURCESAT) LISS III images of different
water levels from R.L. 552.30 m to R.L. 540.40 were Water spread area at regular interval
analysed in this study. A graph of elevation Vs area (see figure - 3) was
plotted and a best fit polynomial equation of second
order was derived for the graph as below
y = 0.039x2 + 1.72x + 26.35
with Coefficient of co-relation, r =
0.9996
where x = Elevation difference in meters (measured
above R.L. 540.40 m)
y = Water spread area in Mm2
Using this equation, the Water Spread Areas at
regular interval of elevation have been worked out
Calculation of Reservoir Capacity

Computation of reservoir capacity at different
Fig. 2 Mula reservoir satellite image
elevations has been done using following prismoidal
formula
Steps involved in digital image analysis
V = H/3 x (A1 + A2 + SQRT (A1 x A2))
The basic concept is to find out the water-spread area
Where V- Reservoir capacity between two
from satellite images of different water levels. The
successive elevation H1 and H2
difference between aerial spread of water between
H- Elevation difference (H2 – H1) between two
current year and earlier years is the aerial extent of
successive contours
silting at these levels. The methodology for
A1 and A2 are water spread areas at elevation H1 and
estimation of live capacity of reservoir using remote
H2
sensing consists of following major heads:
 Digital data base creation The capacity at different levels worked out by DGPS
 Estimation of water-spread area bathymetric survey and by remote sensing survey
 Calculation of reservoir capacity were clubbed together to estimate total revised
 Comparison with previous surveys capacity from reservoir bed level to FRL. A table
 Estimation of live capacity loss due to giving revised area and storage capacity at different
sedimentation elevations at an interval of 1 meter is given below. In
this table area and capacity from R.L. 511.00m to
Image processing and analysis R.L. 541.00m are as per Bathymetric survey and area
It comprised of following activities. and capacity above R.L. 541.00m up to FRL 552.30
a) Image Geo-referencing m are as per Remote sensing survey.
b) Study area extraction from the satellite images
c) Water spread area extraction Sedimentation assessment
After pre-processing of images i.e. geo-referencing The quantum of sedimentation is estimated by
and extracting area of interest from all the images, deducting present capacity from original capacity.
the geo-referenced subsets were digitally analysed Following table gives various parameters derived
using standard image analysis software. In the from the study.

Table -1 Table -2
Date of pass Elevation in m. Area in Mm2 Elevation in Revised Area Revised
meters in Mm2 Capacity
9-5-2007 540.40 25.92
In Mm3
15-4-07 543.12 32.33 511.00 0.0028 0.00
22-3-2007 544.95 34.63 515.00 0.54 0.771
519.00 2.00 5.442
7-2-2006 546.29 37.99
523.00 4.60 18.579
26-2-2007 546.86 38.73 527.00 7.41 42.410
2-2-2007 547.89 41.89 532.00 12.14 90.856
533.00 13.13 103.488
4-1-2008 548.75 43.14
534.00 14.16 117.129
26-12-2005 549.14 43.87 535.00 15.36 131.881
9-1-2007 550.00 46.41 536.00 16.69 147.901
537.00 18.26 165.370
11-12-2007 550.13 46.59
538.00 20.26 184.622
17-11-2007 551.95 51.9 539.00 22.35 205.918
24-10-2007 552.30 52.31 540.00 24.53 229.35
541.00 27.40 255.299
542.00 29.20 283.598
543.00 31.09 313.737
Elevation Vs Satellite measured water spread area curve between R.L.540.40 m to 544.00 33.05 345.798
R.L.552.30 m
60 545.00 35.09 379.861
546.00 37.21 416.002
50
547.00 39.40 454.299
Water s p rea d A rea in Mm ---->
548.00 41.67 494.832

2
40
549.00 44.03 537.677

30
550.00 46.46 582.913
20
551.00 48.96 630.617
552.00 51.55 680.869
10
552.30 52.34 696.448
0
0 2 4 6 8 10 12 14 Table - 3
Elevation difference in meters measured above R.L. 540.40 m ---->
Details Original Survey
Survey conducte
Fig. 3Elevation Vs Satellite estimated water spread year d in year
area (1972) (2008)
Gross Capacity
between R.L. 511.00 738.02 696.448
m to R.L. 552.30 m
Revised Elevation - Capacity curve for Mula reservoir and its comparison with original
555
capacity in Mm3
550
Silt deposited
545 between original 41.572
and latest surveys in
E levation in m eters ---->
540 C apacity by DG PS S urvey
535 C apacity by R emote s ens ing

Original C apacity
Mm3
530
525
Period in years -- 36
520
Rate of silt 1.154
515
510
deposited between
505
two surveys in
0 100 200 300 400 500
Gross storage capacity in Mm3 -->

600 700 800
Mm3/year
% to original 5.633
Fig: 4 Elevation Vs Original and Revised Storage capacity 0.156
Capacity % Annual loss
Rate of siltation in 5.074
Ha-
m/100sq.km/year

RESULTS & DISCUSSIONS
The sedimentation study of Mula reservoir by Hybrid
Technique of combined DGPS bathymetric survey
and remote sensing survey covered entire reservoir
portion and hence siltation in reservoir at different
elevations could be estimated. The present study
revealed that the Gross storage capacity of reservoir
is reduced by a margin of 41.572 Mm3 which is about
5.633% of the total capacity. This implies that
sedimentation in the reservoir in a span of 36 years is
41.572 Mm3. Percentage annual loss in capacity is of
the order of 0.156 only which is not significant. The
live storage capacity of the reservoir is reduced by
30.25 Mm3 which is now 579.09 Mm3 against
original live storage capacity of 609.34 Mm3 The
rate of siltation, which is 5.074 ha-m/100sq.km./year,
is also within the norms generally adopted by CWC
(6.54 - 7.15 ham/100 sq.km /year) for major
reservoirs in Maharashtra
CONCLUSIONS
The Hybrid technique of DGPS bathymetric survey
in combination with remote sensing survey was
found to be very useful for assessing total
sedimentation in reservoirs. It is faster, economical,
less laborious and reliable also. It is recommended
that such sedimentation studies may be repeated at 5-
10 years interval for assessing the true contents of the
reservoir and keeping the content table of reservoir
updated.
REFERENCE
Technical report of M.E.R.I., Nashik on Capacity
evaluation and sedimentation assessment of Mula
reservoir by hybrid technique -Year 2010.

ISBN 978-81-921121-0-7
REVIEW ON USE OF RECYCLED AGGREGATES IN CONCRETE
G.D. Ransinchung R.N

Asstt. Professor, Department of Civil Engineering, IIT-Roorkee, 247667, gdranfce@iitr.ernet.in
Kumar Praveen
Professor, Department of Civil Engineering, IIT-Roorkee, 247667,pkaerfce@iitr.ernet.in
Chauhan Prakash Arun
M.Tech. Student, Department of Civil Engineering, IIT-Roorkee, 247667, arunppce@iitr.ernet.in
ABSTRACT: Use of recycled aggregate in concrete can be useful for environmental protection and economical
terms. Recycled aggregates are the materials for the future. The application of recycled aggregates has been
started in many construction projects in the countries like European, American and Asian. As on today, recycled
aggregates have not been experimented in India on a wide scale due to lack of awareness and limitations exist
on the part of formulating the guidelines. Therefore, more research work has to be carried out especially in de-
veloping country like in India. There is significant potential for growth of recycled and secondary aggregates as
an appropriate and “green” solution to the anticipated increased world – wide construction activity. This paper
herein reviews the influence of recycled aggregates on mechanical and durability properties of concrete.
Keywords: Recycled aggregates; Recycled aggregate concrete; Waste materials; Construction & demolition; Durability
INTRODUCTION
Any construction activity requires several materials 5. Creation of building waste resulting from manmade
such as concrete, steel, brick, stone, glass, clay, mud, disaster/waste.
wood, and so on. However, the cement concrete re-
mains the main construction material used in con- PLASTICS WOOD 5%
struction industries. For its suitability and adaptability OTHERS 2% 1%
with respect to the changing environment, the con- METAL 2%
crete must be such that it can conserve resources, pro-
tect the environment, economize and lead to proper
utilization of energy. To achieve this, major emphasis BRICKS
must be laid on the use of wastes and by-products in AND TILES
cement and concrete used for new constructions. The 25% CONCRETE
utilization of recycled aggregate is particularly very 65%
promising as 75 per cent of concrete is made of ag-
gregates etc. The enormous quantities of demolished
concrete are available at various construction sites,
which are now posing a serious problem of disposal in
urban areas. This can easily be recycled as aggregate
and used in concrete. Research & Development activi- Fig.1 Distribution of Waste Materials
ties have been taken up all over the world for proving The recycling and reuse of construction & demolition
its feasibility, economic viability and cost effective- wastes seems feasible solution in rehabilitation and
ness. new constructions after the natural disaster or demoli-
Figure 1 shows the percentage of recycling material tion of old structures. This becomes very important
available in this world as per Material classes in Cali- especially for those countries where national and local
fornia’s overall disposed waste stream, 2003. (Casca- policies are stringent for disposal of construction and
dia Consulting Group, Inc., 2004). demolition wastes with guidance, penalties, levies etc.
The main reasons for increase of volume of demoli-
tion concrete / masonry waste are as follows:- INDIA STATUS
1. Many old buildings, concrete pavements, bridges There is severe shortage of infrastructural facilities
and other structures have overcome their age and like houses, hospitals, roads etc. in India and large
limit of use due to structural deterioration beyond quantities of construction materials for creating these
repairs and need to be demolished. facilities are needed. The planning Commission allo-
2. The structures, even adequate to use are under cated approximately 50% of capital outlay for infra-
demolition because they are not serving the structure development in successive 10th & 11th five
needs in present scenario. year plans. Rapid infrastructural development such
3. New construction for better economic growth. highways, airports etc. and growing demand for hous-
4. Structures are turned into debris resulting from ing has led to scarcity & rise in cost of construction
natural disasters like earthquake, cyclone and floods materials. Most of waste materials produced by de-
etc.
molished structures disposed off by dumping them as

land fill. Dumping of wastes on land is causing short- recycled material. Cost of disposal of waste from con-
age of dumping place in urban areas. Therefore, it is struction industry to landfill has a direct bearing on
necessary to start recycling and re-use of demolition recycling operations. Low dumping costs in develop-
concrete waste to save environment, cost and energy. ing countries also acts as a barrier to recycling activi-
Central Pollution Control Board has estimated current ties. Imposition of charge on sanitary landfill can in-
quantum of solid waste generation in India to the tune duce builders and owners to divert the waste for recy-
of 48 million tons per annum out of which, waste cling. Some of these issues act as barriers in promot-
from construction industry only accounts for more ing more widespread use of recycled aggregate and
than 25%. Management of such high quantum of concrete made with recycled aggregate.
waste puts enormous pressure on solid waste man-
agement system. In view of significant role of recy- Lack of Appropriately Located Recycling Facilities
cled construction material and technology in the de- Construction and demolition waste is generated in
velopment of urban infrastructure, based on the sur- small quantities at locations, which could be widely
vey conducted by few organisations on’ Utilization of separated. Therefore, portable equipment is needed,
Waste from Construction Industry’ targeting housing which can be used and set up close to a demolition
/building and road segment. The total quantum of site. Transporting waste over large distances makes
waste from construction industry is estimated to be 12 the proposition of using C&D waste uneconomical.
to 14.7 million tons per annum out of which 7-8 mil- Lack of such plants is a major barrier for ‘Newco-
lion tons are concrete and brick waste (Singh, S.K. et mers’ in the field of C&D (construction and demoli-
al.,1997). According to findings of survey, 70% of the tion) waste management. Commissioning of appro-
respondent have given the reason for not adopting priately located recycling crusher units in a pilot plant
recycling of waste from Construction Industry is “Not can help in lowering barriers against recycling of con-
aware of the recycling techniques” while remaining struction & demolition waste (Akash Rao et al., 2006)
30% have indicated that they are not even aware of
recycling possibilities. Further, the user agencies/ Absence of Appropriate Technology
industries pointed out that presently, the BIS (Bureau There are very few commercially viable technologies
of Indian Standards) and other codal provisions do not for recycling construction and demolition wastes, and
provide the specifications for use of recycled product methods that can be used to crush C&D waste on a
in the construction activities (use of recycle aggregate commercial scale are urgently required. In fact, when
in concrete –a paradigm shift) the technology is established, other issues such as
In view of above, there is urgent need to take up the quality control of raw material and finished product,
following measures:- etc. can be taken up.
1. Sensitization/ dissemination/ capacity building to-
wards utilization of construction & demolition waste. Lack of Awareness
2. Preparation and implementation of techno-legal Lack of awareness towards recycling possibilities and
regime including legislations, guidance, penalties etc. environmental implications of using only fresh-mined
for disposal of building & construction waste. aggregates are the main barriers due to which C&D
3. Delineation of dumping areas for pre-selection, waste is disposed only in landfills. Creating awareness
treatment, transport of RCA (Recycle Concrete Ag- and dissemination of information relating to the above
gregate). barriers and the properties of concrete made with re-
cycled aggregate are essential to mobilize public opi-
4. National level support on research studies on RCA.
nion and instill confidence in favor of the recycling
5. Preparation of techno-financial regime, financial
option. There is a need to create a market for recycled
support for introducing RCA in construction including
products by involving the construction industry and
assistance in transportation, establishing recycling
encouraging them to use recycled materials in
plant etc.
projects.
6. Preparation of data base on utilization of RCA.
7. Formulation of guidelines, specifications and codal Lack of Government Support
provisions. A lack of government support and commitment to-
8. Preparation of list of experts available in this field wards development of recycling industry is often
who can provide know how and technology on totality seen. Developing appropriate policy supported by
basis. proper regulatory framework can provide necessary
9. Incentives on using recycled aggregate concrete impetus. It will also help in data compilation, docu-
subsidy or tax exemption. mentation and control over disposal of waste material.
BARRIERS IN PROMOTING USE OF RA AND Lack of Proper Standard

RAC Apart from the specifications of RILEM (Réunion
Acceptability of recycled material is hampered due to Internationale des Laboratoires d’Essais et de Re-
a poor image associated with recycling activity, and cherches sur les Matériaux et les Constructions), JIS
lack of confidence in a finished product made from (Japan Industrial Standard) and those used in Hong

Kong, only very limited codal specifications/standards sourced from the two public filling areas. The mixes
regarding use of recycled aggregates are available. In were varied by replacement of natural coarse and fine
fact, use of concrete with 100% recycled coarse ag- aggregate by recycled aggregates up to 100% by
gregate for lower grade applications is allowed in weight, with or without the incorporation of fly ash.
Hong Kong, though for higher grade applications The test results showed that the replacement of coarse
(above M35 concrete), only 20% replacement is al- and fine natural aggregates by recycled aggregates at
lowed, and the concrete can be used for general appli- the levels of 25 and 50% had little effect on the com-
cations, except in water retaining structures. In Japan, pressive strength of bricks and blocks, but higher le-
JIS has drafted a Technical Report; TRA 0006 “Re- vels of replacement reduced the compressive strength.
cycled Concrete-Using Recycled Aggregate” to pro- Olorunsogo & Padayachee (2002) studied durability
mote the use of concrete made with recycled aggre- indexes such as the chloride conductivity, oxygen
gate Development of relevant standards for recycled permeability and water absorptivity; of three different
materials would provide producers with targets and concrete mixes containing 0, 50 and 100%, RA moni-
users an assurance of quality of material. Standards tored at ages 3, 7, 28 and 365 days. They reported that
formulated in the above-mentioned countries can be a overall, durability quality of RA concrete reduced
guideline for development of specifications with increases in the quantities of recycled aggregates
An effort has been made in this paper to present a that were included in a mix and the quality improved
summary of the use of various recycled aggregates in with the age of curing. They explained this phenome-
concrete and brief the salient properties of recycled non by the fact that cracks and fissures created in re-
aggregates related to mechanical and durability. cycled aggregates during processing render the aggre-
gate susceptible to ease of permeation, diffusion and
ADVANTAGE OF USING RECYCLED CON- absorption of fluids.
CRETE PRODUCTS
1. It’s an accepted source of aggregate into new con- TECHNIQUES TO IMPROVE THE PROPER-
crete by ASTM (American Society for Testing and TIES OF RECYCLED AGGREGATES BY PRE-
Materials) and AASHTO (American Association of SOAKING TREATMENTS
State Highway and Transportation Officials). Vivian W.Y. Tama et al. (2006), have conducted an
2. It’s currently being used in concrete and asphalt experiment using three acidic solvents namely hy-
products with better performance over comparable drochloric acid (HCL), sulphuric acid (H2SO4) and
virgin aggregates. phosphoric acid(H3PO4) with concentration of 0.1
3. It provides superior compaction and constructabil- mole [if concentrated acid is used, dilute it with distill
ity. water with concentration of about 0.1 mole].
4. It’s higher yield recycled aggregates are lighter After the pre-soaking treatments, concrete specimens
weight per unit of volume, which means less weight made from the treated RA in the forms of 100mm
per cubic yard, resulting in reduced material costs, sized cube, 100mm×500mm beam, and 100mm di-
haul costs, and overall project costs. ameter cylinder according to British Standard and
5. It weighs 10-15% less than comparable virgin quar- ASTM Standard (ASTM C293-00, 2001; BS 1881:
ry products (concrete). Part 116, 1983, BS 1881: Part 121, 1983) are prepared
6. It offers a way to reduce landfill waste streams. to assess their compressive strength, flexural strength
and static modulus of elasticity. Table 1 shows the test
WORK DONE ABROAD ON RECYCLE AGRE- result for compressive strength, flexural strength and
GATE static modulus of elasticity for various proportions
Ajdukiewicz & Kliszczewicz (2002) reported that recycle aggregate in concrete.
properties of the original concrete influence the me-
chanical properties of recycled aggregate concrete and The mechanical properties of recycled aggregate con-
it is possible to obtain recycled concrete with higher crete have recorded marked improvements after the
compressive strength than the original one. The mix pre-treatments. The above table shows that 21.84%
design of recycled concrete is similar to the proce- improvement with 20% RA (Recycle Aggregate)
dures for concrete with natural aggregate. Corrections substitution is recorded for compressive strength at 7
to water content are necessary to obtain proper wor- days curing, 22.90% improvement with 25% RA
kability, but the changes in the water/cement (w/c) substitution recorded for flexural strength at 14 days
ratio may be relatively small. curing, and 20.48% improvement with 30% RA subs-
Pooneal (2002) in their study, used recycled aggre- titution for modulus of elasticity when using the HCl
gates from construction and demolition (C&D) wastes approach.

Table 1: Pre-Soaking Treatments Method (Days of Curing)
Mortar HCl MortarH2SO4 MortarH3PO4
Properties of recycled
aggregate concrete 7 14 28 7 14 28 7 14 28
Compressive strength (MPa)

5% 4.11 1.49 3.25 17.11 6.52 6.71 5.88 3.6 3.21
10% 0.08 0.66 0.66 2.43 8 3.39 8.45 4.86 3.15
15% 4.81 3.89 2.36 5.94 3.67 2.97 4.05 1.16 7.84
20% 21.84 5.2 10.08 1.99 9.4 10.9 2.44 0.4 0.07
25% 0.18 1.94 6.67 1.49 2.03 12.79 0.47 2.83 14.01
30% 16.51 3.11 5.17 13.19 2.36 1.33 8.6 2.89 1.26
Flexural strength (MPa)

5% 12.35 7.92 4.35 11.76 6.2 2.68 18.82 12.74 3.51
10% 16.03 8.37 2 11.81 7.77 5.1 10.55 17.13 18.58
15% 9.41 0.18 1.15 12.69 3.33 1.48 12.47 8.5 1.64
20% 5.42 10.84 1.24 2.01 4.94 1.77 3.41 7.79 4.61
25% 5.07 22.9 4.44 8.46 13.66 0.82 2.11 17.44 1.64
30% 3.07 7.71 2.61 6.56 15.21 0.98 3.48 10.75 0.33
Static modulus of elasticity (N mm2)−1
5% 5.68 4.85 7.7
10% 3.05 15.37 7.41
15% 4.19 2.42 1.28
20% 7.33 7.96 9.09
25% 2 0.06 5.85
30% 20.48 9.78 10.82
Fig. 3 Interfacial zone [after pre-soaking in H2SO4

solution]
Fig. 2 Interfacial zone (after pre-soaking in HCL solu-
tion)
Interfacial Zone between Cement Paste and Re-

cycled Aggregate after Presoaking Treatment
The pre-soaking treatment approach should be consi-
dered as cost effective as it can directly remove weak
links of recycled aggregate and the old cement mortar
attached with at a relatively low cost. The improve-
ment of interfacial zone between cement paste and
recycle aggregate after adopting this method has been
shown in the Figures 2, 3 & 4. Although it needs in-
vestment to implement these pre-soaking approaches,
the overall quality for RA is greatly improved and be Fig. 4 Interfacial zone (after pre-soaking in H3PO4
as competitive as the normal aggregate. solution)

REFERENCES
Alexander MG, 1996. The effects of ageing on the in
ter facial zone in concrete. Interfacial transition
zone in concrete: state-of-the-art report 1996, p.
150–74.
ASTM C293-00.Standard test method for flexural
strength of concrete (using simple beam with cen-
ter-
point loading). American Society for Testing and Mate-
rials, United States, 2001.
Barra, M, Vazquez, E. Properties of concretes with re-
cycled aggregates: influence of properties of the ag-
gregates and their interpretation, Sustainable con-
struction: use of recycled concrete aggregate: pro-
Fig. 5 Poorer new interfacial zone for normal ap- ceedings of the international symposium organized
proach (without pre-soaking treatment) by the concrete technology unit, University of Dun-
dee and held at the D epartment of Trade and Indus-
The only limitation is that since the major composi- try Conference Centre, Lo ndon, UK on 11–12 No-
tions of cement mortar includes calcium oxide (CaO), vember 1998, London: Thomas Telford; 1998. p.
silicon dioxide (SiO2), aluminium oxide (Al2O3) and 19–30.
Building Contractors Society of Japan. Study on re-
iron oxide (Fe2O3). The reaction under the acidic envi-
cycled aggregate and recycled aggregate concrete,
ronment (HCl,H2SO4andH3PO4) produces
BuildingConstruction Society of Japan, Committee
Ca3(PO4)2·3H2O and CaHPO4,2Fe(PO4)·3H2O and on Disposal and Reuse of Concrete Construction-
FeHPO4generated from CaO and Fe2O3 with H3PO4, Waste, Summary in Concrete Journal,
are unable to clean up the most cement mortar at- 1978;16(7):18–31.
tached and limit the benefits gained from the pre- Mindess S, Young F, Darwin D. Concrete 2003:1–664.
treatment approach with an optimal strength im- Poon CS, Kou SC, Lam L. Use of recycled aggregates in
provement of 14.01% (with 25% RA substitutions at molded concrete bricks and blocks. Construction
28 days curing), 18.82% (with 5% RA substitutions at Building Mater 2002; 16:281–9.
7 days curing) and 10.82% (with 30% RA substitu- Rao A., Experimental Investigation on Use of Recycled
tions) for compressive strength, flexural strength and Aggregates in Mortar and Concrete. Thesis submit-
the static modulus of elasticity, respectively. ted to the Department of Civil Engineering, Indian
Institute of Technology Kanpur; 2005.
CONCLUSIONS RILEM 121-DRG-Specifications for concrete with re-
Based on the literature review, the following conclu- cycled aggregates, Mater Struct. 1994; 27:557–559.
sions have been drawn: Sharma, P.C., Singh, S.K. and Nagraj, N. (1998), “Fu
 Use of recycled aggregates from demolished con- ture of Recycled Aggregate Concrete in India,” Na-
crete wastes has been considered as an effective tional Seminar on New Materials and Technology in
Building Industry, July 24-25, Vigyan Bhawan,New
method for saving the limited landfill spaces and
Delhi, pp. IV-197-IV- 205.
natural resources.
Singh, S. K. and P. C. Sharma (1998)”Recycling and Re
 Laboratory test results showed that the replacement use of Building Waste in Constructions- A Review,”
of coarse and fine natural aggregates by recycled All India Seminar on Concrete for Infrastructural
aggregates at the levels of 25 and 50% had little ef- Development, Roorkee, pp 317-329.
fect on the compressive strength of bricks and Singh, S.K., Sharma, P.C., and Nagraj, N. (1997),
blocks, but higher levels of replacement reduced the “State- of-Art Report on Recycled Aggregate
compressive strength. Concrete,” SERC Report, Ghaziabad.
 Its higher yield recycled aggregates are lighter Tam CM, Deng ZM, Zeng SX. Evaluation of construc-
weight per unit of volume, which means less weight tion methods and performance for high rise public
per cubic yard, resulting in reduced material costs, housing construction in Hong Kong. Build Environ
haul costs, and overall project costs. 2002; 37:983 91.
 Pre-soaking treatment on recycled aggregates in Vivian W.Y. Tama, C.M. Tam and K.N. Le Received 21
hydrochloric acid (HCL), sulphuric acid (H2SO4) November 2005; received in revised form 26 May
and phosphoric acid (H3PO4) solutions have record- 2006; accepted 31 May 2006
ed marked improvements in mechanical & durabili- Web Site:www.acpa-southwest.org/environ.htm
Use of recycled concrete and advantage of recycled
ty properties of concrete.
concrete aggregate materials
 Based on the scanning electron microscopy Website:www.nbmcw.com/.../concrete/.../576-use-of-
study, strong interfacial zone exhibits between recycled-aggregates-in S. K. Singh, Scientist, Struc-
presoaked aggregate and hydrated cement paste. tural Engineering Division, Central Building Re-
search Institute, Roorkee and P. C. Sharma.

ISBN 978-81-921121-0-7
EFFECT OF LOADING FREQUENCY ON PERFORMANCE OF RC

BEAM-COLUMN CONNECTIONS
C. Marthong
Post-graduate Student, IIT Guwahati, Pin-781039, Assam, India, Email: commarthong@gmail.com
A. Dutta
Professor, Civil Engineering Department, IIT Guwahati, Pin-781039, Assam, India, Email: adutta@iitg.ernet.in
S.K. Deb
Professor, Civil Engineering Department, IIT Guwahati, Pin-781039, Assam, India, Email:skdeb@iitg.ernet.in
ABSTRACT: This paper reports an experimental programme aiming to explore the performance of deficient
RC beam-column connections subjected to excitations of different frequencies. Scaled exterior beam-column
connections with typical deficiencies namely: (1) beam weak in flexure (2) beam weak in shear and (3) column
weak in shear were considered. Altogether, six tests were conducted where specimen corresponding to each
deficiency type were subjected to a cyclic displacement histories with two excitation frequencies of 0.025 Hz
and 1.0 Hz. Important parameters related to seismic capacity such as ultimate strength, stiffness degradation,
energy dissipation, ductility of the connections were evaluated. The performances of the connections under two
different loading frequencies were compared in term of the above parameters. The analysis of results revealed
that the above parameters considered for the study exhibited slightly higher values with higher loading
frequency, which is attributed due to strain rate effect. Further, the damage patterns and failure of the
connections showed greater damage during higher loading frequency compared to those for lower loading
frequency.
Keywords: Beam-column connections, deficiency type, excitation frequency, seismic parameters, inertial
effect.
INTRODUCTION frequencies are substantially lower than those

Connections are defined as a common point of corresponding to the frequencies of actual seismic
intersection of columns and beams and provide excitation. Very few studies on dynamic test of RC
resistance to the applied external loads due to the beam-column connections may be seen in the
bending moment encountered at the joints. Therefore, literature [Agbabian et al.,(1994) and Dhakal et al.,
connections play an important part in structures. In (2005)]. Aiming to narrow this gap, the study targeted
the past, numerous reinforced concrete frame to explore the performances of RC exterior beam-
structures collapsed due to severe earthquake. Post- column connections subjected to excitations of higher
earthquake investigations into damaged structures frequency. This study become more important in
generally showed that in many cases, damages of RC moderate and low seismicity region where some
frame structure were localized in beam-column structures may not have been designed for lateral
connections (Ghobarah et al.,1997) which might have loads and high frequency excitations such as severe
led to partial or total collapse of the building. Further, earthquake, explosion and construction induced
it was observed that the exterior connection which is vibrations may pose more threat than the low-
confined by only two or three framing beams had frequency seismic actions. The scope of this
suffered more in comparison to the interior ones. experimental study is limited to testing of six two-
Several experimental studies have been carried out to third scale exterior connections with three typical
evaluate the behavior of connections under cyclic deficiencies. Out of six specimens, each three
loading (Murty et al. 2003). However, no record specimens were subjected with two loading
could be found where the comparative studies frequencies (0.025 Hz and 1.0 Hz) respectively.
covering various deficiencies like beam-column
connections with beam weak in flexure, beam weak in EXPERIMENTAL PROGRAM
shear and column weak in shear were addressed Selection of specimens
altogether. Thus, a holistic approach was initiated to Figure 1a shows the deflected nature of a frame under
cover different deficient cases of the connections so as the influence of lateral loading. The points of contra-
to gather a comprehensive knowledge about the flexure lie at the mid-points of beams and columns for
behaviour of these connections. such frame. A free body diagram of an isolated beam-
Most of the RC beam-column connection tests column assembly in its deformed position is shown in
conducted in the past are of quasi-static nature Figure 1b. A typical full scale residential building
[Hakuto et al., (2000), Shiohara (2001)]. Obviously, it with floor to floor height as 3.3 meters and the beam
is advantageous because the quasi-static cyclic testing of 3.0 meters effective span was considered. The
allows a careful monitoring of the specimen connection was scaled down to two-third size for
behaviour during the test and the strain-rate effects do experimental investigation. This study was
not affect the material behaviour. However, these

concentrated on a sub-structure consisting of an
external beam-column connection.
(b)
(a) (b)
Fig.1 Frame under lateral loading for (a) Deflected
shape and (b) Isolated exterior connection
Description of specimens
In the present study, two-third size exterior RC beam-
column connections with three typical deficiencies
namely: (1) beam weak in flexure (BWF) (2) beam
weak in shear (BWS) and (3) column weak in shear
(CWS) were considered.
The detailing of all specimens considered in this study
are shown in Figure 2. The beam for BWF specimen
was designed as under reinforced section following (c)
the provisions of standard codal provisions and
guidelines [IS 456 (2000), IS 1893 (I)(2002), Jain et Fig. 2 Reinforcement detailing for (a) BWF (b) BWS
al. (2006)]. Strong column-weak beam principle was and (c) CWS specimens
adopted for the design of the connection. The cross
section of the column was 200 mm x 200 mm and that Test set-up and loading procedures
of the beam was 200 mm x 240 mm. BWS specimen Test set-up as shown in Figure 3 was used for the
was exactly similar in all respects to that of BWF testing, where column was placed in horizontal
except the shear reinforcement provided in beam were position and beam in vertical position. Both ends of
reduced to make the beam weak in shear. Target cube the column were placed on fabricated roller supports
strength of 30 N/mm2 was used for both BWF and in order to simulate the appropriate boundary
BWS specimens. For CWS specimen, the cross condition. Constant axial load of 10% of gross
section of the column was reduced than earlier cases capacity of column was applied to the column end
while the cross section of beam was increased to abutting against an ‘A’ frame. The lateral load was
make the beam-column as a strong beam-weak applied on the beam tip by a displacement-control
column. The main reinforcements in column were loading system through a 250 kN capacity dynamic
maintained similar to those of earlier cases, while actuator having a maximum displacement range of 
same was increased in beam. In order to ensure the 125mm. The displacement amplitudes were increased
shear weakness of the specimens, spacing of lateral with one loading cycle at every amplitude of
ties in the column was increased. Concrete of displacement history. A displacement history is
comparatively weaker grade than earlier cases (target shown in Figure 4.
cube strength of 25 N/mm2) was used.
Fig. 3 Schematic diagram of test set up

(a)

75 frequency test. However for specimens under cyclic
displacement of higher frequency, the damaged were
Displacement (mm)
50 localized at the weakest shear zone. Severe crushing
25 of concrete and necking of main reinforcement were
0
observed. The joint region was practically
undamaged. This is because the damage was mostly
-25 concentrated on the beam part without transferring to
-50 the relatively stronger joint region.
-75
CWS specimen under both loading types exhibited
0 100 200 300 400 500 600 700 800 similar extent of damage (Fig. 7). However, because
Time (Sec) of efficient load-transfer occurring at cyclic
(a) displacement of lower frequency, the cracks on
specimens are well propagated toward the weakest
75 shear zone in the column region.
Displacement (mm)
50
25
0
-25
-50
-75
0 2 4 6 8 10 12 14 16 18 20
Time (Sec) (a) (b)
(b)
Fig. 5 Failure modes of BWF specimen under (a)
Fig. 4 Loading history for (a) Lower frequency Lower and (b) Higher loading frequency
(b) Higher frequency
TEST RESULTS AND COMPARISONS
General behaviour
Crack initiation and propagation were monitored as
closely as possible in all the specimens.
Cracks for BWF specimens were widely distributed at
the joint region for specimens under cyclic
displacement of lower loading frequency. In contrast,
(a) (b)
specimens under cyclic displacement of higher
frequency, the major damages were localized at the Fig. 6 Failure modes of BWS specimen under (a)
beam-column joint interface. This indicates that the Lower and (b) Higher loading frequency
loading frequency is related to the transfer of forces
between the reinforcing bars and the concrete.
Further, hairlines cracking were developed during the
initial stages of loading both at the joint region and at
the joint interface of specimens in both loading cases.
However, because of more efficient load-transfer
occurring at the lower loading frequency test,
additional cracking progressively developed at the
joint region (Fig. 5).
(a) (b)
BWS specimens under cyclic displacement of higher
frequency also showed different failure pattern in Fig. 7 Failure modes of CWS specimen under (a)
comparison to those specimens under cyclic Lower and (b) Higher loading frequency
displacement of lower frequency (Fig. 6). Cracks
were distributed both at joint region and beam part for
specimens subjected with displacement of lower
frequency. This behaviour was also attributed to the
efficient load-transfer occurring at the lower loading

Hysteretic response 50
50 Higer
Beam tip load (kN)

(a) BWF (f = 0.025 Hz)
40 frequency 30
Beam tip load (kN)
20 10
0 -10
-70 -50 -30 -10 10 30 50 70
-20
-30
-40
-50 -50
-70-60 -40 -20 0 20 40 60 70 Beam tip displacement (mm)
Beam tip displacement (mm)
(a)
50
40 (b) BWF (f=1.0 Hz) 50
Higher
Beam tip load (kN)
frequency 30
Beam tip load (kN)

20
Lower
0 frequency 10
-20
-60 -40 -20 -10 0 20 40 60
-40
-50
-30
-70-60 -40 -20 0 20 40 60 70
Beam tip displacement (mm)
Fig. 8 Hysteretic responses at (a) lower and (b) higher -50
loading frequency test Beam tip displacement (mm)
Hysteretic responses of all the connection were (b)
plotted from the recorded data. Figure 8 show one of
the typical responses. The hysteretic response under a 40
Higher
higher loading frequency test is not as smooth as 30
frequency
Beam tip load (kN)
compared with the loops obtained from lower loading 20

Lower
frequency test. This behaviour may be attributed to
frequency 10
the development of acceleration in the direction
opposite to that of the displacement being applied 0
(Dhakal et al., 2003). This opposite inertia force -40 -30 -20 -10-10 0 10 20 30 40
renders the load cell reading to drop sharply around -20
the peaks of the displacement cycles. Important
-30
parameter related to seismic capacity such as ultimate
strength, stiffness degradation, energy dissipation, -40
ductility etc. were evaluated from these responses. Beam tip displacement (mm)
Comparisons of the above parameters with those (c)
obtained from lower loading frequency test were
made. Fig. 9 Envelope of hysteresis loops for (a) BWF
(b) BWS and (c) CWS specimens
Response cycles and peak loads
The envelope curves of hysteresis loops for all Stiffness degradation
specimen types under lower and higher loading Stiffness is calculated as the slope of the line joining
frequencies were plotted. These plots are shown in the peak of positive and negative loads at a given
Figure 9. Comparing the curves at each displacement cycle (Naeim and Kelly, 1999). The plots of stiffness
level, a similar trend in attaining the loading carrying variation versus drift angle for all specimens are
capacities were observed in both loading cases. shown in Figure 10. Drift angle is defined as the ratio
However, the envelope of hysteresis loops of of beam tip displacement to the length of the beam.
specimens under higher loading frequency show Comparing all the curves, the trend of degradation
slightly higher load carrying capacity in both push and was very similar in all cases studied. At the beginning
pull directions. Table 1 summarized the load carrying of the test, specimens under cyclic displacement of
capacities of all the specimens. The maximum load higher frequency show slightly higher stiffness values
obtained by averaging the peak push and pull load than those of lower loading frequency test. This
was considered as ultimate load carrying capacity. behaviour may be attributed due to strain rate effect.
However, at the end of the test stiffness attained by
specimens in both loading cases are well comparable.
The percent reductions in stiffness in both loading

cases in all specimens types were evaluated. frequency test dissipated its energy about 2 times
Comparing drift by drift levels, it has been observed higher than that of specimen under lower frequency
that the stiffness degradation rate of specimens under test. Similarly, BWS and CWS specimens under
higher loading frequency with increase in lateral higher loading frequency dissipated their energy about
movement is little higher as compared to those of the 1.6 and 1.3 times higher than that of specimen under
lower loading frequency. lower frequency test.
15 25
Higher frequency test Higher frequency
Stiffness (kN/mm)
12 Lower frequency
Cumulative energy
Lower frequency test 20
dissipated (kN-m)
9 15
6 10
3 5
0 0
0 1 2 3 4 5 6 0 1 2 3 4 5 6 7 8
Drift angle (%)
Drift angle (%)
(a) (a)
12 18
15
Cumulative energy
dissipated (kN-m)
10
Stiffness (kN/mm)
Lower frequency test Lower frequency

8 12
6 9
4 6
2 3
0 0
0 1 2 3 4 5 0 1 2 3 4 5 6
Drift angle (%) Drift angle (%)
(b) (b)
12 4
Cumulative energy
10
dissipated (kN-m)
Stiffness (kN/mm)
Lower frequency test 3 Lower frequency

8
6 2
4
1
2
0 0
0 1 2 3 4 0 1 2 3 4
Drift angle (%) Drift angle (%)
(c) (c)
Fig. 10 Stiffness variation versus drift angle for (a) Fig. 11 Cumulative energy dissipation for (a) BWF
BWF (b) BWS and (c) CWS specimens (b) BWS and (c) CWS specimens
3.1.1 Energy dissipation
The ability of a structural element to resist an Displacement ductility
earthquake load depends to a large extent on its Displacement ductility for all specimens were
capacity to dissipate its energy. The plots of energy calculated from the respective envelope curves
dissipations versus drift angle are shown in Figure 11. (Shannag et al., 2002) which can be explained in
The plot shows that higher amounts of energy were Figure 12. For computation, the ultimate displacement
dissipated by specimens under cyclic displacement of (du) was set at a displacement corresponding to 20 %
higher frequency. The energy dissipated by each drop of peak load. A relatively comparable ductility
specimen at the ultimate drift level are presented in values were achieved by the specimens both under
Table 1. BWF specimen under higher loading lower and higher loading frequency (Table 1).

axial column force. Earthquake Engineering and
Structural Dynamics, 23, 859-876.
Dhakal, R. P., and Pan, T. C. (2003). “Characteristic
of high-speed cyclic test of beam-column joint.”
ACI, Structural journals 100(2), 188-196.
Dhakal, R. P., Pan, T. C., Irawan, P., Tsai, K. C.,
Lin, K. C., and Chen, C. H. (2005). “Experimental
study on the dynamic response of gravity-designed
reinforced concrete connections.” Engineering
Structures, Elsevier 27, 75-87.
Ghobarah, A., Aziz, A., and Biddah, T.S (1997).
Rehabilitation of reinforced concrete frame
connections using corrugated steel jacketing, ACI
Fig. 12 Procedures for ductility calculation Structural Journal, 4(3), 283–294.
Hakuto, H., Park, R. and Tanaka, H. (2000). Seismic
Table 1 Capacity comparisons load tests on interior and exterior beam-column
joints with substandard reinforcing details. ACI
Control Specimens Structural Journal, 97(1), 11-25
Specimens IS 456 (2000), Plain and reinforced concrete - Code of
with loading Load Energy Practice, Bureau of Indian Standard, New Delhi.
Ductility
frequency capacity dissipation IS 1893 (I) (2002), Criteria for earthquake resistant
(du/dy)
(kN) (kN-m) design of structures - Part 1: General provisions
BWF: Low 35.995 12.34 9.4 and buildings, Bureau of Indian Standard, New
BWF: High 40.755 24.67 9.4 Delhi.
BWS: Low 37.080 10.14 8.33 Jain, S.K. and. Murty, C.V.R (2006). Proposed draft
BWS: High 41.070 16.51 7.69 provision and commentary on ductile detailing of
CWS: Low 31.230 2.45 6.67 RC structures subjected to seismic forces,
CWS: High 35.905 3.21 7.25 Document No.:: IITK-GSDMA-EQ11-V-3.0
Note, low= lower loading frequency, high= higher ::IITK- GSDMA-EQ16- V2.0.
loading frequency and ∆y= yield displacement Murty, C.V.R., Rai, D.C., Bajpai, K.K., and Jain, S.K.
(2003). Effectiveness of reinforcement details in
CONCLUSIONS exterior reinforced concrete beam-column joints
Six scale RC exterior beam-column connections with for earthquake resistance. ACI structural journal,
three typical deficiencies were tested. The two 100 (2), 149-156.
displacement histories were characterized by two Naeim, F. and Kelly, J.M (1999). Design of seismic
loading frequencies. On the basis of their comparison isolated structures from theory to practice, John
it was found that (i) seismic parameter viz. ultimate Willey & sons, Inc.
Shannag, M.J., Barakat, S. and Abdul-Kareem, M.
strength, energy dissipation, stiffness etc. exhibited
(2002). Cyclic behavior of HPFRC-repaired
slightly higher values with higher loading frequency reinforced concrete interior beam-column joints,
in all specimens types, which is attributed due to Materials and Structures, 35(6), 348-356.
strain rate effect (ii) the higher loading frequency test Shiohara, H. (2001). New model for shear failure of
cause greater damaged and produced fewer and more RC interior beam-column connections. Journal of
localized cracks on the specimens than those observed Structural Engineering, ASCE, 127 (2), 152-160.
at the lower frequency test. This indicates that the
applied loading frequency influence transfer of forces
between reinforcing bars and concrete (iii) the load
cell reading at the loading point would be affected by
inertial forces associated with high-speed cyclic test
(iv) among all deficient specimens, BWS specimens
showed a poor performance during high speed cyclic
test.
REFERENCES
Agbabian, M.S., Higazy, E. M., Abdel-gaffar, A. M.,
and Elnashai, A. S. (1994). Experimental
observations on the seismic shear performance of
RC beam-column connections subjected to varying

ISBN 978-81-921121-0-7
STUDIES ON THE TREATMENT OF VEGETABLE WASTE WITH HIGH

SOLID ANAEROBIC DIGESTERS
S. Venkatesh
Assistant Professor (Science and Humanities-Chemistry, Vel Tech High Tech Dr.Rangarajan and Dr.Sakunthala Engineering
College, Avadi, Chennai, Tamilnadu, India, E-mail: venkateshsrinivasamurthy@yahoo.com, venkat_iomau@hotmail.com
ABSTRACT- Solid waste constitutes a major environmental pollution in urban and semi-urban areas of industrially
advanced as well as developing countries of the world. Anaerobic digestion of biodegradable solid wastes is being
considered as an effective method of both management of solid wastes and as a renewable source of energy in the
form of biogas. Anaerobic digestion is the bacterial fermentation of organic wastes ion the absence of free oxygen.
Since the vegetable waste forms the major portion of organic waste and generation of biogas seem to be a viable
waste management option. The present study focuses, on the anaerobic digestion of vegetable waste using high solid
digestion (HSD) of different design. The performance of three types of HSD operated with vegetable waste as
substrate was evaluated. The first type digester was with solid and liquid phase within the digester unit. The second
type was a single unit without any phase separation. The third type of digester was with a separate solid and liquid
phase units. Two laboratory scale digesters of each type (volume -1.1 litres) were fabricated using 2 inch diameter
PVC pipe. In the first type, each digester unit was divided into two chambers using PVC perforated disc of 6mm
diameter. The first type digester was with solid and liquid phase within the digester unit. The second type was a
single unit without any phase separation. The third type of digester was with a separate solid and liquid phase units.
Two laboratory scale digesters of each type (volume -1.1 litres) were fabricated using 2 inch diameter PVC pipe. In
the first type, each digester unit was divided into two chambers using PVC perforated disc of 6mm diameter. The
digested cow dung slurry (250 ml) from the biogas plant was used as inoculum. Initially, during the start up period
there was a steady rise in biogas yield in irregular fashion in all the digesters (1056-1606 L kg-1 VS/d) from 2nd to 7th
week of operation. This increase in gas production rates may be mainly due to favorable temperature and production
of CO2 during the initial stages of operation. However the gas yield decreased during the 8th week and steadily
increased from 9th week onwards in all the digesters. The digesters from the 8th week were fed with 100.8 g fresh
weight once in every week throughout the period of study. The performance of type 1 digester in terms of biogas
yield (782.47 L kg-1 VS/d – 9th week) is better than type II digester (561.8 L kg-1 VS/d) but lesser than type III
(801.02 L kg-1 VS/d) digester. The performance of type III digester was 8.9% higher than type 1 and 23.1% more
than type II digester. The highest biogas yield in type III digester was found to be statistically significant at 99%
confidential level.
Key Words: Anaerobic digestion, High solid digestion, Multiphase digestion, Solid waste management
INTRODUCTION
Anaerobic fermentation is one of the frontiers of converts the product of the first group into hydrogen,
environmental engineering which has witnessed great carbon dioxide, and acetic acid. The third group
breakthrough in the recent years. Anaerobic digestion consists of two physiologically different groups of
is bacterial fermentation of organic wastes in the methane forming bacteria, one converting hydrogen
absence of free oxygen. The fermentation when and carbon dioxide to methane and the other forming
carried out for producing energy, leads to the methane from decarboxylation of acetate [1, 2, 3].
breakdown of complex biodegradable organics in a
four stage process to form an end product called Rapid urbanization all over the world has created a
biogas which can be used a fuel or can be converted serious problem of solid waste disposal. Solid wastes
to electricity with the help of suitable generators constitute a major environmental pollution problem
(Figure 1). Three physiological groups of bacteria are in urban and semi-urban areas of industrially
involved in the anaerobic conversion of organic advanced as well as developing countries of the
materials. The first group of hydrolyzing and world. Anaerobic digestion of biodegradable solid
fermenting bacteria converts complex organic wastes is being as an effective method of both
materials such as carbohydrates, proteins and lipids management of solid wastes and as a renewable
to fatty acids, alcohols, carbon dioxide, ammonia and source of energy in the form of biogas [4]. Anaerobic
hydrogen. The second group of acetogenic bacteria bioconversion of organic feed stocks, such as

municipal solid wastes (MSW), aquatic biomass etc., separation or multi-stage digestion. The concept of
has the potential of producing considerable energy phase separation involves operating the anaerobic
and a high-grade fertilizer/ soil conditioner. digestion process in distinct phases. This is also
Historically, research on high solid anaerobic termed as three phase or triphasic process. In some
fermentation has focused on the single charge cases the first and the second phase are operated
(batch), non mixed reactor concept, with recirculation together while the methane phase is run separately,
of effluent [5]. In these designs, gas production such processes are called two-stage or diphasic
occurs either in a single stage or leached acids may process. The most notable aspect of multiphase
be circulated to a second methanogenic stage. With process is its overall speed. For example, in
non mixed systems, rates of gas production are conventional digester it takes 15-20 days for 60-70%
generally much slower then in liquid, mixed systems conversion of biomass into methane, in multi phase
[6]. digestion the same result is achieved in less than half
of this time, since hydraulic retention has direct
In India, the main problematic solid wastes, besides bearing on the digester size and consequently on the
MSW originate from food processing industries, process economics [5]. The reduction in conversion
agro-industrial and agricultural residues. As time in multi phase digestion leads to a cost
conventional biogas plant fail to treat such wastes advantage. This is also most suitable and efficient
efficiently the high solid digestion technology is process for phytomass digestion. In the present study,
being tried to solve the solid waste management the anaerobic digestion of vegetable waste using high
problems since early eighties. One approach which solid digestion (HSD) of different design was
holds great promise in solving the technical problems observed.
associated with the anaerobic digestion is phase
Hydrogen
and Carbon
dioxide
Biomass and Simple Organics Higher Organic
Organic wastes acids Methane
Acetic acid
Stage 1 Stage 2 & 3 Stage 4 (Methane

(Hydrolysis (Acid Phase) Phase)
Phase)
Figure 1: Different phases involved in anaerobic

digestion
b) Inocula
MATERIALS In high solid digesters, the effluent slurry form a
natural biogas digester was used as inocula. The
a) Vegetable waste slurry was mixed with double distilled water (1:4
The vegetable waste was chopped manually using ratio), stirred well, allowed to settle for about four
knife to approximately 2-3 cm pieces for charging the hours, decanted and filtered through cloth. These
digesters. The uniformly chopped vegetable waste inocula were used to feed the digesters [7]. The
were analyzed for total solids, volatile solids, inocula were analyzed for total solids, volatile solids,
moisture content, carbon and nitrogen (Table 1) COD (Table 2)
Table 1: Characteristics Of Vegetable Waste Used As Feed

Parameters Results
Total solids 7.93%
Volatile solids 6.74%
Moisture content 92.07%
Nitrogen 336 mg/Kg dry wt.
Carbon 0.506%

Table 2: Chemical Analysis Of Inocula Used For Seeding
Inocula Total solids Volatile solids C.O.D
Digested cow dung slurry 4700 mg/L 3600 mg/L 11200 mg/L
c) Reactors In all the digester types vegetable wastes was used as

(i) Digester with phase separation within the single a substrate. Uniformity in operational parameters
unit such as loading rate, temperature, recycling of
Two laboratory scale digesters of each type (volume - bioleacheate was maintained in all the digester types.
1.1 litres) were fabricated using 2 inch diameter PVC The digesters were charged initially with 25 g of
pipe (Figure 2a). Each digester unit was divided into spent material and 6.3 g fresh weight of freshly
two chambers using PVC perforated disc of 6mm chopped vegetable waste (0.5 g dry weight). The
diameter. The perforations were made for the digesters were fed subsequently with 6.3 g (fresh
movement of the leacheate from the solid bed to the weight), once weekly for first three weeks prior to
liquid phase during recycling process. The upper start. After 3 weeks, the digester was charged with
chamber was used as solid phase and the lower 50.8 g (fresh weight) and left for acclimation for the
portion was used as liquid phase. The volume of solid next 21 days. On the 21st day they were fed with
phase compartment was 800 ml, while the volume of 100.8 g (fresh weight) once weekly which was
the liquid phase chamber was kept as 300 ml continued throughout the period of study. The
respectively. The biogas production from solid phase digested cow dung slurry (250 ml) from the biogas
and liquid phase chambers were measured plant was used as inoculum. The bioleacheate was
individually. All the digesters were sealed airtight to circulated over the solid bed in all the types of
prevent the escape of gas. digesters using electrically operated peristaltic pump
for 20 minutes twice daily at constant flow rates. In
(ii) Digestion without any phase separation all the above digester the bioleacheate drawn from
Two laboratory scale digesters of each type (volume - the bottom (liquid phase) was sprinkled over the top
1.1 litres) were fabricated using 2 inch diameter PVC of the solid bed using electrically operated peristaltic
pipe (Figure 2b). Unlike the above kind of HSDs pump. The liquid was sprinkled with the help of glass
there was no compartmentalization in the reactor sprayer with four arms for spraying. The digester was
unit. operated in a batch mode. Once the digester was
filled up the n the digested phytomass equivalent to
(iii) Digestion with separate solid and liquid phase one third volume of the reactor was removed from
unit the bottom portion of the digester. The single feed
Two laboratory scale digesters of each type (volume - outlet was provided at the side to remove the digested
1.1 litres) were fabricated using 2 inch diameter PVC waste. All the digesters were connected to a gas
pipe (Figure 2c). The volume of solid phase collecting system with rubber tubing. The gas
compartment was 800 ml, while the volume of the produced was measured by the graduated plastic
liquid phase chamber was kept as 300 ml cylinders. The gas produced was measured twice a
respectively. Both the units were connected by a day at twelve hour interval. The gas collecting system
rubber tube with flow regulator valve. was filled with acidified water (pH- 2) in order to
prevent the absorption of CO2 and H2S from the
METHODOLOGY biogas.
Digester Operation

might encourage the growth of the facultative
fermenting and acetogenic bacteria other than
methanogens results in high production of CO2 [7,8].
However the gas yield decreased during the 8th week
and steadily increased from 9th week onwards in all
the digesters. The digesters from the 8th week were
fed with 100.8 g fresh weight once in every week
throughout the period of study. There was a slight
drip in biogas yield on the 13th week in all the three
digester types due to the partial removal of the
digester contents from the digester unit.
The type II digester without phase separation was

found to perform with lowest efficiency interms of
biogas yield. This could be due the fact that both
acidogens and methanogens grow in a consortium
Fig. 2 Different High Solid Digester Unit within the single unit. As the concept of phase
separation clearly indicates that both acidogens and
methanogens vary in their requirements –namely pH,
RESULTS AND DISCUSSION type of feed, temperature etc. The methanogens are
The results of the digester performance in terms of very sensitive to the above mentioned factors while
biogas yield are given in Table 3. Initially, during the the acidogens are hardier [9,10].The growth rate of
start up period there was a steady rise in biogas yield acidogens may be higher while that of the
in irregular fashion in all the digesters (1056-1606 L methanogens may be poorer in a single digester unit
kg-1 VS/d) from 2nd to 7th week of operation. This without any phase separation resulting in poor biogas
increase in gas production rates may be mainly due to yield (Figure 3) than digester with phase separation.
favorable temperature and production of CO2 during This is further evident from the fact that digester with
the initial stages of operation. Also during the initial phase separation within the single digester unit (type
periods the population of acidogens is expected to be I) performed better than digester without phase
higher in proportion than the methanogens favouring separation (type 2) while the digester with separate
CO2 production. Secondly during the initial periods solid and liquid phase (type 3) performed better than
the availability of empty head space filled with air all the digester type interms of biogas yield.
and the air trapped in between the feed material
Table 3: Performance of Different High Solid Digester with Vegetable Waste as Feed
Period Fresh Type I digester Type II Type III digester

(Days) wt. Loading rate digester
(g/d) (g/d) Biogas yield* (L Kg-1 Total Biogas yield (L Kg-1 VS/d)
VS/d) gas
Total Volatile Solid Liquid Total yield Solid Liquid Total
.Solids Solids phase phase phase phase
1-7 4.47 0.35 0.29 224.13 210.34 434.47 431.03 245.1 220.34 465.44
8-14 0.9 0.071 0.06 750 666.66 1416.66 1250 767 684.6 1452.66
15-21 0.9 0.071 0.06 833.3 683.33 1516.63 1183 840 697.65 1537.65
22-28 0.9 0.071 0.06 850 716.66 1566.66 1316.6 867 740.66 1607.66
29-35 2.4 0.19 0.16 687.5 593.75 1281.25 1031.2 717 615.05 1332.05
36-42 2.4 0.19 0.16 781.25 687.5 1468.75 1231.25 797 699.5 1496.5
43-49 2.4 0.19 0.16 818.75 756.25 1575 1281.25 829.75 777.2 1606.95
50-56 14.4 1.142 0.97 345.36 335.05 680.41 500 361.85 351.54 713.39
57-63 14.4 1.142 0.97 376.28 365.97 742.25 513 396.9 376.28 773.18
64-70 14.4 1.142 0.97 400 382.47 782.47 561.8 409.27 391.75 801.02
71-77 14.4 1.142 0.97 403.09 387.62 790.71 585.56 411.34 402.06 813.4
78-84 14.4 1.142 0.97 415.46 396.9 812.36 617.52 448.45 423.71 872.16
85-91 14.4 1.142 0.97 409.71 388.65 798.36 622.26 434.02 422.68 856.7
92-99 14.4 1.142 0.97 423.71 407.21 830.92 685.56 479.39 448.98 927.83
100-105 14.4 1.142 0.97 441.23 413.4 854.63 680.41 484.53 453.6 938.13
*The biogas yield is presented as an average of seven days

advantage of phase separation is that the liquid
Bi ogas yi el d as a function of ti me
1800 effluent would be lesser in VFA content and when
Biogas yield (L Kg-1 VS day -1)
1600 recycled into the solid phase using peristaltic pump

1400
1200 will not alter the VFA content of the solid phase
1000 digester. Hence the pH level within the solid phase
800
600 digester unit will be quite stable. The stable pH or
400 less altered ph is quite advantageous for the
200
0 facultative anaerobes and acidogens present in the
digester to perform well in terms of biogas yield.
7
1
9
5
1
5
0-
-6
-9
-4
-2
-3
-7
10
Digester without phase
43
57
71
85
15
29
0-
10
seperat ion
Days Digester with phase
seperat ion within single unit
Digester with separate solid
The statistically significant data of biogas yield were
Figure 3: Biogas Yield From High Solid Digester As assessed by using students –t-test for all the three
A Function Of Time types of digesters. The digester with separate solid
and liquid phase was found to be performing better
The performance of type III digester was 8.9% higher than all the other types of digester (Table 4).The
than type 1 and 23.1% more than type II digester. highest biogas yield was found to be statistically
Therefore separating the solid and liquid phase in a significant at 99% confidence level.
high solid digester does improve the performance of The removed spent phytomass from the digester were
the digester interms of biogas yield. The phase analyzed for organic nitrogen. The organic nitrogen
separation perhaps helps in providing optimal was found to be 25% more than that of the organic
conditions for the respective bacterial groups there by nitrogen content originally present in the feed (Figure
enhancing their growth and activity. Moreover, by 4).
separating the phases as separate entities or units The increase in the organic nitrogen can be attributed
(type 3) digester, the flow rate, concentration of to the microbial action and the digested material has
volatile fatty acid (VFA) in the bioleacheate can be good fertilizer value. The spent feed with high
controlled in order to facilitate the growth of organic nitrogen content thus can be further used as
methanogens. Additionally by providing inert support fertilizer.
media or by stirring the contents of the liquid phase
the activity of methanogens in the liquid phase can be CONCLUSION
enhanced which in turn increase the biogas yield. Thus, our studies on anaerobic digestion of
The concept of phase separation was clearly stressed vegetable waste indicate that high solid
in our studies as both acidogens and methanogens digestion coupled with the concept of phase
vary in their requirements –namely pH, type of feed, separation is the better option –rather viable and
temperature etc. The methanogens are very sensitive efficient option for phytomass treatment.
to the above mentioned factors while the acidogens
are hardier [9, 10]. Therefore separating the solid and ACKNOWLEDGEMENT
liquid phase in a high solid digester does improve the The author thanks the Centre for Pollution
performance of the digester interms of biogas yield Control and Energy Technology (CPCET-
because the growth rate of acidogens may be higher
Pondicherry University) for their support in the
while that of the methanogens may be poorer in a
single digester unit without any phase separation and research activity.
may result in poor biogas yield . The another
Table 4: Comparison of Performance of Different Types Of HSD’s In Terms of Biogas Yield

Loading rate (g/d) Days Biogas yield (L Kg-1 VS/d) t value S*/NS**
0.97 20 Type I- 860.12 Type II- 679.19 80.78 S
0.97 20 Type I- 860.12 Type III- 937.72 -39.08 S
0.97 20 Type II- 679.19 Type III- 937.72 -104.33 S

500 acids with subsequent conversion to methane
450
and manure” In proceeding of the twelfth
dry wt)
400
350 International Conference on Solid waste

-1
Nitrog en co ntent (mg kg
300 Technology and Management, Philadelphia,

250
200
USA, Nov 17-20, 4C1-4C8, 1996.
150 [10] S.A. Abbasi, and E.V. Ramasamy,
100 “Biotechnological pollution control systems:
50
0
the technologies most appropriate for the third
Fresh phytomass spent phytomass world countries” University press Ltd.,
Hyderabad, 1997.
Fig 4: Comparison of Organic Nitrogen Content
Present In Fresh And Spent Phytomass
REFERENCES
[1] M.P.Bryant, M.J. Wolin, and R.S. woife,
“Methanobacillus omlianskii, a symbiotic
association of two species of bacteria”.
Archives Microbiology, Vol. 59, 20-31, 1967.
[2] R.A. Mah, D.M.Ward, L. Baresi, and T.L.
Glass, “Biogenesis of methane”. Ann. Rev.
Microbiology, Vol.31, 309-34, 1977.
[3] B.R. Nagar, and C.Tietjan, “A status report on
the use of biogas technology for increasing fuel
and soil productivity”. Journal of Scientific
and Industrial Research, Vol. 37, 445-448,
1978.
[4] R.M. Shasikant, and K.U. Abande, “Anaerobic
digestion of vegetable solid waste as a method
of solid waste management” Journal of the
IPHE, Vol. 1, 92-99, 1995.
[5] S. Ghosh, “Solid phase digestion of low
moisture feeds” Biotechnol. Bioengg. Suppl.
Vol. 14, 365-376, 1984.
[6] J.L. Gaddy, and E.C. Clausen, “Solar energy
research Institute” (SERI), pp-231-269, 1985.
[7] H.N. Chanakya, Sushma Borgaonkar, G.
Meena, and K.S. Jagadish, “Solid phase biogas
production with garbage or water hyacinth”
Bioresource technology, Vol. 46, 227-
231,1993.
[8] M. Shyam, and P.K. Sharma, “Solid-state
anaerobic digestion of cattle dung and agro-
residues in small capacity field digesters”
Bioresource technology, Vol.48, 203-207,
1994.
[9] S.A. Abbasi, and E.V. Ramasamy, “Utilization
of biowaste solids by extracting volatile fatty

ISBN 978-81-921121-0-7
PROTOTYPE DEVELOPMENT FOR A BUS ARRIVAL TIME

PREDICTION SYSTEM UNDER INDIAN TRAFFIC CONDITIONS
Akhilesh Koppineni, Krishna Chaithanya
Under graduate student, Dept. of Civil Engg., IIT Madras
K. Sidharth
Under graduate student, Dept. of Mechanical Engg., IIT Madras
Lelitha Vanajakshi, (Corresponding Author),
Assistant Professor, Dept. of Civil Engg., IIT Madras, Chennai – 36. Ph. 44 2257 4291, email: lelitha@iitm.ac.in
ABSTRACT: This study focuses on developing a prototype for a robust real time bus arrival prediction
system, which can be used under Indian traffic conditions. A prediction algorithm reported to be performing
accurately under Indian traffic conditions using a model based approach is modified and tested for identifying
the best prediction technique to be adopted for the prototype development. The algorithm uses Kalman Filtering
Technique for the estimation of bus arrival time and was tested for varying traffic conditions during different
times of the day. The system uses real time data from the GPS units installed in public transport buses and
predicts the arrival time of buses. The results were compared with the actual observations from the field. The
development of prototypes for field implementation and field demonstrations involved integration of the output
from the real time algorithm with the information display systems. Prototypes were developed for a regular
variable message sign at bus stops, web based information system and Kiosk applications at bus stops. This
information can also be conveyed to registered users through SMS. The issues to be tackled included reliability,
security, scalability and transferability. For example, the system developed should not have restrictions on the
number of buses, and bus-stops along the route. The application should also be transferable to any route with
varying lengths. All these were tackled and prototypes were developed and demonstrated which can be used for
field implementation of bus arrival time prediction system under Indian traffic conditions.
INTRODUCTION 1990; Thakuriah et al. 1992; VanLint et al. 2002;

India, on its way to economic growth, is facing drastic Jeong & Rilett 2005; Vanajakshi & Rilett 2004; Kwon
changes in urban transportation. Increase in vehicle et al. 2000; Sen et al. 1991; Ishak & Al-Deek 2002),
ownership, and inefficient public mode of transport most of which are developed for homogeneous traffic
are leading to severe traffic congestion in most of the conditions and may not perform efficiently under
urban areas. There are several approaches that can be Indian traffic conditions. There are not many studies
adopted to tackle this problem. Construction of more reported under heterogeneous traffic conditions for
infrastructures is one of the basic solutions, which short-term travel time prediction (Ramakrishna et al.
may not be always possible. Also, there is limitation 2006; Vanajakshi et al. 2008) and none of them
to increase infrastructure and it may never be able to carried out field implementation. Also, the limited
meet the vehicular growth rate. Various traffic existing bus arrival prediction systems implemented
management measures can be another option and in Indian cities as pilot studies (Wikipedia) are not
Intelligent Transportation Systems (ITS) is one of the based on any accurate prediction methods and hence
advanced options under this category. The major are not able to provide accurate and reliable
objective of ITS is to evaluate, develop, analyse and information. The present study will be an attempt in
integrate new technologies and concepts to achieve this direction for the prototype development of a bus
traffic efficiency, improve environmental quality, save arrival prediction system that is quick in processing
energy, conserve time, and enhance safety and and results in accurate prediction and is presented
comfort for drivers, pedestrians, and other traffic with associated customization issues.
groups. One functional area of ITS namely Advanced
Public Transportation Systems (APTS) concentrate on The major objective of the study is to develop
making the public transport system more attractive by prototypes for the field implementation of a passenger
using more advanced technologies. One such information system which will provide real time bus
application is the provision of bus arrival time arrival time information both pre trip and at bus stops
information to the passengers to make their journey under Indian traffic conditions. The specific
more efficient. Providing reliable bus arrival objectives include:
information to passengers will definitely be an 1) Application Development: An algorithm proposed
incentive to shift from private to public mode of by Vanajakshi et al. (2008) will be modified for
transport, which will help in relieving congestion. integrating with real time data and information display
for field implementation. A robust application would
Numerous algorithms for bus arrival time prediction be coded in Python to implement the arrival time
exist varying from model based, data driven, prediction algorithm for field application. The
statistical and empirical methods (Hoffman & Janko application needs to be highly scalable in order to

ensure that it can be easily expanded to include new process. The data after filtering can be used for the
routes and buses as and when required. application. In the present study, the selected
2) Application testing: The application will be tested approach is a model based prediction algorithm which
for its accuracy during different scenarios that can uses Kalman filtering technique. In this algorithm,
happen in the field such as normal flow, congestion, data from two previous buses, PV1 and PV2 are used
incidents etc. to verify the reliability of the as input to predict the arrival time of the current bus,
application. which is called test vehicle, TV. Thus, the real time
3) Prototype development: A display board prototype data received from three consecutive buses are used
shall be developed for the display of information for the bus arrival time prediction (ATP). PV1, PV2
related to predicted bus arrivals and other details like for a selected TV should be from the same date folder,
current bus location for various bus-stops via a Kiosk and should have the same routeID. Also since the
application. Same information will be made available ATP has to be made for the next bus-stop to which
through a web site for users to access it pre-trip. Also, TV shall be arriving next, it is necessary that both
a map application will be developed to show the PV1 and PV2 should have already crossed this bus-
current location of the buses in a map. Demonstration stop. If this condition is not satisfied then previous
of the above three will be carried out using data from trips are considered. Once the three datasets of PV1,
5C route in Chennai. PV2 and TV have been identified, they are fed into
the ATP algorithm which is discussed in the next
STUDY ROUTE AND DATA COLLECTION section
Data for this project is being collected using
GPS/GPRS units installed in the buses. Data from METHODOLOGY
these units are sent to a dedicated server once in every The prediction algorithm adopted in the present study
10 seconds as 2 packets of 5 sec each. This data is based on the algorithm developed by Vanajakshi et
packet when received by the server is appended to a al. (2008). The base algorithm is discussed first
.csv (comma separated value) file. Data from each bus followed by the possible modifications.
for the day is stored in a different file. The file names
are of the format ‘DDMON-BusID’ eg: ‘10Nov- Base Prediction Method
iit07’, and are stored in a separate folder for every The base algorithm predicts the arrival time/travel
year. Files are archived daily and new files are created time of the vehicle of interest, called as Test Vehicle
to store the incoming data everyday at 00:00:00. The (TV), using data from two previous buses traveling in
real-time data received in the server is immediately the same route, which are called as Probe Vehicle1
used in order to make a real-time prediction of the (PV1), and Probe Vehicle 2 (PV2). To start with, the
arrival time of the bus. The raw data contains route under consideration is divided into N
information on ‘deviceID’, ‘Clock time’, ‘latitude’ subsections and it is assumed that the evolution of
and ‘longitude’. From this data, the information of travel time of two consecutive subsections is related
interest in the present study namely travel time can be as
easily estimated. x(k  1)  a (k ) x(k )  w(k ), (1)
where, x(k) is the travel time for covering the kth
The 5C bus route between Taramani - Broadway was subsection, a(k) is a parameter which relates the travel
the routes selected for the project. The 5C route is time taken in the (k+1)th subsection to the travel time
about 16.8 kilometers; and takes about 60 to 70 taken in the kth subsection and w(k) is the process
minutes for the bus to travel from one terminus to the disturbance associated with the kth subsection. The
other in each direction. There are 7 buses in 5C route measurement process is assumes to be given by
with headway of around 30 minutes and there are 10 z (k )  x(k )  v(k ), (2)
bus-stops along the route. where, z(k) is the measured travel time of the kth
subsection and v(k) is the measurement noise. It is
Data Filtration and Analysis assumed that both w(k) and v(k) are Gaussian zero
All the data that is received at the server is not mean white noise signals. The prediction scheme
necessary for the bus Arrival Time Prediction (ATP). consists of the following steps:
Hence, this raw data needs to be filtered in order to 1) The parameter a(k) is calculated for each
input only the required data into the algorithms. The subsection using the travel time data for PV1
unfiltered data consists of the location details of the thorough
bus for the entire duration for which it is being
a(k) = xpv1(k+1)/xpv1 (k), (3)
monitored. For example, we do not need to keep track
where, xpv1(k) is the travel time taken by PV1 to cover
of the data being sent by the bus while it is parked in
the kth subsection.
the bus depot at night, or while it is kept waiting at the
2) The Kalman filter based recursive algorithm is then
terminus during lunch etc. Such data needs to be
used for predicting the travel time of the test vehicle.
filtered out and only the essential data has to be fed
into the algorithm in order to get the required results.
A program is developed for automating this filtering

This base method assumes the process disturbance co- Group 1 and Group 3 one can see that in cases where
variance Q and measurement co-variance R to be the difference between the travel times of PV1 and
constant throughout the simulation; however the PV2 exceeding d in a particular subsection the
values of Q and R can be made to vary to reflect the prediction is carried out solely based on the data
accuracy and reliability of data obtained from PV1 provided by the vehicle having the lower travel time
and PV2 as in the next method. of the two.
Simple Adaptive Prediction Method Once the initial conditions have been assumed, similar
The simple Adaptive Prediction Method presented in steps as in base method are applied recursively for
this subsection computes the value of Q and R at subsections k =2 to N, with a(k) as :
every step by utilizing the travel time information x pv1 (k ) x pv 2 (k )
from previous subsections. For this, equation (1) and a (k  1)     (1   )  ,
(2) can be rearranged as x pv1 ( k  1) x pv 2 (k  1)
(11)
w(k )  xtv (k  1)  a(k ) xtv (k ) , (4)
The value of α for these three different cases is: Group
v( k )  xtv ( k )  x pv 2 ( k ) , (5) 1: α=0,Group 2: α=0.5,Group 3: α=1.
where, xtv (k ) is the travel time for the test vehicle in The scheme was implemented and compared with the
the kth subsection base method and the results are discussed in the next
section.
Then one may compute Q(k) and R(k) as:
k
1 2 IMPLEMENTATION AND RESULTS
Q( k ) 
S  w(i )  w (k )
ik S
, (6)
In this section implementation of the algorithms
discussed in the previous section are discussed and a
S
comparison of the performance is carried out to
 w(i)
i k  S
choose the one that can be used for the real time field
where, w ( k )  , (7) implementation. Following this, the selected
S algorithm will be used for the development of the
and S is a predetermined constant number. complete application that can be used for automated
real time field implementation.
Similarly, R(k) can be estimated using
k
1
R (k )   v(i)  v (k ) 2 , (8) Implementation
S i k  S Implementation of base, adaptive and enhanced base
This approach was explored in the present study and method is carried out and is detailed in this section.
the performance is compared with the base method. While implementing, once the TV starts from a
terminus, it is assigned a route and its arrival
Enhanced Base Prediction Method prediction started, provided two previous buses are
Another modification that was explored in the present already in the route or covered the route. Thus, the
study for better prediction performance was the use of corresponding ‘PV1’ and ‘PV2’ datasets for the
weights for travel time values. This is called as current TV trip need to be identified as discussed
Enhanced Base Prediction Method. The method takes below.
a weighted average of the travel times of PV1 and
PV2 as the measured travel time for that subsection. PV1, and PV2 for a selected TV should be from the
The weights are estimated based on the relative travel same date folder, and should have the same routeID.
Hence, if TV’s tripID is ‘t’ then PV1 and PV2 are so
time values of PV1 and PV2. We define v(k ) as,
selected that their tripIDs are ‘t-1’ and ‘t-2’
v( k )  x pv1 ( k )  x pv 2 ( k ) , (9) respectively. Also since the ATP has to be made for
If v > d, then subsection k belongs to Group 1 the bus-stop to which TV shall be arriving next, it is
If -d≤ v ≤ d, then subsection k belongs to Group 2 necessary that both PV1 and PV2 should have already
If v < -d, then subsection k belongs to Group 3 crossed this bus-stop. If this condition is not satisfied
where, d is a predetermined threshold value. In the then previous trips are considered. Once the three
present study, the value of d is taken as datasets PV1, PV2 and TV have been identified, they
N N are fed into the algorithm.
 x pv1 (i)
i 1
x
i 1
pv 2 (i )
The performance of three algorithms discussed in the
d  , (10) earlier section was compared to select the most
N N suitable algorithm for real time bus arrival time
which is the sum of the means of the travel times of
prediction. Accuracy of these methods is compared
PV1 and PV2. By studying the prediction method for

during different times of the day namely; the morning Table 2: MAPE for Base and Enhanced Base
peak period, afternoon off-peak period and evening methods for 200 m subsections During Evening Peak
peak period. One week’s data from 6th March to 12th Period
March, 2011 from the 5C route is used for Base Method Enhanced Base
Date
comparison. MAPE Method MAPE
It was found that there are two distinct peak periods 6th Mar, 2011 25.90 24.88
during the day, one in the morning from around 7:30
AM - 10:30 AM and another in the evening from 7th Mar, 2011 27.10 25.81
around 4:30 PM – 9:00 PM. Also it was observed that 8th Mar, 2011 28.92 30.34
during afternoons the travel times are relatively low 9th Mar, 2011 35.22 33.17
and also consistent. Hence, the time period between
10:30 AM – 4:30 PM is considered as off-peak period 10th Mar, 2011 39.24 38.20
for this study. The algorithm requires at least two 11th Mar, 2011 36.19 38.22
buses (for PV1 and PV2) to have traversed the same
12th Mar, 2011 31.66 32.08
route before TV. Based on this and other constraints,
the first TV is usually taken after 7:30 AM and the
last trip before 9:30 PM. Hence, the ATP algorithm A similar comparison was carried out using the
was ran for trips between 7:30 AM and 9 PM. Enhanced Base Prediction method too.The MAPE
The accuracy of the prediction of the algorithms is values for 200 m subsections for the Base and
separately analyzed for the above mentioned three Enhanced Base method during evening peak are given
time periods. This would help in identifying the better in Table 2. Similar results were obtained for off peak
predictor during each of these time periods.The and morning peak. It can be seen that even in case of
accuracy of the prediction schemes are quantified Enhanced Base Prediction method the reduction in the
using the Mean Absolute Percentage Error (MAPE) MAPE value is not very significant. In fact in some
given by cases the Enhanced Base Method gives higher
MAPEs when compared to Base method. One
N xtvm ( k )  xˆ ii ( k ) possible explanation could be that the choice of α =
1 xtvm ( k )
* 100 0.5 might not be suitable for this study. It is possible
MAPE = , (12) that value of α might be different for each route,
N depending on the characteristics of the route. A route
where, xtvm(k) is the measured travel time of the test specific optimized α value may increase the
vehicle in the kth subsection, and N is the total number performance. Further studies need to be carried out to
of subsections. investigate the route characteristics which can affect
the optimum value of α.
The implementation was carried out by dividing the
entire route into subsections of 200m each and MAPE It was concluded from the results that out of the three
in predicted travel times is calculated for each of this model based algorithms, there is no significant
sub sections. Results obtained using all three improvement in the prediction performance due to the
prediction methods are discussed below. The MAPE modifications and hence the Base Prediction Method
values for 200 m subsections for the Base and Simple will be adopted in this study. The development of the
Adaptive methods during morning peak are given in prototype application using this method is discussed
Table 1. Similar results were obtained for off peak in the following section.
and evening peak. As can be seen from the Table 1,
by using the Simple Adaptive method, no significant
reduction in the MAPE values was observed. PROTOTYPE DEVELOPMENT
Table 1: MAPE for Base and Simple Adaptive Based on the comparison of performance of the base
methods for 200 m subsections During Morning Peak algorithm and the modified algorithms presented in
Period the previous chapter, the base method was selected for
Simple final implementation due to its simplicity and
Base Method relatively comparable performance with the modified
Date Adaptive
MAPE versions. The algorithm was modified for real time
Method MAPE
processing and issues were addressed to get the
6th Mar, 2011 32.00 31.83
desired output in real-time. The present study also
7th Mar, 2011 23.71 23.15 studied various information dissemination options.
8th Mar, 2011 25.56 25.47 Mainly three different methods were found to be
suitable for the present application, namely Bus-Stop
9th Mar, 2011 31.06 28.65 Variable Message Signs (VMS), Kiosk at bus stops
10th Mar, 2011 27.56 27.70 and Web Based information system. Prototype for
11th Mar, 2011 34.62 35.78 each are tested so that it can be directly used for field
implementation and are discussed below.
12th Mar, 2011 26.68 26.90

Bus Stop Variable Message Signs (VMS) the present study, a similar Information Dissemination
VMS are the most popularly adopted means for Unit (IDU) was developed. The IDU is a suitably
passenger information systems. These display units customized simple computer unit which can either be
are available in various sizes and has enough used as a display board at the bus-stops or can be used
flexibility in terms of display messages since they can as an interactive kiosk unit as per the requirement.
be controlled remotely through GPRS. These display Each IDU consists of a simple low-end computer with
units have an in-built GPRS module, which can a GPRS module attached to it for data transfer. The
receive information from the server in real time. GPRS module acts as a client to the central server
Information is transmitted from the server at regular where all the data is processed and sent to the IDU.
intervals (usually in less than a minute) and this The application running on the IDU runs a separate
information is displayed at the bus-stop. Usually the instance for each bus-stop and queries the central
LED matrix only displays numerical and text server for updated bus-stop data, every 10 seconds.
messages. The display can be made in any languages.
The IDU can be placed at any bus stops and can be
Most of the metropolitan transport corporations which used for showing the current location of the bus.
are trying bus arrival information system are using Another option for IDU is to make it interactive and
these display units. Many of them shows the current dynamic, where map application can be run. The main
time, route number of the bus, type of the bus (deluxe idea here is to display the dynamic location of the bus
in this case) and also the final destination of the bus on a map. In Fig. 3 such a Map Display Bus-Stop IDU
route along with the expected arrival time of the next application is shown. This application shows the
bus. Fig. 1 shows one such sample. current location of different buses in the vicinity of
the bus-stop and their arrival times at the bus-stop.
This information is updated once every 10 seconds.
Fig. 1: MTC’s display format
The format of display suggested in the present study

is slightly different to make it more simple and
informative. In order to keep the information provided
short and simple, a modified format is suggested as
shown in Fig. 2.
Route Expected Time of Time Now

Number Arrival
5C Within 15 mins 04:42 PM
Fig. 3: Map Display Bus-Stop IDU application
Fig. 2: Modified display format Web Based Information Service
Apart from the above mentioned two methods,
Kiosk at Bus Stops providing bus arrival details through a website is also
A computer kiosk houses a computer terminal that a popular way of information dissemination to the
often employs custom kiosk software designed to commuters. Usually it is in the form of an interactive
function flawlessly while preventing users from website where one can select the desired route and bus
accessing system functions. Computerized kiosks may stop to get the real-time bus arrival and location
store data locally, or retrieve it from a computer details. Through the website a commuter can check
network. Interactive kiosks can be used at bus-stops to the arrival details from the comfort of one’s office or
provide commuters with useful information. The user home, and can reach the bus-stop close to the
needs to enter the particulars of the information he/she predicted arrival time. Once this information is
is seeking. Suppose a commuter wants to know the available to the commuter, it saves a great deal of time
arrival time of a particular bus number at the bus-stop, for the commuter, who can spend the intermediate
then he/she would be required to enter the route time in a more productive way than waiting at the
number and the bus-stop of interest. Once this is done, bus-stop. Figure 3 shows one such example of a web
the kiosk fetches the required information from the service.
central server. All this is done in real time and hence Similar application was also developed in the present
each query gives the latest available information. study which will inform the commuters on their
They are interactive and more personalised when expected wait time at the bus stops through a web site.
compared to the bus-stop display units. As a part of

However, it will be more advantageous if this like India are different with its heterogeneity and lack
information can be provided to them before they start of lane discipline and hence predicting the bus arrival
the trip to avoid the wait time at bus stops. A web time is much more challenging. Only a limited
based application was developed for this purpose, number of studies have been reported for such traffic
which can be used by commuters to check the conditions and none of them carried out field
expected arrival time of the route of choice in the bus implementation. Also, a few transportation agencies
stop they are interested. The web site is interactive in India have already tried to implement the bus
and provides the users with the current location of the arrival prediction systems in the country. However, all
bus and it’s ETA at any chosen bus-stop. A Google of them mainly concentrated on the technology
Maps based website was developed for this for easy demonstration of feasibility rather than the reliability
access to information for the commuter. Fig. 4 below of the information being provided. In most cases the
shows a screenshot of the same. The website provides accuracy of arrival time prediction is poor, and the
the flexibility of choosing the bus-stop at which the primary reason for this may be that the algorithms
user wishes to board the bus. Unlike kiosk display the used for prediction were not the best one. In most of
website provides the estimated arrival time at the bus- the cases, the prediction algorithm details are not
stop even when the ETA is greater than 15 minutes. available since it was implemented by an outside
This change was incorporated to help the user decide agency. Most of the above implementations were
when she/he needs to go to the bus-stop to board the outsourced and used methodologies developed for
desired bus. western countries which have homogenous traffic, and
may not be suitable in the heterogeneous Indian traffic
conditions. The present study was an attempt in this
area of developing an accurate real time bus arrival
prediction system under Indian traffic conditions.
GPS data collected from buses running in route
numbers 5C in Chennai, India were used. In order to
have a real time automated application, it was
required to develop an automated data collection
filtration and analysis system without manual
intervention. A model based prediction scheme was
used and the Kalman Filtering Technique was adapted
to estimate and predict the travel time of buses. Three
prediction schemes were tested and the best one was
Fig. 4: Google Maps based web application
used for implementation in the real-time application.
Numerous issues were identified and addressed during
The expected arrival time of the selected route for the
the implementation of the model in real-time. Issues
selected bus stop for chosen time of the day can also
such as effect of traffic jams, overtaking among the
be provided to users through SMS. For this kind of a
buses on the same route, bus breakdown, abrupt
service the users are required to subscribe beforehand.
changes in bus routes, etc. which are specific to Indian
Once registered, users get regular SMS during
conditions are taken into consideration.
different times of the day as per their choice.
The project also developed prototype systems for
SUMMARY AND CONCLUSIONS
information dissemination which can be readily used
One of the major factors responsible for the success or
for field implementation. Prototypes for three
failure of any public transport service is its reliability.
different information dissemination modes were
One way of improving the reliability of service is to
developed namely, bus stop VMS display, Kiosk
provide the passengers with accurate information
display at bus stops and web based application. A
regarding the service. In case of the bus service,
more advanced kiosk system was also developed
information regarding current bus location, arrival
considering specific requirements of users. It is more
time at bus stops, etc. will help to reduce the
economical and provides greater information in a
uncertainty and waiting time involved in the use of
cognitively ergonomic format to aid the commuter’s
bus service system. This may help in attracting more
decision making skills. This information on
passengers to the public transport, which in turn will
predicted bus arrival time will be useful to the
help to reduce congestion in the road network. Hence,
commuters to make the trips more efficient. For the
techniques for accurate forecasting of bus arrival
transit authorities such information will be useful for
times have become a major area of research in recent
planning the routing and scheduling activities.
years.
A majority of the studies in the area of bus travel ACKNOWLEDGEMENT

time/arrival time prediction are from homogeneous The authors acknowledge the support for this study as
traffic conditions. The traffic conditions in countries part of the project CIE/10-11/168/IITM/LELI by

Ministry of Urban Development, Government of
India, through letter No. N-11025/30/2008-UCD.
REFERENCES
Hoffman C., Janko J. (1990), Travel Time As A Basis
Of The LISB Guidance Strategy, Proc. IEEE
Road Traffic Control Conf., 1990.
Http://En.Wikipedia.Org/Wiki/Ahmedabad_Brts.
Http://En.Wikipedia.Org/Wiki/Delhi_Bus_Rapid_Tra
nsit_System.
Ishak S., Al-Deek H. (2002), Performance Evaluation
Of Short Term Time Series Traffic Prediction
Model, ASCE J. Transp. Eng., 128, 6, 490-499.
Jeong R., Rilett L.R. (2005), Bus Arrival Time
Prediction Model for Real-Time Applications,
Transportation Research Board, National
Research Council, Washington, D.C.
Kwon J., Coifman B., Bickel P. (2000), Day-To-Day
Travel Time Trends and Travel Time Prediction
from Loop Detector Data, Transportation
Research Board, National Research Council,
Washington, D.C.
Ramakrishna, Y., Ramakrishna P, Sivanandan R
(2006), Bus Travel Time Prediction Using Gps
Data, Map India.
Sen A., Liu N., Thakuriah P., Li J. (1991), Short-
Term Forecasting Of Link Travel Times: A
Preliminary Proposal. ADVANCE. Working
Paper Series, No. 7.
Thakuriah P., Sen A., Li J., Liu N., Koppelman F.S.,
Bhat C. (1992), Data Needs For Short Term
Link Travel Time Prediction, Urban
Transportation Center, University Of Illinois,
Chicago.
Van Lint J.W.C., Hoogendoorn S.P., Van Zuylen H.J.
(2002), Freeway Travel Time Prediction with
State Space Neural Networks. Transportation
Research Board, National Research Council,
Washington, D.C.
Vanajakshi L., Rilett L.R. (2004), A Comparison Of
The Performance Of Artificial Neural Networks
And Support Vector Machines For The
Prediction Of Vehicle Speed, IEEE Intelligent
Vehicles Symposium, Parma, Italy.
Vanajakshi L., Subramanian S.C., Sivanandan, R.
(2008), Travel Time Prediction Under
Heterogeneous Traffic Conditions Using Global
Positioning System Data From Buses. IET
Intelligent Transport Systems, 3(1), 1–9.

ISBN 978-81-921121-0-7
CORROSION MONITORING TECHNIQUES FOR EVALUATION OF

PERFORMANCE OF STEEL IN CONCRETE STRUCTURE
Jitu Kujur
Research Scholar, Department of Civil Engineering, IIT Delhi, New Delhi-110016, Email: jk08ranchi@gmail.com
B. Bhattacharjee
Professor, Department of Civil Engineering, IIT Delhi, New Delhi-110016, Email: bishwa@civil.iitd.ernet.in
ABSTRACT: Corrosion of steel in concrete is an electrochemical process. Electrochemical techniques have

now become important tools for study of rebar corrosion in laboratories together with their application to real
life structures. The reason for the frequent use of electrochemical techniques in practice is the increase number
of structures requiring an assessment of their residual health condition due to the premature deterioration caused
by reinforcement corrosion. In addition it is necessary to evaluate the short-term performance of different
structures to know their early age behavior by performing various corrosion tests at initial periods, for prognosis
of possible deterioration so as to estimate the projected service life. When the information regarding the
performance of RC structures against chloride or carbonation induced corrosion has to be obtained within a
restricted time, the application of various electrochemical tests has now become an important aspect of
durability studies in concrete. The commonly used techniques are potential measurement, polarization resistant
measurement, electrochemical impedance measurement etc.
Keywords: Corrosion, Potential, Polarization, Impedance
INTRODUCTION
Concrete is the mostly used construction material in Half cell potential measurement
the world due to its specially property such as lower Half cell potential measurement is commonly used
life cycle cost, mould ability , robustness and can be method for the routine inspection of corrosion activity
designed for a desired quality. Once concrete is used in reinforced concrete structures [4].However
in construction, it must show a minimum performance quantification and reliable prediction of the corrosion
level when exposed to the environment over a rate requires the use of other electrochemical
specified period of time, but it is not maintenance free techniques. The experimental set up for half cell
and requires regular maintenance. Corrosion of potential measurement is shown in the fig. 1. Couple
reinforcement is one of the main durability problems of anodic and cathodic reaction takes places on the
in RC structure which occurs due to carbonation or surface of the same corroding bar and there is a flow
presence of chloride at rebar level [1]. Passive layer of current from cathodic to anodic site within the bar.
formed on steel surface due to high alkalinity of the Corrosion activity and Half cell potential
concrete is destroyed due to presence of chloride at measurement guideline are given in ASTM C 876 [5].
rebar level or decrease in pH of concrete due to The half-cell potential is measure of electrode
carbonation, the atmospheric carbon dioxide reacts potential, represents the potential difference between
with hydration product of the cement forming calcium the steel and the adjacent concrete electrolyte and thus
carbonate then lowering of pH takes place. Once a property of the steel/concrete interface.
passive layer is destroyed corrosion of reinforcement
takes places. Product of corrosion occupy greater
volume than original material thereby exerting
pressure on surrounding concrete resulting in
cracking, spalling and delamination in RC structures
[1-3]. Durable and cost effective repairs can be
performed only on reliable information on the level
and rate of deterioration. Half cell potential
measurement is a powerful rapid and non destructive
technique both in condition assessment and in repair
work. Once information regarding the indication of
corrosion occurrence is obtained, the rate of corrosion
may be measured through the electrochemical
technique such as linear polarization technique and Fig. 1 Half-cell potential measurement
AC Impedance spectroscopy.
The half cell potential value is measured with
CORROSION MONITORING TECHNIQUES: reference to standard reference electrode. As per the
ASTM C 876, the probability of corrosion in terms of

potential with reference to copper–copper sulfate
electrode (Cu/CuSO4) is given as follows;
If the potentials over an area are more positive than -
200mV, then there is a greater than 90% probability
that no rebar corrosion is occurring in that area.
If the potential over an area are in the range of -
200mV to -350mV, then the corrosion activity of the
steel reinforcement in that area is uncertain.
If the potentials over an area are more negative than -
350mV, then there is a greater than 90% probability
that reinforcing steel corrosion is occurring in that
area.
Measurement of half-cell potential is comparatively
Fig. 2 Experimental setup for LPR measurement
simple. A reference electrode (e.g. saturated calomel
electrode) is used in measurement of half cell
Linear polarization technique test with IR
potential. ACM field machine is connected between
compensation is done with guard ring arrangement on
the corroding rebar and the reference electrode kept
the carbonated concrete prism specimen.
on the concrete surface as shown in fig 1. The factors
Experimental set up for LPR measurement is shown
affecting potential measurements are reference
in fig. 2. Concrete prism specimen is prepared as per
electrode position, Temperature, moisture content,
the guideline given in ASTM G109 [8], the size of
presence of cracks, high resistive layers of concrete
the concrete prism specimen is kept as
and stray currents etc [6].
300x30x150mm with two steel specimens embedded
at bottom with cover of 25mm and one steel specimen
Linear polarization resistance technique (LPR)
centrally at top with clear cover depth of 5mm.
Time taken after corrosion initiation to reach the
Conditioning and preparation of steel specimen is
defined serviceability limit is commonly termed as
done as per the guideline given in ASTM 109. The
propagation time. This propagation period mainly
polarized surface area of the reinforcing steel is taken
depend on the rate of corrosion of reinforcement in
to be that area lying under the circle intersecting the
concrete where as the availability of moisture and
midpoint between the two sensor electrodes [6]. Here
oxygen at the surface of the steel in concrete support
the confinement of the current is achieved to fixed
the corrosion process. The Linear Polarization
area of the reinforcing steel by the use of a second
Resistance (LPR) technique introduced by Stern and
auxiliary guard ring electrode surrounding the inner
Geary (1957) is very popular method of monitoring
auxiliary electrode. For linear polarization resistance
corrosion rate in laboratory as well as field application
measurement, the working electrode i.e. the steel
and mainly field applications are based on this
reinforcement in the prism was polarized to ±20mV
technique. In this technique, a three electrode
from the equilibrium potential at a scan rate of 0.1mV
configuration along with sensor electrode is used to
per second. Before performing the test, the conducting
determine the polarization resistance, Rp, and hence
sponge was wetted with soap solution and place on
the corrosion rate. In LPR measurement, the
the surface of the prism specimen to have proper
reinforcing steel is polarized by the application of a
electrical contact with the guard ring. Assembly of
small perturbation from the equilibrium potential
guard ring electrode, auxiliary electrode and reference
through an auxiliary electrode. Polarization resistant
electrode is then placed above the wetted sponge. The
(Rp) value is related to Icorr by means of a constant,
electrical connection necessary are made for working
known as Stern Geary constant.
electrode and guard ring to ACM field machine. The
Therefore
instrument is capable of processing the data and
B plotting the output automatically as supported by
I corr  --
RP software connected to PC. However the guard rings
(1) technique nowadays is well recognized and a number
Using B = 26 mV, the maximum error is 2. of commercial devices are available in the market.
Uncertainty in B value can involve in the Stern
formula [7]. AC Impedance technique
AC impedance spectroscopy also sometimes known
as the electrochemical impedance spectroscopy.
The principle involved in this technique is that an
electrochemical interface can be viewed as a
combination of passive electrical circuit elements
namely resistance and capacitance as shown in figure
3. The AC impedance technique has been widely used
for determination of corrosion rate of the steel
reinforcement in concrete [9, 10, and 11]. The AC

impedance technique involves the application of a only the concrete resistance, Rc. At very low
small-amplitude sinusoidal potential to the working frequency, Z” (ω) again disappears leaving a sum of
electrode at a number of discrete frequencies. At each Rc and Rp . Thus the radius of the semi circle is Rp/2.
one of these frequencies, the resulting current A typical experimentally obtained Nyquist plot for
waveform will exhibit a sinusoidal response that is out RC made with Portland Pozzolana cement, w/c
of phase with the applied potential signal by a certain ratio=0.45, and TMT tempcore steel is shown in fig.
amount and has current amplitude that is inversely 4.
proportional to the impedance of the interface.
ASTM G106 and ASTM G 3 explain about AC
Impedance measurement technique. Let Z1 be the
interface impedance.
1 1 j
  ---- (2)
Z1 R p X c
Rp , Xc, and C are polarization resistant, capacitance
and double layer capacitance respectively and Xc
=1/ωc
Then Impedance of the above electrical equivalent
circuit is given by the following expression;
Rp CR p2
Z ( )  Rc  j
(1   2 C 2 R p2 ) (1   2 C 2 R p2 )
Fig.4 Nyquist plot for PPC/0.45/TMT Tempcore
-(3) steel
Z  0 ( )  Rc  R P -- (4)
A guard ring electrode is used which consists of
Z   ( )  Rc -- (5) auxiliary electrode, reference electrode, guard ring
and sensor electrode
Rp Due to sophistication in measurement, AC impedance
Z ( )  a 
2 -- (6) spectroscopy technique is more frequently used in
laboratory studied rather than in field surveys. Further
often it is used as a powerful tool to understand the
behavior of the steel/concrete interface and provides
information about the mechanism of corrosion
reaction at the steel/concrete interface and the rebar
corrosion kinetics etc [9, 11]. In AC impedance
spectroscopy, not only is the equipment more
expensive and sophisticated, but the time needed to
obtain an Rp value is much longer than in the case of
DC techniques. This is due to, in the case of low
corrosion rates, the necessity to reach frequencies in
the range of 10-5 Hz or lower, which means a period
of several hours for obtaining a single value [63].
Tafel plot
Fig: 3 Equivalent circuits In this technique, a potential scan is applied to the
AC Impedance test is done on the carbonated concrete working electrode starting from the open circuit
prism specimen who is used in LPR test. Conditioning potential (corrosion potential Ecorr and extending to
of concrete specimen and experimental arrangement few milli volts (250mV) either in anodic or cathodic
are similar to the LPR test as shown in fig.2. The direction [12]. The applied potential forces the
difference is the technique used; a sinusoidal voltage specimen to assume a potential other than Ecorr. The
of amplitude 5mV was applied in the frequency range current measured in this case is the difference
of 0.01HZ to 30000 Hz is applied to all the between the anodic and cathodic currents and is a
specimens. The impedance behavior of an electrode non-zero quantity. The potential is plotted against the
may be expressed in Nyquist plots of Z” (ω) measured current, and the resulting potential current
(imaginary component) as a function of Z’ (ω) (real diagram is referred to as the Tafel plot as shown in
component) or in Bode plots of impedance and θ fig.5. This plot exhibits a linear region of the plot with
(phase angle) versus log (ω). The Nyquist plot shows Ecorr gives Icorr. Tafel constants are not required for
a semicircle with frequency increasing in a corrosion rate calculations in this technique, but these
counterclockwise direction. At very high frequency, can be used in the linear polarization technique for
the imaginary component Z” (ω) disappears, leaving

determination of the corrosion rate. This method may attack. The average corrosion rate may be obtained as
damage the reinforcing steel/concrete interface since a follows:
large over potential is applied.
K .W
Corrosion rate = --- (8)
A.T .D
Where K is a constant, T is the exposure time, A is the
surface area, W is the mass loss, and D is the density
of the corroding metal. No instantaneous corrosion
rates can be measured in this technique, but only a
mean during the period of test. Obviously, the
gravimetric method is the most accurate method to
quantify corrosion attack in laboratory experiments.
Although this method is very time-consuming and
only applicable to the laboratory studies, it is a useful
tool to check the accuracy of the electrochemical
techniques that are able to measure the corrosion rate
quantitatively, such as the Linear Polarization and AC
Impedance techniques.For the purpose of comparison,
using Faraday’s law, the corrosion rate in mm/year
obtained from gravimetric (mass loss) measurement
Fig: 5 Polarization curve showing Tafel slope was converted to corrosion current density (µA/cm2)
by assuming uniform corrosion occurred over the
In Tafel slop method, corrosion rate is determined by steel surface by the following equation [6]
substitution of Tafel slope values (βa and βc) into
Butler – Volmer equation given as;
  2.303   2.303  
i  icorr exp   a   exp  c  -
  a   c  0.00327 × ×
= (9)
- (7) ×
Main disadvantage of using Tafel’s extrapolation
technique is that it is an electrochemically destructive Where Icorr = corrosion current density in µA/cm2, a =
technique, as application of external activation atomic weight of iron, i.e. 55.84 amu, n = no. of
(voltage or current) may alter the electrochemical electrons exchanged in corrosion reaction, i.e. 2 for
behavior of the system significantly. Rebar condition iron, D = density of steel as above.
can be interpreted on the basis of corrosion current Gravimetric (mass loss) measurement is used as a
density, and following guidelines have been suggested destructive test, generally conducted in laboratory.
as in table 1 and are widely accepted for the rebar However it serves as the most reliable reference
condition assessment. method. It is a simple, although time consuming
technique for determination of corrosion rate.
Corrosion rate obtained by different non-destructive
electrochemical techniques and instruments differs
from each other [15]. According to Soleymani &
Ismail (2004) the corrosion rates by LPR technique
are 10-30% higher than those by Tafel plot technique.
In another study, on the comparison of the corrosion
rate obtained by AC impedance with previously
reported results by linear polarization and Tafel plot
technique, Ismail & Ohtsu (2006) observed that the
values obtained by AC impedance technique are
lower than Tafel technique by 5-20% while the values
Gravimetric mass loss method obtained by linear polarization technique are 20-30%
This technique is a destructive method, which consists higher than those by Tafel plot technique. The
in weighing the specimen bars before and after being corrosion rates determined by LPR with guard ring
introduced in the concrete to be tested .The detail test can be around one-seventh of those measured by LPR
procedures of preparing, cleaning, and evaluating without guard ring [15]. Pradhan &Bhattacharjee
corrosion test specimens are described in ASTM G1- (2009) have observed that the average variation in the
03 [14]. In simple way it is the difference in weight values of corrosion rate obtained by linear
(gravimetric loss) is a quantitative average of the polarization resistance (LPR) technique with guard
ring arrangement with those obtained by gravimetric
method is about ±6% and with those obtained by AC

impedance spectroscopy is 9%. Thus most reliable reinforcement in concrete exposed to chloride
method for corrosion rate is the gravimetric environments, American Society for Testing and
measurement and to improve upon the reliability of Materials, Philadelphia, PA, 1999.
non-destructive electrochemical tests which are Ismail, M. and Ohtsu, M. (2006). Corrosion rate of
simple and repetitive, their results must be compared ordinary and high-performance concrete
with those obtained through the gravimetric subjected to chloride attack by AC impedance
measurement. spectroscopy, Construction and Building
Materials, 20(7), 458–469.
CONCLUSION Andrade, C., and Alonso, C. (1996). Corrosion rate
Half-cell potential is suitable for routine inspection of monitoring in the laboratory and on-site,
rebar corrosion initiation. LPR technique is simple Construction and Building Materials, 10(5),
and most widely used in laboratory as well as in field 315-328.
application for measurement of rate of corrosion of Gu, G. P.; Beaudoin, J. J.; and Ramachandran, V. S.,
steel in concrete. AC Impedance is sophisticated (2001), “Techniques for corrosion investigation
technique and time consuming although it gives in reinforced concrete,” Handbook of Analytical
reliable result. Tafel’s extrapolation technique is an Techniques in concrete science and
electrochemically destructive technique, as Technology, V. S. Ramachandran, J. J.
application of external activation (voltage or current) Beaudoin, Noyes publications, New Jersey,
may alter the electrochemical behavior of the system 441-504.
significantly therefore special care is taken in Gelany, M. A. E. (2001). Short-term corrosion rate
adopting this technique. Corrosion rate obtained by measurement of OPC and HPC reinforced
different non-destructive electrochemical techniques concrete specimens by electrochemical
and instruments differs from each other. Gravimetric techniques, Materials and Structures, V.34(241),
techniques serves as the most reliable method can be 426-432.
used as reference method to study the performance of Revie, R. W. and Uhlig, H. H. (1985). Corrosion and
steel in concrete. corrosion control, third edition, John Wiley &
Sons, New York.
REFERENCE ASTM G1-03. (2003). Standard Practice for
Schiessl, P. (1988). Corrosion of steel in concrete,” Preparing, Cleaning, and Evaluating Corrosion
Report of Technical committee, 60-CSC Test Specimens, American Society for Testing
RILEM, Chapman and Hall, New York. and Materials, Philadelphia, PA.
Sarja, A. and Vesicari, E. (1996). Durability design of Liu, Y. and. Weyers, R. E. (2003). Comparison of
concrete structures,” Report of RILEM technical guarded and unguarded linear polarization CCD
committee 130-CSL, E & FN SPON. devices with weight loss measurements, Cement
Broomfield, J.P. (2006). Corrosion of steel in and Concrete Research, 33(7), 1093–1101.
concrete, Understanding, investigation and Ismail, M. E. and Soleymani, H. R. (2004).
repair, E & FN SPON; An imprint of Chapman Comparing corrosion measurement methods to
& Hall. assess the corrosion activity of laboratory OPC
Bhattacharjee, B. and Pradhan, B. (2009). Half cell and HPC concrete specimens, Cement and
potential as an indicator of chloride-Induced Concrete Research, 34(11), 2037–2044.
rebar corrosion Initiation in RC, Journal of
Materials in Civil Engineering, ASCE, 21(10),
543-552.
ASTM C876 (Reapproved 1999). Standards Test
Method for Half Cell Potentials of uncoated
Reinforcing Steel in Concrete, American
Society for Testing and Materials, Philadelphia,
PA.
Bhattacharjee, B. and Pradhan, B. (2009).
Performance evaluation of rebar in chloride
contaminated concrete by corrosion rate,
Construction and Building Materials, 23 (6),
2346–2356.
Bungey, J. H. and Millard, S. G. (1996). Testing of
Concrete in Structures, Published by Blackie
Academic & Professional, an imprint of
Chapman & Hall, ISBN 0-203-48783-4.
ASTM G109-99a (Reapproved 2005). Standard test
method for determining the effects of chemical
admixtures on the corrosion of embedded steel

ISBN 978-81-921121-0-7
ASSESSMENT OF CONCRETE STRENGTH USING FLYASH AND

RICE HUSK ASH
Satish D. Kene
Research Scholar, Dept. of Civil Engg., YCCE, Nagpur-10, Maharashtra, Email: satishkene2285@gmail.com
Pravin V. Domke
Research Scholar, Dept. of Civil Engg., YCCE, Nagpur-10, Maharashtra, Email: pravindomke@rediffmail.com
Sandesh D. Deshmukh
Research Scholar, Dept. of Civil Engg., YCCE, Nagpur-10, Maharashtra,
Email:sandesh_deshmukh31@rediffmail.com
R.S.Deotale
Assistant Professor, Dept. of Civil Engg., YCCE, Nagpur-10, Maharashtra, Email: rsdeotale1@rediffmail.com
ABSTRACT:In the ancient period, construction work was mostly carried out with help of mudstone from
industry. Fly ash is a by-product of burned coal from power station and rice husk ash is the by –product of
burned rice husk at higher temperature from paper plant. Considerable efforts are being taken worldwide to
utilise natural waste and bye –product as supplementary cementing materials to improve the properties of
cement concrete. Rice husk ash (RHA) and Fly ash (FA) is such materials. RHA is bye-product of paddy
industry. Rice husk ash is a highly reactive pozzolanic material produced by controlled burning of rice husk. FA
is finely divided produced by coal-fired power station. Fly ash possesses pozzolonic properties similar to
naturally occurring pozzolonic material. The detailed experimental investigation done to study the effect of
partial replacement of cement with RHA and FA on concrete. In this paperI started proportion form 30% FA and
0% RHA mix together in concrete by replacement of cement ,last proportion taken 0% FA and 30% RHA, with
gradual increase of RHA by 1% and simultaneously gradual decrease of FA by 1%. It is observed that though
the strength of RHA concrete goes on decreasing after the 15% addition of RHA, the composition of 10% RHA
+ 20% FA gives maximum strength results as well as shows the potential to be used as useful material for
different building materials
Keywords – Admixture, Concrete, Compressive Strength, Fly Ash, Multiple Regression Analysis, Rice husk
Ash.
INTRODUCTION wet concrete are carried out to ensure uniform quality

Concrete as is well known is a heterogeneous mix of concrete only. Strength is not a measurable at that
cement, water and aggregates. The admixtures may stage with the available technology. Therefore the
be added in concrete in order to enhance some of the concrete samples are to be cured for 28 days in
properties desired specially. In its simplest form, normal method to arrive at the compressive strength
concrete is a mixture of paste and aggregates. Various and for necessary follow up action. It is not only
materials are added such as fly ash, rice husk, difficult to dismantle the suspected portion of
admixture to obtain concrete of desired property. The concrete at such a stage but also expensive in terms of
character of the concrete is determined by quality of time and money. Predicting the strength at the
the paste. The key to achieving a strong, durable manufacturing stage, however, is yet to receive due
concrete rests in the careful proportioning, mixing and attention of engineers. Hence, any new approach that
compacting of the ingredients. The detailed is capable of predicting reliably the compressive
experimental investigation done to study the effect of strength of hardened concrete based on the properties
partial replacement of cement with RHA and FA on of the ingredients and the wet concrete will be helpful
cement. In this project I started proportion form 30% to practicing engineers. Besides, such tests could be
FA and 0% RHA mix together in concrete by performed with the same ease as the workability tests.
replacement of cement ,last proportion taken 0%FA RHA has two roles in concrete manufacture, as a
and 30% RHA. Numerous tests are performed on wet substitute for cement, reducing the cost and weight of
concrete such as workability tests such as compaction concrete in the production of low cost building
factor test and slump test. The tests on hardened blocks. The workability of RHA concrete has been
concrete are destructive test while the destructive test found to decrease but FA increases the workability of
includes compressive test on concrete cube for size concrete so RHA and FA mix together in concrete to
(150 x 150 x 150) mm, Flexural strength on concrete improve the workability of concrete. The work
beam (500 x 100 x100) and split tensile strength on presented in thispaper reports an investigation on the
concrete cylinder (150 mm ø x 300mm) as per IS: 516 behavior of concrete produced from blending cement
– 1959, IS: 5816 – 1999 and IS: 516 – 1959 with FA and RHA.
respectively.In actual practice, test on workability of

Regression analysis method for predicting the  To provide economical construction material.
7,14,28,90 day’s compressive strength of concrete is  Provide safeguard to the environment by utilizing
presented in this project. The proposed method is waste properly.
aimed at establishing a predictive relationship
between properties and proportions of ingredients of Materials and Methods
concrete, compaction factor, weight of concrete cubes
and strength of concrete The work presented in this paper reports an
investigation on the behaviour of concrete produced
The objectives and scope of present study are. from blending cement with RHA and FA. The
physical and chemical properties of RHA, FA and
 To find the optimum mix design with regards to OPC were first investigated. Mixture proportioning
the amount of water, RHA, FA and cement ratio. was performed to produce high workability concrete
 To investigate the physical properties of the RHA (200- 240 mm slump) with target strength of 32.1
and FA– density (lightweight), strength (bending Mpa (M25) for the control mixture. The effect of
and compression), water absorption and moisture RHA on concrete properties was studied by means of
content. the fresh properties of concrete and the mechanical
 To study the relative strength development with properties. I.e. Compressive strength, tensile splitting
age of (RHA + FA) concrete with control strength, flexural test was studied as the time
concrete. dependent property.
 Use of industrial waste in a useful manner.
 To conduct compression test on (RHA+FA) and A. Cement
control concrete on standard IS specimen size
(150 x 150 x 150) mm. The cement used was Ordinary Portland cement (43
 To conduct Flexural test on (RHA+FA) and Grade) with a specific gravity of 3.15. Initial and final
control concrete on standard IS specimen size setting time of the cement was 50 min and 365 min,
(100 x 100 x 500) mm. respectively. Its chemical composition is given in
 To conduct split tensile test on (RHA+FA) and Table 1.
control concrete on standard IS specimen size
(150 mm ø x 300mm) mm.
Table 1: Following are the Chemical properties of cement (OPC), Fly ash and Rice husk ash
Materials SiO2 Al2O3 Fe2O3 CaO MgO LOI SO3 K2O Na2O3
Cement 19.71 5.20 3.73 62.91 2.54 0.96 2.72 0.90 0.25
Fly ash 40 25 6 20 3.71 3.0 1.74 0.80 0.96
Rice husk 78.21 ( SiO2+ Al2O3+ Fe2O3) 0.99 4.89 -------- -------- --------- ---------
ash =82.64
B. Rice Husk Ash shown that RHA resulting from the burning of rice
husks at control temperatures have physical and
Rice husk ash used was obtained from Ellora Paper chemical properties that meet ASTM (American
Plant located in Tumsar Bhandara .The Specific Society for Testing and Materials).Standard C 618-
gravity of rice husk ash is 2.10 and bulk density is 94a. Studies have shown that to obtain the required
0.781 g/cc RHA, produced after burning of Rice husk particle size, the RHA needs to be grown to size 45
(RH) has high reactivity and pozzolanic property. µm – 10 µm.
Indian Standard code of practice for plain and
reinforced concrete, IS 456- 2000, recommends use of C. Fly Ash
RHA in concrete but does not specify quantities.
Chemical compositions of RHA are affected due to Fly ash used was obtained Koradi Power Plant
burning process and temperature. Silica content in the Nagpur. Fly ash is one of the residues generated in the
ash increases with higher the burning temperature. As combustion of coal. Fly ash is generally captured
per study by Houston, D. F. (1972) RHA produced by from the chimneys of power generation facilities,
burning rice husk between 600 and 700°C whereas bottom ash is, as the name suggests, removed
temperatures for 2 hours, contains 90-95% SiO2, 1- from the bottom of the furnace.
3% K2O and < 5% unburnt carbon. Under controlled
burning condition in industrial furnace, Studies have

In the past, fly ash was generally released into the Good quality river sand was used as a fine aggregate.
atmosphere via the smoke stack, but pollution control The fineness modulus, specific gravity and dry
equipment mandated in recent decades now require density are 2.32, 2.68 and 1690 kg/m3. Coarse
that it be captured prior to release. It is generally aggregate passing through 20mm and retained 10mm
stored on site at most US electric power generation sieve was used.Its specific gravity and dry density
facilities. Depending upon the source and makeup of was 2.7 and 1550 kg/m3.
the coal being burned, the components of the fly ash
produced vary considerably, but all fly ash includes E. Chemical Admixture
substantial amounts of silica (silicon dioxide, SiO2)
(both amorphous and crystalline) and lime (calcium A commercial AC- Green Slump-GS-02 black cat
oxide, (CaO). Fly ash is commonly used to Chemical Limited and Glenium- AG-30 JP – BASF
supplement Portland cement in concrete production, Const. Chemical Limited type hyper plasticizer was
where it can bring both technological and economic used to maintain the workability of fresh concrete.
benefits, and is increasingly finding use in synthesis The dosage of hyper plasticizer was kept constant in
of geopolymers and zeolites. mass basis; it was 1%-1.6% of cement weight. The
aim of keeping the amount of plasticizer constant is to
D. Aggregate neglect, if any, the influence of plasticizer on the
properties of hardened concrete.
EXPERIMENTAL PROGRAMME D. Mixture Proportioning
Experimental programme comprises of test on The mixture proportioning was done according the
cement, RHA, FA, cement concrete with partial Indian Standard Recommended Method IS 10262-
replacement of cement with RHA and FA. 1982.The target mean strength was 32.1 Mpa for the
OPC control mixture, the total binder content was
A. RICE HUSK ASH 435.45 kg/m ,fine aggregate is taken 476kg/m and
coarse aggregate is taken 1242.62kg/m the water to
1) Normal Consistency = 17% binder ratio was kept constant as 0.44, the
2) Initial and Final Setting time = 195min. Superplasticizer content was varied to maintain a
and 265min. slump of (200-240 mm) for all mixtures. The total
3) Compressive Strength = 11 N/mm2 mixing time was 5 minutes, the samples were then
4) Specific Gravity = 2.09 casted and left for 24 hrs before demoulding They
B. ORDINARY PORTLAND CEMENT were then placed in the curing tank until the day of
testing Cement, sand, Fly ash, Rice husk ash and fine
OPC 43 grade cement is used for this whole and coarse aggregate were properly mixed together in
experimental study. The physical test results on OPC accordance with British Standard Code of Practice
are as follows. (BS 8110)19 in the ratio 1:1.1:2.85 by weight before
water was added and was properly mixed together to
1) Normal consistency = 22%
achieve homogenous material. Water absorption
2) Initial Setting time = 30 min.
capacity and moisture content were taken into
3) Final Setting Time = 10 hrs.
consideration and appropriately subtracted from the
4) Specific Gravity = 3.15
water/cement ratio used for mixing. Muthadhi et al. 9
C. TEST ON CONCRETE
reported the blending of rice husk ash (RHA) in
An M25 mix is designed as per guidelines in IS cement is recommended in most international building
10262, 1982 based on the preliminary studies codes now. Hence, cement was replaced in
conducted in the constituent materials. Tests on fresh percentages of 0, 1,2,3,4, 5 up to 30% with rice husk
concrete are obtained as follows. ash and fly ash and 150 × 150 × 150mm3, Beam and
Cylinder moulds were used for casting. Compaction
1) Slump Test=55mm of concrete in three layers with 25 strokes of 16 mm
2) Vee-Bee = 13sec. rod was carried out for each layer. The concrete was
3) Compaction factor =0.95 left in the mould and allowed to set for 24 hours
4) Flow Test =78 %. before the cubes were de moulded and placed in
curing tank. The concrete cubes were cured in the
tank for 7, 14, 28 and 90 days.
E. Testing methods 516 – 1959 ,the testing done for flexural strength of
beam as per IS : 5816 - 1999 and the testing done for
Testing is done as per following IS code. The testing split tensile strength of cylinder as per IS : 516 –
done for compressive strength of cubes as per IS : 1959.

Table No. 2 Table for compressive strength, Flexural strength, Split tensile strength
MIX STRENGTH AFTER CURING IN DAYS in N/mm2
MIX PROPORTION FLEXUR SPLIT

SR. 14 AL TENSILE
7DAYS 28 DAYS 90 DAYS
NO. RICE H. ASH DAYS STRENGT STRENGT
FLYASH BY
BY H N/mm² H N/mm²
% CEMENT
%CEMENT
1 Control mix 24.56 26.78 40.52 45.21 10.58 6.5
2 30 0 19.11 24.52 36 40.44 10.58 3.4
3 29 1 30.66 35.86 33.77 36.00 10 3.25
4 28 2 20 25.66 37.77 42.67 11 3.39
5 27 3 22.22 26.32 36 40.89 11.25 3.67
6 26 4 35.11 43.55 44 48.89 12 3.67
7 25 5 43.55 44 45 53.33 13.75 3.53
8 24 6 42.66 44 44 49.78 12.5 3.4
9 23 7 35.55 42.22 44.44 53.33 11.25 3.53
10 22 8 40.44 42.66 44.03 51.56 10.98 3.58
11 21 9 33.77 40.88 42.66 52.44 8.5 3.67
12 20 10 26.66 28.44 28.88 35.11 9.2 3.25
13 19 11 25.77 35.11 35.55 40.89 8.32 3.11
14 18 12 26.22 35.55 36 40.89 7.5 3.53
15 17 13 28.44 32 35.55 43.11 7.0 3.25
16 16 14 33.33 36 40 42.22 6.3 3.21
17 15 15 31.11 33.77 34.22 39.56 6.2 3.11
18 14 16 25.77 28 32 39.56 5.8 2.82
19 13 17 27.11 29.33 33.33 43.56 5.7 2.68
20 12 18 26.66 29.33 29.77 34.67 5.4 2.97
21 11 19 25.77 28.88 31.11 36.00 5.5 2.94
22 10 20 27.11 29.77 30.66 33.78 5.25 2.91
23 9 21 25.33 27.55 28.88 34.67 4.25 2.85
24 8 22 28.44 32 37.77 38.62 4.12 2.68
25 7 23 26.22 28.88 29.77 33.78 4.08 2.82
26 6 24 24.88 27.55 30.22 34.67 5.0 2.81
27 5 25 23.55 24.88 25.33 30.67 4.36 2.81
28 4 26 24.88 26.22 29.33 31.56 4.08 2.82
29 3 27 29.33 31.55 32.44 38.22 3.85 2.68
30 2 28 18.22 20.44 23.11 26.22 3.65 2.54
31 1 29 19.55 20.44 21.33 25.78 3.65 2.54
32 0 30 18.33 20 20.44 25.78 3.45 2.26

Fig. 1 Compressive Strength of Cubes at 90 Day Fig.2 Compressive Strength of Cubes at 28 Day
Fig. 4 Compressive Strength of Cubes at 7 Day

Fig. 3 Compressive Strength of Cubes at 14 Day
Fig. 6 Split Tensile Strength of Cylinders at 28 Days

Fig. 5 Flexural Strength of Beams at 28 Day
Fig.8 Relation between Deflection in mm and

Fig. 7 Relation between Workability of concrete and %
% of (FA+ RHA)
of (FA+ RHA)

9 28.14 29.77 -1.63
METHODOLOGY OF INVESTIGATION 10 27.31 30.22 -2.91

The data of 30 trials of mixes are used for the 11 25.42 25.33 0.09
analysis. From the data of 30 trials with adding
plasticizer equations are developed ‘Multiple 12 26.13 29.33 -3.2
Regression and correlation Analysis’ was applied to 13 30.55 32.44 -1.89
derive the equations. In Multiple Regression Analysis,
various formulae were developed, by varying the 14 22.67 23.11 -0.44
input parameters to predict the 7, 14, 28 and 90 days
strength of concrete cube. Selection of following
15 22.37 21.33 1.04
equations with different inputs, which would help the 16 22.11 20.44 1.67
user to predict the strength of concrete cube with
available data / input parameters, is based on the Table 4: Summary of actual strength, predicted
results of analysis and the validation of formula. strength with superplasticizer 28 days’ strength using
four variables
Equation with plasticizer
1) 28 Days’ Strength Using Four Variables: (Sample ACTUAL
No. 1-15) PREDICTED
SR STRENGT
STR28 = -0.170 x RHA + 74.364 x AD.WT. – 0.617 STRENGTH DIFF.
x AD.QU. – 1.202 x STR7 + 2.192 x STR14...... (1) NO H
in N/mm2
2) 28 Days’ Strength Using Four Variables: (Sample in N/mm2
No. 16-31) 1 30.77 36 -5.23
STR14 = -0.111 x RHA + 2.245 x 10-3 x AD.QU. + 2 41.75 39.77 1.98
0.642 x STR7 + 0.453 x STR28...... (2)
3 32.20 37.77 -5.57
Table 3 Summary of actual strength, predicted 4 33.15 36 -2.85
strength with superplasticizer 28 days’ strength using
four variables. 5 44.26 44 0.26
PREDICTED ACTUAL
SR 6 40.12 45 -4.88
STRENGTH STRENGTH DIFF.
NO 7 38.50 44 -5.5
in N/mm2 in N/mm2
8 40.30 44.44 -4.14
1 32.67 34.22 -1.55
9 36.22 44 -7.78
2 27.56 32 -4.44
10 39.16 44 -4.84
3 28.23 33.33 -5.1
11 27.29 28.88 -1.59
4 28.40 29.77 -1.37
12 34.62 35.55 -0.93
5 28.06 31.11 -3.05
13 35.88 40 -4.12
6 28.60 30.66 -2.06
14 31.26 35.55 -4.29
7 27.41 28.88 -1.47
15 32.98 36 -3.02
8 33.28 37.77 -4.49
CONCLUSIONS
Based on the results presented above, the percentage of RHA. Because RHA is highly
following conclusions can be drawn: porous material.
1) Compressive strength increases with the 3) The workability of RHA concrete has been
increase in the percentage of Fly ash and found to decrease with increase in RHA
Rice Husk Ash up to replacement (21%FA replacement.
and 9% RHA) of Cement in Concrete for 4) It was found that rice husk when burned
different mix proportions. produced amount of silica (more than 80%).
2) Concrete requires approximate increase in For this reason it provides excellent thermal
water cement ratio due to increase in insulation.

5) Rice husk ash contains more silica, and aggregate and normal weight concrete.
hence we prefer rice husk ash use in concrete However, the work can be extended to the
than silica fume to increase the strength. concrete of light weight and heavy density
6) Through Rice husk ash is harmful for human 12) The maximum 90 days compressive strength
being, but the cost of rice husk ash is zero was obtained with 23% fly ash 7% rice husk
and thus we prefer RHA use in concrete as ash mix.
compared to silica fumes. 13) The maximum 28 days split tensile strength
7) The workability of RHA concrete has been was obtained with 25% fly ash 5% rice husk
found to decrease but FA increases the ash mix.
workability of concrete so RHA and FA mix 14) The maximum 28 days flexural strength was
together in concrete to improve the obtained again with 25% fly ash and 5% rice
workability of concrete. husk ash mix.
8) Rice Husk Ash can be used with admixtures, 15) The transition zone gets improved and
plasticizers, and super plasticizers, for densified with the use of ternary mix
increasing the strength of concrete with concretes containing rice husk ash and fly
partial replacement of cement. ash.
9) In the presented work many factors are 16) Due to the high specific surface area of the
considered, which are believed to affect RHA, the dosage of superplasticizer had to
strength of concrete. MULTIPLE be increased along with RHA fineness to
REGRESSION ANALYSIS is effectively maintain the desired workability.
used as a predictive tool. 17) The mechanical properties in terms of
10) Regression analysis as is well-known, gives flexural and tensile strength have been
explicit formula, which we can be directly significantly improved with the addition of
used to predict the strength of concrete. RHA.
Prediction of strength of concrete cube with 18) Rate analysis shows that as the percentage of
regression analysis is easy and handy. The RHA and FA added on the concrete the cost
formulae can predict only 7 days, 14 days, 28 goes decrease up to 29%.
days and 90 days strength of the same 19) RHA when added in the concrete reduces the
concrete cube. weight of the concrete up to 15% after 90
11) Present work is aimed at developing days of curing.
predictive tool with respect to normal density
REFERENCES Standards, New Delhi.

D.V. Reddy, Ph. D, P.E. Professor and Director of Shetty M. S. "Concrete Technology" S. Chand & Co.Ltd,
the Centre for Marine Structures & 1997.
Geotechnique P.E, Department of Civil Alireza Naji Givi , Suraya Abdul Rashid , Farah Nora
Engineering, Florida Atlantic University and A. Aziz , Mohamad Amran Mohd Salleh
Marcelina Alvarez, B.S. Structural Engineer, developed “Contribution of Rice Husk Ash to
URS Corporation, Boca Raton developed the Properties of Mortar and Concrete:A
“Marine Durability Characteristics of Rice Review”.
Husk Ash Modified Reinforced Concrete”. Badorul Hisham Abu Bakar, Ramadhansyah
“Rice Husk Ash – Properties and its Uses: A PutrajayaC and Hamidi Abdulaziz developed
Review” by A. Muthadhi, R. Anita and Dr. S. “Malaysian Rice Husk Ash – Improving the
Kothandaraman . Durabilityand Corrosion Resistance of Concrete:
G. A. Habeeb, M. M. Fayyadh Department of Civil Pre-review”.
Engineering, Faculty of Engineering, University A dimensional analysis method for predicting the 28-
of Malaya, Malaysia developed “Rice Husk day compressive strength of 53-grade cements is
Ash Concrete: the Effect of RHA Average presented in the paper by Dr. D. R. Pathak, N. P.
Particle Size on Mechanical Properties and Deshpande (J. Mat. in Civ. Engrg.
Drying Shrinkage”. (November/December 2005).
International Conference, Waste and Byproducts as Dao Van Dong- Doctor, Pham Duy Huu- Professor,
Secondary Resources for Building Materials, Nguyen Ngoc Lan- Engineer developed “Effect
13-16 April, 1999, New Delhi, India. of rice husk ash on properties of high strength
IS 10262 -1981 “IS Method of Mix Design”, Bureau concrete”.
of Indian Standards, New Delhi Dr Kedar Mohan Paturkar developed “Effect of
IS 516 -1959 “Methods of Tests for strength of cement replacement by silica fume and fly ash
concrete”, Bureau of Indian Standards, New
Delhi Books
IS 456 -2000 “Code of Practice for Plain and M. S. Shetty, Concrete Technology, 5th ed., S.Chand
Reinforced Concrete”, Bureau of Indian & Co.Ltd, 2002.

ISBN 978-81-921121-0-7
PILE FOUNDATIONS DESIGN AND CONSTRUCTION
M. Sujan
Junior Engineer (Civil), Engineering section, The English and Foreign Languages University, Hyderabad 500 007,
Email:sujan_be@rediffmail.com
ABSTRACT: When a good bearing stratum is not available near the ground surface or at relatively shallow
depths, the loads of super structures have to be 'transmitted" to firm strata capable or catering such loads though
may be at much depth. Such foundations made are called Deep foundations. Thus deep foundations are those
foundations where the soil support is found at appreciable depth below the main structure. The deep foundations
may be Pile foundations or wells and caissons. A pile is relatively small diameter shaft which is driven into the
ground or otherwise introduced into the soil by suitable means so as to support all the loads. Wells and caissons
are usually installed by excavation of sub-soil. Thus in case of wells and caissons, visual inspection of the firm
stratum on which they rest, may be made.
INTRODUCTION METHODOLOGY
A shallow foundation is usually provided when the Classification of Piles
soil at a shallow depth has adequate capacity to Piles may be classified in a number of ways based on
support the load of the superstructure. However, in different criteria:
situations where the soil at shallow depths is poor, in a) Function or action
order to transmit the load safely, the depth of b) Composition and material
foundation has to be increased till a suitable soil c) Instillation
stratum is met. In view of increased depth, such
foundations are called deep foundations. The deep Classification Based On Function or Action
foundation is recognized has pile foundation. pile
a) End-Bearing Piles: used to transfer the pile
foundations are usually adopted for the multi-storeyed
tip to a suitable bearing stratum, passing soft
building or for bridges across drains, etc. where
soil.
moderate to heavy loading are expected.
b) Frication piles: Used to transfer loads to a
depth in a frictional material by means of
STATEMENT OF THE PROBLEM
skin friction along the surface area of the
Necessity of Pile Foundatio
pile.
a) The pile foundation is used in the following
c) Tension: uplift piles are used to anchor
conditions:
structure subjected to uplift due to
b) When the strata at or just below the ground
hydrostatic pressure or to overturning
surface is highly compressible and very week to
moment due to horizontal forces.
support the load transmitted by the structure.
d) Compaction piles: used to compact loose
c) Pile foundation is required for the transmission of
granular soils in order to increase the bearing
the structural loads through deep water to a firm
capacity. Since they are not required to carry
stratum.
any load, the material may not be required to
d) Pile foundation is required when the soil
be strong. The pile tube, drive to compact the
condition are such that a wash out, erosion or
soil, is gradually taken out and sand is filled
scour of soil may occur from the underneath a
in its place thus forming a sand pile.
shallow foundation.
e) Anchor piles: used to provide anchorage
e) Piles are used for the foundation of some
against horizontal pull from sheet piling or
structures, such as transmission towers, off-shore
water.
plate forms, which are subjected to uplifts.
f) Fender piles: used to protect water front
f) Piles are used in the case of the expansive soils
structure against impact from ships or other
such as black cotton soils, which swell and shrink
floating objects.
as the water content changes piles are used to
g) Sheet Piles: Commonly used as bulkheads, or
tenser the load below the active zone.
cut-offs to reduce seepage and uplift in
Functions of Piles:- hydraulic structures.
The function of the piles is to transmit that load to the
soil of higher shearing strength at sufficient depth and
Classification Based On Material and Composition
act as compression or tension members when the a) Timber Piles: - These are made of timber of
structure is subjected to the wind and wave forces in sound quality. Length may be up to about 8
addition to vertical direct thrust. m; spacing is adopted for greater lengths.
Diameter may be from 30to 40cm. Timer
piles perform well either in fully dry

1
condition or submerged condition. Maximum Fig 1.0:- Uses of piles
design load is about 250 KN.
b) Steel Piles: - Steel Piles are generally either
in the form of thick pipe or rolled steel H-
sections. Pipe steel are driven into the ground
with their ends open or closed. Piles are
provided with a driving point or shoe at the
lower end.
c) Concrete Piles: - Cement Concrete is used in
the construction of the Concrete piles.
Concrete piles are either precast or cast in
situ.
d) Composite Piles: - Composite piles are made
of two materials. Composite pile may consist
of the lower portion of steel and upper
portion of cast in situ concrete. As it is
difficult to provide joint between two
dissimilar materials, composite piles are
rarely used in practice.
Classification Based on Instillation
1. Driven Piles: - Timber, steel or precast

concrete piles may be driven into portion
either vertically or at an inclination. Pile
hammers and piles driving equipment are Table: 21.0 Typical Lengths and Capacities of various
used for the driving piles. Pile Types
2. Cast in situ piles: - Only concrete piles can Approximate
Pile Length(M)
be cast in situ. Holes are drilled and these are design load (KN)
Pile Type
filled with the concrete. They may be straight Usual Usual
Maximum Maximum
bored piles or may be under reamed with one range range
or more bulbs at intervals. 150-
Timber 10-18 30 300
3. Driven and cast in situ piles: - These piles are 200
formed by driving a casing with a closed Driven
300-
bottom end into soil. The casing is later filled Recast 10-15 30 900
600
with concrete. The casing may or may not be Concrete
withdrawn. Driven
500-
prestressed 20-30 60 900
Use of Piles 600
Concrete
Cast in
The important ways in which piles are used are as 300-
situ 15-25 40 900
follows; 750
Concrete
Concrete
a) To carry the vertical compressive load, 45 600-
cast in situ 15-25 9000
b) To resist uplift or tension forces, Large dia 3000
bulb piles
c) To resist horizontal or inclined loads. 300-
Composite 20-40 60 2000
900
300- 2500-
Steel Pile 20-40 Unlimited
1000 10000
3.3 Pile Driving
The operation of forcing a pile into the ground is

known as pile driving. The oldest method and the
1
Source: C.Venkatramaiah,Geotechical Engineering, New Age
International Publishers, New Delhi 2008, P.654
2
Gopal Ranjan., A.S.R.Rao, ”Basic and Applied soil Mechanics”, New
Age International (p) Limited, Publishers, Page.553.

most widely used even today are by means of a d) Nature of expansive soil: - The expansive
hammer. The equipment used to lift the hammer and nature of the soil should be checked. This
allow falling on to the head of the pile is known as the can be checked by the differential free swell
pile driven. The pile hammers are the following types.
test
a) Drop hammer Table: 2.0 Degree of Expansiveness
b) Single acting hammer (steam or pneumatic)
Degree of
c) Double acting hammer (steam or pneumatic) DFS9%)
Expansiveness
d) Diesel hammer (internal combustion) Low < 20
e) Vibratory hammer
Moderate 20 -3705
High 37.5-50
Very high  50
Piles in Sand
The driven pile into sand derives most of its load

carrying capacity by bearing. A typical load
settlement curve for a pile embedded in sand is as
show in Figure-3.0. The settlement load is therefore
based on settlement value rather than from
considerations of shearing failure.
3
Fig 2.0:- Setup of Pile Driver
Fig 3.0:- Load settlement curve in sand
Data required for the Pile Design
Piles in Clay
a) Soil exploration data: - An extensive filed
The allowable load taken by a single pile in clay can
test should out before taking up the pile be determined by a static load test. The load
construction on the site. There should be settlement curve is show in figure4.0. The failure of
sufficient bore holes made on site to study of the soil occurs at points x. the allowable load Qa is
the soil properties on site. The laboratory obtained by applying a factor of safety to the failure
tests should also be carried out to study the load Qd.
permeability, shear test and cohesive Qd =Qa/F.S
properties of the soil.
A rough estimate of the load carrying capacity Qd of a
b) Depth of water table: -The location of the frication pile may be obtained from the following
water table should be carefully studied and relationship.
possible variations in the water table during Q
the rainy and dry seasons should be Where B = Diameter of the pile
considered in the design the pile length. L = Embedded Length of the pile
c) Chemical properties of soil: - The ingredients C = Unit cohesion of the soil
of the soil should be studied. The chemical For a single pile, C should be replaced by Ca
Properties of soil be evaluated and studied so Where Ca = m*c
The value of m ranges from 0.40 for stiff clays to 1.0
that there is no loss to piles after construction
for the soft clays
of them
3
Source: C.Venkatramaiah,Geotechical Engineering, New Age
International Publishers, New Delhi 2008, P.655

Figure 4.0:- Load Settlement curve of a pile in clay
CONCLUSION
a) A pile is a relatively small diameter shaft
which is driven or installed into the ground.
b) Piles are useful to transferring the load
through the poor soil or water to a suitable
bearing stratum by means of end bearing.
c) Piles are used to resist horizontal loads as in
the case of foundation for retaining walls,
bridge abutments and wharves.
d) Piles sometimes use to resist the uplift loads.
REFERENCES
C.Venkatramaiah,Geotechnical Engineering, New
International Publishers, Delhi Page No.651-656
Dr.K.R.Arora.”Soil Mechanics and foundation
Engineering:, Standard Publishers Distributors.,
PageNo.671-675.
Dr.B.C.Pumia., Ashok Kumar Jain., Arun Kumar
Jain, ”Soil Mechanics and Foundations”, Laxmi
Publicayions (p) LTD.
Gopal Ranjan., A.S.R.Rao, ”Basic and Applied soil
Mechanics”, New Age International (p) Limited,
Publishers, Page No.546-552.
Karuna Moy Ghosh.”Foundation Design in Pratice”,
PHI Learning Private Limited PageNo.136-140.
Satyendra Mittal, “Pile foundations design and
construction”, CBS Publishers & Distributors
PVT.LTD, Page No.1-26.

ISBN 978-81-921121-0-7
MODELING QUEUING & DISCHARGE BEHAVIOUR OF

HETEROGENEOUS TRAFFIC AT MANUALLY CONTROLLED
URBAN INTERSECTION
Ghadiyali G.A.
Highway/Traffic Engineer, SAI Consulting Engineers Pvt.Ltd., Ahmedabad-380054, Email: ghadiyaligaurang@gmail.com
Shah M.K.
Senior Engineer, MaRS Planning & Engineering Services Pvt. Ltd., Ahmedabad-380054, Email: shah.megha16@yahoo.com
Dr. Katti B.K.
Professor Emeritus, Department of Civil Engineering, Sardar Vallabhbhai National Institute of Technology, Surat, India
ABSTRACT: Industrialisation has enormously built pressure on urban agglomerations as it is one of the basic
roots of urbanisation. Urban areas are the worst sufferers wherein urban intersections as a part of the
transportation system have become most vulnerable nodes to affect traffic quality under heterogeneous
conditions. The traffic on urban roads in Indian metropolitan cities consists of motorized as well as non-
motorized vehicles. This traffic being heterogeneous in nature, the static and dynamic behaviour of each vehicle
is different. Also their interaction between each other at an intersection is quite different compared to that at a
midblock section. This has a significant influence on the performance of the intersection in terms of vehicle
handling capacity and discharge capacity. Owing to the fact that effect of mixed traffic at an urban intersection
is an important measure to assess the performance of the intersection, an attempt has been made in this paper to
analyse the queuing and discharge behaviour of traffic at a 4-legged urban intersection of Surat city. Vehicle
queuing analysis along with macro and micro level observations of discharge of vehicles has been done at an
approach with video graphic technique. Further, discharge rate of mixed traffic has been modelled which takes
into consideration different attributes such as share of 2-wheelers and 4-wheelers in total traffic, red time at an
approach and queued vehicles during red period. Also the formation and dissipation of vehicle queue during red
and green period has been respectively noted which reveals the behaviour of mixed traffic at an urban
intersection. From the analysis, it gets revealed that the effect of share of 2-wheelers in the queue at the
intersection is quite contradictory compared to that share of 4-wheelers. The influence of queued vehicles at an
approach during red period in the traffic is also quite noteworthy which is reflected in the model itself. This
forms a significant base for the development of discharge rate model. With increase in percentage 2-wheelers,
the discharge rates are expected to increase and thereby the consumption of green time of queued vehicles is
expected to decrease. In view of this, discharge modification factor regression model has been developed. The
methodology and model so developed forms a strong foundation in designing a signal system with varied
composition of traffic and different lane configurations.
INTRODUCTION percentages. Heterogeneity not only exists in terms of

Most of the metropolitan cities in India are facing vehicles but also in case of road users and pedestrians.
serious traffic problems such as congestion,
bottlenecks, reduction in speed, higher travel time, Intersections are an important part of an urban
increased pollution levels, etc. The developing cities roadway and they have a significant effect on the
are sufferers of these problems as they possess operation and performance of the traffic system.
diversified traffic characteristics. The characteristics Urban intersections are the most important nodes as
of traffic flow include speed of vehicles, their they connect various transport linkages in any urban
concentration, and density of flow and these are transportation system. These linkages contribute
governed by a variety of factors attributable to road traffic to intersections from different agglomerations
features, the vehicle performance characteristics and of the city wherein traffic gets accumulated at a single
road user behaviour. Traffic phenomena are complex spot. As a result, delay in travel time is experienced
and nonlinear, depending on the interactions of a large by vehicle users due to inefficient performance of the
number of vehicles. Due to the individual reactions of intersection, lack of traffic discipline, heterogeneity of
human drivers, vehicles do not interact simply traffic, inter-modal interaction of traffic, etc. Even the
following the laws of mechanics, but rather show discharge rates and the clearance rates of each traffic
phenomena of cluster formation and shock wave mode vary according to their acceleration rates. This
propagation,[citation needed] both forward and in turn increases the interaction between different
backward, depending on vehicle density in a given modes at the intersection and imparts delay to the
area. Most of the Indian cities (Metropolitan cities) overall traffic accumulated at an approach.
comprise of heterogeneous traffic composition, with
two-wheeler and four-wheeler as the leading Urbanization is a determinant as well as result of
economic development. It is a natural consequence of

economic changes that take place as a country zone of Surat and its approaches lead to major
develops. commercial and residential zones. Considerable
delays and substantial formation of queue takes place
An effort has been made in this paper to develop a at two major approaches of the intersection during
generalized model for discharge rate of traffic under morning peak and evening peak.
heterogeneous traffic environment to have a clear
understanding of the behaviour of each mode in the Fig 1 shows the bird view of intersection under study
traffic and its influence on the overall discharge rate. and Table 1 describes the geometric features of
approaches of Ghod-dod intersection.
NEED OF THE RESEARCH STUDY
It is necessary to understand the behavioural pattern
of each traffic mode as well as the mixed traffic at an
intersection, as the effect of each mode’s behaviour on GHODDOD INTERSECTION
mixed traffic is quite diverse & noteworthy. The
models regarding mixed traffic discharge rate
developed in the western countries are based on
homogeneity of traffic. PCU factors have been
developed for Indian conditions, but they only deal
with static characteristics of the vehicles. These
factors only convert different types of vehicles into an
equivalent passenger car unit, which does not deal
with the dynamic behaviour of the vehicle in the
traffic stream. Also, earlier studies do not take into
account the discharging pattern and the queue
building & dissipation pattern of mixed traffic at the
intersection.
Considering the above issues, an effort for developing

a discharge rate model based on heterogeneity of Fig 1 Bird View of Ghod-dod Intersection
traffic conditions at the intersection is indispensable
which encompasses most of the attributes representing Table 1 Approach Geometric Features of Ghod-dod
mixed traffic in a better way. intersection
Approach Geometric Features
OBJECTIVE OF THE RESEARCH STUDY Parle Point 4-lane divided with 9.15 m
The main objectives of the research study are as Approach carriageway width on both side and 1
follows; m wide median.
Bhatar 4-lane divided 8.5 m carriageway
- To understand the queuing phenomenon and Approach width on both side and 1 m wide
discharge behaviour of mixed traffic at the median.
intersection & Athwa 2 Lane undivided, 8.5 m carriageway
- To develop a model of discharge rate at a Approach width.
manually controlled urban intersection under Jogger’s 4-lane divided 9 m carriageway width
Indian heterogeneous traffic conditions. Park on both side with 1.2 m wide median.
Approach
STUDY AREA
Surat city has seen an unprecedented growth in last DATA COLLECTION & ANALYSIS
four decades recording one of the highest growth rates Study of traffic volume, traffic composition, turning
in the country and a 10-fold population rise. There are traffic, discharge rate of mixed traffic in different
about 125 intersections in Surat, out of which 35 phases of successive cycles at an approach, STOP &
intersections are provided with traffic signals and 90 GO time observations, etc. was done through video-
intersections are uncontrolled. Before selecting the graphic technique in morning and evening peak
study location, a brief observation of various urban periods. Macro as well as micro level discharge rate
intersections was done in which some intersections study and queuing phenomenon behavioural study has
were provided with traffic signals and some were been done to have a correct idea about the discharge
controlled by traffic police. rate of mixed traffic at the intersection.
Ghod-dod intersection of Surat city has been selected Following Fig 2 shows the volume of traffic on all the
as the study intersection which is operated by police approaches of Ghod-dod intersection. The volume
during morning and evening peak hours. It is a 4- from Bhatar approach during morning as well as
legged urban intersection falling in the South-West evening peak periods dominates with a range of 1450-

1500 vph. Least traffic volume is recorded on Athwa
Approach.
The percentage of 2W is dominating with a range of

50-55 % on each approach, followed by 4W with 20-
30 % and 3W with 15-20 % contribution to the traffic.
The contribution of other modes of transport is almost
negligible. This can be seen in Table 2 below.
Total Traffic Volume Morning Period

1600
Evening Period
Volume (vehicles per 5 min)
1400
1200
1000
800
Fig 3 Morning Intersection Traffic Flow
600
400
200
0
Bhatar Parle Jogger’s Park Athwa
Time (min)
Fig 2 Traffic Volume Scenario at Ghod-dod

Intersection
Table 2 Summary of Traffic Composition at Ghod-

dod Intersection
Morning Peak %
Mini/Std.
Approach 2W 3W 4W LCV NMT
Bus
Bhatar 51.61 18.25 21.34 0.20 0.13 8.40
Parle Point 54.22 17.31 25.83 0.59 0.59 1.47
Jogger’s
54.79 16.70 26.56 0.29 0.29 1.37 Fig 4 Evening Intersection Traffic Flow
Park
Athwa 51.98 14.88 31.45 0.00 0.38 1.32
Evening Peak %
Mini/Std. VEHICLE QUEUING ANALYSIS
Approach 2W 3W 4W LCV NMT Vehicle queuing at an approach is a function of red
Bus
Bhatar 50.37 19.18 23.97 0.20 0.54 5.54 time given to that approach as well as the arrival rate.
Parle Point 51.93 23.02 20.89 0.19 0.48 3.19 Generally queuing of vehicles is a common
Jogger’s phenomenon observed at any urban intersection, but it
52.98 17.90 27.67 0.41 0.00 1.03 is a complex problem to analyse. It is of core
Park
importance to study the building of queue at the
Athwa 52.29 15.17 26.70 0.47 0.16 5.06
intersection and the dissipation of queue thereafter.
The building of queue leads to formation of queue
The turning movement survey at the site depicts that
length which ultimately results in increase of delay at
the majority of traffic flow is towards the Bhatar
the intersection. So, queuing phenomenon plays a
approach during morning and evening period. This is
major role in deciding the efficiency of an
observed for the reason that most of the commercial
intersection.
activities are concentrated on the ring road and also
Bhatar being part of old walled city, residential
Queue Building Phenomenon
pockets are also located.
Fig 5 indicates the number of vehicles queued in a red
period of 40 seconds. The red period is divided in
Following Fig 3 & Fig 4 provide the intersection
slots of 3 seconds and the number of vehicles joining
traffic flow during morning and evening peak hour at
the queue is determined from video graphic data. It
Ghod-Dod intersection.
can be seen that the number of vehicles joining the
queue is almost uniform.

Queue Building
45 Macro Level Observations
40 40
Cumulative no. of vehicles queued 40
38 Discharge rate is an important factor which plays a
34 35
35 32 vital role in deciding the overall efficiency of any
29 urban intersection. It has a major impact on the cycle
30
25 22 length, delay and the formation of queue on the
19 approach therein. The number of vehicles discharged
20 17
15 11
13 from an approach in the respective green times
10 7 provided is noted and the average discharge rate of the
4 mixed traffic is determined thereafter. This is done
5
0
through the extraction of data from the video-graphic
footage. The number of vehicles discharged from the
Bhatar approach is taken under consideration to find
Red Time Slots (sec) the average mixed traffic discharge rate.
Table 3 describes the green time, the traffic
Fig 5 Queue Building Phenomenon discharged during that green time and the computed
discharge rate of Bhatar approach.
Queue Decay Phenomenon
Queuing decay is a phenomenon in which the number Table 3 Bhatar Approach Discharge Rate
of queued vehicles dissipates with respect to time. It is Calculations
a general phenomenon that the queued vehicles start Traffic
clearing off the approach as soon as the green is Green Discharge
Cycle discharged
indicated. The number of queued vehicles starts Time Rate
(No) during green
clearing off in green period. Not the entire green (sec) (veh/sec)
(sec)
period is utilized by the queued vehicles, provided the
approach volume is less. If the flow reaches the 1 18 58 3.22
saturation flow, the green time is not sufficient to 2 19 53 2.79
clear off the queued vehicles and hence queue builds 3 16 61 3.81
up in an increasing manner. Following graphical 4 18 63 3.5
representation Fig 6 represents the decay of queued 5 20 79 3.95
vehicles for a given number of queued vehicles of 40 6 29 86 2.97
and green time of 18 seconds. It is observed that the 7 18 68 3.78
vehicles which are queued during red period clear off
8 18 55 3.06
the approach within 11 seconds, whereas the rest of
9 20 67 3.35
the green period is utilized by vehicles which are
joining later. The maximum decay rate is of 4.67 per 3 10 14 43 3.07
seconds during 6-9 seconds. Queue decay rate varies
from 2.33 to 3. Average queue decay rate is 3 veh./sec
for 2 lanes.
Fig 7 Discharge Rate scenario on Bhatar Approach

Fig 6 Queue Decay Phenomenon
VEHICLE DISCHARGE STUDY

The maximum and minimum discharge rates observed
Discharged Vehicles for G = 20 sec
on this approach are 3.95 and 2.79 vehicles per second
respectively. This discharge rate is calculated for an 16 15
14 Intermediate Green Time
effective width of 8.5 m. It can be seen that the
Frequency of Vehicles Discharged

14
average discharge rate being 3.35 vehicles per second
12 11 11
and the overall discharge rate lies within the range of
10
2.80 to 4.00 vehicles per second. 8
8
Micro Level Observations 6 5
Also the discharge of vehicles is observed for three 4 3
different values of green time. For this purpose out of 2

the ten cycles on the Bhatar approach, three cycles 0
namely the maximum green time, minimum green 0-3 3-6 6-9 9-12 12-15 15-18 18-20
time and an intermediate green time is selected to Green Time- Slots of 3 sec
observe the behaviour of discharged traffic.
Fig 8 depicts the number of vehicles discharging in Fig 9 Discharged vehicles in slots of 3 seconds for G
every three seconds time for a green time of 18 = 20 seconds
seconds. It is observed that the maximum number of
vehicles discharging is 14 whether the minimum
number of vehicles discharging is 5.
Discharged Vehicles for G = 29 sec
Frequency of vehicles discharged 14 13
Discharged Vehicles for G = 18 sec 12
Frequency of Vehicles Discharged
12 Maximum Green Time

16 10 10 10
14 Minimum Green Time
13 10
14 8 8
8 7
12 10
9 5
10 6
7
8 4 3
5
6
2
4
0
2
0-3 3-6 6-9 9-12 12-15 15-18 18-21 21-24 24-27 27-29
0
0-3 3-6 6-9 9-12 12-15 15-18 Green Time -Slots of 3 sec
Green Time -Slots of 3 sec
Fig 10 Discharged vehicles in slots of 3 seconds for G
Fig 8 Discharged vehicles in slots of 3 seconds for G = 29 seconds
= 18 seconds
MODELING OF MIXED TRAFFIC
It can be clearly seen that the number of vehicles DISCHARGE RATE
discharged increases gradually from the initiation of Based on the data collected, a discharge rate model
green time up to a certain period. Then after it has been developed considering effect of various
stabilizes and the number of discharging vehicles attributes on the mixed traffic discharge rate.
drops at the end of green time. Regression modeling technique is used to develop this
model. Various attributes like the arrival rate of
Similarly, Fig 9 states the number of vehicles vehicles, the red time given to an approach and the
discharged for a green time of 20 seconds. It can be percentage of two wheelers and four wheelers are
seen that the maximum and minimum number of found to have a significant effect on the mixed traffic
vehicles discharged in a time of 3 seconds is 15 and 3 discharge rate. Bhatar approach is selected to develop
respectively. the model, wherein ten cycle timings are taken and
corresponding to these cycles, the vehicles discharged
In Fig 10 it can be seen that during the slot of 6-9 in the green time, the red time and the vehicles queued
seconds the discharged traffic suddenly drops. This is during this red are noted.
because of the influence of a standard bus in the
traffic during that time slot. The bus takes a longer Following is the discharge rate regression model
time to clear off the stop line, which in turn affected developed.
the vehicles queued behind it.
Y=1.111+0.009x1–0.0077x2+0.015x3– 0.0264x4
(1)

R2 = 0.812 dynamic effect of traffic mode and its individual
where, effect into the traffic mix at the intersection. Also one
Y = Mixed Traffic Discharge Rate (veh/sec/lane) to one interaction of each traffic mode with the other
x1= Share of two wheelers in mixed traffic (%) in the traffic mass is not reflected if mixed traffic is
x2 = Queued vehicles during red time converted into homogeneous traffic.
x3 = Red Time (sec) Owing to these conditions, the discharge rate model
x4 = Share of four wheelers in mixed traffic (%) so presented in this paper takes into account the effect
of traffic heterogeneity and fluctuations in traffic
The model is developed for a particular range of composition. The independent variables such as
composition of traffic modes and lane configuration. composition of two and four wheelers in mixed
The composition of two wheelers varies between 50- traffic, red period and vehicles queued in respective
60 %, that of four wheelers between 20-30 %, that of red times have a significant influence on mixed traffic
three wheelers between 15-25 %, and the rest of the discharge rate. This model can be applied to urban
share of traffic mode is composed of six wheelers and intersections of developing cities possessing
NMT. Also the model is developed based on the lane heterogeneous traffic.
configuration of Bhatar approach, i.e. it is two lane The discharge rate model for mixed traffic so
roads with effective carriageway width of 8.5 meters. developed can be further utilized to develop delay
model as a measure of performance of any urban
It can be interpreted from the above discharge rate intersection. It can also form a strong base for
model that the discharge rate of mixed traffic determination of optimum signal cycle lengths at
increases with increase in the percentage of two intersections.
wheelers in the composition, while discharge rate
decreases with rise in the share of four wheelers. ACKNOWLEDGMENTS
Increase in discharge rate because of two wheelers is I am grateful to my post graduate research guide Dr.
due to higher acceleration rates and smaller B.K. Katti for his invaluable contributions towards
dimensions. On the other hand, four wheelers posses this publication.
lower acceleration rates compared to two wheelers
which thereby reduce the overall discharge rate of REFERENCES
mixed traffic. As the number of queued vehicles Allsop R. E. (1971). Delay-Minimizing Settings for
increases, vehicles find difficult to clear the approach Fixed-Time Traffic Signals at a Single Road
in specific time due to one to one interaction in the Junction, J. Inst, Maths; Appl. 8, 164-185.
queue. Due to this vehicles consume more time to D S Reddy, Dr C S R K Prasad, (August 2008).
discharge from the approach, thereby reducing the Design and Evaluation of Isolated Traffic Signal
overall discharge rate of mixed traffic. As red time using New Minimum Delay Cycle Length
increases, queued vehicles increase. With increase in Equation, IE (I) Journal; CV Vol. 89.
queued vehicles, the joining vehicles in the queue also Ghadiyali G.A. & Dr. B.K. Katti, (July 2010).
increase. Because of rise in the joining queue vehicles Development of Fixed Time Signal Design Model
the discharge rate goes up. Under Heterogeneous Traffic Conditions. A
dissertation submitted to SVNIT, Surat.
The Coefficient of determination (R2) value of the Sam Yagar, (1975). Minimizing Delay at a Signalized
model is 0.812 (df = 9, α = 0.95), which indicates a Intersection for Time-Invariant Demand Rates,
strong relation between the attributes and the Transportation Research; Volume 9, 129-141,
independent variable. Also the correlation between the Pergamon press (Great Britain).
dependent attributes is developed and it proves that L.R. Kadiyali, Traffic Engineering & Transport
there is a very less relationship amongst them. The Planning, Khanna Publishers.
discharge rate model developed is further utilized to Greenshields B. D., Shapiro D. and Ericksen E. L.
calculate the delay at the intersection. Based on this (1947). Traffic Performance at Urban Street
delay, various delay models for have been developed Intersections. Technical Report-1, Yale Bureau of
for different lane configurations and different volume Highway.
levels. The discharge rate model so developed has M. Hossain, (Nov. 2000). Estimation of Saturation
been further amended for different composition of Flow at Signalised Intersections of Developing
traffic, lane configurations and right turning traffic by Cities: A Micro-Simulation Modeling Approach,
developing discharge modification factor. Transportation Research Part A: Policy & Practice,
Volume 35, Issue 2, Pages 123-141.
CONCLUSION
It has been a common practice to convert mixed
traffic volume into homogenous traffic adopting
certain PCU conversion factors at the intersection.
These factors only take into account the static
behaviour of traffic mode, but do not encounter

ISBN 978-81-921121-0-7
SENSITIVITY ANALYSIS OF SHALLOW FOUNDATION RESPONSES

Prishati Raychowdhury
Assistant Professor, Department of Civil Engineering, Indian Institute of Technology Kanpur, India, 208016, Email:
prishati@iitk.ac.in
Sumit Jindal
Former Graduate student, Department of Civil Engineering, Indian Institute of Technology Kanpur, India, 208016, Email:
jndlsmt@gmail.com
ABSTRACT: The responses of shallow foundation largely depend on the selection of soil and model
parameters. Since the response of foundation significantly controls the behavior of the overall structure
supported by it; and consequently several design decisions, it is extremely important to identify and characterize
the relevant parameters. Furthermore, it is also important to identify the sources and extent of uncertainty of soil
and model input parameters, along with the effect of their uncertainty on the shallow foundation response. This
work intends to investigate the effect of soil and model parameter uncertainty on the response of shallow
foundation-structure systems resting on dry dense sand. The soil-foundation-structure interface (SFSI) in this
study is modeled using Beam-on-Nonlinear-Winkler-Foundation (BNWF) concept, where the SFSI is assumed
to be an assembly of discrete, nonlinear elements composed of springs, dashpots and gap elements. The
sensitivity of both soil and model input parameters on shallow foundation responses, such as moment, shear,
settlement and rotation are investigated using First-Order Second-Moment (FOSM) analysis and Monte Carlo
simulation through Latin hypercube sampling technique. It is observed that soil and model parameters have
significant effect on the footing responses. The force demands (moment and shear) are largely sensitive to
friction angle (~60% relative variance); whereas the settlement is sensitive to friction angle, shear modulus and
Poisson's ratio (~30% relative variance for each parameter).
INTRODUCTION additional information. Epistemic uncertainty results

Uncertainty refers to situations in which the outcome from lack of information and shortcomings in
of an event or the value of a parameter may differ measurements and calculations Lacasse and Nadim
from true value of that parameter. Uncertainty plays (1996). Epistemic uncertainty can usually be reduced
an important role in response of the soil-foundation by acquisition of more information or improvements
systems. These uncertainties may arise from in measuring methods.
geotechnical and structural material properties, input
loadings and modeling methods. For a structure with LITERATURE REVIEW
substantial soil-foundation-structure-interaction In last few decades, significant research has been
(SFSI) effects, geotechnical uncertainties may play carried out for proper understanding the behavior of
crucial role in predicting the overall system response. structure due to uncertainty in soil parameters. In an
early work, Lumb (1966) showed that the soil
Most soils are naturally formed in many different parameters can be modeled as random variables
depositional environments; therefore it shows confirming to the Gaussian distribution within the
variation in their physical properties from point to framework of probability theory. Ronold and Bjerager
point. The soil properties exhibit variations even (1992) observed that the model uncertainties are an
within an apparently homogeneous soil profile. Basic important in reliability analysis for prediction of
soil parameters that control the strength and stiffness stresses, capacities, deformation etc., in structure and
of the soil-foundation system are cohesion, friction foundation systems. Chakraborty and Dey (1996)
angle, unit weight, shear modulus and Poisson’s ratio studied the stochastic structural responses considering
of soil. These soil parameters can be delineated using uncertainty in structural properties, soil properties and
deterministic or probabilistic models. Deterministic loadings using Monte Carlo simulation technique.
models use a single discrete descriptor for the Lutes et al. (2000) evaluated the response of a
parameter of interest, whereas probabilistic models seismically excited structural system with uncertain
define parameters by using discrete statistical soil and structural properties. Ray Chaudhuri S, Gupta
descriptor or probability distribution (density) (2002) investigated the variability in seismic response
function. Uncertainty in soil properties can be of secondary systems due to uncertain soil properties
formally grouped into aleatory and epistemic through a mode acceleration method. Foye et al.
uncertainty (Lacasse and Nadim, 1996) Aleatory (2006) described a thorough study for assessment of
uncertainty represents the natural randomness of a variable uncertainties by defining the probability
property and is a function of spatial variability of the density functions for uncertain design variables in
soil property. This type of uncertainty is inherent to load resistance factor design (LRFD). Na et al. (2008)
the variable and cannot be reduced or eliminated by investigated the effect of uncertainties of geotechnical

parameters on gravity type quay-wall in liquefiable pult = 0.5γKp Df2 (2)
condition using tornado diagram and First-Order
Second-Moment (FOSM) analysis. Raychowdhury where pult is the passive earth pressure per unit length
(2009) studied the effect of soil parameter uncertainty of footing and Kp the passive earth pressure
on seismic demand of low-rise steel building coefficient. For the t-x material, the sliding resistance
supported by shallow foundations on dense silty sand tult is determined using classical Mohr-Coulomb
with considering a set of 20 ground motions. failure criteria:
Raychowdhury and Hutchinson (2010) carried out the
sensitivity analysis of shallow foundation response to tult = Wg tanδ + cAf (3)
uncertain input parameters using simplified FOSM
and Tornado Diagram methods. where tult is the frictional resistance per unit area of
foundation, Wg the weight on the foundation from the
SCOPE OF THE STUDY structure, δ the angle of friction between foundation
This article focuses on studying the effect of and soil, which typically varies from 1/3 ϕ to 2/3 ϕ,
uncertainty in soil and model parameters on the Af the surface area of the foundation.
response of shallow foundation supported shearwall
building. To incorporate the nonlinearity at the soil- The vertical and lateral stiffness, Kv and Kh are
foundation interface, a Beam-on-Nonlinear-Winkler- calculated using expressions given by Gazetas (1991)
Foundation (BNWF) approach is adopted. The as follows:
uncertainty analysis is carried out using simplified
First-Order Second-Moment method and Latin
hypercube sampling technique.
(4)
MODELING OF SFSI
In this paper, a Beam-on-Nonlinear-Winkler- (5)
Foundation (BNWF) approach is used to model the In order to investigate the effect of uncertain input
nonlinear soil-structure-interaction of shallow parameters on the shallow foundation response, a
foundations subjected to lateral loads. The BNWF single shearwall structure supported by strip footing
model includes a system of closely spaced resting on dense dry sand of relative density 80% is
independent, mechanistic vertical and lateral elements considered for this study. The dimensions of footing
consisting of nonlinear springs, dashpots and gap are 1.0 m wide, 0.25 m in height and depth of
elements (Fig. 1). The vertical springs (q-z elements) embedment is 0.5 m. Shearwall dimensions are 0.5m
are intended to capture the axial and rotational in length, 0.2 m in width and the height of wall is 5.0
behavior of the footing, whereas the lateral springs, t- m. A monotonic loading is applied at the top of the
x element and p-x element are intended to capture the structure and responses are evaluated in terms of
sliding and passive resistance, respectively. The maximum absolute values of moment, shear, rotation
meterial models were originally developed by and settlement demands.
Boulanger et al. (1999) and modified by
Raychowdhury and Hutchinson (2008). This model is It is evident from the above discussion that the
capable of reasonably capturing the experimentally strength and stiffness of the spring elements of a
observed behavior for various shallow foundation particular size of footing are functions of basic soil
conditions. For more details regarding the BNWF properties such as friction angle, shear modulus,
modeling, one can look into Raychowdhury and Poisson's ratio and unit weight. Therefore, these soil
Hutchinson (2008). properties are expected to influence the response of
the footing-structure system. However, in addition to
The backbone curves are thus characterized by a these soil properties, few model
linear elastic region, then an increasingly growing parameters such as spring spacing, ratio of spring
nonlinear region (Fig. 2). For q-z material, ultimate stiffness along the length etc. (as shown in Fig. 3)
vertical capacity qult is calculated based on general may also alter the foundation responses.
bearing capacity equation given by Terzaghi (1943):
(1)
where qult is the ultimate vertical bearing capacity per

unit area of footing, c the cohesion, γ the unit weight
of soil, Df is the depth of embedment, B the width of
footing, and Nc, Nq, Nγ are bearing capacity factors
calculated after Mayerhof (1963). For the p-x
material, the passive resistance pult is calculated as:

Fig. 1 Schematic diagram of structure with BNWF
model for shallow foundations
UNCERTAINTY ANALYSIS
Fig. 2 Behavior of material models: (a) q-z element
In order to perform uncertainty analysis, two different
(b) p-y element and (c) t-z element
techniques are adopted: First-Order-Second-Moment
(FOSM) Method and Latin hypercube method. Below
is a brief description of both methods.
First-Order-Second-Moment (FOSM) Method

The FOSM method is used to perform simplified
sensitivity analysis to evaluate the effect of variability
of input variables on each response variable. This
method uses a Taylor series expansion of the function
Fig. 3 Vertical spring distribution along the footing to be evaluated and expansion is truncated after the
length linear first order term. It is assumed that the
relationship between the response variables and the
SELECTION OF UNCERTAIN PARAMETERS uncertain input parameters are assumed to be linear or
Based on the discussion provided in the previous
low-to-moderately nonlinear.
section the parameters shown in Table 1 are chosen
for the uncertainty analysis. Table 2 provides the
The response of the foundation is considered as a
range of values of the parameters for dense dry sand
random variable Q, which has been expressed as a
of relative density 80%. These values are based on function of the input random variables, Pi (for
EPRI (1990) and Harden et al. (2005). It is assumed i=1,...,N ) denoting uncertain parameters and Q given
that all uncertain input parameters are random by,
variables with a Gaussian distribution, having no
negative values. The upper and lower limits of the
(6)
random variables are assumed to be in 95th and 5th
percentile of its probability distribution. The Pi has been characterized by its mean p, and variance
corresponding mean () and standard deviation () p 2. Now, Q can be expanded using a Taylor series as
can be calculated as follows:
and
where LL and LU are the lower and upper limits

respectively and k depends on the probability level (7)
(e.g. k=1.645 for a probability of exceedance = 5%).
The assumed correlations among the uncertain
parameters are provided in Table 3. Considering only the first order terms of Equation 7,
and ignoring higher order terms, Q can be
Table 1 Uncertain parameters considered in this study approximated
as:
Taking expectation of both sides of Equation 6., the

mean of Q can be expressed as:
Table 2 Detail of the uncertain parameters
(9)
Utilizing the second order moment of Q as expressed
in Equation 7, the variance of Q can be derived as:
Table 3 Correlation matrix of the parameters

that there are no ties) where i = 1,...,N, draw an N x K
matrix containing ranks corresponding to the target
covariance matrix. The Latin hypercube sample Pij (i
= 1,...,N; j = 1,...,K) by
(12)
where F-1 j is the inverse cumulative distribution

where Pi,Pj denotes correlation coefficient for random function of variable j. This method is based on the
variables Pi and Pj. The partial derivative of h(P1, rank correlations among the input variables defined by
P2,..,PN) with respect to Pi has been calculated Iman and Conover (1982) which is based on Cholesky
numerically using the finite difference method decomposition of the covariance matrix. Latin
(central) as follows: hypercube sampling in this study has been done by
using MATLAB software, for that covariance matrix
is calculated by using standard deviation (Table 2)
(11) and correlation coeffcient (Table 3) between the any
Latin hypercube method two input parameters. The above-defined simple
For probabilistic analysis of engineering structures shearwall structure is considered for this analysis. The
having uncertain input variables, Monte Carlo response of this soil-foundation structure system is
Simulation (MCS) technique is widely used, as it is an dependent on the six independent input, normal
accurate approach to generate a number of equally distributed variables defined in Table 1 and Table 2.
likely random realizations. In order to decrease the
number of realizations required to provide reliable In order to find out the correct sample size pushover
results in MCS, Latin hypercube sampling (LHS) analysis are carried out using 10, 20,...,100, 200 and
technique is used in this study. LHS is a type of 300 number of samples. Figure 4 shows the plot
stratified MCS and has been widely used in the recent sample size versus the mean responses normalized by
past to carry out sensitivity and uncertainty analysis value corresponding to a sample size of 300. From
(Iman and Conover, 1982). It provides a very efficient Figure 4, it can observed that the mean of the
way of sampling parameters from their multivariate responses tend to converge as the sample size
distributions for estimating mean and standard increases. At the sample size of 100 the response of
deviations of response variables. It follows a general the system has almost converged. Therefore, for six
idea of a Latin square in which there is only one independent and normally distributed variables, a
sample in each row and each column. In Latin sample size of 100 is used.
hypercube sampling method, in order to generate a
1.06
sample of size k from n variables, the probability moment
Normalised mean response
distribution of each variable is divided into segments 1.04

Shear
with equal probability. Then the samples are chosen Rotation
Settlement
randomly in such a way that each interval contains 1.02
one sample. The value of each parameter is combined

with the other parameter in such a way that all 1
possible combinations of segments are sampled.

0.98
Finally we get m samples, where the samples cover 0 50 100 150 200 250 300 350
the n intervals for all variables. Thus the sampling Sample size
scheme does not require more samples for more Fig. 4 Convergence test for Latin hypercube sampling
dimensions (variables). method
In this study, to find out the uncertainty in the RESULTS AND DISCUSSION
response from the uncertainty in the input is studied In order to evaluate the effect of soil and model
by generating a sample for input variables with Stein's parameter uncertainty on the response of the shallow
(1987) approach. Stein (1987) introduced a method foundation, four response parameters are chosen:
for describing a Latin hypercube sample from absolute maximum moment |Mmax|, absolute
dependent variables. The procedure for generating maximum shear |Vmax |, absolute maximum rotation
LHS realization is as follows: For sample of size N |max| and absolute maximum settlement |Smax|. A
and K variables, obtain an independent and identically monotonic loading is applied at the top of the
distributed sample S1,...,SN such that each Si has the structure and responses are and forces and
correct joint distribution (F) when N is large. Now, displacements are obtained. The analysis is done
define M the matrix with K columns and N rows. Let using finite element software OpenSees (Open System
qj be the vector with the ranks of Sj for j = 1,...,K, for Earthquake Engineering Simulation). Figure 5
now replace each elements of jth column of M by qij shows a typical response of the system in terms of
(assume that each component of S is continuous so shear-drift and moment-rotation behavior for two

extreme values of friction angle. It can be observed
that friction angle has effect on the post-yield 0.603 φ' (deg)
behavior and the peak response. However, the initial
0.234 v
stiffness remains unchanged.
Parameters
0.155 Gs (Mpa)
Similarly, the analysis is carried out for varying each
0.004 Re (%)
parameter at a time while keeping other parameters
fixed at their mean values, and FOSM analysis is 0.004 Rk
carried out to find out the sensitivity of each 0.001
Ss (%)
parameter on the responses. Figure 6 to 10 show the
results of FOSM analysis for moment, shear, rotation 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
and settlement, respectively. It can be observed that Relative variance of moment
for moment and shear, friction angle is most Fig. 6 Results of FOSM analysis: relative variance for
important parameter (60% relative variance). moment
Poisson's ratio and shear modulus are moderately
important (about 23% and 16%, respectiely), and 0.603
model parameters have negligible effect (less than φ' (deg)
5%). However, model parameter stiffness intensity 0.234

v
ratio, Rk seems to have great effect on the rotational
Parameters
0.155
Gs (Mpa)
demand (~67%). Settlement is observed to be affected
by all three soil parameters (friction angle, shear 0.004
Re (%)
modulus and Poisson's ratio) almost equally (~30%) 0.004
Rk
for each parameter. Model parameters does not affect
0.001
this response much. Ss (%)
Table 4 shows the result obtained from Latin 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Relative variance of shear

hypercube method. The response is presented in term
Fig. 7 Results of FOSM analysis: relative variance for
of the mean and coefficient of variation (Cv) of each
shear
demand parameters. It can be observed from this table
that with 3%, 16%, 15%, 54%, 48% and 30% Cv of
friction angle, Poisson's ratio, shear modulus, end
length ratio, stiffness intensity ratio and spring 0.117 φ' (deg)
spacing input parameters, respectively can result in 0.050 v

moderate variation in demand parameters with Cv as
Parameters
16%, 16%, 22% and 24% for the absolute maximum 0.034 Gs (Mpa)
moment, shear, rotation and settlement demands, 0.001 Re (%)

respectively. Note that all responses are more
Rk
sensitive to the soil parameters than the model 0.663
parameters. Friction angle is the most sensitive 0.136 Ss (%)

amongst all input parameter; as with a 3% Cv results
in significant variation in the response variables. 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Relative variance of rotation

rotation
φ' (deg)
0.264
0.307
v
Parameters
0.296 Gs (Mpa)
Fig. 5 Response of shearwall-footing system 0.110 Re (%)
0.021 Rk
0.002 Ss (%)
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Relative variance of settlement

settlement

Table 4 Variability of response parameters foundations, Technical Report 2005/04, Pacific
Earthquake Engineering Research Center.
Iman, R. L. and Conover, W. J. (1982). A
distribution free approach to including rank
correlation among input variables. Bulletin of the
Seismic Society of America, 11(3), 311-334.
CONCLUSIONS Jindal S. (2011). Shallow foundation response
The effect of uncertainty in soil and model parameters analysis: a parametric study, Master’s Thesis,
on the soil-foundation system response has been Indian Institute of Technology Kanpur, India.
studied in this article. The soil-foundation system has Lacasse, S. and Nadim, F. (1996). Uncertainties in
been modeled using BNWF concept and the characterizing soil properties. Uncertainty in the
uncertainty analyses are carried out using FOSM and Geologic Environment: From Theory to Practice,
Latin hypercube method. It has been observed that for Proceedings of Uncertainty 96, Madison,
moment and shear, friction angle is most important Wisconsin, July 31-August 3, 1996, New York,
parameter (60% relative variance); Poisson's ratio and USA, ASCE Geotechnical Special Publication
shear modulus are moderately important (about 23% No. 58; 49-75.
and 16%, respectively); and model parameters have Lumb, P. (1966) The variability of natural soils.
negligible effect (less than 5%). The rotational Engineering Structures, 3, 74-97.
demand (~67%) is largely dependent on stiffness Ronold, K. O. and Bjerager, P. (1992) Model
intensity ratio. The settlement demand is almost uncertainty representation in geotechnical
equally sensitive to friction angle, shear modulus and reliability analyses, Journal of Geotechnical
Poisson's ratio (~30% variance for each parameter). Engineering, 118(3), 363-376.
The results from Latin hypercube method indicate that McKay, M. D., Beckman, R. J., and Conover, W. J.
a coefficient of variation of 3% in friction angle (1969). A comparison of three methods for
results in 16%, 16%, 22% and 24% for the absolute selecting values of input variables in the analysis
maximum moment, shear, rotation and settlement of output from a computer code, Technometrics,
demands, respectively, indicating that this parameters 21(2), 239-245.
have great effect on each response variables. It can Meyerhof, G. G. (1963). Some recent research on
finally be concluded that soil parameters such as the bearing capacity of foundations, Canadian
friction angle, shear modulus and Poisson's ratio may Geotechnical Journal, 1(1), 16-26.
have significant effect on the response of foundation. Na, U. J., Ray Chaudhuri, S., and Shinozuka, M.
Therefore, selection of these parameters should be (2008) Probabilistic assessment for seismic
considered critically when designing a structure with performance of port structures, Soil Dynamics
significant soil-structure interaction effect. and Earthquake Engineering, 28, 147-158.
Raychowdhury, P., and Hutchinson, T. C. (2008).
REFERENCES Nonlinear material models for winkler-based
Boulanger, R. W., Curras, C. J., Kutter, B. L., shallow foundation response evaluation,
Wilson, D. W., and Abghari, A. (1999). Seismic GeoCongress 2008, Characterization,
soil-pile-structure interaction experiments and Monitoring, and Modeling of GeoSystems, March
analyses, Journal of Geotechnical and 9-12, New Orleans, LA, ASCE Geotechnical
Geoenvironmental Engineering, 125(9), 750-759. Special Publication No. 179, 686–693.
Chakraborty, S. and Dey, S. S. (1996). Stochastic Ray Chaudhuri, S. and Gupta, V. K. (2002)
finite-element simulation of random structure on Variability in seismic response of secondary
uncertain foundation under random loading, systems due to uncertain soil properties,
International Journal of Mechanical Sciences, Engineering Structures, 24(12), 1601-1613.
38(11), 1209-1218. Raychowdhury, P. (2009). Effect of soil parameter
EPRI (1990). Manual on Estimating Soil Properties uncertainty on seismic demand of low-rise steel
for Foundation Design. Electric Power Research buildings on dense sand, Soil Dynamics and
Institute, Palo Alto, California. Earthquake Engineering, 29, 1367-1378.
Foye, K. C., Salgado, R., and Scott, B. (2006). Raychowdhury, P. and Hutchinson, T. C. (2010)
Assessment of variable uncertainties for Sensitivity of shallow foundation response to
reliability based design of foundations, Journal of model input parameters, Journal of Geotechnical
Geotechnical and Geoenviornmental and Geoenvironmental Engineering, 136(3), 538-
Engineering, 132(9), 1197-1207. 541.
Gazetas, G. (1991). Foundation Engineering Stein, M (1987). Large sample properties of
Handbook, Fang, H.Y. (Ed). Van Nostrand simulations using latin hypercube sampling.
Rienhold. Technometrics, 29(2), 143-151.
Harden, C. W., Hutchinson, T. C., Martin, G. R., Terzaghi K. (1943). Theoretical Soil Mechanics, J.
and Kutter, B. L. (2005). Numerical modeling of Wiley, New York.
the nonlinear cyclic response of shallow

ISBN 978-81-921121-0-7
DUST DEPOSITION CAPACITY OF CERTAIN ROADSIDE PLANTS

IN AIZAWL, MIZORAM: IMPLICATIONS FOR ENVIRONMENTAL
GEOMAGNETIC STUDIES
Prabhat Kumar Rai
Senior Assistant Professor, Department of Environmental Science,School of Earth Science and Natural Resource
Management, Mizoram University, Tanhril, Aizawl-796004, Mizoram, Email: prabhatrai24@gmail.com
ABSTRACT: Particulate pollution poses serious threat to environment as well as human health particularly in
urban areas. Present research work, performed during the period of 2010-11, investigated the seasonal variation
in dust /particulate deposition capacity for certain roadside plant leaves in a peri-urban area of Aizawl,
Mizoram. Leaves from roadside trees form natural, widely distributed collection surfaces for pollutants. The
recorded plants were Hibiscus rosa-sinensis, Psidium guajava, Mikania micrantha, Eupatorium odoratum,
Mangifera indica and Bougainvillea spectabilis. Apart from vehicular dust generation other anthropogenic
sources are soil erosion, mining and stone quarrying activities prevailing particularly in peri-urban and rural
regions of Aizawl. The dust deposition capacity was recorded maximum during the winter season followed by
summer and least in rainy season. Among different pants investigated the decreasing order of dust deposition
capacity was Mangifera indica> Psidium guajava> Eupatorium odoratum> Hibiscus rosa-sinensis>
Bougainvillea spectabilis> Mikania micrantha, which may be attributed to morphology, physiology and
biochemistry of plant leaves tolerance to dust/particulates. Since dust deposited on plant leaves may be derived
from iron impurities in fuels or rock dust with magnetite, aforesaid results may be further studied by performing
environmental geomagnetic studies of plant leaves and an overview of the concept is presented in this paper. In
nutshell, biomagnetic monitoring of particulate pollution may biomonitor the extent of dust/particulate pollution
in an effective and systematic way.
INTRODUCTION The particles of dust that deposit from the atmosphere

and accumulate along roadsides are called road dust
Besides social and economic problems, the particles and originate from the interaction of solid,
development model of the so-called Third World has liquid and gaseous metals (Akhter & Madany, 1993;
caused serious environmental degradation and Faiz et al., 2009). Dust pollution in the atmosphere,
challenges in the field of civil engineering. It is well particularly of pollutant particles below 10 mm
known that air pollution represents a threat to both the (PM10), is of current concern worldwide due to
environment and human health, and it is estimated adverse health effects associated with their inhalation
that millions of tons of toxic pollutants are released (Calderon-Garciduenas et al., 2004; Morris et al.,
into the air each year. Rapid pace of industrialization 1995; Oberdorster, 2000; Pope et al., 2004; Faiz et al.,
and urbanization has given birth to dust/particulate 2009). Indeed, particulate matter in dust is thought to
matter pollution, impact of which may be correlated be the most harmful pollution component widely
with urban planning as well as topography of the present in the environment, with no known level at
particular region. Further, the implication of the which adverse health effects occur (Bealey et al.,
intake of dust particles with high concentration of 2007).
heavy metals poses potentially deleterious effects on
the health of human beings (Faiz et al., 2009). Sources of particulate pollution
Particulate matter/dust pollution Sources of particulate pollution may be natural or

anthropogenic in nature. Emission sources may
Solid matter, which is composed of soil, include natural processes such as wildfires, volcano
anthropogenic metallic constituents, and natural eruption and dust storms. Man-made pollutions
biogenic materials, is called dust (Ferreira-Baptista & encompass combustion processes used for heating,
DeMiguel, 2005). The particulates belong to the class power production, industry, and traffic vehicles.
of poorly soluble particles that also encompasses
carbon black, coal mine dust, and titanium dioxide HUMAN HEALTH IMPACTS OF
(Borm et al., 2005; Moller et al., 2008). PARTICULATE POLLUTANTS
Measurements of the particulate matter in ambient air
are usually reported as the mass of particles with an In general, particulate matter comprises polycyclic
aerodynamic diameter that is less than 2.5 µm (PM2.5) aromatic hydrocarbons (PAH) and volatile organic
or 10 µm (PM10) (Zhu et al., 2006). Aforesaid particle compounds (VOCs), which may have deleterious
sizes are emphasized in view of their pertinent health impact on human health. Associations between urban
impacts. pollutants and respiratory and cardiovascular

problems, and still a greater incidence of certain In view of the abovementioned deleterious impacts of
cancer types have already been established (Lester & particulate matters, it is quite obvious to investigate
Seskin, 1970; Saldiva et al., 2002; Lin et al., 2003). the feasible and eco-sustainable control technologies.
Constituents of particulate pollutants emanating from MONITORING AND ECO-SUSTAINABLE

vehicular emissions have been demonstrated to cause MANAGEMENT OF PARTICULATES
genotoxicological impact on plants as well as humans.
For instance, although polycyclic aromatic Dust particles can be removed from the atmosphere
hydrocarbons (PAHs) are relatively chemically inert by dry, wet or occult deposition (NEGTAP, 2001).
compounds, however, through metabolic activation to Dry deposition is the removal of pollutants by
electrophilic derivatives (e.g. diolepoxides, quinones, sedimentation under gravity, diffusion processes (i.e.,
conjugated hydroxyalkyl derivatives) these are Brownian motion) or by turbulent transfer resulting in
capable of covalent interaction with nucleophilic impaction and interception (Beckett et al., 2004).
centres of DNA (Schoket, 1999). These adducts of However, existing technologies for abatement of
PAH to DNA cause base pair substitutions, frameshift particulates are not cost-effective.
mutations, deletions, S-phase arrest, strand breakage
and a variety of chromosomal alterations. (Schoket, Biomonitoring of particulates
1999).
Environmental contamination and human exposure
In an integrated way, PAH and VOCs lead to the with respect to dust pollution have dramatically
formation of bulky DNA adducts (Moller et al., increased during the past ten years (Faiz et al., 2009).
2008). The genotoxic effect of pollutants on the Internationally, the quantification and valuation of
ecosystem, including the build-up of resistant species, pertinent ecosystem services have drawn much
is also of considerable concern (Ma et al., 1994; attention since the last decade (Nowak, 1994; Taha,
Grant, 1998). The potential genotoxic effects on 1996, 1997; Beckett et al., 1998; McPherson et al.,
human health by such vapor phase chemicals include 1997, 1998, 1999; Nowak et al., 1998, 2000, 2002,
malignant cell formation, the accumulation of 2006; Rosenfeld et al., 1998; Scott et al., 1998;
heritable abnormal genes within the population, heart Akbari et al., 2001; Akbari, 2002).
disease, aging and cataracts (Grant, 1998). The effects
of toxic compounds, and the subsequent genotoxic Besides the application of advanced technologies in
effects on plants, are of particular importance as prevention and amelioration, the option of employing
plants comprise a large portion of our biosphere and natural mechanisms to effect ecological abatement is
constitute a vital link in the food chain (Grant, 1998; occasionally adopted by design but commonly
Rajput & Agrawal, 2005). Estimating air genotoxicity contributes by default. Although, there are many
is therefore crucial to evaluating risk to the conventional (physical and chemical) devices for
environment and public health. assessment of air pollution, however, plant systems
allow the direct assessment of the of air stressors.
Particulate matters also consist of heavy metals, Biomonitoring of dust pollution and its biochemical
carbon core and pollen, which further interacts with impact has drawn attention of both national and
gaseous pollutant resulting in the formation of DNA international scientific community. The main
lesions. There has been considerable concern on the advantage of the plants use as bio-monitor is that
pulmonary effects of particulates less than 2.5 µm plants are wide-spread providing a high density of
(PM2.5) or 10 µm (PM10), as they can reach the alveoli sampling points (Moreno et al., 2003). Moreover, the
and translocate to the circulation, whereas particles of most economical and reasonable method for
larger size deposit mainly in the upper airways and biomonitoring heavy metal levels in the atmosphere is
can be cleared by the mucociliary system (Oberdo¨ using plants (Çelik et al., 2005). Among trees,
rster et al., 2005; Moller et al., 2008). Further, it has evergreen species are better traps for particles than
increasingly being realized that generation of reactive deciduous ones because of their longer leaf longevity,
oxygen species (ROS) and oxidative stress is an which can accumulate pollutants throughout the year
important toxicological mechanism of particle (Gratani & Varone, 2006, 2007).
induced lung cancer (Knaapen et al., 2004; Risom et
al., 2005). The fraction of particulate matter contains In this regard, the air cleansing capacity of urban trees
a number of constituents that may increase the presents an alternative approach to foster an
generation of ROS by a variety of reactions such as integrated approach to the sustainable management of
transition metal catalyses, metabolism, redoxcycling urban ecosystems (Nowak et al., 2002, 2006;
of quinones, and inflammation. Particulate matter, Randolph, 2004). Biological monitors are organisms
thus, can generate oxidative damage to DNA, that provide quantitative information on some aspects
including guanine oxidation, which is mutagenic of their environment, such as how much of a pollutant
(Kasai, 1997; Moller et al., 2008). is present (Martin & Coughtrey, 1982). Moreover, in
urban area higher plants are mostly suitable for

monitoring dust pollution as lichens and mosses are of roadside plant leaves were performed, in addition
often missing (Al – Alawi & Mandiwana, 2007). to a series of work, cited above related with European
countries (Matzka & Maher, 1999; Maher, 2009).
Therefore, vegetation is an efficient sink for dust Magnetic minerals derived from vehicular combustion
originating from diverse sources (Fowler et al., 1989). are mainly maghemite and metallic iron grains having
Dust particles from the air are mainly adhered to the a size range of 0.1–0.7 µm (Pandey et al., 2005;
outside of plants. This is in contrast to air polluting Maher, 2009). This grain size is particularly
gases and very small particles (<0.1µm) which are dangerous to humans because of its ability to be
assimilated for an important part via the stomata into inhaled into the lungs. Iron often occurs as an
the leaves (Fowler et al., 1989). The use of different impurity in fossil fuels during industrial, domestic, or
plant materials as biomonitors of anthropogenic vehicle combustion; carbon and organic material are
contamination is discussed in detail in Markert lost by oxidation and the iron forms a non-volatile
(1993). residue, often comprising glassy spherules (due to
melting). These spherules are magnetic, with easily
Likewise, in the past decade there has been increased measurable magnetization levels. Also, combustion-
interest, in many parts of the world, in the study of related particles in vehicles, via exhaust emissions and
tree leaves as bio-accumulators of trace elements/ abrasion/ corrosion of engine and/or vehicle body
metals present in dust, in the surroundings of material can generate nonspherical magnetite particles
industrial facilities (Helmisaari et al., 1995; Nieminen (Pandey et al., 2005; Maher, 2009).
& Helmisaari, 1996; Bussotti et al., 1997; Giertych et
al., 1997; Mièieta & Mur´ın, 1998; Rautio et al., 1998 Magnetic biomonitoring (Matzka & Maher, 1999;
a,b,c,d) and in urban environments (Alfani et al., Maher, 2009) seems to be a valuable means both to
1995, 1996a,b, 1997; Monaci & Bargagli, 1997; gain significantly enhanced spatial resolution for
Monaci et al., 2000), although few studies have been pollutant data, and test proposed particulate
made of rural and background areas (Loppi et al., source/health linkages. The deposition of pollution
1997; Ukonmaanaho et al., 1998; Cˇ eburnis & particles on tree leaf surfaces has been shown to result
Steinnes, 2000). Leaves are potentially efficient in easily measurable magnetic properties, including
receptors and biomonitors of particulate pollution, as magnetic remanence (i.e. the magnetisation remaining
they provide a large total surface for particle after a sample has been placed in and then removed
collection, numbers of samples and sample sites can from an applied dc field) and magnetic susceptibility
be high (i.e. hundreds), and, in pollution contexts, the (the magnetisation induced when the sample is placed
leaves themselves are insignificantly magnetic. in a small ≈ twice the Earth’s magnetic field-ac field).
Further, tree leaves also preclude sampling problems
associated with the use of artificial particle collectors Magnetic properties of leaves (Muxworthy et al.,
(including power requirements). 2002; Moreno et al., 2003; Urbat et al., 2004; Pandey
et al., 2005; Maher et al., 2008; Mitchell & Maher,
BIOMAGNETIC MONITORING WITH 2009; Maher, 2009) have been used to identify the
ROADSIDE PLANT LEAVES spread of pollution derived from vehicular emission.
Leaves with large surface areas per unit of weight,
A new approach, biomagnetic monitoring, may favourable surface properties (a waxy coating), and a
provide a robust means to achieve measurement and long lifespan, such as conifer needles or evergreen
sourcing of PM10 at unprecedented levels of spatial tree leaves, are considered to be good accumulators of
resolution and is applicable all around the world particulate matter from the atmosphere (Freer-Smith
(Maher, 2009) (e.g. Shu et al., 2001, in China; et al., 1997; Alfani et al., 2000).
Gautam et al., 2005, in Nepal; Pandey et al., 2005, in
India; Chaparro et al., 2006, in Argentina; Kim et al., Methodology used in studying the magnetic
2007, in Korea; Szönyi et al., 2008, in Europe). properties of roadside plant leaves
Henceforth, in view of aforesaid, present proposed
paper provides an overview on biomagnetic For geomagnetic studies, leaf samples are magnetised
monitoring of roadside plant leaves as an integral part (at room temperature) with incremental, pulsed dc
of the present work. fields of 20, 50, 100 and 300 milliTesla (mT) and 1
Tesla (T), using a magnetometer (molspin pulse
Further, magnetic biomonitoring studies of magnetiser). The resultant isothermal (i.e. room
plant leaves may act as proxy of ambient particulate temperature) magnetic remanences (IRMs) are
pollution. Magnetic biomonitoring of pollutants by generally measured using a cryogenic/other applicable
measurements taken from roadside tree leaves is magnetometer. Figure 1, explains the concise
potentially efficient, as samples are abundant and methodology for studying the magnetic properties of
hundreds of samples can be collected and analyzed tree leaves. The 2D magnetization values of plant
within few days. In Singrauli Industrial region leaves may be correlated with the dust content and
(Pandey et al., 2005) and hilly areas of Nepal other constituents of dust (e.g. heavy metals), hence,
(Gautam et al., 2005), magnetic biomonitoring studies act as an indicator/proxy of particulate pollution.

In India, Indian Institute of Geomagnetism and its reduced while tolerant species can thrive and
regional centers are providing different categories of dominate the vegetation
magnetometers (Molspin MINISPIN Spinner, AGICO
JR-6 high sensitivity dual speed Spinner and Molspin Likewise, a number of studies have been done on the
Vibrating sample magnetometer (VSM) for hysteresis pollution effects on different aspects of plant life such
loop) in order to investigate the different phenomenon as overall growth and development (Gupta & Ghouse,
linked with environmental geomagnetism (Figure 2). 1987; Saquib et al. 1992; Misra & Behera, 1994;
Pandey et al., 1999; Prusty et al., 2005), foliar
morphology (Gupta & Mishra, 1994; Trivedi and
Singh, 1995; Somashekar et al., 1999; Singh and
Sthapak, 1999; Farooq et al., 2000; Pal et al., 2000;
Shrivastava & Joshi, 2002), anatomy (Zafar, 1985;
Arjunan et al., 1993; Garg et al., 2000; Singh, 2000),
and biochemical changes (Balsberg-Pahlsson, 1989;
Pandey & Sinha, 1991; Vyas et al., 1991; Budharaja
& Agrawal, 1992; Tiwari & Patel, 1993;
Krishnamurthy et al., 1994; Senapati & Misra, 1996;
Joshi et al., 1997; Pandey et al., 1999; Garty et al.,
2001; Mashitha & Pise, 2001; Gavali et al., 2002).
Pollution effects have been found to link the
physiological response of plants to acceleration in the
process of senescence (Lee et al., 1981; Kohert et al.,
1986). One of the overt manifestations of plant
senescence is gradual disappearance of chlorophyll
and associated yellowing of leaves that may be linked
with a consequent decline in the capacity for
Fig. 1 Diagrammatic representation of methodology photosynthesis (Mandal & Mukherji, 2000).
used in studying the magnetic properties of dust
loaded roadside plant leaves (Source: Mitchell and Study area
Maher, 2009; Maher, 2009).
The Aizawl district (210 58’ -240 35’ N Latitude and
BIOMONITORING OF DUST/PARTICULTES 920 15’ -930 29’ E Longitude; 850 meters above mean
WITH CERTAIN ROADSIDE PLANTS IN sea level), capital of Mizoram, is heavily entrapped
AIZAWL, MIZORAM: A CASE STUDY with heavy vehicles load (Mizoram State Transport
data, 2006) resulting in increased quantity of
In Indian cities airborne particulate matter seems to be vehicular emission particularly dust/particulates in
a very serious problem (Agarwal et al., 1999). urban areas. Rapid urbanization and continuously
Vegetation is an important tool to remove particulates expanding population has been the major cause of
from the atmosphere (Simonich and Hites, 1994). increase in the number of vehicles and hence
Several studies also revealed that leaves are sensitive particulates. Sampling sites were taken at Ramrikawn
and highly exposed parts of a plant and may act as area (2km stretch connecting Mizoram University).
persistent absorbers of dust in a polluted environment Samplings were conducted quarterly (2010-11) i.e.
(Maiti, 1993; Samal & Santra, 2002). They act as during rainy, winter and summer season.
pollution receptors and reduce dust concentration of
the air (Maiti, 1993; Nowak, 1994; Singh, 2000; Mizoram Pollution Control Board (2005)
Samal & Santra, 2002). The capacity of leaves as dust data also demonstrated the increasing level of
receptors depends upon their surface geometry, particulates. One preliminary study in Aizawl
phyllotaxy, epidermal and cuticular features, leaf (Lalrinpuii & Lalramnghinglova, 2008) recorded the
pubescence, and height and canopy of trees (Nowak, level of suspended particulate matter (SPM) and
1994; Singh, 2000; Singh et al., 2002). However, respirable particulate matter (RPM) above permissible
many plants are very sensitive to air pollutants, and limit of National Ambient Air Quality Standards
pollutants can damage their leaves, impair plant (NAAQS). Moreover, following a gap of 15 years, the
growth, and limit primary productivity (Ulrich, 1984). Ministry of Environment and Forests {(MoEn), Press
Henceforth, assessment of biochemical parameters is release, 18th November} has announced the
extremely relevant in order to identify tolerant plants. notification of the Revised National Ambient Air
The most obvious damage occurs in Quality Standards 2009 in the official Gazette.
the leaves and associated biochemical parameters. Therefore, in present context of new standard
Growth and reproduction in some plants may be PM2.5, it is pertinent to characterize the air quality of
impaired and the population of sensitive species is ecologically sensitive regions like Aizawl, Mizoram.
Also, Aizawl city is not well-planned (Lalrinpuii &

Lalramnghinglova, 2008) which may further increase In the light of abovementioned, it was quite
the air pollutants concentration due to lack of space obvious for me to extensively monitor and examine
required for their dispersal. the extent of dust deposition on roadside plant leaves.
Table 1 Seasonal variation in dust deposition capacity for certain roadside plants in Aizawl, Mizoram
.
Dust deposition capacity {(mg/cm2) ; average of three values}
Season Hibiscus Psidium Mikania Eupatorium Mangifera Bougainvillea

rosa- guajava micrantha odoratum indica spectabilis
sinensis
Rainy 0.0022 0.042 0.017 0.014 0.057 0.011
Winter 0.184 0.362 0.151 0.209 0.407 0.162
Summer 0.092 0.286 0.089 0.131 0.296 0.087
Methodology investigated the decreasing order of dust deposition

capacity was Mangifera indica> Psidium guajava>
The recorded plants were Hibiscus rosa-sinensis, Eupatorium odoratum> Hibiscus rosa-sinensis>
Psidium guajava, Mikania micrantha, Eupatorium Bougainvillea spectabilis> Mikania micrantha, which
odoratum, Mangifera indica and Bougainvillea may be attributed to morphology, physiology and
spectabilis. From each plant at least 6-12 young biochemistry of plant leaves tolerance to
leaves (in triplicate) was selected for the study. The dust/particulates.
upper dorsal surface of all these leaves was cleaned
using a fine brush (in-situ). The leaf selected as a)
control was covered with a transparent polythene bag
having perforations to facilitate evapo-transpiration.
All the leaves were left for specific period to allow
dust to accumulate on their surface. After completion
of the allotted period, the surface of all these
previously selected leaves were cleaned using a fine
brush and the dust will be collected in preweighed
tracing paper with maximum care. The leaves were
cut from the petiole for further analysis in laboratory.
The individual leaf area (in cm2) was calculated by
graph paper.
The dust samples was weighed using an electrical

balance and the amount of dust was calculated using
the equation
Where, W is dust content (mg/cm2),
w1 is initial weight of tracing paper,
w2 is final weight of tracing paper with dust, and
a is total area of the leaf (cm2) (Prusty et al., 2005). b)
Results Fig. 2 Magnetometers a) Molspin MINISPIN Spinner

b) AGICO JR-6 high sensitivity dual speed Spinner at
The dust deposition capacity was recorded maximum Indian Institute of Geomagnetism
during the winter season followed by summer and
least in rainy season. Among different pants

CONCLUSION Arjunan, M.C., Gunamani, T., Ponnammal, N.R.
1993. Poll. Res. 12, 197–206.
Since dust deposited on plant leaves may be derived Balsberg-Pahlsson, A.M. 1989. Effects of heavy-
from iron impurities in fuels or rock dust with metal and SO2 pollution on the concentrations of
magnetite, aforesaid results may be further studied by carbohydrates and nitrogen in tree leaves. Can. J.
performing environmental geomagnetic studies of Bot. 67, 2106–2133.
plant leaves and an overview of the concept is Bealey, W.J., Mcdonald, A.G., Nemitz, E., Donovan,
presented earlier in this paper. The 2D magnetization R., Dragosits, U., Duffy, T.R., Fowler, D. 2007.
values of plant leaves may be correlated with the dust Estimating the reduction of urban PM10
content and other constituents of dust (e.g. heavy concentrations by trees within an environmental
metals). Thus, the investigation of magnetic properties information system for planners. Journal of
of roadside plant leaves act as an indicator or proxy Environmental Management 85, 44–58.
for particulate matter pollution. Henceforth, Beckett, K.B., Freer-Smith, P.H., Taylor, G. 1998.
biomagnetic monitoring of particulate pollution may Urban woodlands: their role in reducing the effects
biomonitor the extent of dust/particulate pollution in of particulate pollution. Environmental Pollution
an effective and systematic way. 99, 347–360.
Borm, P.J., Schins, R.P., Albrecht., C. (2004).
ACKNOWLEDGEMENTS Inhaled particles and lung cancer, part B:
paradigms and risk assessment, Int. J. Cancer 110,
The author is thankful to Department of 3–14.
Biotechnology (DBT), Ministry of Science & Budharaja, A., Agrawal, S.K. 1992. Acta Ecol. 14,
Technology, Government of India for financial 92–98.
assistance wide research project BT/PR- Bussotti, F., Cenni, E., Cozzi, A., Ferretti, M. 1997.
11889/BCE/08/730/2009 to Prabhat Kumar Rai. The impact of geothermal power plants on forest
Further, thanks are due to Dr. Onkar Nath Tiwari for vegetation. A case study at Travale (Tuscany,
kind cooperation in this regard. The author is thankful Central Italy). Environ. Monit. Assess. 45, 181–
to Lalita L.S. Panda (DST-inspire fellow) and Miss 194.
Malsawmthangi Pachuau for their assistance in data Cˇ eburnis, D., Steinnes, E. 2000. Conifer needles as
collections mentioned in the case study. biomonitors of atmospheric heavy metal
deposition: comparison with mosses and
REFERENCES
precipitation, role of the canopy. Atmos. Environ.
Agarwal, A., Narain, S., Sen, S. 1999. State of India’s 34, 4265–4271.
environment. In: Agarwal, A. (Ed.), The Citizens’ Caldero´ n-Garciduenˇ as, L., Reed, W., Maronpot,
Fifth Report. Part I: National Overview. Centre for R.R., Henriquez-Roldan, C., Delago- Chavez, R.,
Science and Environment, New Delhi, pp. 167– Calderon-Garciduenas, A., Dragustinovis, I.,
206. Franco-Lira, M., Aragon- Flores, M., Solt, A.C.,
Akbari, H., 2002. Shade trees reduce building energy Altenburg, M., Torres-Jardon, R., Swenburg, J.A.
use and CO2 emissions from power plants. 2004. Brain inflammation and Alzheimer’s-like
Environmental Pollution 116, S119–S126. pathology in individuals exposed to severe air
Akbari, H., Pomerantz, M., Taha, H. 2001. Cool pollution. Toxicologic Pathology 32, 650–658.
surfaces and shade trees to reduce energy use and Çelik, A., Kartal, A.A., Akdog˘an, A., Kaska, Y.
improve air quality in urban areas. Solar Energy 2005. Determining the heavy metal pollution in
70, 295–310. Denizli (Turkey) by using Robinio pseudoacacia
Akhter, M.S. & Madany, I.M. 1993. Heavy metals in L. Environmental International 31, 105–112.
street and house dust in Bahrain,Water, Air and Chaparro, MAE, Gogorza, CSG, Chaparro, MAE,
Soil Pollution 66, 111–119. Irurzun MA, Sinito. A.M. 2006. Review of
Al – Alawi, M.M., Mandiwana, K.L., 2007. The use magnetism and heavy metal pollution studies of
of Aleppo pine needles as bio-monitor of heavy various environments in Argentina. Earth, Planets
metals in the atmosphere. Journal of Hazardous and Space 58, 1411-1422.
Materials 148, 43–46. Faiz, Y., Tufail, M., Javed M.T., Chaudhry, M.M.,
Alfani, A., Baldantoni, D., Maisto, G., Bartoli, G., Siddique, N. 2009. Road dust pollution of Cd, Cu,
Virzo de Santo, A. 2000. Temporal and spatial Ni, Pb and Zn along Islamabad Expressway,
variation in C, N, S and trace element contents in Pakistan. Microchemical Journal 92, 186–192
the leaves of Quercus ilex within the urban area of Farooq, M., Arya, K.R., Kumar, S., Gopal, K., Joshi,
Naples. Environmental Pollution 109, 119–129. P.C., Hans, R.K. 2000. Industrial pollutants
Alfani, A., Bartoli, G., Rutigliano, F.A., Maisto, G., mediated damage to mango (Mangifera Indica)
Virzo de Santo, A., 1996b. Trace metal crop: A case study. J. Environ. Biol. 21, 165–167.
biomonitoring in the soil and the leaves of Ferreira-Baptista, L., DeMiguel, E. 2005.
Quercus ilex in the urban area of Naples. Biol. Geochemistry and risk assessment of street dust in
Trace Elem. Res. 51, 117–131.

Luanda, Angola: a tropical urban environment. Knaapen, A.M., Borm, P.J., Albrecht, C., Schins, R.P.
Atmospheric Environment 39, 4501–4512. 2004. Inhaled particles and lung cancer. Part A:
Freer-Smith, P.H., Holloway, S., Goodman, A. 1997. mechanisms. Int. J. Cancer 109, 799–809.
The uptake of particulates by an urban woodland, Kohert, R.J., Amundsun, R.G., Lawrence, J.A., 1986.
site description and particulate composition. Environ. Poll., Series A. 41, 219–234.
Environmental Pollution 95 (1), 27–35. Krishnamurthy, R., Srinivas, T., Bhagwat, K.A.,
Garg, S.S., Kumar, N., Das, G. 2000. Indian J. 1994. J. Environ. Biol. 15, 97–106.
Environ. Prot. 20, 326–328. Lalrinpuii, H., Lalramnghinglova, H. 2008.
Garty, J., Tamir, O., Hassid, I., Eshel, A., Cohen, Y., Assessment of air pollution in Aizawl city.
Karnieli, A., Orlovsky, L. 2001. Photosynthesis, Current Science 94 (7), 852-853.
Chlorophyll Integrity, and Spectral Reflectance in Lee, E.H., Bennat, J.H., Heggestad, H.E. 1981.
Lichens Exposed to Air Pollution. J. Environ. Retardation of senescence in red clover leaf discs
Qual. 30, 884–893. by a new antiozonant, N-[2-(OxO-1-
Gautam, P., Blaha, U., Appel, E. 2005. Magnetic imidazolidinyl) ethyl]-N-phenylurea. Plant
susceptibility of dust-loaded leaves as a proxy of Physiol. 67, 347–350.
traffic-related heavy metal pollution in Kathmandu Lester, B.L., Seskin, E.P. 1970. Air pollution and
city, Nepal. Atmospheric Environment 39, 2201- human health. American Association for the
2211 Advancement of Science 169, 723-733.
Gavali, J.G., Saha, D., Krishnayya, K., 2002. Diff Lin, C.A., Pereira, L.A., Conceiç~ao, G.M.S., Kishi,
erence in sulphur accumulation in eleven tropical H.S., Milani Jr., R., Braga, A.L.F., Saldiva, P.H.N.
tree species growing in polluted environments. 2003. Association between air pollution and
Indian J. Environ. Health 44, 88–91. ischemic cardiovascular emergency room visits.
Giertych, M.J., De Temmerman, L.O., Rachwal, L. Environmental Research 92, 57-63.
1997. Distribution of elements along the length of Loppi, S., Nelli, L., Ancora, S., Bargagli, R. 1997.
Scots pine needles in a heavily polluted and Passive monitoring of trace elements by means of
control environment. Tree Physiol. 17, 697–703. tree leaves, epiphytic lichens and bark substrate.
Grant, W.F. 1998. Higher plant assays for the Environ. Monit. Assess. 45, 81–88.
detection of genotoxicity in air polluted Ma, T.H., Cabrera, G.L., Chen, R., Gill, B.S., Sandhu,
environments. Ecosystem Health 4, 210-229. S.S., Valenberg, A.L., Salamone, M.F. 1994.
Gratani, L., Varone, L. 2006. Carbon sequestration by Tradescantia micronucleus bioassay. Mutat. Res.
Quercus ilex L. and Quercus pubescens Willd. and 310, 221-230.
their contribution to decreasing air temperature in Maher, B.A. 2009. Rain and Dust: Magnetic Records
Rome. Urban Ecosystems 9, 27–37. of Climate and Pollution. Elements 5, 229-234.
Gratani, L., Varone, L. 2007. Plant crown traits and Maher, B.A., Mooreb, C., Matzka, J. 2008. Spatial
carbon sequestration capability by Platanus variation in vehicle-derived metal pollution
hybrida Brot. in Rome. Landscape and Urban identified by magnetic and elemental analysis of
Planning 81, 282–286. roadside tree leaves. Atmospheric Environment 42,
Gupta, A.K., Mishra, R.M. 1994. Effect of lime kiln’s 364–373.
air pollution on some plant species. Poll. Res. 13, Maher, Barbara A. and Dennis, P. F. 2001 Evidence
1–9. against dust-mediated control of glacial-
Gupta, M.C., Ghouse, A.K.M. 1987. The effect of interglacial changes in atmospheric CO2. Nature
coal smoke pollutants on growth yield and leaf 41 (6834), 176-180.
epidermis features of Abelmoschus esculenus Maiti, S.K., 1993. Indian J. Environ. Prot. 13, 276–
Moench. Environ. Pollut. 43, 263–270. 280.
Helmisaari, H.-S., Derome, J., Fritze, H., Nieminen, Mandal, M., Mukherji, S., 2000. Changes in
T., Palmgren, K., Salemaa, M., Vanha-Majamaa, chlorophyll context, chlorophyllase activity, Hill
I., 1995. Copper in Scots pine forest around a reaction, photosynthetic CO2 uptake, sugar and
heavy metal smelter in South-Western Finland. starch contents in fi ve dicotyledonous plants
Water Air Soil Pollut. 85, 1727–1732. exposed to automobile exhaust pollution. J.
Kasai, H. 1997. Analysis of a form of oxidative DNA Environ. Biol. 21, 37–41
damage, 8- hydroxy-20-deoxyguanosine, as a Markert, B. 1993. editor. Plants as biomonitors.
marker of cellular oxidative stress during Indicators for heavy metals in the terrestrial
carcinogenesis. Mutat. Res. 387, 147–163. environment. Weinheim: VCH
Kim W, Doh S-J, Park Y-H, Yun S-T 2007. Two-year Verlagsgesellschaft; 645 pp.
magnetic monitoring in conjunction with Martin, M.H., Coughtrey, P.J. 1982. Biological
geochemical and electron microscopic data of Monitoring of Heavy Metal Pollution: Land and
roadside dust in Seoul, Korea. Atmospheric Air. Applied Science Publishers, New York.
Environment 41, 7627-7641. Mashitha, P.M., Pise, V.I., 2001. Biomonitoring of air
pollution by correcting the pollution tolerance

index of some commonly ground trees of an urban Urban Forest Climate Project. USDA, Forest
area. Poll. Res. 20, 195–197. Service, Gen. Tech. Rep. NE-186, pp. 63–81.
McPherson, E.G., Nowak, D., Heisler, G., Grimmond, Nowak, D.J., Civerolo, J.C., Rao, S.T., Sistla, G.,
S., Souch, C., Grant, R., Rowntree, R. 1997. Luley, C.J., Crane, D.E., 2000. A modeling study
Quantifying urban forest structure, function, and of the impact of urban trees on ozone. Atmospheric
value: the Chicago Urban Forest Climate Project. Environment 34, 1601–1613.
Urban Ecosystems 1, 49–61. Nowak, D.J., Crane, D.E., Stevens, J.C., 2006. Air
McPherson, E.G., Scott, K.I., Simpson, J.R. 1998. pollution removal by urban trees and shrubs in the
Estimating cost effectiveness of residential yard United States. Urban Forestry and Urban
trees for improving air quality in Sacramento, Greening 4, 115–123.
California, using existing models. Atmospheric Nowak, D.J., Crane, D.E., Stevens, J.C., Ibarra, M.,
Environment 32, 75–84. 2002. Brooklyn’s Urban Forest, USDA, Forest
McPherson, E.G., Simpson, J.R., Peper, P.J., Xiao, Q. Service, Gen. Tech. Rep. NE-290, 109pp.
1999. Benefit-cost analysis of Modesto’s Nowak, D.J., Dwyer, J.F., 2000. Understanding the
municipal urban forest. Journal of Arboriculture benefits and costs of urban forest ecosystems. In:
25, 235–248. Kuser, J.E. (Ed.), Handbook of Urban and
Mièieta, K., Mur´ın, G. 1998. Three species of genus Community Forestry in the Northeast. Kluwer
Pinus suitable as bioindicators of polluted Academic/Plenum Publishers, New York, pp. 11–
environment. Water Air Soil Pollut. 104, 413–422. 25.
Misra, R., Behera, P.K., 1994. Poll. Res. 13, 203–206. Nowak, D.J., McHale, P.J., Ibarra, M., Crane, D.,
Mitchell, R., Maher, B.A. 2009. Evaluation and Stecens, J., Luley, C., 1998. Modeling the effects
application of biomagnetic monitoring of traffic- of urban vegetation on air pollution. In: Gryning,
derived particulate pollution. Atmospheric S., Chaumerliac, N. (Eds.), Air Pollution Modeling
Environment 43, 2095-2103. and its Application XII. Plenum, New York, pp.
Møller, P. et al. 2008. Air pollution, oxidative damage 399–407.
to DNA, and carcinogenesis. Cancer Letters 266, Nowak, D.J., Rowntree, R.A. (Eds.), Chicago’s
84–97. Urban Forest Ecosystem: Results of the Chicago
Monaci, F., Bargagli, R. 1997. Barium and other trace Urban Forest Climate Project. USDA, Forest
metals as indicators of vehicle emissions. Water Service, Gen. Tech. Rep. NE-186, pp. 63–81.
Air Soil Pollut. 1000, 89–98. Oberdo¨ rster, G., 2000. Toxicology of ultrafine
Monaci, F., Moni, F., Lanciotti, E., Grechi, D., particles: in vivo studies. Philosophical
Bargagli, R. 2000. Biomonitoring of airborne Transactions of the Royal Society of London, A
metals in urban environments: new tracers of 358, 2719–2740.
vehicle emission, in place of lead. Environ. Pollut. Oberdo¨ rster, G., Obero¨ rster, E., Oberdo¨ rster, J.
107, 321–327. 2005 Nanotoxicology: an emerging discipline
Moreno, E., Sagnotti, L., Dinare` s – Turell, J., evolving from studies of ultrafine particles,
Winkler, A., Cascella, A., 2003. Biomonitoring of Environ. Health Perspect. 113, 823–839.
traffic air pollution in Rome using magnetic Pal, A., Kulshreshtha, K., Ahmad, K.J., Yunus, M.,
properties of tree leaves. Atmospheric 2000. Changes in leaf surface structures of two
Environment 37, 2967–2977. avenue tree species caused by auto exhaust
Morris, W.A., Versteeg, J.K., Bryant, D.W., pollution. J. Environ. Biol. 21, 15–21.
Legzdins, A.E., Mccarry, B.E., Marvin, C.H., Pandey, D.D., Nirala, K., Gautam, R.R., 1999. Indian
1995. Preliminary comparisons between J. Environ. Ecoplan. 2, 43–46.
mutagenicity and magnetic susceptibility of Pandey, D.D., Sinha, C.S., 1991. Environ. Ecol. 9,
respirable airborne particulate. Atmospheric 617–620.
Environment 29, 3441– 3450. Pandey, J., Agrawal, M. 1993. Protection of Plants
Muxworthy, A.R., Schmidbauer, E., Petersen, N. Against Air Pollutants: Role of Chemical
2002. Magnetic properties and Mössbauer spectra Protectants. J. Environ. Manage. 37, 163–174.
of urban atmospheric particulate matter: a case Pandey, S.K., B.D. Tripathi, S.K. Prajapati, V.K.
study from Munich, Germany. Geophysical Mishra, A.R. Upadhyay, Rai, P. K., Sharma, A.P.,
Journal International 150, 558-570. 2005. Magnetic properties of vehicle derived
Nieminen, T., Helmisaari, H.-S. 1996. Nutrient particulates and amelioration by Ficus infectoria: a
translocation in the foliage of Pinus sylvestris L. keystone species. Ambio: A Journal on Human
growing along a heavy metal pollution gradient. Environment 34 (8), 645-647.
Tree Physiol. 16, 825–831. Pope, C.A., Hansen, M.L., Long, R.W., Nielsen, K.R.,
Nowak, D.J., 1994. Air pollution removal by Eatough, N.L., Wilson, W.E., Eatough, D.J., 2004.
Chicago’s urban forest. In: McPherson, E.G., Ambient particulate air pollution, heart rate
Nowak, D.J., Rowntree, R.A. (Eds.), Chicago’s variability, and blood markers of inflammation in a
Urban Forest Ecosystem: Results of the Chicago panel of elderly subjects. Environmental Health
Perspectives 112, 339–345.

Rajput M, Agrawal M. 2005 Biomonitoring of air Singh, P., Sthapak, J., 1999. Reduction in protein
pollution in a seasonally dry tropical suburban area contents in a few plants as indicators of air
using wheat transplants. Environ Monit Assess pollution. Poll. Res. 18, 281–283.
101, 39–53. Singh, R.B. 2000. Impact of stone crusher dust
Randolph, J. 2004. Environmental Land Use Planning pollution on tomato (Lycopersicum esculantum) in
and Management. Island Press, Washington, DC, the Sonbhadra District of U.P. J. Environ. Pollut.,
664pp. 7, 235-239.
Rautio, P., Huttunen, S., Lamppu, J. 1998a. Effects of Singh, R.B., Das, U.C., Prasad, B.B., Jha, S.K., 2002.
sulphur and heavy metal deposition on foliar Poll. Res. 21, 13–16.
chemistry of Scots pines in Finnish Lapland and Somashekar, R.K., Ravikumar, R., Ramesh, A.M.,
on the Kola Peninsula. Chemosphere 36 (4–5), 1999. Impact of granite mining on some plant
979–984. species around quarries and crusher of Bangalore
Rautio, P., Huttunen, S., Lamppu, J. 1998b. Seasonal district. Poll. Res. 18, 445–451.
foliar chemistry of northern Scots pine under Szönyi M, Sagnotti L, Hirt AM 2008. A refined
sulphur and heavy metal pollution. Chemosphere biomonitoring study of airborne particulate matter
37 (2), 271–278. pollution in Rome, with magnetic measurements
Rautio, P., Huttunen, S., Lamppu, J. 1998c. Element of Quercus Ilex tree leaves. Geophysical Journal
concentrations in Scots pine needles on radial 173, 127-141.
transects across a subartic area. Water Air Soil Taha, H., 1996. Modeling the impacts of increased
Pollut. 102, 389–405. urban vegetation on the ozone air quality in the
Risom, L., Møller, P., Loft, S. (2005) Oxidative South Coast Air Basin. Atmospheric Environment
stress-induced DNA damage by particulate air 30, 3423–3430.
pollution. Mutat. Res. 592 119–137. Taha, H., 1997. Modeling the impacts of large-scale
Rosenfeld, A.H., Romm, J.J., Akbari, H., Pomerantz, albedo changes on ozone air quality in the South
M. 1998. Cool communities: strategies for heat Coast Air Basin. Atmospheric Environment 31,
islands mitigation and smog reduction. Energy and 1667–1676.
Buildings 28, 51–62. Tiwari, T.N., Patel, M.K., 1993. Indian J. Environ.
Saldiva, P.H., Clarke, R.W., Coull, B.A., Stearns, Prot. 13, 93–95.
R.C., Lawrence, J., Murthy, G.G., Diaz, E., Trivedi, M.L., Singh, R.S., 1995. Reduction in protein
Koutrakis, P., Suth, H., Tsuda, A., Godleski, J.J. contents in a few plants as indicators of air
2002. Lung inflammation induced by concentrated pollution. Poll. Res. 14, 269–273.
ambient air particles is related to particle Ukonmaanaho, L., Starr, M., Hirvi, J.-P., Kokko, A.,
composition. American Journal of Respiratory and Lahermo, P., Mannio, J., Paukola, T., Ruoho-
Critical Care Medicine 165, 1610-1617. Airola, T., Tanskanen, H. 1998. Heavy metal
Samal, A.K., Santra, S.C., 2002. Indian J. Environ. concentrations in various aqueous and biotic media
Health 44, 71–76. in Finnish Integrated Monitoring catchments.
Saquib, M., Ahmad, Z., Zaheer, A., Malabar, A.A., Boreal Environ. Res. 3, 235–249.
1992. J. Environ. Biol. 13, 145–148. Ulrich, B. 1984. Eff ects of air pollution on forest
Schoket, B. 1999. DNA damage in humans exposed ecosystems and waters: The principles
to environmental and dietary polycyclic aromatic demonstrated at a case study in Central Europe.
hydrocarbons. Mutation Research 424, 143–153. Atmos. Environ. 18, 621–628.
Scott, K.I., McPherson, E.G., Simpson, J.R., 1998. Urbat M, Lehndorff E, Schwark L. 2004.
Air pollutant uptake by Sacramento’s urban forest. Biomonitoring of air quality in the Cologne
Journal of Arboriculture 24, 224–233. conurbation using pine needles as a passive
Senapati, M.R., Misra, P.K., 1996. Impact of auto sampler —part 1: magnetic properties. Atmos
exhaust lead pollution on vegetation. Poll. Res. 15, Environ 38, 3781–92.
109–111. Vyas, D., Krishnayya, N.S.R., Bedi, S.J., 1991. Indian
Shrivastava, N., Joshi, S., 2002. Effect of automobile J. Environ. Health 33, 260–263.
air pollution on the growth of some plants at Kota. Zafar, M., 1985. Poll. Res. 4, 59–61.
Geobios 29, 281–282. Zhu, Y., Kuhn, T., Mayo, P., and Hinds, W. C. 2006.
Shu J, Dearing JA, Morse AP, Yu L, Yuan N 2001. Comparison of daytime and nighttime
Determining the sources of atmospheric particles concentration profiles and size distributions of
in Shanghai, China, from magnetic and ultrafine particles near a major highway. Environ.
geochemical properties. Atmospheric Environment Sci. Technol. 40, 2531–2536.
35, 2615-2625
Simonich, S., Hites, R., 1994. Importance of
vegetation in removing polycyclic hydrocarbons
from the atmosphere. Nature 370, 49–51.

ISBN 978-81-921121-0-7
DISASTER MITIGATION AND MANAGEMENT FOR FLOOD

CONTROL USING GEOINFORMATICAL TOOLS
M.B.Kumthekar
Professor and Head, Civil Engineering Department, Government College of Engineering, Karad
P.K.Deshpande
Asst. Professor, Geology Department, Walchand College of Engineering, Sangli
R.D.Padhye
Lecturer, Applied Mech. Deptt. Walchand College of Engineering, Sangli
S.B.Thalange
P.G.Student,Civil Engineering Department, Government College of Engineering, Karad
ABSTRACT: The aim of this work is to provide the necessary disaster mitigation and management plans along
the banks of river for the flood management and mitigation. The area has seasonal monsoon rain which causes
flood and spread of water and affect the same settlement area along the bank length, hence there is need of better
management and mitigation. Sustainable developments and management of the area is influenced by topography
of the area and geomorphologic features. The flood spread and its effect to affected area can be studied out by
using Geoinformatical tools by which we can generate the flood effecting area layers and can propose or suggest
the required sustainable infrastructure along the effecting area. The manual estimation of geomorphic
parameters is tedious and cumbersome process. However through this attempt of work, it has been found that
integration of RS and GIS allows reliable, most accurate and updated database tool for handling special data,
very useful in deriving geomorphometric parameters which are essential for planning of future infrastructural
constructions, suggested here, for sustainable development, mitigation and management. In the present study
gemorphometric analysis has been carried out and the RS Data have been interpreted. The Geoinformatic tools
are to be used as valuable and powerful decision support system for effective national planning, efficient
resource allocation and infrastructure development planning and management. The future scope of this work is
to develop the sustainable structure which will direct helps in controlling, mitigating and managing the flood.
The effect has been to suggest the flood related construction activity that may mitigate the frequent flood hazard
condition along the banks of Koyana river near Sangamnagar, in district Satara of Western Maharashtra.
INTRODUCTION graphically displayed for analysis. By overlaying or

Flood is a natural disaster in this country. However intersecting different geographical layers, flood prone
human activities in many circumstances change flood areas can be identified and targeted for mitigation or
behavior. Activities in the flood plain and catchment stricter floodplain management practices.
such as land clearing for urbanization and agriculture. Remote Sensing can be very effective for flood
At present, one of the ways to study and to management in the following ways as
understand the flood behavior, is by generating the Detailed mapping that is required for the production
flood extent or flood risk map. However, to analysis a of hazard assessment maps and for input to various
river system requires tremendous amount of data such types of hydrological models.
as rainfall distribution, river properties and most Developing a larger scale view of the general flood
important the flood plain topography. Geoinformatic situation within a river basin with the aim of
tools are avaible to handle the processing of such identifying areas at greatest risk and in the need of
problem as and also for generating flood layers for immediate assistance.
various scenarios. Further analysis such as flood Remote sensing and GIS technique has successfully
damage assessment can be carried out for mitigation, established its application in following areas of flood
infrastructural planning and management by making management such as flood inundation mapping, flood
suggestive proposals for sustainable infrastructural plain zoning and river morphological studies.
development in flood hazard zones.
Morphology Study
River morphology is concerned with the
FLOOD MANAGEMENT SYSTEM
structure and form of rivers including channel
Geoinformatic Tools
configuration, channel geometry, bed form and profile
Advancements in the Remote Sensing technology and
characteristics. Various flood control structural
the Geographic Information Systems (GIS) help in
measures such as construction of embankments,
real time monitoring, early warning and quick damage
gabion wall, retaining wall, raising of villages,
assessment of flood disasters. A Geographic
selective dredging etc. have been implemented in past
Information System is a tool that can assist floodplain
to reduce the impact of the flood disaster on human
managers in identifying flood prone areas in their
life and property. It is essential to monitor the
community. With a GIS, geographical information is
embankments regularly to identify the vulnerable
stored in a database that can be queried and

reaches. Conventional method of river surveying is Geomorphologic and Geotechnical setup. It is known
time consuming during flood condition and to be the worst affected area by the several natural
expensive. Most of the flood prone rivers in India calamities like flood, drought, disastrous seismicity;
change their course after every flood wave eroding in chronological past and affected by severe monsoon
river banks. Satellite remote sensing based which have caused many injuries, diseases, epidemics
morphological studies are quite useful in following even the mass casualties, those adding to loss of
aspects: health, economy and social burden on the
To identify the changes in river course over a time development of Maharashtra.
period.
To identify the erosion prone areas along the river The Koyana Hydropower project which is associated
course. with Koyana Dam and a typical “Shivajisagar
To study the efficacy of flood management Reservoir” having about 70 Km North-South
structures. orientation in its catchment sited on the forehead of
The river configuration and flood control Western Ghats. This project is also called as
works maps can be effectively used to identify the “Bhagyalaxmi” of Maharashtra state as its water is
vulnerable river reaches and status of the flood control being used not only for electric power generation but
embankments/spurs so that necessary measures can be also for irrigation and water supply.
taken accordingly avoid disaster. The bank erosion
maps can be used for planning bank protection works. During the South-West monsoon Koyana catchment
The study of river configuration will be useful to receives very heavy precipitation. Its typical North-
understand the behavior of the river and can be used South orientation allows the heavy over land flow
for laying physical models. collection in the catchment area. Its results in very
rapid inflow and increase in the rate of filling of
FLOOD MANAGEMENT reservoir. And that is why in early monsoon period;
Providing absolute protection to all flood prone areas even water level is much as below the HFL there is
for all magnitude of floods of different probabilities of need of water discharge from dam, several time. It
occurrence is neither practically possible nor always results in the flood situation on banks of
economically viable. Hence, a practical approach in Koyana River. In the settlement like Sangamnagar
flood management is to provide a reasonable degree and Karad and in continuation affects area of Sangli
of protection, against flood damage at economic costs and Kolhapur District.
through a combination of structural and non-structural
measures. Geoinformatics or Remote Sensing and GIS technique
Structural Measures has now emerged as a very fast, accurate and reliable
The main thrust of flood protection programme tool to develop the sustainable infrastructure and
undertaken in India so far has been in the nature of plans for the disaster mitigation and management
structural measures like: programmes. There is need now of judicious
(i) Embankment, flood walls, construction management that will help the new
(ii) Gabion wall, infrastructural development based on the RS and GIS
(iii) Retaining wall, for disaster mitigation and management plan. The
(iv) Dams and reservoirs generated spatial and nonspatial data and information
(v) Diversion of flood waters. regarding river basin, population, existing
Non-Structural Measures infrastructure, inundation and damage assessment etc.
Flood plain zoning will prove to be useful in future planning. The work
Flood proofing of increasing the height of bridges, construction of
Flood plain zoning aims at disseminating Gabion wall to avoid erosion and water spread.
information on a wider basis so as to regulate Rehabilitation of flood prone population etc. can be
indiscriminate and unplanned development in flood planned and may be proposed on the basis of present
plains and is relevant both for unprotected as well as study. This will help to the civil engineers working in
protected area. At present the watershed development Krishna river basin down stream to the confluence of
and management programmers are being implemented river’s Krishna and Koyana at Karad.
under centrally sponsored schemes of Ministry of
Agriculture, Ministry of Rural Areas and Employment Sangamnagar Bridge is located at latlon 17021’05.9”
& Ministry of Forests. Flood proofing measures help N to 73046’10.18” E. Here at Sangamnagar, the bridge
greatly in mitigation of distress and provide is old and submersible type as shown in figure ‘01’.
immediate relief to the population. When there is discharge of more then 25000 cusecs
bridge under goes the water. On right side and on
upstream side of bridge the Kajoli Nadi meets to
CASE STUDY
Koyana River and there is a large deposit of fine to
ost of the part of Western Maharashtra that is
course unsorted Pleistocene deposit on right bank of
best known to be the Deccan volcanic province as a
Koyana and also downstream and upstream of bridge
typical Geographical location, Climatic condition,

as shown in figure ‘02’ and figure ‘03’. The
Sagamnagar bridge which is next to Helwak, on left
bank of Koyana river, it has precipitous slope and
state highway-78 passing. The Sagamnagar bridge
connects nearly 60-65 villages on right bank, if the
bridge goes under water it breaks the communication
with nearly all 60-65 villages; the important villages
are like Kadoli, Chafer, Nirgam, Rohine, Mirgaon,
etc. The transportation through boat is the only way to
go on left bank side of Koyana River during flood
condition. The boat facility is available at Maneri
only. The rejuvenation deposits are found in river
channel and along the banks. The farming is done on Fig.02
right bank by locals, but large area of farm lands has
been washed away during the flood of 2005 along the
bank. As the ground on right banks has gently slope
there is sudden spread of flood water. The population
of 20000 to 25000 is been affected during the flood.
The building shown in figure ‘04’ has been collapsed
due to water pressure, it is located on left side bank of
river. The bridge is too old and height is also less
which is submerged many time when discharge more
then 25,000 cusecs. The road bed and pier foots have
eroded during the flood and bridge rail has been
broken along some length of bridge. There is need of
proposing the new bridge parallel to old bridge with Fig.03
sufficient height of length 250 meter because during
flood the local people on right bank have to go Maneri
village for travelling towards left bank by the boat and
thus the new bridge is proposed. The ground slope is
level slope to precipitous slope on right bank of
Koyana river which have the chances of water spread
and soil erosion. Therefore to avoid this situation
there is a need of proposing the sustainable gabion
structure along bank length. Artificial Pond is
proposed at Kajoli River by blocking the water at the
confluence of Kajoli River and Koyana River which
will create the seasonal business of fresh fish and
Fig.04
create revenue for local community. Thus, there is a
sufficient scope of pisciculture at this particular
stretch of Koyana river.
METHODOLOGY OF GENERATING LAYERS
The Gabion wall is the Sustainable structure which
The Software ILWIS, develop by ITC Netherlands
prevents water to spread as well as from soil erosion
has been used in present study. After proper
and washout of farm land on right bank of Koyana
Geocoding and Georeferancing of IRS 1D PAN Data
river is being proposed which help for flood
the sub-map consisting of the study area was
mitigation and management. The height of gabion
extracted. With reference to flood hazard zonation
wall should be six meter and its length 220 meter on
map provided by the Koyana dam authority the flood
upstream side and 500 on downstream side of bridge.
hazard zonation map has been digitized in value
domain for 2,55,000 cusecs. Similarly the segment for
the minimum discharge of 2,200 cusecs have also
been digitized.
After interpolation this segment layer the flood hazard
zonation map was ready that was further sliced into
‘8’ different flood hazard zones as shown in Figure
‘07’. Using the ASTER GDEM as the DTM the
steropair of PAN imagery have been developed which
is viewed as Anaglyph Figure ‘06’. After adding
layer, of flood zone’s, over the Anaglyph the
Fig. 01 assessment about the inundation area, vulnerability

assessment of the flood inundated may possible that
has be carried out.
LANDSAT seven TM Data has also been used for the
supervised classification of flood hazard zone. Band-
4, Band-3, and Band-2 where used to develop the
standard FCC of study area. Also the by using a Band
rationing technique the area was classified into water,
vegetation and barrier land Figure ‘08’. This layer can
be studied along with the flood zonation map and the
Elevation Data derived from ASTER GDEM Figure
‘09’. Finally Slope Aspect Sliced Map and Young’s
Slope Classification was done for detail study as
shown in figure ‘10’ and figure ‘11’.
After the brain storming on the generated Remote
Fig.07
Sensing and GIS layers the further construction
activities were digitized and quantified in software
environment. After considering various layer
combinations it is observed that there is a need of
application of flood management and mitigation
techniques. That will not only reduce the flood hazard
but also support the sustainable infrastructure
development along the river bank. The suggested new
bridge and Gabion wall as shown in Figure ‘12’ will
help in infrastructural development along with flood
management and mitigation plans. Suggesting gabion
wall help to resist the sever soil erosion and protect
the agriculture land along side river near
Sangamnager along with acting as barriers for the
spread of water for mitigation. Fig. 08
Fig.05
Fig. 09
Fig.06

various GIS segment and polygons. In the present
study the flood hazard zonation map has been
correlated with the 3D Anaglyph of PAN Data and
other layer’s derived from ASTER Data.
On the bases of these study the infrastructure
development plan and flood related management and
mitigation activity have been proposed. It is, thus,
concluded that infrastructural plans and mitigation
activities can be done with the help of Geoinformatic
tools which is a very useful, efficiency and effective
Data Base for planning engineer.
Fig. 10 REFERENCES:
A.M. Zende, P.K. Deshpande, M.B. Kumthekar, K .S.
Wagh, “Sustainable Developments Proposed In
Kerala River Watershed Of Western
Maharashtra, Based On Morphometric Analysis
& Remote Sensing Data Interpretation, In GIS
Environment” FACE 08, 2nd National
Conference on Focusing on Advances in Civil
Engineering, TKM College of Engineering,
Kollam, Kerala, India.21 – 23rd February 2008.
Deshpande P.K., Kumthekar M.B. (2004)
‘Geomorphometric analysis for sustainable
water resources planning and management using
R.S. & GIS’. PUBSWRPM, BITS Pilani.Oct-30,
pp. 199-208.
Fig. 11 R.K.Khanna, C.K.Agrawal, Pravin Kumar, “Remote
Sensing And GIS Applications In
FloodManagement” Central Water
Commission,New Delhi.
www.astergdem.com.
www.itc.nl.
Fig. 12
RESULT AND DISCUSSION
For the disaster mitigating and management which is
most important activity for sustainable development is
easily possible with the help of Geoinformatical tools.
Integration of Remote Sensing along with
Geographical Information System (GIS) allows most
reliable, accurate and updated database for
infrastructure development. It has really been
emerged as a very powerful tool helpful in
formulation of proposals of Sustainable infrastructural
planning, mitigating and management for
development in various types of land areas.
ILIWS (Intergrated Land and Water Information
System) is related the very effective software that is
able handle the Remote sensing Data as well as

ISBN 978-81-921121-0-7
DYNAMIC ANALYSIS OF LAMINATED COMPOSITE FOLDED PLATE

FOR DIFFERENT BOUNDARY CONDITIONS
Sourav Chandra
PG student,Department of Civil Engineering,Jadavpur University, Kol-32, e-mail: chandra.sourav2009@gmail.com
Sreyashi Das
Asst. Professor, Department of Civil Engineering,Jadavpur University , Kol-32,e-mail: palsreyashi@gmail.com
A. Guha Niyogi
Professor, Department of Civil Engineering,Jadavpur University, Kol-32, e-mail: agn_ju@yahoo.com
ABSTRACT: A 9-noded Lagrangian plate bending finite element that incorporates first-order transverse shear
deformation and rotary inertia is used to predict the free and forced vibration response of laminated composite folded
plate structures. A 6 x 6 transformation matrix is applied to transform the system element matrices before assembly. An
additional drilling degree of freedom is appended with the usual five degrees-of-freedom per node to use the
transformation. The finite element results presented here show good agreement with the available semi-analytical
solutions and finite element results. Parametric studies are presented for free and forced vibration response for
laminated composite folded plates, with reference to crank angle and different boundary conditions.
INTRODUCTION u(x,y,z) = u0(x,y)+zθx(x,y)

To civil engineers laminated composite folded plate v(x,y,z) = v0(x,y)+zθy(x,y)
structures are of particular interest as roofing material due w(x,y,z) = w0(x,y)
to their light weight and extensive roof-spanning and
capability as compared to flat plates. Folded plates are θx = w,x + φx
implemented in cladding factory roofs also. Sometimes θy = w,y+φy (1)
the structure may have to resist high wind or seismic load
too. The conventional concrete roofs provide necessary Where u, v and w are the displacements of a general point in
strength and stability but the weight is high. Low cost the x, y and z directions, respectively. The parameter u0, v0 are
laminated composites comprising of epoxy matrix are the in plane displacements and w0 is the displacement of a
very efficient in withstanding the dynamic loads due to point on the middle plane along z direction. θx and θy are the
their superior weight-specific strength and stiffness.They total rotations of the normal about y and x axis respectively,
are non-biodegradable and hence, maintenance cost is whereas φx and φy are the constant average shear
low. Published works on composite folded plates are deformations about y- and x-axes.
scanty, and hence this research.
In the present analysis the first order shear deformation FINITE ELEMENT FORMULATION:
theory (FSDT) is adopted to analyze the laminated In this study, nine noded Lagrangian rectangular plate
composite folded plates. A shear correction factor 5/6 has elements having five degrees of freedom per node are used
been applied to take account of non uniform distribution for the finite element solution of the laminated folded plates.
of transverse shear strain along the thickness of the The interpolation function of the displacement field (δ) is
laminate. In the free vibration analysis, the subspace defined as
iteration technique [1] is adopted for extraction of the
⎧ ⎫
eigen pairs. In the forced vibration analysis, the damping ⎪ ⎪
is neglected. Newmark’s explicit integration technique [1] = =∑ ` (2)
is adopted for the transient analysis. ⎨ ⎬
⎪ ⎪
Research on dynamic analysis of laminated composite ⎩ ⎭
folded plate structures are reported by Guha Niyogi et al Where, Ni are the Lagrangeian interpolation functions[1], δi is
[2] using FSDT. Lee et al. [3] made use of third order the nodal displacement u0i, v0i, woi, θxi, and θyi and n is the
plate deformation theories to solve the laminated plate total number of nodes per element.
problems. S. Pal and A. Guha Niyogi [4] studied the
effect of free vibration and transient response for The stiffness matrix of the plate element assumes the form
cantilever stiffened laminated composite and sandwich K 
T
folded plate with different arrangements of stiffeners e   B  D BdA

Ae
(3)
using folded plate formulation. The review of literature Where,
reveals that dynamic behavior of laminated folded plate
for free and transient vibration for various boundary
   B  i . (4)
conditions has not been investigated until now and hence {} being the strain vector, and {i} the nodal displacement
this study is carried out. vector. [B] is the strain-displacement matrix, and [D] is the
stiffness matrix given by
THEORETICAL FORMULATIONS:
The displacement field for FSDT can be expressed as

 Aij Bij 0  corresponding to the θz terms are set to zero, while a very
small positive number, about 1000 times smaller than the
D   Bij Dij 0  (5)
smallest leading diagonal, is introduced in the stiffness matrix
 0 0 Alm  at the corresponding leading diagonal term [1]. Similarly
Where, 45×1 load vector is expanded to 54×1 by inserting zero at
N zk k corresponding locations.
Aij , Bij , Dij    Q  (1, z, z Governing equation for undamped free and forced vibration
2
ij
)dz, i, j  1, 2,6 (6a)
k 1 z
k 1 are summarized as follows
N zk [ ′] ̈ + [ ′]{ } = 0
and, Akm     (Qlm ) k dz, l , m  4,5,   5 / 6 (6b) and
k 1 z
k 1
[ ′] ̈ + [ ′]{ } = [ ′] (10)
Here, Qij are the elements of off-axis stress-strain
relations. Qijk relates stresses and strains in kth layer by
the relation,
NUMERICAL RESULTS AND DISSCUSSIONS:
i k = Qijk jk. i,j=1,2,6. A computer program in FORTRAN, based on the finite
1, 2, and 6 denote x, y and xy respectively and 1, 2, element model described in the earlier section, is used to get
and 6 denote x, y and xy, respectively. the dynamic solution of laminated composite folded plate
Whereas, structures. Free and forced vibration responses are shown for
lk = Q lmk mk, l, m = 4,5 and  is the shear correction variation in crank angle and different boundary conditions.
factor taken as 5/6. Symmetric and anti-symmetric cross ply laminate is used for
The mass matrix of the plate element is given by the analysis. The validation of the code developed and the
(7) mesh convergence study has been provided elsewhere [4].
M e   N   N dA
T
Ae
[ρ] being the density matrix functions. The integration in Case study 1: Effect of Crank Angle and boundary
every case is carried out over the area of the plate conditions on Free Vibration of Folded Plate
element. Generally, a 3-point Gauss quadrature is adopted Free vibration analysis of a single fold plate structure, made
to compute the bending stiffness of the elements, whereas up of E-glass Epoxy composite has been carried out for
a 2-point integration is applied to calculate the shear variation in crank angle and boundary conditions. The
stiffness and mass matrix. property of the E-glass Epoxy material is given below.
E1=60.7 ×109 Nm-2, E2=24.8×109 Nm-2, G12 = G13 =
FINITE ELEMENT TRANSFORMATION: G23=12.0×109 Nm-2, ν12=ν21=0.23, ρ=1300Kgm-3
A transformation [2] is prepared to relate local The length of the plate is taken as 1.8 m, each inclined side
displacement u, v, w, θx, θy, θz and global displacement u′, being 1.2m [Figure 3].
v′, w′, θ′x, θ′y, θ′z with positive direction of displacements
and rotations, shown in Figures 1 and 2
{ } = [ ]{ ′ } (8)
1.2 m 1.2 m
y,v y‫׳‬,v‫׳‬
x,u
α
z,w z‫׳‬,w‫׳‬ 1.8 m
Figure 1: Transformation of translations from local (x) to
global (x‫ )׳‬axes Figure 3: Geometry of ONE fold plate structure
y,θy
y‫׳‬,θ‫׳‬y
x‫׳‬,θ‫׳‬x The thickness used in the analysis is 0.02 m. The crank angle
α is varied as 90o, 120o and 150o. (0o/90o/0o/90o) fibre angle is
used for this study. Four different boundary conditions used
x,θx for the analysis are as follows,
z,θz 1. All four edges are clamped (CCCC)

z‫׳‬,θ‫׳‬z 2. One inclined edge clamped and others three edges
Figure 2: Transformation
The element matrices are of rotationsinfrom
expressed local
global (x) to
coordinate free (CFFF)
global
system(x‫)׳‬from
axeslocal coordinate using transformation matrix 3. Two straight edges clamped and two inclined edges
as free (CCFF)
[K′]e = [T]T [K]e [T], 4. Two inclined edges clamped and two straight edges
[M′]e = [T]T [M]e [T], free (FFCC).
{ ′} = [ ] { } (9)
Before applying the transformation, the 45×45 stiffness First five natural frequencies of single folded plate structure
and mass matrices are expanded to 54×54 by inserting θz for crank angle α = 90°, 120° and 150° with four boundary
drilling degree of freedom at each node of a finite conditions are tabulated in Table 1.
element. In mass matrix the off-diagonal terms
4 50.98 50.93 50.67
5 56.05 56.11 56.13
CCFF 1 48.67 48.69 48.70
2 57.75 57.78 57.78
Case study 2: Effect of Fibre Orientation on free
3 73.70 73.52 72.77
Vibration of Folded Plate
Table 2 shows the natural frequencies of one Fold plate 4 80.36 80.13 79.15
structure for symmetric cross ply laminate (0o/90o/90o/0o ). 5 92.55 92.58 92.59
All other data remains same as case study 1. From the FFCC 1 45.85 45.85 45.85
table we can see that for cases CCCC or, CCFF, the 2 47.75 47.74 47.68
natural frequency decreases when compared to 3 82.99 83.01 83.02
(0o/90o/0o/90o), whereas, for other two cases, (FFCC and 4 102.40 102.12 100.73
CFFF), natural frequency increases for fibre angle 5 132.67 132.67 132.68
(0o/90o/90o/0o). Hence the dynamic response of laminated
composite folded plate does not only depend on the fibre
Case study 3: Transient responses of ONE fold plate
orientation, but also on the boundary conditions. Hence,
structure for different boundary conditions Figure 5
any generalized recommendation is very difficult without
represents comparison of transient response of the laminated
undergoing numerical experiments.
composite folded plate structure for a fixed crank angle α =
120o at the center point of the ridge line under uniformly
Table 1. Natural frequencies (Hz) of single folded
composite plate with fiber orientation of (0/90/0/90) applied step loading of intensity 1.0x105 N/m2, acting normal
to all the faces, outward and upward for various boundary
Boundary Mode Crank Angle conditions. The fibre angle is taken as (0o/90o/0o/90o). From
Conditions No. 90° 120° 150° the figure it is observed that, the deflection is higher for the
CCCC 1 79.62 79.71 79.76 case where two straight edges are clamped. The cases CCCC
2 103.95 103.76 102.77 and CCFF show a periodical nature of displacement in the
3 147.78 147.87 147.94 graph. This periodical nature is not much pronounced for
4 163.09 162.96 162.20 other two cases.
5 263.31 264.04 264.46
CONCLUSION
CFFF 1 10.48 10.48 10.49
This paper is concerned with the analysis of laminated
2 17.54 17.49 17.27 composite folded plate structures. For the numerical
3 43.14 43.15 43.17 simulation of the problem finite element technique has been
4 47.24 47.20 47.02 used. Various boundary configurations have been considered
5 62.86 62.99 63.06 in the present study. A set of new results for composite folded
CCFF 1 57.83 57.91 57.95 plate structures with various crank angles and boundary
2 65.61 65.70 65.75 conditions are presented. The first order shear deformation
3 87.54 87.25 85.99 theory is found to work well for folded plates where the five
regular degrees of freedom are appended with a sixth drilling
4 93.18 92.85 91.33
degree of freedom. It has been shown that dynamic response
5 94.88 95.00 95.07 of laminated composite ONE fold plate structure largely
FFCC 1 40.13 40.13 40.13 depends on the boundary condition. The crank angle does not
2 42.95 42.93 42.88 play any significant effect in free vibration analysis of ONE
3 85.75 85.85 85.90 fold plate structure. However, it is difficult to speculate mode
4 110.22 109.84 107.81 shapes without undergoing numerical experiments.
5 114.97 114.97 114.98
REFERENCES
Bathe, K. J., 1996, Finite Element Procedures, Prentice Hall
Table 2. Natural frequencies (Hz) of single folded of India Pvt. Ltd., New Delhi.
composite plate with fibre orientation of (0/90/90/0) Guha Niyogi, A. Laha, M. K.and Sinha, P. K. (1999). Finite
Element Vibration Analysis of Laminated
Composite Folded Plate Structures, Shock and Vibration, 6:
Boundary Mode Crank Angle
273–283.
Conditions No. 90° 120° 150° Lee and Lee (1995). Vibration Analysis of Anisotropic Plates
CCCC 1 76.68 76.70 76.70 with Eccentric Stiffeners, Comput. Struct, 57(1):
2 95.16 94.99 94.29 99–105
3 159.39 159.41 159.42 Pal. S, Guha Niyogi. A, Application of folded plate
4 169.83 169.68 168.99 formulation in analysing stiffened laminated composite
5 229.76 230.00 230.11 and sandwich folded plate vibration, Journal of
CFFF 1 11.08 11.09 11.09 Reinforced Plastics and Composites, 27(7), pp. 693-710,
2 16.56 16.53 16.37 2008.
3 47.07 47.08 47.09
MODE 1 MODE 2 MODE 3
CCCC
CFFF
CCFF
FFCC
Figure 4: First three mode shape for laminated composite folded plate with crank angle =90o and fibre orientation =
(0/90/0/90) for different boundary conditions
Figure 5: Transient response of laminated composite folded plate(0/90/0/90) for various boundary conditions

ISBN 978-81-921121-0-7
CONCRETE BEAMS STRENGTHENED WITH NEAR SURFACE

MOUNTED AFRP RODS
V.G. Kalpana
Assistant professor, Department of Civil Engineering, Coimbatore Institute of Technology, Coimbatore-641014, Tamilnadu.
Email-kalpana@cit.edu.in
K. Subramanian
Professor and Head, Department of Civil Engineering, Coimbatore Institute of Technology Coimbatore-641014, Tamilnadu.
Email-hodcivil@cit.edu.in
ABSTRACT: Strengthening with fibre reinforced polymer (FRP) composites is a widely used method for
retrofitting or increasing load bearing capacity of structural members. FRP reinforcement is available in
different forms such as laminates, sheets, bars, strips, grids and prestressing tendons to serve a wide range of
purposes. This research work focuses on use of FRP bars as near surface mounted (NSM) reinforcement for
strengthening of concrete structures. Out of four steel reinforced concrete beams three beams were strengthened
with near surface mounted AFRP bars and one as control beam .All the beams were tested under monotonic
static two point loading. Three different diameters of 5.5mm, 7.5mm, and 10mm AFRP rods were used as NSM
rods. From the experimental results load deflection behavior and load carrying capacity of NSM strengthened
rods with that of control beams were investigated. The mode of failure and the pattern of cracks formed were
also observed and presented in this paper.
INTRODUCTION strengthened with NSM AFRP rods on the tension

The use of Near Surface Mounted (NSM) fibre side surface. All the beams were of 1200mm in length
reinforced polymer rods is a promising technology for with a rectangular cross section of 150mm x 100 mm
increasing flexural and shear strength of deficient as shown in Figure 1.
reinforced concrete structures. Among the various
strengthening techniques that have been developed
and applied to strengthen deteriorated RC structures, a
number of applications using FRP reinforcements
have significantly increased recently. FRP
reinforcements are bonded to concrete surfaces by
Fig. 1 Beam reinforcement and cross section details.
adhesives, experience debonding failure at the
The beams consists of 2nos 8mm diameter
interface between FRP reinforcements and concrete.
compression steel reinforcements, 2nos 10mm
Most research works focused on investigating the
strengthening effects and failure modes of Externally diameter tension steel reinforcements and 2legged
Bonded Reinforcements (EBR) system. The problem 8mm diameter steel stirrups of 80mm c/c spacing. The
flexural strengthening was done by making a groove
of premature failure of EBR system may be solved by
of width and depth equal to 1.5times the diameter of
increasing the interface between FRP and concrete
NSM rod at the bottom of the beam .Then groove was
which leads to the introduction of NSM technique.
filled partially with epoxy resin and NSM AFRP rod
The NSM system for concrete structure using steel
was mounted into the grooves as shown in Figure 2.
reinforcement already began in 1940s. However, the
The remaining space is filled with resin and allowed
corrosion of the steel reinforcement and the poor
the beam for curing at room temperature for 7 days.
bonding performance of the grouting material needs
alternate as NSM +FRP system. Three diameters such as 5.5mm, 7.5mm and 10mm
NSM AFRP rods were used in this study.
RESEARCH SIGNIFICANCE
Table 1 Particulars of test specimens
This paper reports the test results of an experimental
investigation carried out on reinforced concrete beams Sl. Strengthening
Beam ID Type of specimen
strengthened in flexure using NSM AFRP Rods. This No Pattern
study focused on evaluating the effect of NSM 1 CB Control beam -
reinforcements on strength gaining. The modes of AFRP rod 5.5mm
failure of strengthened beams and the load deflection 2 AFRP5.5 NSM
diameter 1no.
behaviour are also examined in this study. AFRP rod 7.5mm
3 AFRP7.5 NSM
diameter 1no.
EXPERIMENTAL OVERVIEW AFRP rod 10mm
4 AFRP10 NSM
diameter 1no.
Test specimens
Totally four RC beams listed in Table 1 were used in
this study, in which one as control beam and three are

and its propagation are recorded. The load at initial
crack, ultimate load, type of failure etc., were also
carefully observed and recorded.
Fig.2 Beams strengthened with NSM rods
Materials for Concrete Beam

The Portland Pozzolana Cement conforming to IS:
4031 was used for the preparation of test specimens.
The fine
aggregate used in this experimental investigation was
Fig.3 Experimental setup
natural river sand confirming to zone II of IS: 383 –
1970. Crushed granite aggregates particles passing
ANALYSIS OF TEST RESULTS
through 20mm and retained on 4.75mm I.S sieve was
Experimental results presenting the relation load-
used as natural aggregates which met the grading
deflection at midspan of NSM reinforced beam and
requirement of IS: 383 – 1970. Fe415 TMT bars are
control beam are given in Figure 4.It can be observed
used for steel reinforcements. The mix proportion for
that the maximum load achieved by NSM
M30 grade of concrete is 1: 0.98: 2.38 with w/c ratio
5.5mm,7.5mm and 10 mm diameter AFRP were 81%,
of 0.385.The properties of AFRP rod which is used as
105% and 69% higher than unstrengthened beam. The
strengthening material is tabulated in Table 2.The
stiffness of strengthened and control beams are same
adhesive used to bond the AFRP rod to the beam is
until the formation of first crack. The failure occurred
epoxy and its properties are shown in Table 3.
due to splitting of epoxy resin and concrete as shown
in Figure 5.The comparison chart for ultimate loads of
Table 2 Properties of AFRP Rod
four beams was shown in Figure 6.
Tensile
Ultimate Tensile
Diameter Elastic
load strength
(mm) modulus
(kN) (Mpa)
(Gpa)
5.5 28.47
7.5 57.83 60 1400
10 97.86
Table 3 Properties of Epoxy

Property, Unit Resin Hardener
Araldite
Type Aradur, 140
GY 257
Commercial form Liquid Liquid
Quantity, (pbw) 100 50 Fig.4 Load deflection for control beam and NSM
Density at 25oC AFRP strengthened beams
1.15 0.98
(g/cm3)
Specific gravity 1.80 2.00
Experimental Setup
The flexural capacity of the RC Beam specimen was
obtained by means of two point bending test; beams
were being simply supported over a span of 1.2m.
The load was applied by means of a hydraulic jack
and distributed through steel I-section. The test setup Fig.5 NSM failed beams
is shown in Figure.3. At each load increment the
observations such as deflection, crack development

Fig.6 Ultimate load of tested beams
CONCLUSIONS
The present study was to determine the effect of NSM
strengthening technique for beams. Performance of
NSM technique was found to be very effective with
no occurrence of delamination, debonding or fracture
of FRP. It has been seen that NSM specimens utilized
the AFRP reinforcement more efficiently.
REFERENCES
De Lorenzis.L., Nanni.A., and La Tegola.A., (2000)
Strengthening of Reinforced Concrete Structures
with Near Surface Mounted FRP Rods , bibl.
International Meeting on Composite Materials,
PLAST 2000, Milan, Italy, May 9-11.
Jung.W.T., Park.Y.H., Park.J.S., Kang.J.Y., and
You,Y.-J. (2005), Experimental Investigation on
Flexural Behavior of RC Beams Strengthened by
NSM CFRP Reinforcements, ACI special
publications, SP-230-46, 1, 795-806
Kishi.N., Mikami.H., Kurihashi.Y. and Sawada.S.,
(2005), Flexural Behavior of RC Beams
Reinforced with NSM AFRP rods, Proceedings of
the International Symposium on Bond Behaviour of
FRP in Structures, BBFS 2005, Chen and Teng
(eds) ,2005 International Institute for FRP in
Construction, 337-342.
Ranković.S., Folić.R. and Marina
Mijalković.M.,(2010), Effects of RC Beams
Reinfocement using Near Surface Mounted
Reinforced FRP Composite, Facta Universitatis
Architecture and Civil Engineering, 8(2), 177 –
185.
Yost. J.R., Gross. S.P., Dinehart. D.W. and
Mildenberg. J.J., (2007), Flexural Behavior of
Concrete Beams Strengthened with Near-Surface-
Mounted CFRP Strips, ACI Structural Journal,
104(4), July-August, 430-437.

ISBN 978-81-921121-0-7
DAMAGE ANALYSIS OF STEEL STRUCTURES

Pranav P. Pande
IV Sem, M.E. (CASDD),Ujjain Engineering college, Ujjain , Email-prof.pande@gmail.com
ABSTRACT:In the most general terms damage can be defined as changes introduced into a system that adversely
effect the current or future performance of that system. Implicit in this definitions is the concept that damages is not
meaningful without a comparison between two different states of the system one of which is assumed to represent
the initial and often undamaged state. This discussion is focused on the study of damage identification in structural
system and it’s effect or impact on the structure. The need for quantitative global damage detection methods that can
be applied to complex structures had led to the development and continued research of methods that examine change
in the vibration characteristics of the structure. such a system has also been referred as a NDE methodology. During
the past decade, a great deal of research has been conducted in the area of nondestructive damage evaluation of
structural systems via changes in their vibration characteristics. In many studies, the resonant frequencies were used
to identify damage and estimate the amount of damage i.e. size and impact on performance of structure. In the last
years several methods have been developed that use the changes in the modal shapes and vibration frequencies of
a structure as data to evaluate the damage in their structural elements. These methods have the advantage of not
requiring the direct exhibition of these elements and of being able to visualize the complete structure if it is properly
instrumented. In consequence, it is possible to obtain a reduction in the time and cost of the evaluation, as well as a
decrease in the impact of operation of the structure. For the localization and the calculation of the magnitude of the
structural damage, these methods require the analytical model of the structure. These methods are denominated
methods of detection of damage based on models.
INTRODUCTION Cross bracings was provided between the floor

To detect the damages in the different types of steel beams. Cross section of the bridge is shown in figure
structures through computational analysis, different Bridge is made up of three identical section each
types of site and laboratory model of different section have three span the end span have equal
structures are used. These analysis have been carried length of 39.9m and the center span have 49.4m
out using different damage conditions. STAAD & length.
SAP software are used for this computational work.
The following details show the type of structure used
for the particular analysis.
Case study of Steel Bridge Model
Tower Model 220 KV s
Tower Model Lab
Steel Frame Lab
Case study of Steel Bridge Model
The Interstate 40 (I-40) Bridge over the Rio Grande
in Albuquerque, New York Mexico were scheduled Fig-1 Elevation View of the I-40 Bridge that was
to be demolished as a part of a construction project tested
to widen the bridge for increased traffic flow prior to Damage Scenarios
this demolition damage was incrementally applied to The damage that was introduced was intended to
the bridge in a controlled manner. Vibration tests simulate fatigue cracking that has been observed in
were performed on the undamaged bridge and after plate girder bridges. This type of cracking results
each level of damage had been introduced. The from out of plane bending of the plate girder web and
primary purpose of these tests was to study the ability usually begins at welded attachment to the web such
to identify and locate damage in a large structure as the seats supporting the floor beams.Four levels of
based on changes in its measured dynamic response. damage were introduced to the middle span of the
Test Structure Geometry plate girder close to the seat supporting the floor
The I-40 Bridges formerly consisted of twin spans beam at mid span. Damage was introduced by
made up of concrete deck supported by two welded making various torch cuts in the web and flange of
steel plate girders and three steel stringers. Loads the girder. A major drawback of introducing damage
from the stringers were transferred to the plate in this manner is that the torch cuts produces crack
girders by floor beams located at 6.1m intervals. much wider tan an actual fatigue crack. Hence these

cracks do not open and close under the loading that The fig no – 3 shows a variation in the natural
was applied during this study frequency which confirms that there is damage in the
The first level of damage designated E-1, consisted Tower and for locating the member which is
of a 0.61m long,10mm wide cut through the web damaged Damage Index method is used in which the
centered at mid height of the web. Next this cut was variation in the value of Zi clearly shows the damage
continued to the bottom of the web to produce a member.
second level of damage designated E-2. For the third
level of damage, E-3, the flange was then cut halfway EXPERIMENTAL ANALYSIS
in from either side directly below the cut in the web. Introduction
Finally the flange to carry the load at this location. For experimental results, fabricated one 220 KV D/C
The various levels of damage are shown in fig. Laboratory tower and tested in the laboratory. This
Transmission Line Tower 220 KV Site tower is 2.26 m high with 43 nodes and 137
Tower Geometry members. To set the tower in to free vibration we
Total Number of Members-260 applied the impact by wooden hammer and measured
Total Number of joints -81 the response by using FFT Analyzer. Here are some
Nine Different Type of section used photo graphs of laboratory tower set up and testing.
Height of Tower-38.517
Support assumed to be fixed
Fig- 3 Lab Tower photograph

Comparison in Experimental
300
Fig-2 Tower Model site 220 KV 250
Frequency in Hz
200
Undamaged
Table 1 Frequency in damage & undamaged state of 150
Damaged
100
220 KV Tower site
50
Mode no Undamaged Frequency Damage Frequency
1 3.29 3.29 0
1 2 3 4 5 6
2 3.32 3.12
Mode no
3 8.87 8.87
4 10.00 10.00
5 10.54 10.53
6 19.66 15.65 Damage Scenarios
7 19.68 19.42
8 19.96 19.73
9 21.82 20.16 i. Loosening of one of the Bolt
10 22.33 21.81 ii. Cut in the Bracing member.
11 25.38 22.38 iii. Missing of member.
12 28.28 25.36
Table 2 Frequency Analysis Result in Undamaged

30
state of Lab Tower
25
Undamage
Mode Analytical Frequency in Experimental
Frequecy in Hz
20
15
No. Hz Frequency in Hz
Damage in main
10
v ertical member
1 141.948 142.20
5 2 197.313 188.45
0
1 2 3 4 5 6 7 8 9 10 11 12
3 229.101 222.52
Mode 4 250.42 247.89
Fig 3 Comparison of Natural Frequency of 220 5 252.521 252.30
KV for Breakage Leg 6 272.04 271.54

Table 3 Frequency Analysis Result in Damaged state Experimental
of Lab Tower From experimental analysis we can conclude
following statement
Mode No. Analytical Experimental It is not possible to measure the deflection values as
Frequency Frequency the large structure are complicated structures and
in Hz in Hz hence locating the damage member is not possible
1 132.152 140.20 unless deflection values can measure at each joint.
2 187.542 184.25
REFERENCES
3 232.51 221.65 Sooyong Park, Yeon-Bok Kim, (2002),
“Nondestructive damage detection in large
4 248.416 242.89
structures via vibration monitoring”, EJSE, pg.
5 258.61 250.30 59-75.
6 272.028 267.12 Pei-Ling Liu, (1995), “Identification and damage
detection of trusses using modal data”, Journal of
Table 4 Frequency Analysis Results Damaged state Structural Engineering, pg. 599-608.
of Lab Tower G C Yao, K C Chang & G C Lee, (1992), “Damage
diagnosis of steel frames using vibrational
Mode No. Analytical Experimental signature analysis”, Journal of Engineering
Frequency in Hz Frequency in Hz Mechanics, pg. 1949-1961.
H L Chen, C C Spyrakos & G Venkatesh, (1995),
1 141.92 138.25 “Evaluating structural deterioration by dynamic
2 182.323 179.53 response”, Journal of Structural
3 198.033 190.20 Engineering,pg.1197-1204.
John B Kosmatka & James M Ricles, (1999),
4 229.123 225.50 “Damage detection in structures by modal
5 250.427 252.31 vibration characterization”, Journal of Structural
6 252.555 272.51 Engineering, pg. 1384-1392.
Z Y Shi, S S Law & L M Zhang. (2000), “Structural
damage detection from modal strain energy
CONCLUSION
change”, Journal of Engineering Mechanics, pg.
1216-1223.
Computational
Jun Zhao & John T DeWolf, (1999), “Sensitivity
From computational analysis we can conclude
study for vibrational parameters used in damage
following statement
detection”, Journal of Structural Engineering, pg.
Analytical value gives a fairly good idea about any
410-416.
type of damage in any steel structure that may be
Ronaldo C Battista, Rosangela S Rodrigues &
either in frame or tower structure.
Michele S Pfeil, (2003), “Dynamic behavior and
In some case like small damage or cut in the member
stability of Transmission line towers under wind
it is difficult to locate the exact damage member.
forces”, Journal of Wind Engineering and
When a small amount of damage like loosening of
Industrial Aerodynamics, pg.1051-1061.
bolts in is introduced to a relatively large 3-D tower
structure, it is observed that the change in natural
frequency is not significant.
In some cases for locating the damage member by
using Damage Index method instead of deflections
rotational values are giving more clear idea of
damaged members.
It is not possible to model the loosening of bolts in
towers (3-D truss structures) as a semi rigid
connection. As the natural frequency change in case
of semi rigid joint varies between 0 to 1 % only.
Therefore loose joint is modeled as reduction in axial
stiffness of the particular member (i.e. reduction in
load carrying capacity of member.

ISBN 978-81-921121-0-7
STREAM FLOW MODELLING OF BEAS RIVER AT MANALI,

HIMACHAL PRADESH
S. K. Jain
National Institute of Hydrology, Roorkee
S. P Rai
Rajeev S. Ahluwalia
ABSTRACT: Snow/Glacier melt monitoring is important for utilities and other purposes. In western Himalaya,
more than half annual runoff is because of snow/glacier melt. Snowmelt estimation is essential for the regional
planning of water resources. To achieve this objective, data such as the snow cover, rainfall, temperature,
evaporation and discharge data are necessary. For a period of five years (October 2004 to September 2009) the
total water budget of the basin at different sites (Manali, Bhunter-Beas, Bhunter-Parwati) was assessed; rainfall and
temperature and snowcover data is used to estimate the total runoff contributed by different component. In this
study river Beas upto Manali is taken for the estimation of snowmelt runoff. River Beas originating from Beas
Kund and Rohtag pass at the altitude of 3505 meters and 3977 meters respectively, both the stream meet to each
other Palchan, 10 km. above the Manali. The seasonal variations of discharge and snowmelt contribution at Manali
in Beas River have been studied. It was observed that snow and glacier melt runoff on average is more than 50% at
Manali.
INTRODUCTION
All the major South-Asian rivers originate in the world including Himalayas (Ferris and Congalton
Himalayas and their upper catchments are covered with 1989).
snow and glaciers. The Indus, Ganga and Brahmaputra
river systems, originating from the Himalayan region, Estimation of the volume of the water draining from
receive substantial amounts of snowmelt water and are the snow is needed for efficient management of water
considered as the lifeline of the Indian sub-continent. resources. The planning of new multi-purpose
Few scientific evaluation is available for Himalayan projects on Himalayan Rivers further emphasizes the
water resources, firstly, due to an insufficient network need for reliable estimates of snow and glacier runoff
of observations for both precipitation and stream because this provide a more dependable flow. Despite
discharge measurements and secondly, Himalayan their well recognized importance and potential, no
terrain being most rugged and inaccessible. attempts have been made to assess in detail the
Nevertheless, the available estimates show that the contribution of snow and glacier in these rivers.
water yield from high Himalayan basins is roughly However, a few attempts have been made to carry out
double that of an equivalent one located in Peninsular for glacierized river basin in the western Himalayan
India. This is mainly due to inputs from snow and ice region to estimate the snow and glacier contribution
melt contributions. The perennial nature of Himalayan mainly on annual basis (Singh et al., 1995, 2005;
Rivers and the suitable topographic setting of the Singh & Kumar, 1997; Singh & Jain, 2003: Kumar et
region provide a substantial exploitable hydropower al., 2005; Haritashya et al., 2006). Singh et al. (1994)
potential in this region. estimated about 28% as the average contribution of
snow-and glacier-melt in the annual flow of the
The contribution of snow and glacier melt runoff to Ganga river at Devprayag. Singh et al. (1997)
the Himalayan rivers is significant and an estimation estimated about 49% as the snow and glacier
of the amount is necessary for the development, contributed for the Chenab river at Akhnoor. Snow-
planning and management of water resources. and glacier-melt contribution at Bhakra Dam site in
Snowmelt accounts for 50-60% of the annual stream Satluj river was estimated to be 60% (Singh & Jain
flow in many areas of the world, including the Sierra 2002). In a similar study of Beas river at Pandoh dam
Nevada, the Rockies, the Alps, the Andes and the site, the snow and glacier melt contribution was
Himalayan Mountains (Ferris and Congalton, 1989). estimated 35.1% (Kumar & Singh, 2007). However,
Water in its frozen state accounts for more than 80 snow and glacier melt contribution for small scale
percent of the total freash water on the earth (Dozier catchment and its seasonal variation are still not
1989) and is the largest contributor to rivers over studied. Therefore, near real time estimation of snow
major portion of the middle and high latitudes. About cover is of utmost importance for effective
30% of the Earth’s land surface is seasonally covered management of water resources and can serve as a
by snow. Snowmelt accounts for 50 to 80 percent of guideline for reservoir operations. Moreover, planning
the annual stream flow in many mountainous areas of of new hydroelectric projects on the Himalayan

Rivers emphasizes the need for reliable estimation of melt contribution in Beas river at the head of Beas
snow and glacier runoff. Keeping in view the basin at Manali in Himachal Pradesh, India.
importance and scope of such estimates, in this study
an attempt was made to estimate the snow-and glacier
Manali
Fig. 1: Location map of Beas Basin

. IRS data (LISS-III, 23.5-m resolution) for one year
(2008-09) were procured from the National Remote
Sensing Agency (NRSA), Hyderabad, India.
THE STUDY AREA AND DATA USED
River Beas originating from Beas Kund and Rohtag
pass at the altitude of 3505 m and 3977 m
METHODOLOGY
respectively, both the stream meet to each other
Palchan, 10 km. above the Manali. There are several temperature index based snowmelt
models like SNOWMOD, the SSARR Model, the
HEC-1 and HEC-1F Models, the NWSRFS Model,
the PRMS Model, the SRM, the GAWSER Model.
The Snow Melt Runoff (SRM) model is widely used
for snowmelt modeling in Himalayan basin. The SRM
uses snow-covered area as input instead of snowfall
data, but it does not simulate the base flow component
of runoff. In other words, SRM does not consider the
contribution to the groundwater reservoir from
snowmelt or rainfall, nor its delayed contribution to
the stream flow in the form of base flow, which can
be an important component of runoff in the
Himalayan Rivers, and plays an important role in
making these rivers perennial. Almost all the stream
flow during winter, when no rainfall or snowmelt
Physiographical and hydro meteorological data are occurs, is generated from the base flow (Singh and
required for estimation the snowmelt contribution Jain, 2003). The SNOWMOD model (Jain, 2001,
from the basin. Physiographical data represent Singh and Jain, 2003) is unique in this aspect as it
physical features of the basin, including its total area, simulates all components of runoff, i.e. snowmelt
its altitudinal distribution through elevation zones and runoff, rainfall-induced runoff and base flow, using
the areas of these zones, and the altitude of limited data.
precipitation and temperature stations. Hydro
meteorological data include daily precipitation, mean The snowmelt model (SNOWMOD) is designed to
air temperature, snow-covered area and streamflow simulate daily streamflow for mountainous basins
data. Temperature,Rainfall, discharge have collected having contribution from both snowmelt and rainfall.
from Bhakra Beas Management Board (BBMB). For The process of generation of streamflow from such
this study data have been used from 1996 to 2008 to basins involves primarily the determination of the
study the seasonal variation of discharge and snow input derived from snowmelt and rain, and its
and glacier melt contribution. transformation into runoff. For simulating the
streamflow, the basin is divided into a number of
Remote sensing data were used to obtain snow- elevation zones and various hydrological processes
covered area data for the study period to prepare snow relevant to snowmelt and rainfall runoff are evaluated
cover depletion curves. MODIS (500-m resolution-8 for each zone. The model deals with snowmelt and
days composite) for two years (2005/06 and 2008/09), rainfall runoff by performing the following three

operations at each time step: (a) available
7000.0 0
meteorological data are extrapolated to the different 10
6000.0
elevation zones, (b) rates of snowmelt and/or rainfall 20
are calculated at different points, and (c) snowmelt 5000.0 Rainfall 30
Discharge 40
runoff from SCA and rainfall runoff from SFA are 4000.0
50
Temperature
integrated, and these components are routed 3000.0 60
separately with proper accounting of baseflow to the 2000.0 70
80
outlet of the basin. The model optimizes the 1000.0
90
parameters used in routing of the snowmelt runoff and 0.0 100
10-Apr
10-Aug
10-Oct
rainfall runoff.
10-Mar
10-Jul
10-May
10-Nov
10-Dec
10-Jan
10-Feb
10-Jun
10-Sep
The information on Snow cover area (SCA) is
determined from the satellite imageries/digital data. 2006
The satellite data were processed using ERDAS

IMAGINE image processing software. First, snow
Modeling of Streamflow of Beas River
cover area maps were prepared for the study basin and The flow data for the year 2005-06 has been
then the SCA for each elevation zone was estimated considered for calibrating the model whereas the year
and plotted against the elapsed time to construct the 2006-07, 2007-08, 2008-09 have been considered for
depletion curves for the various elevation zones in the validating the model. The efficiency of the model has
basin. In order to simulate daily runoff, daily SCA for
been computed based on the daily simulated and
each zone is required as input to the model. Daily
observed flow values for four years. The values of the
values of SCA were obtained by
model efficiencies are 82%, 81%, 83%, and 80%
interpolating/extrapolating the derived depletion
respectively for the years 2005-06, 2006-07, 2007-08
curves. Because the amount of snowfall/snow-covered and 2008-09.
area and temperature conditions fluctuate from year to
year, snow-covered area and depletion trends also
The performance of the model in preserving the
vary from year to year.
runoff volume has been tested based on the criteria
computed as percentage difference in observed and
RESULTS AND DISCUSSION simulated runoff (D) during the entire years. Their
Relationship of Discharge with Temperature values computed for the year 2005-06 and 2008-09
The runoff data have been analysed with temperature are 6.52% and 5.16% respectively. The comparison of
of the Manali site for the year of 2006. It is found that the daily year is shown in Figure 3(a) and Figure 3
when the temperature is almost 5ºC in the month of (b). The model has capability to separate out the
January, corresponding discharge is 100 m3/sec. This simulated and observed flow hydrographs for the all
runoff is almost constant up to mid of February contributions of rainfall, snow and glacier melt and
month. As the air temperature start to rise the base flow from the simulated flows. From these
discharge also show increasing trend. In the month of figures, it has been observed that the model has
May, river runoff increases up to 2090 c m3/sec due to simulated also the daily flows reasonably well
increase in temperature upto 20-21ºC. As temperature showing generally a good matching with the daily
suddenly falls around 20th June, runoff shows observed flows. The trends and peaks of the daily
decreasing trend. Once again temperature start to rise flow hydrographs for the entire period are very well
the corresponding discharge show increasing trend. In simulated by the model. Percentage difference in
monsoon season, when temperature is approximately volume, model efficiency and contributions of rain
21ºC, and rainfall is around 62 mm the discharge is snow and base flow computed by the model are
going to increase and reaches up to 3159 m3/sec. This summarized in Table 1 for the two years. Most of the
relationship of temperature and discharge clearly high peaks observed in the daily flow hydrographs are
indicate the contribution from the melt is the major generly during the months of June and August. It
component with minor contribution of rainfall and reflects the attribution to the glacier melt in the month
runoff. However, the discharge increasing from 100 of June and due to rain also in the month of August.
m3/sec in the month of January to 2090 m3/sec in the Thus these months are considered as the peak melting
month of May/June is only due to snow melt season in the western Himalayan region. However,
contribution. As temperature decline from the month sometimes the flow resulting due to highly rainfall
of September, discharge also declines. This also reflects the peaks in the daily flow hydrographs.
relationship reveals that temperature is main The simulation of baseflow indicates that the
controlling factor for the variation in runoff of Beas baseflow contribution to the streamflow increases as
river at Manali. the season advances, being at maximum during the
Fig. 2: Relationship of temperature with discharge. peak season and then starts decreasing.
Table1. Difference in volume, model efficiency and

contributions of rain snow and base flow computed
by the model

Year Model Difference
Model
in Rain Snow Base In the present study, an attempt was made through the
Volume Efficiency(%) (%) Flow SNOWMOD model to calculate the contribution of
(%) (%) (%) Snow and glacier melt in the annual flows of the Beas
river at Manali. Results for the contribution of the
2005-06SNOWMOD
6.52 82 6 54 40
snowmelt for the years 2005-06 2006-07, 2007-08
and 2008-09 are 54%, 55%, 54%, and 56% 56%
2006-07SNOWMOD
5.41 81 5.4 55 39.6 respectively. Results from above two techniques
validate the methodology for finding out the
snowmelt contribution.
2007-08SNOWMOD
6.51 83 5.1 54 40.9 However, this technique is used to study the
understand the variation of snowmelt contribution in
2008-09SNOWMOD
5.16 80 5 56 39 the river Beas. Further investigations are required for
more précised quantification of snow and glacier
contribution in Beas river. The isotopic techniques
and modelling will be further used for more detailed
investigations.
80
70
Acknowledgements The authors are thankful to
Observed Discharge
Estimated Discharge Bhakra Beas Management Board (BBMB) for
60 Estimated Rain
providing the data used in this study. Authors are also
D isch arg e in C u m ecs
Estimated Snow
50 Base Flow thankful to director NIH who gave the permission to
40
publish the paper in the workshop.
30
REFERENCES
20 Aizen, V., Aizen, E., Malack, J. and Martma, T.,
10
Isotopic measurements of precipitation on central
Asia laciers southestern Tibetan, northern
0
1 51 101 151 201 251 301 351
Himalayas (Central Tien shan), J.Geophys. Res.,
Time (Daily Bases)
1996, 101, 9185-9196.
Fes A. De scally., Relative importance of snow
Fig. (3a) : Observed and simulated daily streamflow for the Beas accumulation and monsson rainfall data for
River at Manali for the period 2005-2006. estimating annual runoff, Jhelum basin, Pakistan.,
Hydrological Science-Journal des science
Hydrologiques., 39,3, June 1994.
90
Kumar Vijay, Singh Pratap, Singh Vivekanand., Snow
80 Observed Discharge
Estimated Discharge
and glacier melt contribution in the Beas river at
70 Estimated rainfall Pandoh Dam, Himachal Pradesh, India.,
Estimated Snow
Disch arg e in Cum ecs
60 Base Flow Hydrological Sciences-Journal-des Sciences

50 Hydrologiques, 52 (2) April 2007.
40
Nijampurkar, V.N., Rao, K., Sarin, M.and Gergan, J.,
30
Isotopic study on Dokrani Bamak glacier, Central
Himalayn: implications for climatic changes and
20
ice dynamics. J.Geol., 2002, 48,160.
10
Ramesh, R and Sarin, M.M., Stable isotopic study of
0 the ganga river system. J.Hydrol., 1992, 139,49-
1 51 101 151 201 251 301 351
62.
Time (Daily Bases)
S.K.Jain, Pratap Singh, Naresh Kumar, Estimation of
Fig. (3b) : Observed and simulated daily streamflow snow and glacier-melt contribution to the Chenab
for the Beas River at Manali for the period 2008- river, Western Himalaya., Mountain Research
2009. and Development, Vol.17, No. 1, 1997, pp.-49-
56.
CONCLUSIONS Sanjay K. Jain., Ajanta Goswami., A.K.Saraf., Role of
Maximum snowmelt contribution were found between Elevation and Aspect in Snow Distribution in
80% and 88% during July and August from Western Himalaya. Water Resour Manage
convection technique. On annual average snowmelt DoI10.1007/s11269-008-9265-5.
contribution into Beas river at Manali was found, S.P.Rai, Bhism Kumar, and P., Singh., Estimation of
48%, 48%, 50%, 50%, 52%, during the years 2002, contribution of southwest monsoon rain to
2003, 2006, 2007 and 2008 respectively.(rai et al Bhagirathi river near Gaumukh, western
2011 ). Himalayas, using oxygen-18isotope., current
science, Vol. 97, no.2 2, 25 July2009.

ISBN 978-81-921121-0-7
RECENT DEVELOPMENTS IN THE MEASUREMENT OF WETTING

SWCC
C. Malaya
Research Scholar, Civil Engineering Department, Indian Institute of Technology Guwahati, Guwahati – 781039, Email:
m.chetia@iitg.ernet.in
S. Sreedeep
Associate Professor, Civil Engineering Department, Indian Institute of Technology Guwahati, Guwahati – 781039, Email:
srees@iitg.ernet.in
ABSTRACT: The unsaturated soil behavior is of prime importance in many of the geotechnical and
geoenvironmental applications. The characterization of the engineering behavior of unsaturated soil is entirely
dependent on the soil-water characteristic curve (SWCC), which is a graphical relationship between water
content (either gravimetric or volumetric) or degree of saturation and soil suction. For decades, experimental
studies are performed to investigate the relationship between the soil suction and volumetric water content
(SWCC). Most of the reported studies, discuss about drying SWCC due to the simplicity of its measurement. It
must be noted that wetting SWCC is equally important for situations like seepage through unsaturated soil. With
this in view, effort has been made in this paper to critically review the recent research outcomes in the
measurement of wetting SWCC. The advantages and limitations associated with different measurement
procedures are presented. This paper also describes a new approach for developing wetting SWCC with the help
of a column test set up. The methodology has been demonstrated by determining the wetting SWCC of a locally
available sandy soil.
INTRODUCTION set up for studying the wetting behaviour of soil. The

Soil suction (matric suction, ψm, or total suction, ψ) test set up has been demonstrated by determining
and its relationship with water content (or degree of drying and wetting SWCCs of a locally available
saturation), is termed as soil-water characteristic curve sandy soil. Details of the methodology adopted are
(SWCC) and it plays a pivotal role in understanding presented in this paper.
the behaviour of unsaturated soil (Fredlund &
Rahardjo 1993). Theoretical Background
A typical drying (desorption) and wetting (adsorption)
It can be noted that the drying and wetting SWCCs SWCC is presented in Figure 1, which indicates a
are different for the same soil, which is attributed to continuous ‘S’ shaped relationship which is
the phenomenon of hysteresis (Fredlund & Rahardjo hysteretic, that is, for a given water content, higher
1993). It must also be noted that the use of wetting or matric suction exists for drying than wetting process
drying SWCC is case specific, depending on the (Birle et al. 2008).
nature of study. Hysteresis information of soil helps to
select the appropriate SWCC equation to be used for
the SWCC parameter identification. Most of the
researchers have used pressure membrane apparatus
for obtaining wetting SWCC. However, this apparatus
is costly and it does not simulate the field condition.
Further, the boundary conditions cannot be controlled
fully. Gallage & Uchimura (2010) have used Tempe
cells for developing wetting SWCC of sand. But Yang
et al (2004) have found that Tempe cells are not
suitable for obtaining the wetting SWCC on coarse-
grained soils. This calls for the development of an
alternative technique that can be employed for
measuring wetting SWCC, which is cost-effective and
represents more realistic situation existing in the field.
The reported literature clearly highlights the use of
column testing devices used for measuring wetting
SWCC. Therefore, it is important to use column test Fig. 1 Idealized soil-water characteristics curves and
set up for developing wetting SWCC, which is more its details.
realistic.
Following are some of the key points that are relevant
With this in view, the present study deals with a for SWCC:
review on similar type of column studies reported in 1. The volumetric water content at saturation, θs,
the literature. This paper also presents a column test describes the water content at which the soil is

completely saturated. This depicts the initial state
of the drying path. Chapius et al. (2006) have developed a column test
2. The air-entry value (AEV), ψa, is the suction at set up to obtain SWR of coarse materials which have
which air first enters the largest pore present in rapidly decreasing water contents at suctions between
the soil sample during a drying process (Brooks 0 kPa and 10 kPa. The column was 1.83 m high and
and Corey 1964). 103 mm in internal diameter. This column set up can
3. Residual water content (θr) is the minimum water be used only for determining drying SWR. The
content below which there is no appreciable column has outlets (quick connects) to measure
change in θ with ψ. Suction corresponding to θr is suction values by placing tensiometers during the
called residual soil suction, ψr (Yang et al. 2004). drying test. Water content was determined at the end
4. The water-entry value, ψ w, on the wetting SWCC, of the drainage, by collecting soil samples after
is defined as the suction at which the water removing the material progressively from the column,
content increase significantly during the wetting starting from the top. The column test results were
process (Birle et al. 2008). found to be more reliable than Tempe cell. Chapius et
al. (2006) have mentioned that the duration of column
A REVIEW ON COLUMN TEST SET UP test to obtain drying SWR may be several weeks for a
Stauffer & Kinzelbach (2001) have used a column column having a length of 1 m to 2 m. Columns that
device consisting of a 927 mm long vertical plexiglass exceed 2 m may take several months to reach
tube of diameter 53.5 mm for obtaining drying and equilibrium.
wetting SWCC of sand. Four tensiometers and two
gamma ray absorption probes were used to measure Li et al. (2009) have developed a wetting front
suction and water content variations during wetting advancing column testing device for the measurement
and drying test performed in the column experiments. of unsaturated hydraulic conductivities of soil. A 120
For wetting experiment, the water content profile was mm diameter and 1000 mm high acrylic column cell
allowed to establish within seven days, controlled by was used to hold the soil sample and suction/ water
a constant head at the lower inlet of the column. content monitoring sensors at four different locations.
Four miniature-tip tensiometers and four ML2X theta-
Yang et al. (2004) have determined drying and probe soil moisture sensors were installed
wetting SWCCs for five sandy soils. The drying horizontally along the column cell to monitor suction
SWCCs were obtained using Tempe pressure cell. and water content profiles during the infiltration and
The wetting SWCCs could not be obtained using capillary-rise process. The monitored water content
Tempe pressure cell. The water was expected to flow and suction values at each section was used to obtain
back into Tempe pressure cell, but this did not happen the unsaturated soil hydraulic conductivity of the soil.
because the ceramic disk offered a great resistance for The study indicates that a column length of 400-600
the water to flow into the soil sample. The wetting mm and only two water content and suction
SWCC was then obtained using a capillary rise open measuring sections are sufficient for the wetting front
tube (Fredlund & Rahardjo 1993). In this study, tubes advancing method.
of 100 mm in diameter and 500 mm in length were
used to compact the soil in it. The equilibrium Lins et al. (2009) have obtained drying and wetting
conditions in capillary rise open tube were achieved SWCC of sandy soils from steady state and transient
after allowing the soil samples to wet for 64 days. Soil state experiments under various flow rates performed
water content was determined by collecting soil in a column testing device. A column of 780 mm high
samples from various levels in the tube using oven and 305 mm in diameter was used to place a sandy
dried method. The height of a soil specimen above the soil specimen of 540 mm high. Five tensiometers and
water table is assumed to be equal to the matric five 3-rod TDR sensors were installed horizontally
suction head. These data were used to obtain wetting along the soil specimen for measuring suction and
SWCC of the soil. water content respectively. A tensiometer and a TDR
sensor were installed at different depth. The AEV,
Yang et al. (2004) have presented a column testing saturated water content and residual water content of
device for infiltration study where tensiometer- the sandy soil obtained from SWCC of steady state
transducer system and time-domain reflectometry column test were found to be similar with the
(TDR) were placed at different depths to measure transient state column test for loose and dense sandy
suction and water content respectively. The size of the specimens. This indicates that the dynamic flow does
soil column used in this study was 1 m high and 190 not have significant influence on the SWCC of this
mm diameter. It can also be extended to 2 m or 3 m type of soil. SWCC parameters derived from sand
high if necessary. The TDR was placed fully into the column testing device were compared with the SWCC
soil column horizontally to measure the water content parameters obtained from modified pressure plate
more accurately. The SWRs during rain (wetting) and apparatus. The comparison shows that independent of
after rain (drying) were obtained using the column the equipment and the test method, similar SWCC
testing device. parameters were obtained for each density.

Zaradava et al. (2009) have developed a column depicted in Figure 2, essentially consists of one
testing device for performing drainage experiment to ECH2O-TE volumetric water content probe connected
establish suction-water content relationship for to a data logger, one T5 tensiometer connected to a
defining the unsaturated soil hydraulic conductivity. data logger, one computer for collecting experimental
The column consists of a perspex cylinder with an results, a flow regulator and a water reservoir, and the
inner diameter of 260 mm and 1 m in height. Four main perspex column into which soil is compacted.
manometers were installed along the column length One more tensiometer was placed in the same soil
for measuring soil suction during drainage sample to check the internal consistency of suction
experiment. For water content determination, vertical observations. In total the column is 250 mm high and
sand cores were removed using a copper tube 100 mm in diameter. The bottom part of the column
(Clayton & Siddique 1999). consists of a water reservoir. The water reservoir and
the soil sample is separated by a perforated plexiglass
It is noted that column set up was mostly used for plate. A filter paper was placed between the soil
studying drainage properties of unsaturated soil. The specimen and the perforated plexiglass plate to avoid
efficacy of the column device is not fully explored for flushing soil particles into the water reservoir.
obtaining wetting SWCC. In this context it must be
noted that the wetting experiment performed in the
To data logger
same soil sample at the end of the drying test would
represent better the hysteresis. Not many studies are Soil
reported that deals with drying and wetting SWCC on sample
the same soil sample using column set up. Perspex
column T5
EXPERIMENTAL DETAILS tensiomet

To data logger
Water flow
Material Description regulator
A locally available sandy soil was used in this study. ECH2O-TE probe
The soil is characterized for its specific gravity and
grain size distribution by following the guidelines
presented in the literature (relevant parts of IS 2720). Water flow
The details of the characterization are listed in Table
1. Fig. 2 Set up of column testing device
Table 1 Physical properties of the sandy soil One port along the column was used for the insertion
considered in this study of the EC-TE probe. The T5s were placed vertically in
Property Value the soil sample. The EC-TE probe was installed
Specific gravity 2.45 horizontally along the soil specimen and placed at the
Particle size characteristics same depth as the ceramic of the T5s. A flow
Coarse sand (4.75-2mm) 0 regulator was connected at the bottom of the column
Medium sand (2-0.425mm) 1 to control the flow of water from the reservoir in to
Fine sand (0.425-0.075mm) 93 column specimen while performing wetting test.
Silt (0.075-0.002mm) 6
Clay (<0.002mm) 0 Drying and Wetting Water Retention Curves
Using Column Test Set up
Equipment Description The air-dried sandy soil was mixed with required
A T5 tensiometer (UMS GmbH, Munich) and amount of distilled water and compacted into the
ECH2O-TE volumetric water content sensor (Decagon perspex column, by giving different number of blows
Inc., USA), was used to measure suction and to achieve a certain water content (w) and dry unit
volumetric water content, respectively, of compacted weight (γd). The soil sample was then saturated by
soil in the perspex mold. Sensors were secured in ponding it with distilled water and further removing
place to ensure proper contact with the soil. The the excess water on top. After saturating the soil
tensiometer used in this study can measure matric sample, EC-TE and T5 sensors were inserted into the
suction up to 100 kPa and characterized by very short packed soil sample. The soil sample was then allowed
response time. The small dimension of the T5 to air dry and the suction and volumetric water
tensiometer ensures minimal disturbance of the soil content of the sample continuously recorded using the
sample during its insertion. The details of working respective data loggers. The data obtained from T5
methodologies of these instruments are reported in the and EC-TE measurements were then used to plot the
literature (Malaya and Sreedeep 2010). drying SWCC corresponding to the compaction state
of the soil. At the end of drying test performed, the
Description of the Column Testing Device wetting process was simulated by operating the flow
The experimental set up used in the present study, as valve to enter the water in to the soil sample gradually

at a very slow rate. The corresponding changes in Brooks, R.H. and Corey, A.T. (1964). Hydraulic
suction and water content were recorded by the properties of porous media. Hydrology paper no. 3,
corresponding data loggers. When the specimen Dept. of Civil Engineering, Colorado State Univ.,
reached zero matric suction in the wetting process, the Fort Collins, Colorado.
assembly was disconnected. The suction and Chapius, R.P., Masse, I., Madinier, B. and Aubertin,
volumetric water content values recorded during M. (2006). A drainage column test for determining
wetting process were used to obtain wetting SWCC. unsaturated properties of coarse materials.
Geotechnical Testing Journal, ASTM, 30(2), Page
RESULTS AND DISCUSSION count: 7.
The data obtained from the T5 and ECH2O-TE Clayton, C.R.I. and Siddique, A. (1999). Tube
measurements carried out in the column testing device sampling disturbance-forgotten truths and new
during drying and wetting process were used to plot prospectives. Proceedings of institute of civil
the SWCCs corresponding to a particular compaction engineering, 137, pp. 127- 135.
state for the sandy soil, as depicted in Figure 3. It can Decagon Devices (2006). Operator’s manual version
be noted from the figure that the drying and wetting 2 ECH2OTE/ECTM water content, ec and
SWCCs are different. The test results indicate that the temperature sensor manual, U.S.A.
hysteresis effect is significant for the type of soil used Fredlund, D.G. and Rahardjo, H. (1993). Soil
in this study. One of the advantages of this study over mechanics for unsaturated soils. John Wiley and
those reported in the literature is that the volumetric Sons, Inc., New York.
water contents are measured ones and not estimated. Gallage, C.P.K. and Uchimura, T. (2010). Effects of
dry density and grain size distribution on soil-water
0.25 characteristic curves of sandy soils. Soils and
Drying SWCC Foundations, JGS, 50(1), 161-172.
Wetting SWCC IS: 2720: Part III: Sec 2: 1980 Indian Standard on
0.20
Test for Soils, Part III Determination of Specific
Gravity-Sec 2 : Fine, Medium and Coarse Grained
0.15 Soils, Bureau of Indian Standards, New Delhi.
IS: 2720: Part 4: 1985 Indian Standard Methods of

Test for Soils, Part 4 Grain Size Analysis, Bureau

0.10 of Indian Standards, New Delhi.
Li, J., Yang, Q., Li, P., and Yang, Q. (2009).
0.05
Experimental research on soil-water characteristic
curve of remolded residual soils. Electronic
Journal of Geotechnical Engineering, Page count:
0.00 12.
Lins, Y., Schanz, T. and Fredlund, D.G. (2009).
10
0 1
10 10
2
10
3
Modified pressure plate apparatus and column
m (KPa) testing device for measuring SWCC of sand.
Geotechnical Testing Journal, ASTM, 32(5), 1-15.
Fig. 3 Measured drying and wetting SWCCs Malaya, C. and Sreedeep, S. (2010). A study on the
influence of measuring procedures on suction-
CONCLUSIONS water content relationship of a sandy soil. Journal
This study presents a comprehensive review of the of Testing and Evaluation, ASTM, 38(6), Page
recent research outcomes in the measurement of count: 9.
wetting SWCC. The study also presents the details of Stauffer, F. and Kinzelbach, W. (2001). Cyclic
a column test set up for determination of drying and hysteretic flow in porous medium column: model,
wetting soil-water characteristic curves (SWCCs). experiment and simulations. Journal of Hydrology,
The usefulness of the test set up has been Science Direct, 240, 264-275.
demonstrated by determining the drying and wetting Yang, H., Rahardjo, H. and Fredlund, D.G. (2004).
SWCCs of a locally available sandy soil. The main Factors affecting drying and wetting soil-water
advantage of the column set up used in this study is characteristic curves of sandy soils. Canadian
that it simulates closely the wetting condition in the Geotechnical Journal, 41(5), 908-920.
field due to the movement of water. Zaradava, K., Powrie, W. and White, J. (2009). The
determination of the moisture retention
REFERENCES characteristics of waste materials using vertical
Birle, E., Heyer, D. and Vogt, N. (2008). Influence of drainage experiments. 3rd international workshop
the initial water content and dry density on the soil- on hydro-physico-mechanics of landfills,
water retention curve and the shrinkage behavior of Braunschweig, Germany, Page count: 12.
a compacted clay. Acta Geotechnica, Springer, 3,
191-200.

ISBN 978-81-921121-0-7
CHANNEL AND CORED UNIT FOR ROOFING\FLOORING FOR LOW

COST HOUSING: A REVEIW
Harshit Jain
UG Student, Department of Civil Engineering, Institute of Technology, Banaras Hindu University, Varanasi- 221005
Email Id: hjair4617@gmail.com
Rajesh Kumar
Associate Professor, Department of Civil Engineering, Institute of Technology, Banaras Hindu University, Varanasi-
221005
S. B Dwivedi
Associate Professor, Department of Civil Engineering, Institute of Technology, Banaras Hindu University, Varanasi-
221005
ABSTRACT: In Developing countries like India infrastructure plays a very important role in development of
countries. There is need of adequate strategy for development of low cost Infrastructure techniques which are
sustainable as well as economic for satisfying future needs. This paper presents comparative study of precast
channel, cored unit for roofing/flooring for low cost housing. Channel Unit is a single span RCC unit, throng
shaped in section. Cored unit is also single span RCC unit. The cored units have the advantage of lighter dead
load, better insulation due to hollow core and saving in material & shuttering cost. Cost analysis is studies and it
is found that Channel unit cost less as comparison to cored unit. Channel Unit requires less amount of concrete
in comparison to Cored Unit which ensures less emission of Carbon dioxide in environment.
INTRODUCTION and improved housing techniques for various

Social housing is defined as “housing for low income components of building like flooring, roofing,
person that is provided by housing institutions and building materials so that sustainable and economic
that excludes immediate individual ownership. It is solution can be provided to this problem. The need to
not an option for the very poor, but its very nature and examine the cause of the problems associated with the
accessing accommodation for housing institution will implementation of low or medium cost housing to
have to earn a secure income, formally or informally ensure the success of future housing policy in India.
to be able to afford the rental or other periodic This paper presents two roofing/flooring units for low
payment of accommodation to be” (South African cost housing in terms of material consumption, cost,
Revised Draft 2003: 4). carbon dioxide, structural specification need to be
provided in case of pre-cast channel and cored unit
To most individuals, housing represents the largest
single investment item of a lifetime. This is especially PRECAST CHANNEL UNIT
true as family income increases and housing viewed
less as a basic consumption and more as a key to a Channel Units have been used in a number of
secure future. The majority of families in developing buildings all over the country. It is a full span precast
countries like India, China, Pakistan, Malaysia etc., do RCC unit, trough shaped in section (Fig. 1). It can be
not have adequate housing and infrastructure facility. used for floors and roofs supported on suitable
Housing provision for all in any country is very structures like brick/stone walls and RCC beams. It
crucial in order to ensure social- economic stability does not require any intermediate temporary props or
and to promote national development supports, since the unit is strong enough to support the
load, for which it is designed. The outer sides of the
In India, the policy of urban development and housing unit are corrugated and are grooved at the ends to
has come a long way long from 1950’s. The lack of provide shear key action between adjacent units.
housing and basic services was very much evident in
1950’s due immigration of large number of people Nominal width of the unit is varies from 300 mm to
from Pakistan to Indian cities (National Urban 600 mm, its depth from 130 mm to 200 mm and a
Housing and Urban Habitat Policy, 2007). Every minimum flange thickness of 30 mm. The length of
five year plan aims at to reduce the gap between the unit may be adjusted to suit the span to be
demand and supply of infrastructural services to low covered, but the maximum length is restricted to 4.2
income citizens of country. There is a great need to m from stiffness considerations (Clause 9.1.1, IS:
develop economic and sustainable strategy to cope 4326, 1993). Horizontal corrugations are provided on
with this problem of developing nations like India, the two longitudinal faces of the units so that the
China, Pakistan, Malaysia etc., so that low income structural roof/ floor acts monolithic after concrete
families or poor families can have strong and grouted in the joints between the units attains
adequate housing facility. Beside institutional as well strength.
as financial requirement, this issue also need modern

inner side of the outer frame of both the mould must
be properly oiled and placed on levelled platform.
Channel Unit. Figure 1(IS: 14201:1994)
PRECAST CORED UNIT Cored Unit Figure 2 (IS 10297: 1982)

Cored Units developed by Central Building Research
Institute, Rooked have been used in a number of In case of Channel unit the trough frame is oiled and
buildings all over the country. The cored unit is full placed in outer frame and concrete is levelled by
span self-supporting precast RCC structural moving the mould to and fro manually or by machine.
component. It is possible to have a one-way slab over U clamps are used for fixing the position of trough
walls/ beams with this system. As concrete near the frame with outer frame. Web portion is filled with M
neutral axis does not contribute to its structural 20 grade of concrete and after one hour of casting U
strength, it is removed to get a lighter unit. The cored clamps are removed. After 3 hour, outer frame is also
units have the advantage of lighter dead load, better removed and the unit is left undisturbed for next 48
insulation due to hollow core and saving in material hours.
and shuttering cost.
In Cored unit, GI pipes are introduced and concrete is
It is a precast RCC structural component having circu- filled, vibrated and compacted. Pipes should be
lar hollow cores throughout its length (Fig 2). The rotated along their longitudinal axis so as to break the
unit is structurally complete and does not require any bond between pipes and concrete. After one to two
temporary support or propping during construction. It hour, pipes are removed rapidly by rotating them
provides a flush ceiling. along their axis. The freshly cast units are kept
covered with wet gunny bags at the casting yard for
The nominal width of the unit is varies from 300 mm 48 hours and afterwards the units are slightly pushed
to 600 mm with the actual as 295 mm and 595 mm to break bond with floor, lifted carefully from ends
respectively. Though the depth (thickness) of the unit and transported to the curing yard manually or in
will depend upon the span and loading conditions, a trolley in their natural position, i.e. with main
depth of 130 mm can be taken for normal situations. reinforcement at the bottom.
The weight of such a unit will be 46 kg per running
metre. These units can be easily handled manually. If ASSEMBLY OF FLOOR/ROOF
mechanical handling facilities are available, units The top surface of wall or beam supports, as the case
having nominal widths of 600 mm can also be used. may be, is levelled. The units are lifted one by one
The minimum flange/ web thickness of the unit is manually or with the help of chain pulley block or
kept as 20 mm mechanically with a hoist and placed side by side
across the span to be covered. The units are then
CASTING AND CURING aligned and levelled properly. Negative reinforcement
The mould of channel and cored unit for in the case of continuous floor/roof slab is placed in
roofing/flooring can be made from well-seasoned position at support. The in-situ joint concrete is then
wood or steel, but for long life run steel should be cured for 14 days by sprinkling water. In case of floor,
made for mould. Channel unit mould consists of two any flooring treatment such as I.P.S. or mosaic tiles or
parts outer frame and inner through frame. The mould in-situ mosaic may be laid. The flooring shall be laid
for cored unit consists of two longitudinal sections, in bays with the bay lines in the direction of units
two end pieces and two galvanised iron pipes. The coinciding with some joints between units.

STRUCTURAL DESIGN
The channel units may be designed by limit state
method as per IS 456-2000. The units may be
designed either simply supported or continuous
depending on the end conditions. The channel units
are to be designed for two distinct stages of loading as
in case two way slab.
Stage I – Loading (Just after Placing of In-situ

Concrete) (As per IS 456:2000)
At the time of laying the units, the load comprises of
following loading conditions:
 self-weight of unit,
 In-situ concrete in
the joint between the units and also the
incidental live load.
At this stage of loading, it is assumed that the in-situ

joint concrete has not attained any strength to ensure EMBODIED CARBON DIOXIDE
monolithcity and continuity over the spans. Hence, at According to EPA, between 900 and 1100 kg (1984
this stage of loading, the following values of bending and 2425 lbs.) of CO2 is emitted for every 1000 kg of
moment and shear force should be taken: Portland cement produced in the U.S (Environmental
Protection Agency, 2005). Amount of carbon dioxide
Loading Conditions released depends on the fuel type, raw ingredients
used and the energy efficiency of the cement plant
B.M. in the middle of = 1.5(wc /8 + wi /12 + wd (NRMCA, June 2008). About 50% of CO2 is released
interior span /16) l2 as a result of calcination of raw material and
B.M. in the middle of = 1.5(wc /8 + wi /10 + wd remaining part of it is as a result of burning of fossil
end span /12) l2 fuels in kiln to maintain the required temperature in
B.M. at support next to = 1.5(wc /8 + wi /10 + wd the kiln for clinker production. However concrete
end support /12) l2 produces very less amount of carbon dioxide which is
B.M. at support next to = 1.5(wi /9 + wd /10) l2 primarily a function of cement content in concrete
end support According to Gajda Johan, 2001, about 100 to 300 kg
B.M. at any other interior = 1.5(wi /9 + wd /12) l2 of carbon dioxide is embodied for every one cubic
support meter of concrete (2400 kg), depending upon mix
S.F = 1.5(0.5 wc + 0.5 wi + design.
0.6 wd) l
The production of cement requires large amount of
where, fuel which releases large amount of carbon dioxide
into atmosphere. From production to casting to supply
wd = Dead load per unit length due to at construction site require large amount of fossil
flooring/roofing treatment etc. for nominal fuels. The burning of fossil fuels release large amount
width of channel unit (uniformly of carbon dioxide into environment. Cored unit has
distributed). more embodied carbon dioxide than channel unit (Fig
2, Fig. 3).
wi = Super-imposed live load per unit length for
nominal width of one channel unit (uniformly
distributed)
MATERIAL CONSUMPTION
Cement mixed with coarse and fine aggregate in
appropriate ratio in order to have appropriate mix
design characteristics which must be not leaner than
M 20 grade of concrete is used for its casting.
Channel unit need less amount of concrete as
compared to cored unit (Fig. 3). Amount of
reinforcement used in channel and cored unit is
depends upon loading condition, end conditions and
design requirement. Figure 4

Span Building Building Building Building Building No. of Strengthening to be
Category Category Category Category Categories for Storeys provided in Floor/
B C D E Earthquake Roof with Channel
in m Nos. Dia Nos. Dia Nos. Dia Nos. Dia Resisting Features Units
mm mm m mm (Clause 7.1.1)
m B 1 to 3 a
5 or 2 8 2 8 2 8 2 10 4 a, c
less C 1&2 a, b
3&4 a, b, c
NOTES D 1 to 4 a, b, c
1. Span of wall will be E 1 to 3 a, b, c
distance between centre lines of its cross walls or Where
buttresses. a = Tie beam as per i),
b = Reinforcing bars of channel unit and tied to tie
2. The number and
beam
diameter of bars given above pertain to high strength
reinforcement as per ii)
deformed bars.
c = Reinforced deck concrete as per iii)
3. Width of RC band is
Table 2: Strengthening Measures for Floors/ Roofs
assumed same as the thickness of the wall. Wall
with Precast
thickness shall be 200 mm minimum. A clear cover of 20
Channel and Cored Unit (IS: 4326:1993)
mm from face of wall will be maintained.
4. The vertical thickness
mould, mould oil, casting platform, props,
of RC band be kept 75 mm minimum, where two
longitudinal bars are specified, one on each face.
scaffolding, centring and shuttering for haunches,
deck concrete etc. Both the techniques are economic
5. Concrete mix shall be as compared to
of grade M20 of IS 456:2000 or 1:1.5:3 by volume. (in
cast in-situ roof/floor slab. The time taken to complete
accordance with IS 456:2000)
the roofing/flooring using precast units is very less as
6. The longitudinal steel
bars shall be held in position by steel links or stirrups 6
compared to cast in-situ reinforced cement concrete
mm dia spaced at 150 mm apart.
roofing slab.
Table 3: Recommended Longitudinal Steel in Reinforced
Concrete Bands
THE STRUCTURAL SYSTEM
(IS: 4326:1993)
The system consists of precast channel units as

explained earlier. The depth of the unit in this case
will be 130 mm while the width will be 300 mm or
600 mm. A 35-40 mm thick layer of deck concrete
with a nominal reinforcement of 6 mm diameter bars
150 mm c/c at both ways is provided above the units.
This system is recommended for areas where the
rainfall is high or for areas prone to high seismicity.
SEISMIC RESISTANCE MEASURES (AS PER

IS 4326: 1993-SECOND REVISION)
All floors and roofs to be constructed with small

precast components shall be strengthened as specified
Figure 5 for various categories of buildings in following
Table2 and Table 3. The various specification for
COST ANALYSIS strengthening measures are given in details in IS
4326: 1993.
Cost of providing and layering of roof/floor slab using
precast channel and cored unit with deck concrete CONCLUSION
with nominal reinforcement as per design and Channel and cored Units for roofing/flooring can be
requirement with cement concrete of grade not leaner one of possible answer for improved roofing/flooring
than M 20 grade of concrete including casting, techniques which can be used for developing low cosh
staking, curing, erecting and placing the units in housing can be used by Social housing Institutions as
position, cost of well as Government for improving better
infrastructural solutions to low income families.
Channel unit require very less amount of concrete as
compared to conventional slab. It also requires very

less amount of time for completion of work as the
units are precast. Channel and cored units’ cost of
lying their assembly is almost same and economic
than slab. Channel and cored require less amount of
concrete; hence release very less amount of carbon
dioxide content. Channel and cored units can also be
used in high seismic zones after some strengthening
measures.
REFERENCE
AP 42- Compilation of Air Pollutant Emission
Factors, Volume 1: Stationary Point and Area
Sources, Environmental Protection Agency,
Washington, DC, 2005.
Building Research Note No. 52, Precast channel Unit
for floor/roof precast unit, CSIR-Central
Building Research Institute, 2010.
Concrete CO2 Fact Sheet, NRMCA Publication
Number 2PCO2, June 2008.
Gajda and John(2001), Energy use of Single-Family
Houses With Various Exterior Walls, CD026,
Portland Cement Association, Skokie, IL, 49
pages
Housing and Urban Policy in India, Ministry of Urban
Housing and Poverty Avieation, Government of
India, New Delhi, 2007
IS: 456-2000, Plain and Reinforced Concrete Code of
Practice (Fourth edition), Bureau of Indian
Standards, New Delhi.
IS: 14201:1994, ‘Precast Reinforced Concrete
Channel Units for Construction of Floors and
Roofs – Specification’, Bureau of Indian
Standard, New Delhi.
IS: 14215:1994, ‘Design and Construction of Floors
and Roofs with Precast Reinforced Concrete
Channel Units- Code of Practice’, Bureau of
Indian Standard New Delhi.
IS 10297: 1982, ‘Code of Practice For Design and
Construction of Floors and Roofs Using Precast
Reinforced Pre-stressed Concrete Ribbed or
Cored Slab Units’, Bureau of Indian Standard,
New Delhi.
IS: 4326:1993,’Eartquake Resistant Design and
Construction of Building’, Bureau of Indian
Standard, New Delhi, Clause 9.2.1.
South African Draft. South African Revised Draft
2003: 4

ISBN 978-81-921121-0-7
A STUDY ON DUCTILITY OF CONCRETE AND MORTAR WITH

THE ADDITION OF RUBBER POWDER
C. B. K. Rao
Professor, Deptt. of Civil Engg., National Institute of Technology, Warangal
R. Bhaskar
Graduate students, Deptt. of Civil Egng., National Institute of Technology, Warangal
V. Rajendra Prasad
Graduate students, Deptt. of Civil Engg., National Institute of Technology, Warangal
ABSTRACT: Concrete is a brittle material. Degree of brittleness increases with strength. Inclusion of fibre as
reinforcement has been found to be a good solution to improve ductility but it has its limitations. Now the
researchers have focused attention to modify the concrete at the matrix level. Thus engineered cement
composites or ductile concrete has been a subject of interest. Though there is breakthrough in the production of
ultra high strength ductile cement composite concrete, the state of art is still a patented one. Interest aroused all
over in engineered cement composites with a hope to improve ductility of concrete. In this work an attempt has
been made to modify the Portland cement adding rubber powder. Tests are conducted using servo UTM under
strain rate control with facility to record complete post peak behaviour. Rubber powder of size 600 µm in 5%,
15% and 25% of weight of cement were used. Compressive strength, tensile strength and strain energy
absorption of the concrete specimens and cement mortar specimens using cement modified with rubber powder
(CMRP) were studied. Mechanical properties such as compressive strength, tensile strength of CMRP concrete
decreased when compared with normal concrete and but post peak energy absorption for CMRP concrete and
mortar specimens increased.
INTRODUCTION
Concrete is the largest man made material in the ultimate, the matrix cracks began to form a much
world. Art of making is simple though science is more extensive net work. In the mechanics of micro-
illusive to a common man and engineer alike. cracking, the first stage can be addressed by
Necessity of making high strength concrete increased improving bond and the second stage is to be
with the challenge of construction of high raise addressed by bridging the cracks with reinforcement.
structures. Art and Science of making high strength Extensive research on FRC has taken place. Addition
concrete has become a reality with constant research of fibers to concrete has given improvement to
by cement chemistry scientists and engineers. It is strength as well as deformable capacity of concrete.
achieved by dense packing with finer secondary Abundant literature is available on all the above
hydration materials viz., mineral admixtures and low aspects. Though the early works have used long
water binder ratio. Cement concrete by nature is a fibers, observation under SEM shows that arrest of
brittle material and the brittleness increases with cracking has to be addressed at nano level. Current
increase in strength. Failure of concrete is explosive age has entered the stage of engineered cementitious
kind for very high strength concrete. Research to composites.
improve ductility has become paramount interest. Replacement of aggregate with light weight aggregate
Ductility may be defined in several ways. It may be and low modulus material has been experimented to
viewed broadly as structural ductility and material produce light weight concrete for low strength
ductility. Initially ductility problem is addressed by structural applications. Utilization of rubber waste is
confining concrete with lateral ties/ stirrups(1). one such area(3). Waste rubber was shredded into
However, there are practical issues of detailing and pieces and used as aggregate. The rubber powder was
limitation to volume of confining reinforcement. used in mortar in some of the investigations. Rubber
Hence, it is desirable to improve deformability and treated with alkaline material such as NaoH, NaCl
strength of concrete. was studied(4). Decrease in strength of concrete and
In the matter of mechanical behaviour of concrete, mortar with rubber was reported but increase in
the bond between aggregate and the matrix is a deformability of concrete was noticed(5). In the
matter of general interest. Hsu(2) et al observed present investigation, the authors used rubber powder
cracking of plain concrete in compression directly of 600 microns and down to study the deformability
with microscopic and x-ray techniques. Even prior to of cement mortar and concrete with Cement Modified
loading, cracks were observed at the aggregate matrix with Rubber Powder (CMRP)(6),(7).
interface. Beyond 30 percent of ultimate strength, the
cracks began to grow under increasing load and after SIGNIFICANCE OF WORK
50 percent of ultimate load, the bond cracks began to One of the major environmental challenges facing
extend through matrix and at about 75 percent of municipalities around the world is disposal of worn

out automobile tyres. Utilization of waste rubber will pozzalanic admixture. Chemical admixture SP337 is
address the problem to some extent. Improvement in used for improving workability.
engineering properties of concrete may be reaped
simultaneously. COMPRESSIVE STRENGTH, TENSILE
STRENGTH AND DEFORMABILITY
OBJECTIVES OF THE STUDY Universal testing machine with servo controlled
It is proposed to add rubber powder finer than 600 actuator of 1000 kN capacity and stroke length of ±
microns to cement and study the mechanical 80 mm supplied by BIS, Bangalore is used to conduct
properties, energy absorption of cement mortar and compressive strength of cylinders for mortar and
concrete. concrete. The machine can be programmed to operate
at stress rate control or strain rate control. The test
EXPERIMENTAL PROGRAMME has been conducted at the strain rate control of 0.002
Experimental programme consists of casting of mm per second. Force and displacements are
cement mortar and concrete cylinders of size 63 mm continuously recorded at a time interval of 0.1
dia and 126 mm height. Two mixes 1:3 and 1:1.5 are second. There are 4000 to 6000 data points for the
used for cement mortar and two mixes 1:1.5:3 and specimens. The testing was continued till complete
1:0.75:1.5 are used for cement concrete. Ordinary failure. Load vs deflection curves are drawn and are
Portland cement 53 grade confirming to IS 12269 – shown in fig. 1. An observation of the load
1987 obtained in single consignment is used. Specific deformation curves shows that as load increased
gravity standard consistency, initial setting time and deformation increased upto peak load and
final setting time are 3.11, 30 percent, 35 min. and deformation continued to increase but load decreased.
215 min respectively. Standard Ennore sand grades I, Deformation of specimen with rubber powder is more
II and III are mixed in three equal parts thoroughly than that in plain specimen. For each specimen total
and used as fine aggregate. Coarse aggregate is area under the load deformation curve and area after
obtained from local quarry in a single consignment. peak load are computed. Cylinders are tested for split
Aggregate passing through 10 mm sieve and retained tensile strength. Results of both concrete and mortar
on 4.75 mm sieve is taken as coarse aggregate specimens are presented in table 2.
specific gravity is 2.76. Rubber powder is made from An examination of the results shows that compressive
used tires by crushing and grinding under normal strength split tensile strength decreased with addition
temperature. It is finely ground tyre rubber from of rubber powder. For small dosage of 5 percent of
which fabric and steel belt have been removed. rubber powder the average decrease is around 20
Rubber powder is added at 5%, 15% and 25% of percent of strength and for a dosage of 25 percent of
cement and thoroughly mixed with cement (CMRP). rubber powder the decrease in strength is more than
Super plasticizer Conplast SP 337 is used for 50 percent. Similar behaviour is noticed in split
workability while preparing cement mortar and tensile strength of concrete. Compressive strength and
concrete cylinders. split tensile strength decreased with addition of
rubber powder in cement mortar.
RESULTS AND DISCUSSIONS Earlier researches noted improvement in
There are 48 mixes for workability, 24 specimens for deformability of concrete with the addition of rubber.
compressive strength and 24 specimens for tensile To examine the deformation capacity, total area and
strength. Results are given in Table1 and 2 of the area after peak, under load deformation diagram for
experimental work. concrete specimens and mortar specimens are
computed and are given in table 2. An examination of
WORKABILITY TEST the results shows that for concrete 1:1.5:3 water
Cement concrete and mortar mixes are prepared as cement ratio 0.4 and 0.5, total area under the curve
per proposed proportions and the flow table test is increased but for the concrete 1:1.5:3 water cement
conducted on the standard flow table used for mortar ration 0.6 total area decreased with increase in rubber
by giving 15 blows in 15 seconds. Average flow is powder. However, for all the specimens with addition
determined in percentage. The results are given in of rubber load deformation response in the
table 1. It can be noticed from the results that descending portion is more than the ascending portion
addition of rubber powder decreased the percentage for specimens with addition of rubber powder.
flow. There is no specific pattern in the change of Similar exercise for mortar specimens shows that
percentage flow. The mix became stiff when the deformation has not increased with addition of rubber
rubber powder of 15 percent is added. Water added to powder.
the mix is based on the cement quantity (i.e based on Obviously, low strength concrete deforms more than
water cement ratio) only. Rubber powder is not a high strength concrete. To examine whether there is
cementitious material. The rubber powder does not any beneficial effect due to addition of rubber powder
participate in hydration of cement. Amount of rubber to concrete, results of concrete 1:1.5: 3 w/c ratio 0.5
powder may be treated as fines in aggregate but in of plain specimens whose compressive strength is
the present contest, authors viewed it as non 17.85 N / sqmm which is nearer to compressive

strength of concrete 1:1.5:3 w/c ratio 0.4 with 15% (2) Hsu, T.T.C., (1963), “Microcracking of Plain
rubber (whose compressive strength is 16.86 Concrete and the Shape of the Stress-Strain
N/sqmm) is compared. Area under load- deformation Curve”, Journal ACI, Vol. 60, 209-223.
curve after peak load is greater for the specimens (3) Z Li, F Li, J S L LI, (1998), “Properities of
with rubber powder (CMRP) than that without rubber concrete incorporating rubber tyre particles”
powder. It shows that deformable capacity of MCR,50, 297-304
concrete increases in the addition of rubber powder. (4) Aziz Abraham, Wisam Amer Aules, Salwa Hadi
It is the view point that rubber powder may bridge Ahamad, (2010), “ Cement mortar
the crack at micro level. Study under SEM may properitiescontain crumb Rubbertreated with
through some more light on the micro level Alkaline material”,. Modern Applied Science,
mechanisms of crack arrest. vol.4, n0.12.
(5) Morteza Khorrami, A Vafai, Ahmed A K, C S
CONCLUSIONS Desai, M H M Ardakani, (2010), “ Experimental
Cement concrete and mortar mixes become stiff with Investigation on Mechanical Characteristics and
the addition of rubber powder. Compressive and split Environmental Effects on Rubber Concrete “,
tensile strength decreases with the addition of rubber International Journal of Concrete Structures and
powder. Load deformation response shows that the materials, Vol.4, no.1,17-23.
deformation capacity of cement concrete specimen (6) Bhaskar R, (2011), “A study on Deformability of
increased with the addition of rubber powder. Concrete with the addition of rubber powder”,
Dissertation 091001, Structures Division, Dept.
REFERENCES of Civil Engineering, NITW.
(1) S.R, Iyyengar KT, P. Desai, Nagireddy, K, (1970) (7) Rajendra Prasad V, (2011) “A study on
“Stress strain characteristics of concrete confined Deformability of cement mortar with the
in steel binders” Magazine of Concrete Research addition of rubber powder”, Dissertation 091016,
Vol.22, No.22, 173-184, Structures Division, Dept. of Civil Engineering,
NITW.
Table 1: Workability of Concrete and Mortar

% Rubber average flow (%)
powder
Concrete 1:1.5:3 with w/c ratio Concrete 1:0.75:1.5 with w/c ratio
0.40 0.50 0.60 0.30 0.35 0.40
0% 3.5 24.75 54.25 14.5 26.5 54
5% 2.25 9.75 32.25 20 25.5 51.25
15% 2.5 3.75 14.75 5.75 3 21.5
25% 3.75 2 8.5 3.75 4.75 5.75
Mortar 1:3 with w/c ratio Mortar 1:1.5 with w/c ratio
0.40 0.50 0.60 0.30 0.35 0.40
0% 6.5 23 58.5 32 56 95
5% 6.0 20 42.5 20 32.5 76.25
15% 5.75 9.75 21.75 9 6.75 27.5
25% 1.75 4.25 14 1.75 7.5 8

Table 2: Strength and ductility of concrete and mortar
% Tensile Comp. Area under load Tensile Comp. Area under load
Rubber Strength Strength vs deformation Strength Strength vs deformation
Powder (N/sqmm) (N/sqmm) curve (kN-mm) (N/sqmm) (N/sqmm) curve (kN-mm)
Total after Total after
peak peak
load load
C 1:1.5:3, W/C= 0.40 M 1:3, W/C= 0.4
0% 4.01 26.64 59.32 37.24 4.41 21.96 110.31 60.07
5% 3.61 20.75 77.51 41.24 3.21 15.33 84.55 32.61
15% 2.21 16.86 68.17 44.99 2.81 11.75 78.04 53.36
25% 2.20 10.39 52.02 35.04 2.21 10.29 52.78 20.11
C 1:1.5:3, W/C= 0.50 M 1:3, W/C= 0.5
0% 3.61 17.85 73.19 29.72 3.61 21.48 126.8 96.17
5% 3.21 12.31 58.05 43.27 3.01 18.49 71.62 41.28
15% 2.81 9.69 102.82 78.41 2.81 8.38 69.3 49.52
25% 2.21 9.35 76.18 61.47 2.41 7.56 54.82 30.83
C 1:1.5:3, W/C= 0.60 M 1:3, W/C= 0.6
0% 3.41 14.34 54.40 38.07 3.21 19.79 63.34 24.49
5% 2.61 10.08 44.90 26.39 2.81 13.27 35.69 16.29
15% 2.01 7.41 34.74 20.62 2.21 8.16 47.05 28.54
25% 2.0 6.95 45.37 29.44 2.01 6.5 32.08 18.41

Fig. 1 load vs Deformation curves for CMRP mortar and concrete specimens

ISBN 978-81-921121-0-7
EFFECTS OF AGGRESSIVE CHLORIDE ENVIRONMENT ON HIGH

PERFORMANCE STEEL FIBRE REINFORCED CONCRETE
Dhirendra Singhal
Coordinator, Department of Civil Engineering, GZS College of Engineering and Technology, Bathinda, 151 001
Singhald62@rediffmail.com
Bal Krishan
Associate Professor ,Department of Civil Engineering, GZS College of Engineering and Technology, Bathinda, 151001
bkbirdi1969@yahoo.com
ABSTRACT: Conventional concrete has very porous transition zone and numerous inherent micro-cracks. This
leads to its susceptibility when exposed to aggressive environment and further limits its ductility and tensile
strength. However, attempts made in the past have shown that the inclusion of silica fumes improves the transition
zone and the resultant concrete is less permeable. Further, the addition of steel fibres enhances the ductility and
tensile strength of concrete significantly. These additions lead to high performance steel fibre reinforced concrete
provided that the produced concrete also has significant improvement in workability. Improved workability can be
obtained by adding mineral and chemical admixtures. Therefore, it is required urgently to study the durability of
high performance steel fibre reinforced concrete with these inclusions.
For this purpose, high performance steel fibre reinforced concrete of M35 grade was exposed to aggressive chloride
solutions of sodium chloride containing chloride solutions 19 and 38 g/litre for the periods of 7 days, 1 month, 3
months and 9 months. In order to expedite the process of deterioration, cubes of 100mm size were submerged to half
of their depth. M35 grade of concrete was intentionally chosen as it had high cement content and high dose of super-
plasticizer.
Compressive strength and change in weights were measured at different ages which indicate that the produced high
performance steel fibre reinforced concrete was durable too.
INTRODUCTION If concrete deteriorates in aggressive solutions, the

In this study, it was attempted to check the durability of impact of deterioration can be best observed by the
high performance steel fibre reinforced concrete weight of cube and loss in strength as exposure to
(HPSFRC) and as specified above that the same was aggressive solutions leads to leaching and formation of
part of another study on design and analysis of high expansive salts known as Freidel’s salt. Hydrated
performance steel fibre reinforced concrete. Since cement has calcium hydroxide which dissolves in
HPSFRC included fly-ash, fibre and silica fumes, it is aggressive solutions and thus leaches out. This leaching
necessary to discuss the differences which can occur leaves porous concrete and thus weight reduction takes
with these constituents. The permeability had been place. If some expansive salt like Freidel’s salt forms,
found to decrease as much as 90% with the inclusions of the expansion leads to internal cracking in concrete.
silica fumes and fibres (Ganesan and Saker 2005, Singh This results in strength loss. Therefore, observations
et al 2002) were made on compressive strength and weight of cubes
after the ages of 7 days, 28 days, 3 months and 9
Cubes were exposed to aggressive chloride solutions. months.
Aggressive chloride solution chosen was of sodium
chloride (NaCl). NaCl was chosen as its concentration is Experimental Programme
the maximum in sea-water. The solution of NaCl had Portland pozzolana cement confirming to IS: 1489-
chloride salt concentration of 19g/litre and 38g/litre, 1991, fine and coarse aggregates confirming to IS: 383-
respectively. The chloride salt concentration was taken 1970 were used. Steel fibres corrugated in shape with
based on the fact that 19g/litre is average sea-water salt aspect ratio of 67 and length 30 mm were used. Cubes
concentration (Singhal 1993). In order to observe of 100 mm in size were cast of M35 HPSFRC. The
maximum damage and expedite the deterioration details of mix proportions used are shown in Table 1
process, solutions of double the average sea-water salt (Bal Krishan 2011). The cement content 450Kg/m3
concentration was also used. Further, the cubes were included 7.5% fly-ash and 12.5% silica fume. Super-
submerged for half the depth so that oxygen is available plasticizer used was viscocrete 10H of Sikka India
for the corrosion of steel fibres (Biczok 1985). The level Private Limited. In order to attain 100mm slump with
of the solutions was kept constant throughout the curing 1% fibre content the dosage of super-plasticizer was
period so as to maintain salt concentrations. found to be 5%. The cubes were cast of both the

concretes i.e. without fibres and with fibres. Since the HPSFRC was 5% for mix M35, as shown in Table 1.
structural properties of SFRC were significantly Although concrete with M35 grade is supposed to be
affected with the shapes of fibres, plain and corrugated much durable yet it was chosen considering that it
fibres were used. The fibre content 1% (by weight of required higher doze of super-plasticizer. The concrete
concrete) was found optimum as such in HPSFRC, 1% with and without fibres which was cured in water served
fibre content was maintained. Optimum dosages of as reference concrete. These cubes were cured in water
super plasticizer were determined as explained by Basu and in solutions of sodium chloride containing chloride
[1999] and considering high workability in terms of salt concentrations of 19 and 38 g/litre, respectively,
Vee-Bee time (It is fundamentally established that right after the stage of demoulding. Thereafter, the
SFRC responds better to dynamic tests). This was done weights of the cubes were taken at different intervals of
to ensure high workability in the presence of fibres and 7 days, 28 days, 3 months and 9 months, respectively.
in order to satisfy one of the prime requirements of Also, the compressive strength of cubes was checked on
HPSFRC. Therefore, the dose of super plasticizer for compression testing machine at these ages.
Table 1 Detail of Mix Proportions Used

Mix Water Cement Fine Coarse 28 days % Silica % % Super
(Kg/m3) (Kg/m3) Aggregate Aggregate Compressive Fumes Fly plasticizer
(Kg/m3) (Kg/m3) Strength (MPa) Ash
M35 164.59 360.12 399.22 1377.92 45.11 12.50 7.5 5.0

The results of compressive strength over cubes appears in Tables 2, 3, 4 and Figures 1, 2 and 3 is the
obtained have been shown in Table 2 for high average over five specimens.
performance concrete (HPC). Tables 3 and 4 show
the results of HPSFRC with plain and corrugated
fibres, respectively. The compressive strength which
Table 2 Results of Compressive Strength of HPC in Water and Different Salt Solutions
High Performance Concrete
Medium Doze of Compressive Compressive Compressive Compressive

Different Strength at 7 Days Strength at 28 Strength at 3 Strength at 9
Salts (MPa) Days (MPa) Months (MPa) Months (MPa)
Water -- 33.36 46.33 47.11 48.50
NaCl 19gm/l 30.57 43.67 48.54 54.27
NaCl 38gm/l 28.11 40.40 44.66 48.97
Table 3 Results of Compressive Strength of HPC with Plain Steel Fibres in Water and Different Salt Solutions
High Performance Concrete with Plain Steel Fibres
Water -- 33.56 46.00 49.60 52.67
NaCl 19gm/l 33.38 46.56 50.22 54.21
NaCl 38gm/l 30.41 42.89 49.24 53.01

Table 4 Results of Compressive Strength of HPC with Corrugated Steel Fibres in Water and Different Salt
Solutions
High Performance Concrete with Corrugated Steel Fibres
Water -- 35.18 49.34 50.89 52.96
NaCl 19gm/l 33.98 49.09 52.44 56.51
NaCl 38gm/l 33.21 48.61 52.21 54.28
Table 2 shows the results of compressive strength of Tables 3 and 4 are with plain and corrugated fibres,
high performance concrete. It can be observed from respectively. In case of plain fibres, the cubes have
the table that high performance concrete when kept in maintained higher compressive strength at the age of
aggressive solutions had less compressive strength 9 months when compared with high performance
initially up to the age of 28 days than the cubes cured concrete. These cubes suffered no deterioration as the
in water. However, the gain in compressive strength difference in strength with respect to the strength of
for the cubes cured in aggressive solutions was cubes cured in water remained insignificant after the
higher after the age of 28 days and it maintained age of 28 days and the percentage loss in
higher compressive strength in all the solutions till compressive strength did not increase with the curing
the period of 9 months. The above results can also be ages. All the above facts can also be confirmed
confirmed through Figure 1. through Figures 2 to 3.

Change in Weight of Cubes in Water and 7. The weights shown in Tables are the average over
Different Chloride Solutions: The observations three specimens.
made over the weight of cubes appear in Tables 5 to
Table-5 Results of HPC Cube Weights in Water and Different Salt Solutions
Medium Weight Weight % age Weight % age Weight % age Weight % age
at 0 at 7 Increase at 28 Increase at 90 Increase at 270 Increase
Day Days in weight Days in weight Days in weight Days in weight
(gms) (gms) (gms) (gms) (gms)
Water 2383 2436 2.22 2438 2.31 2441 2.43 2437 2.27
NaCl 2354 2428 3.14 2439 3.61 2439 3.61 2412 2.46
19gm/l
NaCl 2311 2386 3.25 2394 3.59 2398 3.77 2406 3.31
38gm/l
Table-6 Results of HPC with Plain Steel Fibres Cube Weights in Water and Different Salt Solutions
at 0 at 7 Increase at 28 Increase at 90 Increase at 270 Increase
Day Days in weight Days in weight Days in weight Days in weight
(gms) (gms) (gms) (gms) (gms)
Water 2305 2434 5.60 2435 5.64 2437 5.73 2430 5.42
NaCl 2259 2387 5.67 2392 5.89 2397 6.15 2398 6.15
19gm/l
NaCl 2340 2467 5.43 2472 5.64 2478 5.90 2477 5.86
38gm/l
Table-7 Results of HPC with Corrugated Steel Fibres Cube Weights in Water and Different Salt Solutions
at 0 Day at 7 Increase at 28 Increase at 90 Increase at 270 Increase
(grams) Days in Days in Days in Days in
(grams) weight (grams) weight (grams) weight (grams weight
Water 2304 2401 4.21 2404 4.34 2403 4.30 2405 4.38
NaCl 2257 2337 3.55 2340 3.68 2340 3.68 2339 3.63
19gm/l
NaCl 2251 2346 4.22 2351 4.44 2352 4.49 2365 5.06
38gm/l

From the Tables 5 to 7, it can be clearly observed that Engineering, Institute of Technology, BHU,
the cubes, whether of HPC or of HPSFRC did not Varanasi.
suffer any loss of weight in any of the aggressive salt Bal Krishan (2011), Design and Analysis of High
solutions through-out the age of curing. This Performance Steel Fibre Reinforced Concrete
confirmed that the leaching phenomenon in Composites, Ph.D. Thesis, Punjab Technical
aggressive environment which leaves porosity and University, Jalandhar
therefore, reduced strength did not take place. The
reason may be that shrinkage cracks are better
arrested in the presence of fibres (Mangat and Azari
1988). Further, high dose of super-plasticizer used
was not detrimental. It can therefore be concluded
from the discussions that the HPC or HPSFRC has
required strength, workability and durability and can
be confidently used for repair and rehabilitation
purposes.
CONCLUSIONS
Based on the experimental programme undertaken
the following conclusions can be derived.
The high doze of super-plasticizer used for producing
high performance steel fibre reinforced concrete with
silica fumes and fly-ash is safe
High performance steel fibre reinforced concrete
maintains better durability in aggressive chloride
environment.
ACKNOWLEDGEMENTS
We are heartily thankful to Elkem India Private
Limited, Navi Mumbai for providing Micro Silica;
Associated Cement Company (ACC) for providing
cement; Stewols & Company, Nagpur for supplying
steel fibres; Sika India (Pvt) Limited, New Delhi and
Jaipur for providing super-plasticizer (Viscocrete)
free of cost.
REFERENCES
Ganesan N. and Sekar T. (2005), “Permeability of
Steel Fibre Reinforced High Performance
Concrete Composites”, Journal of Civil
Engineering Division, The Institution of
Engineers (India), 86, 8-11.
Imre Biczok (1985), Concrete Corrosion Concrete
Protection, Academial Kiado, Budapest.
Mangat P. S. and Azari M. M. (1988), “Shrinkage of
Steel Fibre Reinforced Cement Composites”,
Materials and Structures, 21, 163-171.
Singh A. P., Agrawal R. and Singhal D. (2002),
“Permeability and Strength Characteristics of
Steel Fibre Reinforced Concrete”, Journal of
Ferro-cement, 32, 127-137.
Singhal D. (1993), An Investigation of Chloride and
Sulphate Attack on Steel Fibre Reinforced
Concrete, Ph.D. Thesis, Department of Civil

ISBN 978-81-921121-0-7
THE INFLUENCE OF GEO-ENVIRONMENTAL PROPERTIES ON

MUNICIPAL SOLID WASTE
Sunita Kumari
Research scholar, Department of civil Engg. ,IIT Roorkee
A.K. Nema
Associate Professor, Department of civil Engg. ,IIT Delhi
K.B. Ladhane
Research scholar, Department of civil Engg. ,IIT Roorkee
ABSTRACT: Landfilling is one of the most economic and feasible means of disposing municipal solid waste
(MSW) in India. Major portion of these wastes are disposed in landfills due to lack of other disposal
technologies like aerobic composting, vermiculture, incineration, pelletisation etc. About 5000mt of MSW is
generated every day in Delhi and it will increase about 12,750mt per day by 2015, a recent study done by the
National Environment Engineering Research Institute, Nagpur. Hence a large number of modern landfills are
required to dispose these MSW. Landfills are subjected to large amount of differential settlements as a result of
the biodegradation of MSW. The design and permitting of landfills requires performing comprehensive analysis
of geo-environment properties of waste materials, as the safety and cost of landfills are sensitive to variation in
these properties. Unfortunately, the geo-environment properties of waste materials can vary within broad
ranges, change significantly with time, and are not easily amenable to direct measurement, due to heterogeneity
and hard inclusions. Furthermore, published data is limited and the conditions under which the properties were
measured or back calculated are unclear. As a result, the probabilistic theory has been used to estimate the
landfills settlement (Tang et al.1994), mainly focusing on characterizing the waste material properties. The
objective of the present study is to correlate settlement rates of MSW with biodegradation process and
corresponding influencing factors. A review is presented for the geo-environment properties of MSW with the
available past records and their effect on the settlement of wastes with respect to time.
INTRODUCTION et al. (1977), Sower (1973), Dodt et al (1987), Coduto

The composition of municipal solid waste (MSW) in a and Huitric (1990) and Edil et al (1990).
landfill is heterogeneous. Therefore, estimation of
settlement for MSW is difficult as the mechanism of Estimate of the total settlement of a sanitary landfill
waste settlement is complex. These mechanisms can range from 25% to 50% of the landfills initial
be categorised to the following main processes: thickness (Stearns 1987). Settlement mechanism in
1. Physical and mechanical processes: These refuse landfills are very complex and less understood
include the reorientation of particles, movement of the than in coarse or fine-grained soils (Edil et al. 1990).
fine materials into larger voids, and collapse of void This can be attributed in part to its homogenous
spaces. nature, large particle sizes, compression of waste
2. Chemical Process: This includes corrosion, particles, and the loss of solids due to biodegradation.
combustion, and oxidation. Three main stages of settlement have been identified,
3. Dissolution Process: This consists of namely, initial settlement, primary settlement, and
dissolving soluble substances by percolating liquids secondary compression. In literature, initial
and then forming leachate. settlement and primary compression are sometimes
4. Biological Decomposition: The organics in considered different phenomena, but in the framework
the refuse will decompose with time specially of the present study we will consider an overall
controlled by moisture content and percentage of primary compression. Settlement occurs in essentially
organics in the waste. two distinguished stages: (1) primary compression
Since, the unit weight and void ratio vary with the (δp) (2) secondary compression (δs).
types of waste, composition, depth, method of (δt) = (δp) + (δs)
compaction, and rate of decomposition, among other Primary compression includes physical compression
factors. The rate of decomposition is further of particles (distortion, bending, crushing and particle
complicated by several factors, including the effects orientation) and consolidation (significant for
of time, temperature, and environmental conditions saturated waste bodies). In most waste, physical
[e,g., Jesserberger and Kockel (1991)]. Thus, the compression will occur immediately on application of
waste properties may be considered site-specific. load (i.e. in response to placement of overlying layers
of waste). Therefore, primary compression will occur
Theory of Landfills settlement and biodegradation in a period of a few days to a few weeks and hence
Numerous settlement estimation methods have been can be considered to be short term.
proposed in the literature by Merz and Stone (1962), Secondary compression includes all creep effects (i.e.
Rao mechanical compression under constant stress) and
those relating to degradation (both chemical and

biological). Creep effects include time-dependent the relationship similar to equation (1) and is
particle distortion (i.e. bending, crushing), particle presented below:
reorientation and ravelling. Degradation includes
collapse of containers due to a change in strength (i.e. ΔH=ΔH (EL) = C α (EL) H log ( t2/t1 )
corrosion) and degradation of organic compounds.
Degradation potential of components is discussed by Where, ΔH (EL) is settlement at time t2 after external
Landva and Clarke (1990) and is a key element of any load application, t1 is the time for primary
classification system. Many methods have been compression most of which will occur as the load is
proposed to characterise and predict secondary applied and may continue up to 4 months after
compression. Biodegradation is the main component external load application, H is the thickness of refuse
of secondary compression in MSW landfills. fill, and C α (EL) is the coefficient of secondary
Secondary compression occurs throughout the active compression. Literature review indicates that for
life of the landfills and is usually the main component older refuse fills which have undergone
of total settlement. decomposition for some period of time (typically 10
to 15 years), C α (EL) ranges between 0.01 and 0.07
Primary compression (Sharma and Lewis, 1994).
The initial waste settlement, also called settlements
under self-weight, due physical and mechanical Both C α (sw) and C α (EL) values depend on site
mechanisms typically occurs during 1 to 4 months specific environmental conditions and organic content
period of filling. This is followed by time-dependant of the waste fill. Higher compressibility, i.e., higher
settlement (ΔH) which can be expressed by the Cα values for both cases, will indicate higher organic
following equation: content, higher humidity, and/or higher degree of
decomposition of the waste.
ΔH = ΔH (SW) = C α (SW) H log ( t2/t1 )
Table 1. Refuse compressibility parameters (Owesis
Where, ΔH (SW) is settlement at time t2 after fill and Khera 1986)
placement; t1 is the initial period (typically 1 to 4
Reference Primary Secondary
months) of settlement, H is the thickness of refuse fill,
and C α (SW) is coefficient of secondary compression Sowers(1973) (for e0) 0.1-0.41 0.02-0.07
due to self-weight, typically C α (SW) values range Zoino (1974) 0.15-0.33 0.013-0.03
between 0.1 and 0.4 (NAVFAC, 1983). Converse (1975) 0.25-0.3 0.07
Rao et al. (1977) 0.16-0.235 0.012-0.046
Landva et al. (1984) 0.2-0.5 0.0005-0.029
Oweis and Khera (1986) 0.08-0.217 -
Table 2. Secondary compression parameter for MSW

material (after Oweis and khera, 1998)
Material Cα
Ten year old landfill 0.02

Fifteen year old Landfill 0.24
Fifteen to twenty year old landfill 0.02
More than twenty year Old Landfill 0.04
Old Landfill with high soil content 0.001 - 0.005
Fig. 1 The typical time-settlement data for a landfill There is some field data obtained from long term
under a certain load. (From Grisolia & Napoleoni, settlement monitoring studies to support this
1995) approximation. Oweis and Khera (1998) published
Secondary Compression values of Cα for a range of waste materials obtained
As discussed above, long term settlement is mainly from the literature. Table 2 shows selected values
due to biodegradation and mechanical creep from their summary and demonstrates the problem of
compression. After initial period of primary trying to use one Cα values for the entire period of
compression (typically up to 4 months) the time secondary compression. As the rate of degradation is
dependent secondary compression occurs over a long unlikely to be constant with time, it is not surprising
period of time. This settlement can be expressed by that Cα is not a constant. Gourc et al. (1998) provide

a comprehensive review of available calculation But when the readily degradable soluble substrates are
methods. Fassett et al. (1994) and Manassero et al. exhausted, the overall process is controlled by
(1996) both give useful summaries of secondary hydrolysis. The most abundant carbon sources i.e
compression data. Settlement prediction techniques cellulose and hemi cellulose present in waste are
based on modelling the biodegradation process are insoluble; the majority of the decomposition process
under development and appear promising (i.e is limited by hydrolysis( Halvadakis et al.1983). Most
McDougall and Pyrah, 2001). of the researchers assume that cellulose hydrolysis
occurs by first order kinetics (Chen 1974; McGrowan
Biodegradation in Landfills et al. 1988; El-Fadel et al. 1989; Young 1989) and if
MSW landfill suffers a large amount of long term this amount of solid material is converted into a liquid
settlement that is associated with volume reduction form, it can be easily drained out of a landfill, results
due to decomposition of Organic solids. Wardwell a considerable amount of settlement.
and Nelson (1981) illustrated that the observed The decomposition process results in decreasing the
additional secondary strain due to the decomposition volume of the MSW. The decomposition causes
increased as the amount of organic content increased; larger voids to develop within the landfill. Therefore,
for example, from 7% of strain for an organic content the balance between the voids and the self weight of
of 70%. Based on the results, it is concluded that the the landfill material may be upset due to the
remaining additional long term strain due to the decomposition.
decomposition is very dependent on the amount of
biodegradable organic solids. Settlement due to Factors influencing biodegradation
biodegradation also depends upon the landfills age. Rate of municipal waste decomposition is
In a fresh landfill considerable amount of settlement considerably dependent upon the decomposition
will occur due to presence of Organic matters. On the condition in MSW landfills. Different parameters
other hand, an older MSW landfill no longer has such may be monitored to check the rate of decomposition
organic matter for the decomposition to occur and the of waste.
remaining particles are relatively inert.
In MSW landfills, biodegradation is processed in four
stages in which solid organic particles are solubilised
and converted to methane and carbon dioxide through
methanogenesis. Once they are converted to a liquid
form, the intermediate decomposition products are
drained out of the landfill or are converted to methane
and carbon dioxide through methanogenesis.
Therefore, the decomposition step that is of concern
for geotechnical engineers is the conversion of refuse
organic solids to liquid. Detailed studies of the
degradation process indicate that polymer hydrolysis Fig. 3 Factors affecting Stabilization in landfills (El-
is responsible for this solubilisation (Barlaz et al. Fadel, M., Shazbak, S., Saliby, E., & Leckie, J., 1999)
1990).
(1) Moisture Content – The moisture content in
landfills is highly dependent on several interrelated
factors including: initial composition of the waste;
local climatic conditions; landfill operating procedure
(e.g, use and type of daily cover); effectiveness of any
leachate collection and removal system; amount of
moisture generated by biological processes within the
landfills; and amount of moisture removed with
landfill gases (Mitchell & Mitchell, 1992). The in situ
moisture content of MSW varies in between 25-45
percent in accordance with type of waste. Moisture
content is an important variable as it is associated
with the greatest effect on biodegradation processes
because it provides an aqueous environment that
facilitates the transport of nutrients and microbes
within the landfill. This transport mechanism dilutes
Fig. 2 Major degradative steps during the anaerobic the concentration of inhibitors and enhances micro
decomposition phase (El-Fadel, M., Shazbak, S., organisms access to their substrates and hence
Saliby, E., & Leckie, J., 1999) improves gas generation and stabilization.
In the recent landfill, significant amount of readily
degradable soluble substrate are present, hence rate of
decomposition will be governed by methanogenesis.

(2) Initial Unit Weight & compaction effort at Table 3 Statistical summaries of bulk unit weight
placement data for fresh waste (Fassett et al., 1994)
The unit weight of MSW varies within a broad range Poor Moderate Good
because, by its very nature, its components are highly compaction compaction compaction
variable, and placement procedures as well as 3
environmental conditions can greatly affect its unit Range (kN/m ) 3.0-9.0 5.0-7.8 8.8-10.5
weight. Each landfill operator handles incoming Average (kN/m ) 3
5.3 7.0 9.6
waste differently and achieves different levels of Standard 2.5 0.5 0.8
compaction. Fassett et al. (1994) considered that the
deviation(kN/m3 )
following factors should be recorded along with
measured unit weights: MSW composition including Coefficient of 48 8 8
daily cover and moisture content; method and degree variation (%)
of compaction; the depth at which the unit weight was
measured; and the age of the waste.
On exposure to water, constituent of MSW (i.e. food
Like soil, the unit weight of MSW is also affected by waste, garden refuse, paper, textiles) absorbs water,
compaction effort and layer thickness, the depth of results an increase in unit weight of constituent.
burial (i.e. overburden load) and the amount of liquid These increases in individual particle unit weight are
present (moisture content). Unlike soils, the unit added to the increase in bulk unit weight resulting
weight of MSW varies significantly because of waste from increased leachate in the void spaces between
material’s size and density, state of decomposition particles of waste to produce increases in the bulk unit
and degree of control during placement ( such as weight of waste mass. Therefore, older waste would
thickness of daily cover or its absence). It is believed have higher bulk unit weight than fresh waste.
that initially the unit weight of waste is very much
dependent on waste composition, the daily cover, and (3) pH – In situ pH of MSW is variable, depend upon
the degree of compaction during placement. But with waste compositions. When aerobic degradation
age of waste the unit weight becomes dependent on prevails, pH of leachate is maintained around neutral
the depth of burial, the degree of decomposition and whereas in case of anaerobic degradation pH is
climatic conditions. Generally for design purpose maintained generally acidic side res. In aerobic
average unit weight is acceptable. condition, the organic carbon is removed fast from the
waste by conversion to carbon dioxide, while acid
Degradation of MSW components with organic accumulation is less but in case of anaerobic condition
content will result in a loss of mass, changes in size a small amount of organic carbon is removed while
and alteration of the mechanical properties (i.e. acid accumulation is high.
compressibility and shear strength). It will also (4) Waste composition – Significant variation
change the density of the component. As a waste occurs in waste composition from area to area. In
body degrades, void ratio reduces and hence a volume Delhi itself, a large amount of variation occurs in
reduction occurs. It is generally believed that different areas. The variation is brought out in the
degradation results in an increase in waste density, Table 4.
and hence unit weight irrespective of presence of few Table 4. Zone wise MSW composition (in
field measurements in degraded waste. percentage) at Delhi
MSW is a particulate material, having high void ratio

and a high compressibility, compaction processes will
reduce the voids within an individual component as
well as voids between various components. The unit
weight of compacted waste will depend upon the
waste components, thickness of layer, weight, and
type of compaction plant and the number of times
equipment passes over the waste. Fassett et al. (1994)
conducted a detailed survey of bulk unit weight data
from the international literature. A statistical analysis
of the data is shown in Table 3.
MSW landfill suffers a large amount of long term

settlement that is associated with volume reduction

due to decomposition of Organic solids. Wardwell models for municipal solid waste landfill
and Nelson (1981) illustrated that the observed applications.” Sage Publication on behalf of
additional secondary strain due to the decomposition international solid waste Association, Journal of
increased as the amount of organic content increased; Waste Management and Research 17, pp. 347-
for example, from 7% of strain for an organic content 368.
of 70%. Hence, decomposition is very much Fassett, J. B., Leonards, G. A., and Repetto, P.C.
dependent on the amount of biodegradable organic (1994). “Geotechnical Properties of Municipal
solids. Based on results, it is concluded that if the Solid Waste and their Use in Landfill Design.” A
organic content of waste is higher, settlement due to Special Project Report at Purdue University,
decomposition will be higher or vice-versa where as Fraklin.
for inert materials, settlement will be very less. Gourc, J. P., Thomas, S., Vuillemin, M., 1998.
(5) Temperature – The degradation of MSW is not Proposal of a waste settlement survey
affected by change in atmospheric temperature. In methodology. Proceedings Environmental
MSW landfills, the daily cover is made of soil and Geotechnics, Seco e Pinto (Ed.), Balkema,
waste is generally below 0.5m depth of daily cover, pp.195-200.
hence effect of temperature is not significant. Halvadakis, C.P. (1983) Methanogenesis in solid
waste landfill bioreactors. PhD Thesis,
FUTURE SCOPE OF STUDY Department of Civil Engineering, Stanford
For efficient designing of landfills, following issues University, Stanford, CA, USA.
must be explored for further research in developing Jessberger, H. L., Kockel, R., 1993. Determination
country like India:- and assessment of the mechanical properties of
* Determination of Influence of each geo- waste: materials. Proceedings Sardinia ’93,
environmental properties of waste on other properties. Fourth International Landfill Symposium,
* Development of Universal classification system of Cagliari, Italy, pp. 167-177.
waste for the evaluation of mechanical properties. Komilis, D. P., Ham, R. K. & Stegmann, R., (1999)
* Development of Indian Standard code for designing “The effect of landfill design and operation
various components of Landfill. practices on waste degradation behaviour: a
* Integrated Model should be prepared to predict geo- review.” Sage Publication on behalf of
environmental properties of waste with respect to time international solid waste Association, Journal of
with known values of waste properties. Waste Management and Research 17, pp. 20-26.
Landva, A., Clark, J. I., 1990. Geotechnics of waste
REFERENCES fills. Geotechnics of Waste Fills – Theory and
Barlaz, M. A., Ham, R. K., and Schaefer, D. Practice, ASTM STP 1070, pp.86-106.
M.(1989),”Mass-balance analysis of anaerobically Ling, I. Hoe., Leshchinsky, Dov., Mohri, Yoshiyuki.,
decomposed refuse.” J. Envir. Enrg., ASCE, and Kawabata, Toshinori., (1998) “Estimation of
115(6), 1088-1102. Municipal Solid Waste Landfill Settement”
Chen, W.H. (1974). “Time-settlement behaviour of ASCE, Journal of Geotechnical and
milled refuse.” PhD dissertation, Dept. Of Civil Geoenvironmental Engineering, Volume 124,
Engineering, Northwestern University, Evanton , Issue 1, pp. 21-28.
III. Manassero, M., Van Impe, W.F., Bouazza, A., 1996.
Coduto, D .P. and Huitric, R. (1990). “Monitoring Waste disposal and containment. Proceedings
landfill movements using precise instrument.” Second International Congress on Environmental
Geotechnics of waste fills – theory and practices: Geotechnics, Osaka, Vol. 3, pp. 193-242.
ASTM STP 1070. American Soc. For Testing McDougall, J.R., Pyrah, I.C., 2001. Settlement in
Materials, Philadelphia, Pa., 358-370. landfilled waste: extending the geotechnical
Dixon, N., and Jones, D.R.V. (2005). “Engineering approach. Proceedings Sardinia ’01, Eight
Properties of Municipal Solid Waste.” Elsevier International Waste Management and Landfill
Ltd, Journal of Geotextiles and Geomembranes, Symposium, Cagliari, Italy, pp. 481-490
Vol. 23, No. 3, 2005, pp. 205-233. McGowan, K.C., Pohlald, F. G., Saunders, F.M., and
Dodt, M. E., Sweatman, M. B, and Bergstrom, W. R. Williams, N.D. (1988). “A microbial model of
(1987). “Field Measurement of Landfill surface landfill stabilization.” Proc. 1988 Joint CSCE-
settlements.” Geotechnical Practice for Waste ASCE Nat. Conf., ASCE, New York, N.Y., 704-
Disposal ’87, Ann Arbor, Mich., 13, 406-418. 711.
Edil, T.B.,Ranguette,V.J. & Wuellner, W.W. (1990) Merz, R.C., and Stone, R. (1962).” Landfill
Settlement of municipal municipal refuse. In: settlement rates” J. Public works, 93(9), 103-106.
Landva, A. & Knoeles, D. (eds) Geotechnics of Mitchell, R. A. And Mitchell, J.K., “Stability
Waste Fill- Theory and Practice. ASTM STP Evaluation of Waste Landfills,” Stability and
1070. Performance of Slope and Embankments- II,
El-Fadel, M., Shazbak, S., Saliby, E., & Leckie, J., ASCE Geotechnical Special Publication No. 31,
(1999) “Comprative assessment of settlement 1992.

NAVFAC, US Department of the Navy (1983), “Soil .
Mechanics Design Manual 7.3”, Naval Facilities
Engineering Command, Alexandria, VA.
Oweis, I.S & Khera, R. (1986) Criteria for
geotechnical construction of sanitary landfills. In:
H. Y. Fang (ed.) International symposium on
Environmental Geotechnology. Bethlehem,
Pennsylvania, USA: Envo Publishing Company,
pp.205-222.
Park, I. H., and Lee, R.S., (2002) “Long-term
settlement behaviour of MSW landfills with
various fill ages” Sage Publication on behalf of
international solid waste Association, Journal of
Waste Management and Research 20, pp. 259-
268.
Philadelphia, Pennsylvania, USA: American Society
of Testing and Materials, pp.225-239.
Rao,S.K., Moulton, L.K., and Seals, R.K.(1977).
“Settlement of refuse landfills” Geotechnical
practice for disposal of solid waste materials. Ann
Arbor, Mich., 574-599.
Sharma, D. H., (2000) “Solid Waste Landfills
settlements and Post-closure perspective “ASCE,
Journal of Environmental and Pipeline
Engineering” , pp. 447-455.
Sharma, H.D., and Lewis, S.P. (1994), “Waste
Containment Systems, Waste Stabilization and
Landfill.” John Wiley and Sons, Inc., New York,
Page 588.
Sower, G.F. (1973) Settlement of Waste Disposal
Fills. Proceedings of the Eight International
Conference on Soil Mechanics and Foundation
Engineering. Moscow, Russian Federation: pp
207-210.
Sterns, R.P. (1987). “Settlement and gas control: Two
key post-closure concerns.” Waste Age, 18(3), 55-
60.
Tang, W. H., Gilbert, R. B., Angulo, M. & Williams,
R.S. (1994) Probabilistic observation method for
settlement-based design of a landfill cover. In:
Yeung, A.T. & Felio, G.Y. (eds) Vertical and
Horizontal Deformations of Foundations and
Embankments. Geotechnical Special Publication
40. New York, USA: ASCE, pp. 1573-1589.
Wall, K. D., and Zeiss, Chris., (1995) “Municipal
Landfill Biodegradation and settlement” ASCE,
Journal of Geotechnical and Geoenvironmental
Engineering, Volume 121, Issue 3, pp. 214-223.
Wardwell, R. E., & Nelson, J.D. (1981) Settlement of
Sludge Landfills with Fiber Decomposition.
Proceedings, Tenth International Conference for
Soil Mechanics and Foundation Engineering, Vol.
2, Stockholm, Sweden, 397-401.
Young, A. (1989), “Mathematical modelling of
landfill degradation” J. Chemical Technol. And
Biotechnology,. 46, 189-208.
E-Resources
http://www.envfor.nic.in/divisions/cpoll/delpolln.html
http://www.tribuneindia.com/2007/20070307/delhi
.html

ISBN 978-81-921121-0-7
CEMENT AND POLYPROPYLENE FIBRES TO IMPROVE THE LOAD BEARING CAPACITY

OF PEAT
Behzad Kalantari
Dept of Civil Engg, University of Hormozgan, Bandar Abbas, Iran, Email: behzad996@yahoo.com
Arun Prasad
Dept of Civil Engg, Banaras Hindu University, Varanasi-221005, India, Email: prasada@bhu.ac.in
ABSRACT: This article describes the results of a laboratory study on the stabilization of peat using cement as
binding agent and polypropylene fibres as reinforcement. The samples of peat were observed to have very high
natural water content (>150%). Hence, the samples were cured in air at room temperature. The effectiveness of
cement and fibres in the stabilization of peat was evaluated by performing unconfined compression strength (UCS)
and California bearing ratio (CBR) tests. The stabilized samples were air cured for 28, 90, and 180 days for the UCS
test and 90 days for CBR (soaked and un-soaked) test. The amount of cement used to stabilize peat samples for UCS
test was 15, 30 and 50%. Similarly, the amount of cement used to stabilize peat samples for CBR test was 15, 25,
30, 40 and 50%. Further, the amount of fibres used to stabilize peat for CBR test was 0.1, 0.15, and 0.25%. It was
observed that as the curing time increased, the moisture content decreased and in turn, the water-cement ratio (w/c)
also reduced; thereby the samples hardened and gained strength. Also, the polypropylene fibres acted as
reinforcements, when added to peat with cement, and further increased the strength of peat.
INTRODUCTION This air curing method causes the high water content of
Peat is a soil that contains a significant amount of the stabilized peat soil to gradually decrease with time
organic materials. It is known to deform and fail under a because of the hydration process and resulting in an
small surcharge load and is characterized by low shear increase in strength values.
strength, high compressibility and high water content
(Huat, 2004). The different methods (Huat, 2007) TEST MATERIALS
available to improve the load carrying capacity of poor
Peat soil samples were collected as disturbed and
soils include; transfer the load to a more stable soil layer
undisturbed according to AASHTO T86-70 and ASTM
without improving the properties of the in-situ soil;
D42069 from Kampung, Jawa on the western part of
remove the soft soil and replace it, fully or partially,
Malaysia. Table 1 presents the properties of the in-situ
with better quality fill and; improve the in-situ soil
(field) peat soil, and Fig. 1 shows sieve analysis of the
properties with different techniques of ground
peat used in the study as well.
improvement.
Binding agent used for this study was ordinary Portland
To improve the performance of soil with low bearing
cement, and polypropylene fibres (Fig. 2) was used as
capacity, cement has been used by many researchers
non-chemically reactive additive to reinforce the
(Ismail et al. 2002; Baisha et al. 2005; Kasselouri et al.
stabilized peat (Table 2).
2004; and Yetimoglu, and Inanir 2004). Some studies
have also been carried out by researchers (yetimoghlu
and Inanir 2004; Park and Tan 2005; Tang et al. 2007; EXPERIMENTAL PROGRAM
Sivakumar et al. 2008) to study the influence of fibre In order to examine the effect of cement admixture and
inclusion on the mechanical behavior of cemented soil. polypropylene fibres on the load bearing capacity of
The reports in the literature show that randomly peat soil, index properties tests on the peat soil have
distributed polypropylene fibres can be used to been conducted. The tests include: sieve analysis, water
overcome the drawback of using cement alone such as content, liquid and plastic limits, organic content,
high stiffness and brittle behavior of the stabilized soil. specific gravity, and fibre content. Strength tests used
for the study were UCS, and CBR tests. Peat soil used
for the stabilized samples of UCS, and CBR tests were
In the present model study, peat soil is stabilized with
at their natural or field moisture contents, and therefore
cement as binding agent and also reinforced with
no water was added or deducted from the peat soil
polypropylene fibres as chemically non-active additive.
during the mixing process of peat, cement, and fibres.
Air curing method as described by Kalantari and Huat
(2008) is adopted to cure the samples and to strengthen
the stabilized peat soil by keeping it in normal room air
temperature and humidity without adding water from
the outside.

Table 1 Properties of the peat Unconfined Compressive Strength (UCS)
Properties Standard Values Unconfined Compression Strength tests have been
Specifications conducted on the undisturbed peat soil as well as
Depth of sampling 5 to
100cm stabilized peat soil with OPC and fibres. Disturbed
Moisture content (%) ASTM D2216 198 - 417 samples used for the stabilized peat soil’s UCS tests
In-situ bulk density 10.23 – were the peat soil samples at their natural (field)
(kN/m3) 10.40 moisture content. Specified amounts of OPC and
Specific gravity BS 1337 1.22 polypropylene fibres were added to peat soil, mixed
Classification/Von post H4 –H5
Liquid Limit (%) BS 1337 160
well for their homogeneity. Then the mixture has been
Plastic Index (%) BS1337 N.P placed in three layers in UCS mould having inside
Organic content (%) ASTM D2974 80.23 diameter of 38 mm and L/D of 2. Each layer was given
Unconfined Compression ASTM 2166-66 28.5 10 constant full thumb pressures of approximately 10
Strength, Undisturbed (kPa) seconds as used in Sweden for compacting stabilized
California Bearing Ratio ASTM D1883-73 0.782
(%), Undisturbed peat soil samples in their mould described by Axelsson
et al. (2002), and trimmed at both ends, extracted by
extractor jack, and wrapped in plastic sleeves for the
Table 2 Polypropylene fibres specifications (Sika curing period. Curing periods for the stabilized samples
fibres, 2005) were immediately after mixing, 28, 90, and 180 days.
Property Specification
California Bearing Ratio (CBR)
Base Polypropylene fibres with surface agent CBR tests have been conducted on the undisturbed peat
Color Natural as well as stabilized peat with OPC and fibres. For the
Specific gravity 0.91 g/cm3 stabilized peat with OPC (mixture of peat and cement)
Fibre Length 12 mm the soil samples used were samples at their natural
Fibre Diameter 18 micron-nominal (field) moisture content. Specified dosage of OPC and
Tensile Strength 300-440 (MPa) polypropylene fibres were mixed well with the peat for
Elastic Modulus 6000-9000 (N/mm2)
their uniformity and homogeneity, before placing them
in the CBR mold in five layers as stated by the ASTM
water absorption None
standard.
Softening point 160 °C
The prepared stabilized CBR peat soil samples with
OPC and fibres being at their relevant molds were then
kept in the curing condition. Curing period used for the
stabilized peat soil samples was 90 days, before
conducting the soaked as well as the un-soaked CBR
tests on the samples.
Curing Procedure
In-order to cure the stabilized peat soil samples with
OPC and fibres, air curing technique has been used. In
this technique the stabilized peat soil samples for UCS
and CBR tests were kept in normal air temperature of 30
± 2 ºC and out of reach of water intrusion (outside
Fig. 1 Grain size distribution curve source) during the curing period. Air curing technique is
used to strengthen the stabilized peat samples, instead of
the usual moist (water) curing technique or water
submergence method which has been a common
practice of past experiments for stabilized peat soil with
cement described by Axelsson et al. (2002), and
Duraisamy et al. (2006).
The principle of using air curing technique is that peat
when mixed with cement at its natural moisture content
has more than enough water (water content from198 to
417%) for complete curing process to take place.
Therefore the treated peat with cement samples do not
Fig. 2 Polypropylene fibers need more water (submerging the samples in water) for
the curing process to take place, and the already

contained moisture content of the stabilized peat with order to investigate the CBR values of the soaked
cement can carry on the hydration process, thus stabilized peat soil, a set of CBR samples made of
strengthening the samples without extra water. different dosages of cement and fibres(15, 25, 40, and
50% OPC, with 0.15% of fibres) at soil natural moisture
Mixture Dosages content were cured in air for 90 days, and then soaked in
For UCS tests, each set of samples consist of peat soil water for a period of five weeks. During these five
having natural moisture contents plus 15, 30, and 50% weeks of soaking, the soaked CBR soil samples tested
of Ordinary Portland Cement(e.g. 15% cement means in for weight increase until 100% or complete saturation is
each 100 g of peat having natural (field) moisture revealed or obtained and therefore no weight increases
content, 15 g of ordinary Portland cement added), with recorded then after.
and without Polypropylene fibres. The fibre amount
used for the mixtures was 0.1, 0.15, and 0.25 % ( e.g. RESULTS AND DISCUSSIONS
0.15% fibre means for each 100 g of peat at its in-situ
(field) moisture content, 0.15 g fibres were added). Optimum Dosage Determination Tests of
For CBR (un-soaked, and soaked) tests, each sample Polypropylene Fibres
consists of peat soil with natural moisture content plus According to the results shown on Fig. 3, the mixture
15, 25, 30, 40, and 50% of ordinary Portland cement consisting of peat, cement and 0.15% fibers gave
(e.g. 25% cement means for each 100 g of natural peat maximum CBR (un-soaked and soaked) values. Thus
at its in-situ moisture content 25 g of cement was this dosage rate of fibres was chosen to be the optimum
added), with and without additive ( polypropylene amount to be used during the study.
fibres). The polypropylene fibres amount used for the
stabilized CBR samples was 0.15% (e.g. for each 100 g
of peat at its in-situ moisture content, 0.15 g fibres were
added).
Optimum Polypropylene Fibres Dosage

Determination Tests
Polypropylene fibres is usually used in concrete mixes
to control cracks in hardened concrete (Mulli et al.
2006).The usual dosage recommended for concrete Fig. 3 CBR of stabilized samples
mixes differs from 0.6 to 0.9 kg/m3 ( Polypropylene
Fibres, 2005). Soaking Duration Determination Tests
As for soil stabilization, fibres have been used to According to the result shown on Fig. 4, the least
stabilize clayey soil, and according to Nagu et al. (2008) contained OPC (15% of OPC) stabilized peat soil
study on the strength of stabilized clayey soil reinforced sample reaches its 100% saturation and its constant
with nylon fibres. 0.4% of fibres would provide the weight during the first 4 days and the sample with
max. UCS values for the stabilized clayey soil. highest amount of OPC (50%) reaches its constant
For this study in order to find the optimum percentage weight (100% saturation) within first 6 days of
of fibre contents for the stabilized peat that would saturation or soaking process. Based on the obtained
provide the max. strength, peat samples at their natural result of this test, all stabilized peat soil samples before
moisture contents mixed with different percentages of going under soaked CBR tests were submerged in water
OPC and fibres cured in air for a period of 180 days for at least 6 days.
before being tested for their CBR values.
The samples examined for this purpose consisted of
15% of ordinary Portland cement as well as 0.1, 0.15,
and 0.25% of polypropylene fibres. The sample which
showed the maximum strength value for CBR (soaked,
and un-soaked) was chosen as the optimum fibres
mixture dosage for further study of treated peat with
cement.
Soaking CBR Test Procedure

According to AASHTO T193-63,and ASTM D1883-73,
the soaking period of CBR samples for normal soil is 96 Fig. 4 Number of days for soaked CBR samples versus
hours or four days (Bowles 1978).For this study, in- percentage of weight increase.

Effect of Fibres and Curing technique on the UCS of Therefore, it is possible to conclude that, for the
Treated samples hardening process for concrete mixes to take place, not
Results obtained from UCS tests shown on Fig. 5 more than an average W/C of 0.5 is required and as the
indicate that, addition of OPC to peat soil will increase W/C is increased from 0.5 the compressive strength of
the UCS values. concrete decreases.
According to the results obtained in this study for w/c
versus curing time (age) shown on Fig. 6, w/c values
decrease with increasing curing time. And during 180
days of air curing, the w/c of the stabilized peat soils
never falls below 0.5. This suggests that the initial water
content of the natural peat soil is more than sufficient to
carry on the hydration process, and no need for any
external source of water, and this justifies use of air
(a)
curing technique over moist curing technique for
As the curing time for the stabilized peat soil samples
stabilizing peat.
increases the UCS values increase as well. Also,
addition of polypropylene fibers will increase UCS
values further. Values of UCS increase considerably
after 28 day of curing period. The strength gain
continues through 180 days of curing.
Fig. 6 Water-cement ratio versus age of samples mixed

with cement for UCS tests.
(b) UCS test results of 0 (immediately after mixing), 7, 28,

90, and 180 days for stabilized peat with OPC in Figure
7 show that, as W/C values are decreased through
curing process, the UCS values are increased.
Therefore, Air Curing process cause the stabilized peat
soil samples to gradually lose their moisture contents
and become drier and as the stabilized peat soil become
drier(water content is reduced) through air curing
process, the UCS values increase.
(c)
Fig. 5 UCS versus curing periods of treated samples: (a)
15% OPC, (b) 30% OPC, and (c) 50% OPC with and
without polypropylene fibres.
Effect of Curing age and W/C on the UCS values of

Treated Samples
In general concrete that is made of cement and different
sizes of aggregates hardens by time and gains most of Fig. 7 Water-cement ratio versus UCS of treated peat
its ultimate compressive strength at 28 days of age. One with different amount of cement.
of the principal factors affecting the compressive
strength of concrete is the w/c (weight of water /weight Actually, a considerable strength gain of the stabilized
of cement) of the fresh mixed concrete (Davis et al. peat soil is after 28 days of being cured in air, and the
1983). Usually for normal concrete mixes a w/c of about strength gain continues through 90 and 180 days of
0.5 is suggested, and as w/c for fresh concrete increases curing.
the compressive strength of concrete at 28 days will
decrease (Road note, 1975).

Effect of Air curing technique on the CBR values of Air curing technique of stabilized peat during six
Treated Samples months with 15% of OPC (less than 190 kg/m3), will
The result of CBR tests for stabilized peat samples with increase the UCS of in-situ peat by over 13 folds (from
OPC and fibers after being air cured for 90 days are 28.5 kPa. to 380 kPa.), and increases the UCS value of
shown on Fig. 8. The results indicate that as cement in-situ peat by over 14 folds when 0.15% (less than 2
amount in the mixture is increased, the CBR values are kg/m3) polypropylene fibers added to the mixture of
increased and addition of fibers content will increase the peat and OPC. Also air curing technique of three
CBR values further. Also CBR values are decreased for months for the stabilized peat with 15% OPC with
soaked (saturated) samples compared with the unsoaked addition of 0.15% of polypropylene fibers increase the
samples. Fibers contribute more strength to the lower CBR values of in-situ peat by over 23 folds(from 0.8%
dosage of OPC (15%) mixed with peat samples to 19%) for the un-soaked samples, and by over 8 folds
compared with higher dosage of OPC (50%) in the mix. for the soaked samples as well.
The air curing technique as well as OPC and fibers
used for stabilized peat will increase the general rating
REFERENCES
of the in-situ peat soil from very poor (CBR from 0 to
Axelsson, K., Johansson, S.E. and Anderson, R. (2002).
3%) to fair and good (CBR from 7 to above 20%).
Stabilization of organic soils by cement and
puzzolanic reactions-feasibility study. Swedish Deep
Stabilization Research Centre, Report 3, English
translation, 15-16.
Black, A.J, Sivakumar, V., Madhav R.M., and Hamil A.
G. (2007). Reinforced stone column in weak deposit:
laboratory model study., Journal of Geotechnical
and Geoenvironmental Engineering, 133(9), 1154-
Fig. 8 CBR values of undisturbed peat as well as 1161.
different percentage of OPC and polypropylene fibers Baisha, E.A., Hashim, R., Mahmud, H.B. and
for the stabilized peat cured for 90 days. Muntobar, A.S. (2005). “Stabilization of Residual
soil with rice husk ash and cement,” Construction
Also, visual inspection of CBR samples depict that, and Building Materials. 19(6), 448-453.
Polypropylene fibers not only increases the CBR values Bowles, E.J. (1978). Engineering Properties of Soil and
rather contribute a considerable amount of uniformity Their Measurements, 2nd edition, McGraw-Hill,
and intactness to the stabilized peat soil samples as well, USA.
when compared with the soaked stabilized peat soil British Standards Institution. (1990). Methods of Test
samples having only Ordinary Portland Cement. for Soils for Civil Engineering Purposes. London,
BS 1337.
CONCLUSION Davis, E.H., Troxell, E.G., and Hoauk, F.W. G. (1982).
In this study, peat has been stabilized with OPC as a The Testing of Engineering Materials, 4th edition.
binding agent alone, as well as with, OPC, and McGraw-Hill, USA
Polypropylene fibers as none chemical additive. Air Design of Normal Concrete Mixes. (1975). Road note
curing technique used in the study will cause the No. 4, Department of the Environment, Building
stabilized peat samples to gradually lose their moisture Research Establishment, Her Majesty’s Stationary
content through curing period and become drier. The Office, UK, 5-30.
binding agent and the additive as well as the curing Duraisamy, Y., Huat, B.B.K , Muniandy, R.,and Abdul
technique have proved to increase UCS values of the A.A. (2006). Effect of cement on the compressibility
stabilized peat soil samples at 28, 90, and 180 days. of tropical peat soil.”,GSM-IEM Oktober forum on
The result of UCS tests for stabilized peat soil with OPC Engineering Geology and Geotechnical engineering
with different percentage of polypropylene fibers show , Petalang Jaya, October 13, Malaysia.
that 0.15% is the optimum percentage to provide Engineering and Design Laboratory Testing, Engineer.
maximum UCS values. Curing the stabilized peat soil (1980). No 11102-2-1906 Manual. Department of
samples in air, will cause the w/c of the samples to the Army, USA.
reduce. As the w/c for the stabilized peat soils are Hebib, S., and Farrell, R. E. (2003). Some experiences
reduced through curing period, the UCS and CBR on the stabilization of Irish peats. Can. Geoteh J.,
values increase. Peat at its natural state, when mixed 40:107-120.
with OPC has high initial w/c (w/c > 0.5) and it Huat, B.B.K. (2004). Organic and Peat Soils
maintains enough w/c during the air curing process even Engineering, University Putra Malaysia Press,
through 180 days of curing period for hardening process Malaysia.
to take place.

Huat, B.B.K., and Faisal, H.A. (2007). Ground
Improvement Technology, University Putra Malaysia
Press, Malaysia.
Ismail, M.A., Joer, H.A., Sim, W.H. and Randolph, M.
(2002). “Effect of cement type on shear behavior of
cemented calcareous soil,” Journal of Geotechnical
and Geoenvironmental Engineering. 128(6) , 520-
529.
Kalantari, B. and Huat, B.B.K. (2008). Peat soil
stabilization using ordinary portland cement,
polypropylene fibers and air curing technique, Elec.
J. Geotech. Eng., Bund. J., No. 13:1-13.
Kolias, S.V., Kasselouri, R. and Karahalios, A. (2005)
Stabilisation of clayey soils with high calcium fly
ash and cement, Cement and Concrete Composites,
27(2): 301-313.
doi:10.1016/j.cemconcomp.2004.02.019
Mulli, K.A., Walia, P., and Sharma, N.S. (2006).
Application of polypropylene fiber reinforced
concrete (pfrc) with vacuum processing.”Advances
in Bridge Engineering, Walia International Machines
Corp., New Delhi, India.
Nagu, P.S., Chandarkaran, S., and Sankar, N. (2008).
Behavior of lime stabilized clayey soil, reinforced
with nylon fibers. Proceedings of ’08 International
Conference on Geotechnical and Highway
Engineering, Geotropika, Kuala Lumpar, Malaysia,
May.
Nevile, A.M. (2001). Properties of Concrete, 4th edition,
Longman, Malaysia.
Park, T. and Tan, S.A. (2004). Enhanced performance
of reinforced soil walls by the inclusion of short
fiber,” Geotextiles and Geomembranes, 23:348-361.
doi:10.1016/j.geotexmem.12.002
Sika Fibers, (2005). Technical Data Sheet, Version no.
0010 edition 3, Sika Fiber, Malaysia.
Tang, C.B., Shi, W. Gao, W., Chen, and Cai, Y. (2006).
Strength and mechanical behavior of short
polypropylene fiber reinforced and cement stabilized
clayey soil, Geotextiles and Geomembranes,
25(3):194-202. doi:10.1016/j.geotexmem.11.002
Yetimoglu, T., Inanir, M. and Inanir, O.E. (2004). A
study on bearing capacity of randomly distributed
fiber-reinforced sand fills overlying soft clay,”
Geotextiles and Geomembranes, 23:174-183,
doi:10.1016/j.geotexmem.09.004

ISBN 978-81-921121-0-7
EFFECT OF NaOH MOLARITY ON COMPRESSIVE STRENGTH OF

GEOPOLYMER CONCRETE IN AMBIENT CURING CONDITION
S. V. Joshi
Associate Professor, Shree Ramdeobaba Kamla Nehru Engineering College, Nagpur, Maharastra (India)
M. S. Kadu
Professor & Head, Shree Ramdeobaba Kamla Nehru Engineering College, Nagpur, Maharastra (India)
ABSTRACT: Conventional cement concrete usage around the world is second to water. Ordinary Portland cement
(OPC) is conventionally used as primary binder to produce conventional OPC concrete. The amount of carbon dioxide
released during the manufacture of OPC and the extent of energy required to produce OPC are the matters of
environmental concern and poor availability of power. The efforts are being made on various fronts to address these
issues. Fly-ash, abundantly available byproduct of coal fired thermal power stations, having no binding properties of its
own, is now being widely used as an additive binding material in manufacture of concrete .Fly-ash with alkali activators
can produce effective binding material, geo-polymer, through polymerization process which can be used to develop geo-
polymer concrete (GPC). The characteristics of alkali activator, decides the quality of geo-polymer and therefore affects
the important properties of plastic and hardened concrete. In the present work a mixture of sodium hydroxide solution
and sodium silicate solution is used as alkali activator. The present paper deals with exploring the possibility of using
locally available flyash, from thermal power plants in Nagpur region of the State of Maharashtra (India), to effectively
manufacture geo-polymer concrete and studying the effect of molar concentration of sodium hydroxide in alkaline
activator on compressive strength of Geo-polymer concrete. The laboratory investigations provides that under ambient
curing conditions, locally available low-calcium fly ash is suited for manufacture of geo-polymer concrete and the
compressive strength of geo-polymer concrete increases with increase in molar concentration of sodium hydroxide and
best suited molar concentration is 12 with respect to safety in handling and the cost effectivenessaswell
.
INTRODUCTION decreasing the amount of cement in concrete, and by
Ordinary Portland cement (OPC) is conventionally used as decreasing the number of building elements using cement.
primary binder to produce concrete. Worldwide the In Eastern part of Vidharbha region of the state of
production of cement is increasing about 7% annually. The Maharashtra (India), there are four major thermal power
environmental issues associated with the production of plants established in the vicinity of Nagpur City and ill-
OPC are well known. The extent of energy required to effects of disposal and storage of fly-ash are well known.
produce OPC is only next to steel and aluminum. The The relevance of the study lies in exploring the possibility
production of one ton of cement liberates about one ton of of using abundantly available fly-ash in the manufacture
CO2 to the atmosphere, as the result of de-carbonation of of concrete as an alternative to cement in an attempt to
limestone in the kiln during manufacturing of cement and reduce the emission of greenhouse gases, reduction in
the combustion of fossil fuels (Roy 1999). The contribution energy requirement and disposing off the byproducts in an
of Portland cement production worldwide to the environment-friendly way. The present work is aimed at
greenhouse gas emission is estimated to be about 1.35 evaluating the possibility of using locally available flyash,
billion tons annually or about 7% of the total greenhouse and to study the suitability and effects of alkaline activator
gas emissions to the earth’s atmosphere (Malhotra2002). parameter such as molar concentration of sodium
The global warming is being seriously considered at hydroxide solution, in ambient curing condition in
national and international level. The greenhouse effect temperature range of 35-45 degree centigrade, on the
created by the industrial emissions is increasing the global compressive strength of geo-polymer concrete.
temperature that is resulting in climate changes. Therefore,
any action or attempt made to reduce the effect should be GEOPOLYMERS
encouraged and given more attention. In order to produce Geo-polymers are members of the family of inorganic
environmental-friendly concrete, Mehta (2002) suggested polymers. The chemical composition of the geo-polymer
the use of fewer natural resources, less energy, and material is similar to natural zeolitic materials, but the
minimize carbon dioxide emissions. McCaffrey (2002) microstructure is amorphous instead of crystalline (Palomo
suggested that the amount of carbon dioxide (CO2) et al. 1999; Xu and van Deventer 2000). Unlike ordinary
emissions by the cement industries can be reduced by Portland / pozzolonic cements, geo-polymers do not form
decreasing the amount of calcinated material in cement, by calcium silicate-hydrates (C-S-H) for matrix formation,

but utilize the poly-condensation of silica and alumina microstructure ( Gourley 2003). The spherical shape of fly
and a high alkali content to attain structural strength. ash often helps to improve the workability of the fresh
Therefore, geo-polymers are sometimes referred to as concrete. Its small size particle acts as filler of voids in the
alkali activated alumino silicate binders. Geo- concrete to produce dense and durable concrete.
polymerization involves the chemical reaction of alumino- Devodovits et.al.(2005)claimed that to produce optimal
silicate oxides (Si2O5, Al2O2) with alkali polysilicates binding properties, the low-calcium fly ash should have the
yielding polymeric Si-O-Al bonds. ( J. Davidovits- 1985). percentage of unburned material (LOI) less than 5%,
Fe2O3 content should not exceed 10%, and, reactive silica
Constituents of Geo-polymer should be between 40 – 50%, and 80 – 90% of particles
should be smaller than 45 µm. Bagchi,etal.(2005)provided
Source Materials characterization of fly ash from one of the prominent
Any material that contains mostly Silicon (Si) and thermal powers Plant Koradi Thermal Power Station
Aluminum (Al) in amorphous form is a possible source (KTPS) in Vidharbha region (Maharashtra/India). It is
material for the manufacture of geo-polymer. Several suggested that particles of fly ash under 10 microns
mineral and industrial by-product materials have been (regardless of the type of fly ash) are the ones that
investigated in the past. The calcinated source materials contribute to 7 and 28 day strengths. Particles between 10
such as fly ash, slag, calcinated kaolin, demonstrated a to 45 microns are the ones that slowly react between 28
higher final compressive strength when compared to those days and about one year. The particles above 45 microns
made using non-calcinated materials. Among the by- are considered as inert, as that do not participate in
product materials, only fly ash and slag have been proved pozzolonic reactions even after one year, and behave like
to be the potential source materials for making geo- sand. Particles below 10 microns decide the fly ash
polymers. Fly ash is considered to be advantageous due to reactivity.
its high reactivity that comes from its finer particle size Alkaline activators
than slag. Moreover, low-calcium fly ash is more desirable Alkaline liquid plays an important role in the
than slag for geo-polymer feedstock material. polymerization process (Palomo et al.1999). Reactions
Fly Ash occur at a high rate when the alkaline liquid contains
Fly ash, an abundantly available byproduct of thermal soluble silicate, either sodium or potassium silicate,
power stations, particles are typically spherical and finer compared to the use of only sodium hydroxides (Xu and
than portland cement and lime. The diameter ranges from Van Deventer, 2000). Addition of sodium silicate solution
1 micron to 150micron. Fly ash does not have any to the sodium hydroxide solution to prepare the alkaline
cementing properties by itself. In general, the reactivity of liquid enhanced the reaction between the source material
flyash depends upon chemical composition, fineness and and the solution. In general the NaOH solution caused a
percentage of amorphous or reactive silica present. It also higher extent of dissolution of minerals than the KOH
depends on quality of coal used as fuel. The types and solution (Palomo et al. 1999) .The compressive strength
relative amount of incombustible matter in the coal and the workability of geo-polymer concrete are influenced
determine the chemical composition of fly ash. Fly ash that by the proportions and properties of the constituent
results from burning sub-bituminous coals is referred as materials that make the geo-polymer paste.
ASTM Class C fly ash or high-calcium fly ash. It is Water to Geo-polymer solid ratio by mass also plays
typically contains more than 20 percent of CaO .On the important role in development of concrete. It is the ratio of
other hand, fly ash from the bituminous and anthracite the total mass of water to the total mass of geo polymer
coals is referred as ASTM Class F fly ash or low-calcium solids. The total mass of water is the sum of mass of water
fly ash. It consists of mainly an aluminosilicate glass and contained in the sodium silicate liquid, the mass of water
has less than 10 percent of CaO. The colour of fly ash can in the sodium hydroxide liquid and the mass of extra
be tan to dark grey depending upon the chemical and water, if any added to the mixture. The total mass of geo-
mineral constituents (Malhotra and Ramezanianpour polymer solid is the sum of mass of solids in the sodium
1994). The loss on ignition (LOI) is a measure of unburnt silicate solution (i.e. the mass of Na2O and SiO2), the
carbon remaining in the ash. Fineness of fly ash mostly mass of sodium hydroxide solids and the mass of fly ash.
depends on the operating conditions of coal crushers and
the grinding process of the coal itself. Finer gradation METHODOLOGY
generally results in a more reactive ash and contains less The laboratory investigation was focused at studying the
carbon. Low-calcium (ASTM Class F) fly ash is preferred effect of molar concentration of sodium hydroxide liquid
as a source material than high-calcium (ASTM Class C) on compressive strength of concrete.The methodology
fly ash. The presence of calcium in high amount may adopted for the present study is summarized in the
interfere with the polymerization process and alter the following steps.

geopolymer concrete (Rangan B.V. 2005)
(1) Design of reference cement concrete mix for M25

grade OPC concrete using IS code method. (IS: 10262- Table 1: Comparison of Chemical composition of Fly
1982) Ash
As of now, no standard mix design procedure is Composition Rangan et al flyash for
available for geo-polymer concrete, however, same (2005) present work
codes and procedures used for design of OPC concrete Batch I
can be used for geo-polymer concrete (Rangan 2005).
Therefore, for M25 grade OPC concrete, reference mix SiO2 53.36 58.59
is designed using IS code method (IS:10262-1982) Al203 26.49 35.55
and it is followed for determining the % of mass of Fe203 10.86 1.27
combined aggregate in a reference geo-polymer mix. CaO 1.34 1.66
For M-25 Grade, quantity of material required for 1 MgO 0.77 0.34
cum OPC concrete are as follows SO3 1.70 0.36
Na2O 0.37 0.28
Content Wat Ceme F.A. C.A. K2O 0.80 0.62
er nt TiO2 1.47 1.82
Kg 186 422 600 1115 P2O5 1.43 0.60
Ratio 0.44 1 1.42 2.64 LOI 1.39 0.58
(2) Computation of ingredients of geo-polymer While preparing workable mix in laboratory, it was observed that,
concrete higher water to Geo solids ratio is required to develop workable
geo-polymer concrete. It is almost in the range of 0.23 to 0.30
Aggregate which is higher than that recommended range of 0.16 to 0.24
Geologically, major portion of the Vidharbha region of (Rangan 2005). It may be due to high percentage of SiO2 and
AI2O3, and Low percentage of Fe2O3 available in the KTPS flyash.
Maharashtra is covered with black basalt. Therefore,
coarse aggregates (CA) derived from basalt are used. Super plasticizer
Experimental trials suggested that CA of the size of 14 mm As ,water to geosolid ratio adversely affect the
and less can produce a cohesive mix. A combination of 14 compressive strength of geo-polymer concrete, , appropriate
plasticizers were suggested to be used for improving its
mm (40 %), 10 mm (30 %) and 7 mm (30 %) is proposed workability. The plasticizer suitability-compatibility study is
in the present work. The fineness modulus of CA is 5.20. undertaken on different types of plasticizers . It revealed that
Locally available good quality sand having fineness among the plasticizers tested, sulphonate naphthalene based
modulus 2.46 is used as fine aggregate (FA). The mass of plasticizer complying with IS-9103-1999 and BS 5075 part 3 and
combined aggregates ( CA + FA) is taken as 71% of mass ASTM-C-494 type 6 ,BASF Rheobuild–817 is suitable and used
of concrete in the proportion of 65 : 35 as worked out from as 1% of mass of flyash in the present study.
the mix design. Sample calculation: For M-25 Grade Geo polymer
concrete (1cum) using fly-ash and alkaline liquid,
Alkaline Liquid materials are calculated considering mass density of Geo-
Further,for the laboratory investigations alkaline liquid is polymer concrete as 2400 kg/cum, ratio of alkaline liquid to fly
prepared using commercially available sodium silicate ash as 0.35, and ratio of sodium silicate to sodium hydroxide as
2.5 as follows,
liquid and 8 to16 molar concentration sodium hydroxide
Mass of combined aggregate (CA+FA) =1115+600= 1715 kg,
liquid. .Sodium silicate liquid content Na2O = 15.22 %, (same as M25 grade OPC)
SiO2 = 28.53 % and water = 56.25 %. In alkaline liquid, Mass of alkaline liquid and fly-ash = 2400 – 1715 =685 kg
ratio by mass of sodium silicate to sodium hydroxide is Mass of fly-ash = 685/1.35 = 507 kg,
maintained as 2.5, ratio of alkaline liquid to fly-ash by Mass of alkaline liquid = 685 – 507 = 178 kg .
mass is maintained as 0.35, and water to Geo-polymer Mass of sodium hydroxide liquid = 178/(1 +2.5) = 50.85 kg
solid ratio by mass is 0.252. Mass of sodium silicate liquid = 178 – 0.85 = 127.15 kg,
% of mass of combined aggregate = (1715/2400) x100 = 71.46,
Flyash % of mass of coarse aggregate = 1115/1715) x100 =65,
Flyash is procured from Koradi Thermal Power Station (KTPS)-- % of mass of fine aggregate = (600/1715) x100 = 35,
Nagpur in Vidharbha region (Maharashtra/India). The Table 1 Therefore CA: FA as 65 : 35.
provides the comparison of chemical composition of the fly ash
used in the present study and fly ash used in Australia for

the alkaline liquid can be preserved for 3-4 days in unavoidable
situations.
Table 2: Determination of quantities of ingredients of geo- (6) Alkaline liquid gets solidified on account of formation of
polymer concrete sodium meta silicate. It is observed that the chances of
Ratio of Mass of Mass Mass of Mass Fine Coarse 1% solidification are more in lower ratios of sodium silicate to
alkaline Alkaline of Sodium of agg. Agg. Super sodium hydroxide i.e. upto 2.0. Therefore ratio of sodium
liquid liquid NaOH silicate Flyash in in Kg. plasticizer
silicate to sodium hydroxide is recommended as 2.5.
to in Kg Liquid liquid in Kg. Kg. in Kg.
fly ash In Kg in Kg. (7) For higher molar concentration of NaOH liquid from
12 to 16 M, chances of solidification are more. Therefore
12 M or even less concentration of NaOH liquid is
0.35 178 50.85 127.15 507 600 1115 5.07 recommend even though the corresponding compressive
strength is slightly less.
(8) Higher molar NaOH liquid is costly and equally dangerous
(3) 150 x 150 mm size casting and testing of cubes. to handle in laboratory. Even if a drop of higher molar
In order to study the effect of molar concentration of NaOH liquid concentration sodium hydroxide liquid (i.e.12 M – 16 M) comes
in alkaline liquid, on compressive strength of geo-polymer in contact with body it immediately burns the skin. Therefore all
concrete, cubes of size 150x150 mm are casted for various sorts of precautionary measures needs to be taken while handling
combinations of mix. The specimens are cured in ambient liquid, with higher molar concentration.
condition where temperature varies in the range of 35 - 45 degree ( 9) It can be concluded from the work carried out that the fly ash
centigrade. Specimens are tested after 14 days and 28 days.. A collected from KTPS, located close to Nagpur, a second capital
test samples of three specimens is considered to calculate of Maharashtra, where construction activity is on the hype, can be
average compressive strength of GPC. effectively used as geo-polymer source material.
TEST RESULTS SCOPE OF FUTURE RESEARCH

In line with the objectives of the present work, concrete cubes are
Further research with respect to the effect of water to geo-
casted and tested under ambient curing conditions as per IS -456
and the test results are presented in Table 3.
polymer solids ratio, sodium silicate to sodium hydroxide
Table 3: Effect of NaOH Molar Concentration on ratio by mass, and alkaline liquid to fly ash ratio by mass
Compressive Strength. on important properties of GPC need to be done in detail.
NaO 14 days 28 days Compressive % increase Also detailed investigation is required to decide the
H Compressive Strength in quality of most suitable sodium silicate liquid to prepare
mola Strength compressive
r Strength alkaline liquid with various molar concentration of sodium
Avg.Den Avg. Avg Avg.
conc sity in Comp .Density Comp from 14-28 hydroxide liquid.
entra Kg/m3 Streng in Kg/m3 Strength in days
tion th in MPa
MPa REFERENCES
8 2274 4.75 2307 8.55 80 Bagchi S.S. and Jadhav R.T. et al., (2005).
10 2276 6.32 2310 9.70 53.48 “Characterization of fly ash from Koradi Thermal Power
12 2281 7.95 2316 14.78 85.92
50.61
Plant for its use as a Pozzolona”. Journal IAEM, Vol.
14 2312 13.08 2348.6 19.70
16 2314 14.54 2351 21.60 48.56
32(3), 181–183.
Davidovits, J. (1994a). High-Alkali Cements for 21st
CONCLUSIONS Century Concretes. Paper presented at the V. Mohan
From the laboratory investigation, following conclusion are Malhotra Symposium on Concrete Technology: Past,
drawn: Present And Future, University of California, Berkeley.
(1) The density of geo-polymer concrete varies in the range of Gourley, J. T. (2003). Geopolymers; Opportunities for
2275 -2350 Kg /cum. Environmentally Friendly Construction Materials. Paper
(2) As the molar concentration increased from 8 molar to 16
presented at the Materials 2003 Conference: Adaptive
molar the 14 days and 28 days compressive strength of Geo-
polymer concrete is increased.
Materials for a Modern Society, Sydney.
(3) The rate of increase in compressive strength with increase in Hardjito D. and Rangan B. V., (2005). “Development and
molar concentration was slow for 8 to 12 molar concentration and Properties of Low-Calcium fly ash-based Geopolymer
was substancially steep beyond 12 molar concentration. Concrete”. Research Report GC 1 Faculty of
(4) It is observed that increase in strength from 14 to 28 days for Engineering Curtin University of Technology Perth,
various molar concentration of NaOH are in range of 50 % to Australia.
85%, it is maximum for 12molar concentration. Malhotra, V. M. (1999). Making Concrete "Greener" With
(5) A detailed investigation is required to decide quality of most Fly Ash. ACI Concrete International, 21(5), 61-66.
suitable sodium silicate liquid to prepare alkaline liquid with
Malhotra, V. M., & Mehta, P. K. (2002). High-
various molar concentrations of sodium hydroxide liquid. So that

Performance, High-Volume Fly Ash Concrete: Materials,
Mixture Proportioning, Properties, Construction Practice,
and Case Histories. Ottawa: Supplementary Cementing
Materials for Sustainable Development Inc.
McCaffrey, R. (2002). Climate Change and the Cement
Industry. Global Cement and Lime Magazine
(Environmental Special Issue), 15-19.
Palomo, A., M.W.Grutzeck, & M.T.Blanco. (1999).
Alkali-activated fly ashes A cement for the future.
Cement And Concrete Research, 29(8), 1323-1329.
Rangan B.V. (2008). “Mix design and production of fly
ash based Geopolymer”, Indian concrete journal may
2008, 7 – 14.
Roy, D. M. (1999). Alkali-activated cements Opportunities
and Challenges. Cement & Concrete Research, 29(2),
249-254.
Xu, H., & Deventer, J. S. J. V. (2000). The geopolymerisation of
alumino-silicate minerals. International Journal of Mineral
Processing, 59(3), 247-266.

ISBN 978-81-921121-0-7
ECONOMICS AND STUDY UNDER BORDER IRRIGATION ON

GROUNDNUT CROP AT VARIOUS WATER APPLICATION LEVELS
A.B. Rathod
Department of Water Resources Management,Civil Engineering, L. D. College of Engineering
Ahmedabad, E-mail: anilrathod_222399@yahoo.com
B.Jigar Parekh
Department of Water Resources Management,Civil Engineering, L. D. College of Engineering
Ahmedabad . E-mail:shaileshatrivedi@yahoo.co.in
ABSTRACT: The field experiments were conducted for consecutive three years to assess the crop
performance and economics of Border Irrigation system for groundnut (GG-2) during summer season (Feb. to
May). Total Six treatments having different water application levels based on IW/CPE like 0.6, 0.7, 0.8, 0.9, 1.0
and 1.2 were selected. The lowest pod yield of 1278 kg/ha and highest pod yield of 2125 kg/ha could be
obtained at the IW/CPE of 0.6 and 1.0 respectively requiring 531 and 884 ha.mm/ha of irrigation water
respectively. The treatment of IW/CPE of 0.9 was found statistically better one having the highest water use
efficiency of 2.52 kg/ha/mm. The IW/CPE of 0.9 was found most economical water application level. Also, in
case of excess water supply (IW/CPE=1.2) and deficit water conditions (0.6), the border was not that profitable.
INTRODUCTION
(IW/CPE of 0.6, 0.7, 0.8, 0.9, 1.0 and 1.2) were
During the summer season, there is a higher yield replicated 4 times.The water meters were used to
potential of groundnut in Saurashtra and Kutch region measure the volume of water applications. For the
of Gujarat State. However, the biggest constraint is border irrigation, the border slope was 0.7%. The
the limited water resources. One of the critical width of the border was kept as 1.8 m so that total 4
challenges to water resources management is to shift rows of the crop can be in one border. The parshal
from the extensive supply oriented approach to the flume was installed in the supply channel to measure
one focusing upon deficit applications There are the discharge.
several ways of increasing efficiency in irrigation.
One way is changing from surface to pressurized Economics
methods of irrigation and second is to apply deficit The cost of cultivation excluding cost of irrigation (Cc)
water. The extent to second is to apply deficit included cost of various inputs like cost towards land
irrigation is to increase the water use efficiency of a preparation, seeds, seed treatment, fertilizer, sowing,
crop by eliminating irrigations that have little impact agro chemicals, weeding, inter culturing, harvesting,
of yield. The resulting yield reduction may be small threshing, cleaning and packing etc. The cost of seeds,
compared with the benefits gained through diverting fertilizer and agro chemicals were taken following the
the saved water to irrigate additional area or other recommended package of agronomic practices. The
crops for which water would normally be insufficient cost of irrigation (Ci) includes the cost of labour,
under traditional practices. Before implementing a electricity and maintenance required for the irrigation
deficit irrigation programme, it is necessary to know application. The fixed cost (Ct) included the cost of
crop yield responses to water applications. pumping/delivery and irrigation system. It was
assumed that the 7.5 HP pumping system can serve 6
ha area (4ha/season, 2 season/year) for 15 years. Also,
MATERIAL AND METHODOLOGY
it was assumed that the irrigation system could be
Field Experiment : The experiment was carried out at useful for 2 seasons per year. The life of the border
Junagadh Agricultural University campus farm, irrigation system was taken as 10 years. The
Junagadh, Gujarat, India to assess the groundnut crop following expressions were used for assessing the
performance under border system at various water economics.
application levels during summer season (Feb to May)
for consecutive 3 years. The GG-2 variety of Ct = Cc + Ci in which, Ci = Cv + Ct
groundnut crop recommended for this area was Ct =
selected. The row spacing, seed and fertilizer rate
were kept as 0.45m, 100 kg/ha and 25:50:00 (N:P:K)
kg/ha as per agronomic recommendations for this
region. Total six treatments having different water
B/C = NB/Ct or B/C = GB/Ct in which GB =
application levels
Yp X Pp + Yf X Pf and NB = GB - Ct ;
Where,
Ct = Total Cost of cultivation (Rs./season/ha);

Cc = cost of cultivation excluding cost of irrigation treatments having IW/CPE of 0.8 and 0.9 were found
(Rs./ha/season); statistically at par.
Ci = cost of irrigation (Rs./ha/season); Cc = Cost of
cultivation excluding cost of irrigation (Rs./ha/season); The 3 years pooled data showed that the significant
Ci = Cost of irrigation (Rs./has/season); Cf = Fixed differences were observed among the pod yields data
cost of the irrigation (Rs./ha/season); Cv = variable of treatments having IW/CPE of 0.6, 0.7 and 0.8.. The
cost of irrigation water applications (Rs/ha/season); P1 treatments having IW/CPE of 0.8 and 0.9 were found
= Cost of pumping and delivery system, Rs; M = Life statistically at par. The IW/CPE of 1.0 was taken as
of the pumping system, years; P2 = cost of irrigation statistically better one. It was found that the pod yield
system, Rs/ha; N = Life of the irrigation system, years; was decreased for the IW/CPE higher that 0.9. The
I = prevailing rate of interest, fraction. pod yield increased with decreasing rate from the
IW/CPE of 0.6 to 0.9. The observed pod yields at
various seasonal irrigation depths are presented The
RESULTS AND DISCUSSION optimal water requirements were founds as 884 mm
giving maximized pod yield of 2125 kg/has using the
CROP PERFORMANCE
best fitted curve.
Pod Yield
The data as presented in Table-1 indicated that, Fodder Yield
during the first year, the highest yield of The pooled data of three years (Table-1) showed that
1982 kg/ha was under treatment of MST5(1.0) using fodder yield was increased with increased with
seasonal water of 884 mm. The treatments MST4(0.9), increase in the water application level from IW/CPE
MST5(1.0) and MST6(1.2) were statistically at par. of 0.6 to 1.2. The yield under the IW/CPE of 0.6, 0.7,
However, during the second year, the highest pod 0.8 and 0.9 differed significantly. However, the
yield of 2332 kg/ha was observed in the treatment difference between the fodder yield under IW/CPE of
having IW/CPE of 1.0. The lowest pod yield of 1621 0.9 and 1.0 as well as 1.0 and 1.2 were insignificant. It
kg/ha was found in the MST1(0.6) treatment requiring could be seen that the fodder productions increased
seasonal water application of 523 ha.mm/ha. The first with increases in irrigation water inputs. However, the
and second highest treatments MST4(0.9) and rate of increase was low at higher water application
MST3(0.7) and MST3(0.8) treatments. During the levels indicating the lower water use efficiency. The
third year of experimentation, significant differences best fitted curve gave the maximum fodder yield of
in the pod yields were found among the treatments 4667 kg/ha having the optimal water inputs of 1071
having IW/CPE of 0.6, 0.7 and 0.8. However, the mm.
made in growing the summer groundnut using border
Economics was found to be 1.63 at IW/CPE of 1.0 (e) The border
The Table-2 showed that the total cost of cultivation irrigation is not economically viable for deficit and
of summer groundnut crop under border irrigation high water application.
increases with increase in water application levels.
The gross income increased from 20711 Rs./ha in REFERENCES
IW/CPE of 0.6 to 34381 Rs/ha in IW/CPE of 1.0 and Anoynmous (2003). Final Report of ICAR Adhoc
it decreased up to 33378 Rs./ha in IW/CPE of 1.2. Project on Studies on Techno-Economic
The highest benefits cost ratio (gross income to total Feasibilities of Drip Irrigation for Summer
cost) was found to be 1.63 at IW/PE of 1.0. It Groundnut Crop and its Comparison with Micro
increased from 1.13 in IW/PE of 0.6 to 1.63 in Sprinkler and Border Irrigation,Junagadh.
IW/CPE of 1.2. After that, it decreased for the English, M.J. et al. (1990). ASAE Monograph, St.
IW/CPE higher that 1.0. The highest return of 13309 Joseph, MIchigan pp 631-666.
Rs./has was found at IW/CPE of 1.0. The net return Febeiro, C. et al. (2003). Agric. Water Mgmt. 59 :
from water saving over base IW/CPE (1.0) were 155-167.
accounted as Rs. 2463, 5200, 8160 and 11822 Hewitt, T.D. et. al. (1980). Proc. Soil and Crop Sc.
respectively in the IW/CPE of 0.6, 0.7, 0.8 and 0.9 Soc., Fla, 39 : 135 - 140.
respectively. Hexem, R.W. and Heady, E.O. (1978). Water
Production Functions for Irrigated Agriculture,
CONCLUSIONS IOWA State University Press, Ames, IOWA,
(a) The optimal irrigation scheduling for the summer USA.
groundnut under border should be based on IW/CPE Kirda, C.P. et. al. (1999) Crop Yield Response to
of 0.9. (b) The highest possible water use efficiency Deficit Irrigation, Kluwer Academic Publishers.
under border irrigation for summer groundnut crop London.
can be 2.52 kg/ha.mm in this zone. (c) The yield Rank, H.D. et. al. (2003). Proc. Int. Sem for
response to irrigation water for summer groundnut downsizing tech. for rural devpt. RRL
crop can be described well by quadratic from (d) The Bhubneshwar 1: 254-257.
highest internal rate of return (IRR) for investments

Stegman, E.C. et. al. (1980). Design and Operation of Wanjura, F.D., et al. (2002). Agic Water Manage,
Farm Irrig. Systems, ASAE, Washigton. 55:217:237
USDA (1959). SCS National Engg. Handbook,
Washigton, D.C. pp. 8-1 to 8.15.
.
Table 1: Total water applied (ha.mm/ha), pod yield (kg/ha) and fodder yield (kg/ha) obtained under border
irrigation during different years
IW/CPE First year Second year Third year Pooled

Total Pod Fodder Total Pod Fodder Total Pod Fodder Total Pod Fodder
water Yield Yield water Yield Yield water Yield Yield water Yield Yield
Applied (Kg/ha) (kg/ha) Applied (Kg/ha) (kg/ha) Applied (Kg/ha) (kg/ha) Applied (Kg/ha) (kg/ha)
(mm) (mm) (mm) (mm)
0.6 545 977 2333 523 1621 3406 525 1237 2455 531 1278 2731
0.7 625 1153 2929 631 1817 3848 619 1479 3129 629 1483 3302
0.8 713 1407 3611 729 2042 4218 705 1658 3698 716 1702 3841
0.9 784 1837 3961 786 2212 4533 788 1916 4035 786 1988 4177
1.0 884 1982 4554 893 2332 4809 876 2061 4150 884 2125 4504
1.2 1055 1952 4841 1073 2156 4952 1085 1981 4210 1071 2029 4667
SEM - 79.4 110.5 - 57.34 90.8 - 53.01 88.41 - 63.25 96.57
CD - 226.4 315.0 - 163.4 258.9 - 151.1 251.99 - 180.3 275.2
Cv - 14.75 8.83 - 8.19 6.18 - 8.39 7.14 - 10.44 7.38
Table : 2 Cost Benefits (Per Season) Of The Cultivation Of Summer Groundnut Crop Under Border Irrigation
Additional B/C
Total Gross Net Addi.Benefits of
Treatment Total Net Return (With Cost of ratio of
Cost Income Return System over
(IW/CPE) Return of Water Saving) System System
(Rs./ha) (Rs./ha) (Rs./ha) Better treatment
Over Control Adoption
0.6 18248 20711 2463 4113 1650 353 1.13
0.7 19032 24232 5200 7332 2132 255 1.27
0.8 19728 27888 8160 10118 1958 168 1.41
0.9 20288 32110 11822 13369 1547 98 1.58
1.0 21072 34381 13309 13309 As base Nil 1.63
1.2 22568 33378 10810 10810 N.A. N.A. 1.47

ISBN 978-81-921121-0-7
PETROGRAPHIC EXAMINATION OF CORROSION INHIBITORS

ADMIXED CONCRETE
B.N. Singh
Department of Civil Engineering, Institute of Technology, Banaras Hindu University, Varanasi, Email: bnsgranite@yahoo.co.in
S.K. Singh
Ex.M.Tech Student, Department of Civil Engineering, Institute of Technology, Banaras Hindu University, Varanasi
V. Kumar
Department of Civil Engineering, Institute of Technology, Banaras Hindu University, Varanasi
M.A. Quraishi
Department of Applied Chemistry, Institute of Technology, Banaras Hindu University, Varanasi
ABSTRACT: The present paper addresses the petrographic examination of concrete with and without addition of
various dosages of calcium palmitate and calcium nitrite generally used as corrosion inhibitors. Mineralogically and
texturally, these concretes are quite distinct. The concrete samples without corrosion inhibitors are characterized by
the occurrences of microcracks, large-to medium-sized voids and carbonations generally responsible for lowering
the strength and durability of concrete. However, these features are less abundant in the concrete with calcium
palmitate and its mixture with calcium nitrite. The microcracks and voids are infilled with cementitious materials
and C-S-H gel when these corrosion inhibitors are admixed with cement. Further, the formation of thick and thin
rims of hydration and small-to medium-sized grains of calcium hydroxide were observed only in two concrete
samples (PN2 & PN6).
INTRODUCTION with and without corrosion inhibitors (i.e. calcium

Petrography is an important aspect of geology and palmitate and its mixture with calcium nitrite). These
generally dealt with the identification of mineral corrosion inhibitors are generally used to lower the
constituents and textures of the rocks under petrological corrosion rate of steel reinforcement in chloride
microscope. However, the application of petrological environment (Singh, 2009). The available published
microscope in the study of concrete and related works reveal that no detailed petrographic study has
materials has a much shorter history. Petrographic been carried out on the concrete with admixture of
examination of hardened concrete is a quick and well corrosion inhibitors used in the present investigation.
suited method of diagnosing reason for lack of concrete However, Singh et al. (2009) were carried out detailed
durability (Jepsen and Christensen, 1989; John et al. petrographic study on the concrete admixed with
1998). It is well known that stone pieces (i.e. coarse calcium stearate corrosion inhibitor.
aggregate) and sand (i.e. fine aggregate) are the most
important component of concrete. Petrographic MATERIALS USED
examination plays an important role in identification of Coarse and Fine Aggregates: Well graded aggregates
mineral constituents of coarse and fine aggregates as of 10mm & 20mm sized crushed Dalastone and natural
well as the nature of the binding between cement paste sand of Sone River from Chopan, District Sonebhadra
and aggregates of the hardened concrete. Many workers were used. The fineness modulus of the sand was 2.64,
were studied hardened concrete and related building and 6.15 and 7.30 were for 10 mm & 20 mm coarse
materials under petrological microscope to observe the aggregates respectively. The physical properties of the
microstructures (e.g. microcracks, voids etc.), formation aggregates were obtained as per Indian Standards IS:
of gel and extent of carbonation (John et al. 1998; Jana, 383 (1970).
2005; Singh et al. 2007, 2009; G.D. Ransinchung et al. Cement: It was ordinary Portland cement of 43 grade
2008; Lukschova et al. 2008 etc.). Petrography also conforming to IS: 456 (2000).
plays an important role in the identification of alkali Calcium Palmitate and Calcium Nitrate Corrosion
reactive and non-reactive aggregates, formation of Inhibitors:
reaction rims in the expanded/ distressed and Calcium palmitate is a compound of calcium with
deteriorated cores (Sims and Nixon, 2003; Bhatt and palmitic acid. Palmitic acid is a common 16- carbon
Kumar, 2005). saturated fat that represents 10-20% of the normal
The present paper deals with the microstructural dietary fat intakes. Calcium palmitate is one of the
behaviours (e.g. presence or absence of microcracks major components of the gallstones. It is supplied as
and voids), formation of gels and rims and the extent of crystalline solids. It is soluble in the organic solvent
carbonation in the ordinary portland cement concrete chloroform to a concentration of approximately 0.3

mg/ml. Calcium palmitate was prepared in the
laboratory of the Applied Chemistry, I.T., B.H.U.
However, calcium nitrite is commercially available.
Water: Potable tap water was used for preparing
mortar.
METHODOLOGY
The concrete cubes (150 x 150 x 150 mm size) were
prepared in the Structural Engineering Laboratory,
Dept. of Civil Engg., I.T., B.H.U. by using above
mentioned materials as per IS: 456 (2000) mix design
for studying the petrographic features and engineering
Fig.1 : Photomicrograph showing microcracks are
properties. The properties of coarse and fine aggregates
running along the periphery of coarse aggregates and
were determined as per IS: 383 (1970).
mineral aggregates in the blank sample (cement;90
Ten types of concrete cubes [i.e. one Blank sample ‘B’;
days). O.L. × 2.5 X.
3 samples (i.e. CP2, CP3 & CP4 with 2%, 3% and 4%
calcium palmitate by weight of cement); and 6 samples
(i.e. PN1, PN2, PN3, PN4, PN5 & PN6) with various
dosages of calcium palmitate and calcium nitrite
mixtures (1% calcium palmitate + 1% calcium nitrite;
2% calcium palmitate + 1% calcium nitrite; 1% calcium
palmitate + 2% calcium nitrite; 1.5% calcium palmitate
+ 1.5% calcium nitrite; 3% calcium palmitate + 1.5%
calcium nitrite ; and 1.5% calcium palmitate + 3%
calcium nitrite by weight of cement) ]were prepared as
per IS:456 (2000). These concrete cubes were cured for
60 and 90 days. After that concrete cubes were dried
properly for thin section preparation. Thin sections
were prepared from broken chips / pieces of these Fig. 2 : Photomicrograph showing large sized voids
concrete cubes for petrographic examinations. present within the groundmass of the blank sample.
Petrographic studies were carried out under the LEICA Carbonation is also noticed. O.L. × 5 X.
DM LP optical microscope available in the Applied
Geology Lab of the Civil Engg. department, I.T., BHU. Further, it is observed that the microcracks, voids and
carbonation are more prominent in the blank samples B
PETROGRAPHY (Figs.1 & 2) in comparison to concrete samples with
Petrography generally deals with the megascopic and corrosion inhibitors. The blank sample comprises large-
microscopic examinations of rocks and concrete to medium-sized voids (Fig. 2). However, the sizes of
samples. The concrete cubes were broken for studying voids are considerably smaller and infilled (partially
its megascopic characters. It is observed that coarse and fully) with cementitious materials in the samples of
aggregates are angular to sub-angular, triangular and concrete with corrosion inhibitors (Fig. 3).
sub-rounded in shape and enclosed tightly within the
dark-grey coloured groundmass of cementing materials.
Microscopically, the concrete samples with and without
corrosion inhibitors show that coarse aggregates (i.e.
Dalastone), and medium-to coarse-grained minerals
(mostly quartz and feldspars) of fine aggregate are
enclosed within the dark coloured groundmass of
cementing materials. The coarse aggregate constitute
small sized dolomite mineral that interlocked with each
other in comparison to other minerals (e.g. quartz and
feldspar) present in the cementatious groundmass.
Quartz is the most dominating mineral of the fine Fig. 3 : Photomicrograph of sample no. PN1 (cement +
aggregate present in the groundmass. Quartz grains are 1% calcium palmitate + 1% calcium nitride; 90days)
generally medium- to coarse - grained, anhedral to showing small sized voids, partially infilled, voids and
subhedral and rounded to sub-rounded in shape. infilled microcracks. O.L. × 5 X.

On the other hand, the microcracks in the blank samples
are developed at the periphery of the coarse aggregates
and running across the groundmass (Fig. 1). While
microcracks are generally infilled partially and fully
with cementitious materials in the concrete samples
with corrosion inhibitors (Figs. 4 & 5).
Fig. 6 : Photomicrograph of sample no. PN2 (cement +

2% calcium palmitate + 1% calcium nitrite; 60 days)
showing rims around the quartz grains. Large to
medium size grey colored grains are also notice.
Infilled and partially infilled voids are also present.
O.L. × 2.5 X.
Fig. 4 : Photomicrograph of sample no. PN2 (cement +

2% calcium palmitate + 1% calcium nitrite; 60
days)showing thick and thin rims around the coarse
aggregate and mineral grains. Picture also showing
partially infilled microcracks running within the
groundmass.
O.L. × 5 X.
Fig. 7: Photomicrograph of sample no. PN6 (cement +

1.5% calcium palmitate + 3% calcium nitrite; 60 days)
showing partially infilled voids and formation of C-S-H
gel. Picture also show thick and thin rims around the
mineral grains and the periphery of coarse aggregate.
O.L. × 2.5 X.
It is important to mention that rims were formed around

the course aggregate and minerals present only in the
Fig. 5 : Photomicrograph of sample no. PN3 (cement + groundmass of PN2 and PN6 concrete samples (Figs. 4,
1% calcium palmitate + 2% calcium nitite; 90 days) 6 & 7).
showing microcracks infilled with C-S-H gel on the Further, the grey-colored, rounded to sub-rounded,
periphery of the coarse aggregate. Small size voids are angular to rectangular, small- to large-sized grains are
also present. Aggregate and most of the quartz grains frequently observed only within the groundmass of the
showing sharp contact with cementitious materials. PN2 and PN6 concrete samples (Figs. 4, 6 & 7). The
O.L. × 2.5 X. rims and grey colored grains present within the
groundmass of the sample Nos. PN2 and PN6 are very
much similar to the thick and thin rims of hydration and
small- to large-sized grains of calcium hydroxide

reported by Jana (2006; Fig. 4). However, these On the basis of present investigation, it is concluded
features were not observed in the blank sample and rest that these corrosion inhibitors may be used in the
of the concrete samples with corrosion inhibitors. concrete due to better microstructures, formation of
additional C-S-H gel and infilling of pores and
microcracks.
DISCUSSION AND CONCLUSIONS
Petrographic study deals with the mineral constituents
and microstructure / texture of the rocks and concrete REFERENCES
specimens. In the present investigation, concrete with Bhatt, S.C. and Kumar, R. (2005) Relation between
and without corrosion inhibitors (i.e. calcium palmitate mineralogical and physical properties of coarse
and its mixture with calcium nitrite) have been studied and fine aggregates and influence of textural,
under petrological microscope to observe the chemical and dimensional differences on strength
microcracks, voids, formation of gels, and carbonation. of mortar and concrete – Manali (H.P). In : P.S.
The mineral constituents and texture (i.e. size, shape Saklani (ed.), Himalaya (Geological Aspects),
and arrangement of mineral constitutents within the Satish Serial Publishing House, New Delhi, V.3,
groundmass) for both types of concrete samples (i.e. pp. 101-134.
concrete with and without corrosion inhibitors) are G.D. Ransinchung, R.N., Singh, B.N., Kumar, B. and
largely similar. However, microstructurally they are Kumar, V. (2008) Microstructural behaviour of
quite distinct. wollastonite and microsilica admixed concrete for
Microscopic observations reveal that microcracks and rigid pavements. Jour. Highway Research
large to small-sized voids are frequently observed Bulletin, V.1, pp. 13-20.
within the blank sample (Figs 1 & 2). The presence of IS: 383(1970) Specifications for coarse and fine
more microcracks and voids reduce the strength and aggregates from natural resources for concrete.
durability of concretes (John et al. 1998; Jana, 2005). Indian Standard Institution, New Delhi.
On the other hand, the sizes of voids are considerably IS: 456 (2000) Plain and reinforced concrete – code of
smaller and infilled partially and fully with practice. Indian Standard Institution, New Delhi.
cementitious materials in the concrete samples with Jana, D. (2005) Concrete petrography – past, present
corrosion inhibitors (Fig. 3). Microcracks, if present, in and future. In: J.J. Hughes, A.B. Leslie, J.A.
the concrete samples with corrosion inhibitors are Walsh (eds.), Proceeding of the 10th Euroseminar
infilled partially and fully with cementitious materials on Microscopy Applied to Building Materials-
and calcium silicate hydrates (C-H-S) gel (Figs. 4 & 5) Extended Abstract and CD-ROM, Paisley, U.K.,
were formed most probably due to the addition of University of Paisley, on CD-ROM, 22p.
corrosion inhibitors in the ordinary Portland cement Jana, D.(2006) Delayed setting of concrete – A
concrete. petrographic and chemical investigation. In:
Further, it is also observed that intensive carbonation Proceedings of the Twenty-Eighth Conference on
has taken place more frequently in the blank sample Cement Microscopy, Denver, Colorado, U.S.A.,
than the concrete samples containing various dosages of April 30-May 4, 2006, pp. 141-149.
corrosion inhibitors. Carbonation has been noticed Jepsen, B.B. and Christensen, P. (1989) Petrographic
around the voids of blank sample (Fig. 2). It reveals that examination of hardened concrete. Bulletin of the
blank sample would be more porous in comparison to International Association of Engineering Geology,
concrete with corrosion inhibitors. Porosity generally N039 Paris, pp. 99-103.
affect the durability properties of concrete. Carbonation John St., D.A. Polle, A.B. and Sims, I. (1998) Concrete
most commonly takes place due to the reaction of both Petrography, A hand book of investigative
calcium hydroxide and cement hydrates with techniques, published by John Wiley and Sons
atmospheric carbon dioxide (John et al. 1998). Inc. New York, Toranto, 474p.
Further, the formation of rims of hydration and grains Lukschova, S., Prikryl, R. and Pertold, Z. (2008)
of calcium hydroxide observed only in the sample Nos. Petrographic indentification of alkali- silica
PN2 and PN6 (Figs 4, 6, & 7) may be due to addition of reactive aggregates in concrete from 20th Century
slightly higher dosages of calcium palmitate. Otherwise Bridges. Jour. Construction and Building
these rims and grains must also be present in all Materials, V. 23, pp. 734-741.
concrete samples with corrosion inhibitors (i.e. mixture Sims, I. and Nixon, P. (2003) RILEM Recommended
of calcium palmitate and calcium nitrite). These Test Method AAR -1 : Detection of potential
features of PN2 and PN6 are very much similar to the alkali-reactivity of aggregates- Petrography
rims of hydration and grains of calcium hydroxide Method. Jour. Materials and Structures, V. 36, pp.
discussed by Jana (2006). 480-496.

Singh, B.N., Kumari Asha and Kumar, V. (2007) Singh, S.K. (2009) Experimental studies on effect of
Petrographic study of concrete with and without calcium palmitate, mixture of calcium palmitate
microsilica. The Researcher, V.1 (No. 2), pp. 55- and calcium nitrite, and microsilica as corrosion
57. inhibitor in concrete. Unpubl. M.Tech.
Singh B.N., Abhilash, P.P., Kumar, V. and Quraishi, Dissertation, Dept. of Civil Engineering, Institute
M.A. (2009) Microscopic Examination of of Technology, Banaras Hindu University, 83 p.
concrete with and without corrosion inhibitor. e-
Journal Earth Science India, V.2 (II), pp. 94-100.

ISBN 978-81-921121-0-7
DETECTION OF ACTIVE LANDSLIDE AREAS IN HIMALAYAS

USING SMALL BASELINE SUBSET INTERFEROMETRY
A. Bhattacharya
Research Scholar, Earthquake Engineering /IIT Roorkee, Roorkee–247667, Email: atanudeq@gmail.com
Manoj. K. Arora
Professor, Civil Engineering, IIT Roorkee, Roorkee -247667, Email: manoj.arora@gmail.com
Mukat. L. Sharma
Professor, Earthquake Engineering, IIT Roorkee, Roorkee -247667, Email: sharmamukat@gmail.com
ABSTRACT: Himalayas, one of the youngest mountain ranges in the world are tectonically and seismically
very active. The crustal movements along the longitudinal thrusts and transverse faults give rise to earthquakes
and in turn the landslides. The study of landslides has drawn global attention mainly due to increasing
awareness of its socio-economic impact as well as increasing pressure of urbanization on mountain regions.
Therefore, identification and demarcation of landslide active areas is necessary for prevention and proper
planning to mitigate the future damage. However, traditional monitoring techniques are expensive and time
consuming. Differential synthetic aperture radar interferometric (DInSAR) is a newly developed remote sensing
technique, which is useful to measure the displacements at high accuracy due to natural calamities. Improved
access to time series of satellite data has led to the development of several innovative multi-temporal
interferometric algorithms. Among them Small baseline subset (SBAS) algorithm is gaining importance to
measure land displacement with a high degree of accuracy. In this paper, preliminary results of SBAS to identify
potential landslide area in Garhwal and Kumaon Himalayan regions of India have been presented. The field
verifications showed significant landslide activity at position in agreement with SBAS interferometric results.
INTRODUCTION phenomena producing ground displacements. Several

Landslides are a large threat to the population of the examples prove the effectiveness of differential SAR
India Himalayas. Many villages and vital transport interferometry from satellite platforms for measuring
routes are endangered by landslides and losses of lives ground displacements over wide areas due to different
and important infrastructure are not uncommon. The causes. Satellite-borne Earth observation is a useful
landslide incidences in a region have been of serious tool for assessing the stability of mountain slopes and
concern to the society due to loss of life, natural hazard zone mapping. Optical satellite imagery
resources, infrastructural facilities, etc. Therefore a integrated with synthetic aperture radar (SAR) data
detailed mapping of potential landslides is an (Singhroy et al. 1998) has been used to map
important task. However, due to the enormous size of geomorphologic and topographic characteristics that
landslide prone areas, this task cannot be carried out are of relevance for landslide inventory and risk
by ground based methods. Most ground based surface assessment.
displacement measurement techniques provide First InSAR applications to measure slope motion
displacement information on point basis only. As a were reported by Fruneau et al. (1996) for a landslide
result, spatial distribution of displacements in a region (Saint-Etienne-de-Tinee) in southern France with
is not possible. Further, the field based techniques are displacement rates of 1–2 cm per day.
time consuming, cumbersome and costly. This Similarly, Rott et al. (1999) also reported
hampers the planning and monitoring processes. displacement of a landslide above a reservoir in the
In recent years, availability of interferometric Austrian Alps with motion of several centimeters per
Synthetic Aperture Radar (InSAR) data from a variety year. Rizo and Tesauro (2000) demonstrated the
of satellite and airborne remote sensing sensors has potential of DInSAR technology to measure the
gained momentum, as an alternative technique to displacement due to landslide of Randazzo, from the
yield accurate height and surface displacement InSAR data in the month of March and April, 1996. A
information at very high accuracies. InSAR derived displacement rate of 1 cm per day was reported.
displacement measurements are not only feasible but Nagler et al. (2002) used DInSAR for detecting and
also moderate in their costs compared to many monitoring slow movements of mountain slopes on
ground-based methods. With an increasing number of the order of centimeters per year were investigated in
SAR satellites the coverage of the earth's land masses the Austrian and Swiss Alps, using SAR images from
is also improving, especially with ESA's Sentinel the European ERS -1 and ERS -2 satellites. Squarzoni
satellites which will be launched in about a year's et al. (2003) applied DInSAR to detect ground surface
time. One of the main applications of synthetic displacement caused by La Valette landslide located
aperture radar (SAR) interferometry (InSAR) is the in the Ubaye valley, southern French Alps. Fifteen
monitoring of natural hazards, in particular of those differential interferograms realized from ERS-1/2
satellite radar images acquired between 1991 and

1999 were used. Displacement values of the landslide combine interferograms with a short spatial baseline,
have been deduced from the SAR interferometric in order to minimize spatial decorelation.
products interpretation and compared with ground The main aim of this research work was to investigate
laser measurements. The average daily velocity of the the potential of SBAS in estimating the temporal
landslide between 1991 and 1999 derived by behavior of land deformation in the Garhwal and
interferometric analysis was decreased from 1 to 0.4 Kumaon Himalayan regions due to earthquake and
cm per day. A peak velocity of around 2 cm per day landslides. Moreover, future landslide prone areas
was observed in 1996. This site was already were identified with SBAS interferometric technique
investigated with the same technique by Vietmeier et by exploiting the previous SAR images.
al. (1999). The data set processed by Vietmeier et al.
(1999) consisted in three tandem pairs acquired from STUDY AREA
August 1995 to March 1996 using the technique of Himalayan region is geo-dynamically very active.
three-pass interferometry. An average daily velocity Being the youngest mountain system of the world, it
of 1.2 cm was reported. constitutes fragile ecosystem of unstable slopes giving
These studies sufficiently demonstrate the potential of rise to frequent geo-hazards including landslides. The
DInSAR for the measurement of land surface problem is acute in certain regions of the Garhwal
displacement due to landslide at spatial levels. Himalayas. The present research focuses towards an
Though it looks promising as a technique to monitor active landslide area, Mansadevi, Uttarakhand, in
slope motion, several researchers (Klees and Garhwal and Kumaon Himalayan regions of India.
Massonnet 1999; McKean and Roering 2004) The Mansa Devi landslide is located above the city of
highlight some constraints, such as temporal Haridwar. Mansa Devi hill is spread out the area of
decorelation, geometrical decorelation and approximately 2 km. from Mansa Devi main gate to at
atmospheric inhomogeneities, present in conventional the end of Bhemgoda road. The Mansa Devi Hill is
SAR technique. one of the landslide prone areas in Uttarakhand.
These decorelation effects can be reduced by However, it became active only a decade ago and
combining multiple SAR observations using threatens the larger part of the city of Haridwar and a
multitemporal InSAR techniques. Using more than stretch of about 300m of the Haridwar bypass road. It
two SAR scenes leads to redundant measurements, has been found that there are many, small as well as
which can be utilized for more advanced time series large, landslides along the slopes of the Mansa hill
methods (Lauknes et al. 2010). Time series InSAR and ominous for highly built-up areas and populations
algorithm can be briefly classified into Persistent that has been developed along the foot hills and
Scatterer (PS) and Small Baseline Subset (SBAS). adjoining areas near Mansa hill specifically along the
PS algorithm identify stable image pixels, which are Bhemgoda road, Haridwar. The built-up areas are in
coherent over long time intervals, in a stack of danger zone due to landslide in Mansa hill towards
interferograms generated with the same master image the loss of human properties and human being
(Ferretti et al. 2000; Hooper et al. 2004; Kampes himself. The Figure 1 shows the Mansadevi study
2006). On the contrary, SBAS algorithm use SAR area.
image combinations with a short spatial and temporal
baseline to reduce the effects of spatial and temporal DATASET
decorelation (Berardino et al. 2002; Hooper 2008; A total of 21 ERS SAR raw data sets from descending
Lanari et al. 2007; Schmidt & Bürgmann 2003). Each satellite track 291 and frame 2997 were received from
of these algorithms has its advantages and limitations, ESA, covering the Mansadevi region. The SAR
and they have both proved to be effective in images covered the time period between May 1992
successfully estimating deformation time series in and June 1999. Among the images, 15 were from the
various regions (e.g., Hilley et al. 2004; Hooper et al. ERS-1 sensor and 6 from ERS-2. Among the 21
2007; Lauknes et al. 2010). images, 19 images were effectively used in the SBAS
SBAS method, which is used in this study, is processing. Two images, acquired on May 1999 and
optimized for the distributed scattering mechanism, June 1999, were excluded from the dataset because of
which is very common in natural terrain, where very low coherence. Apart from SAR images, an a
several adjacent resolution cells represent the same priori available SRTM DEM’s version3 in geographic
physical deformation phenomenon. Given a data stack latitude and longitude coordinate were used to remove
of multiple SAR images, the distributed scattering the topographic phase. These DEM’s had a grid size
method will give best results when exploiting of 90×90 m and a height standard deviation of ±16
interferometric pairs with as small spatial baseline as meters in plane terrain.
possible. The idea behind the SBAS method is to

decomposition (SVD) method. This allows one to
evaluate the pseudo-inverse of the matrix A that
gives the minimum norm least square solution of
the Equation (3).
DEM Error and Low Pass Displacement Signal

Estimation
Next step of SBAS algorithm is to find out the
error of topographic phase due to the a priori
DEM used in the processing. In order to separate
the DEM error term from Equation (2) one can
apply a displacement model to estimate the DEM
errors and low pass deformation signal. A cubic
Fig. 1: Mansadevi Study Area deformation model given by Berardino et al. (2002),
as given in Equation (4), can be used
SBAS ALGORITHM 1 1
 t i   v .t i  t 0   a .t i  t 0    a .t i  t 0 
2 3
The mathematical concept of SBAS algorithm
assumed N+1 coregistered single look complex (SLC) 2 6
SAR images relative to the same area or same satellite (4)
track and acquired in ordered time (t0…….tN ). One where, v , a and a are the mean velocity, mean
can also assumed that each SAR acquisition should acceleration and mean acceleration variation
interfered at least another image so that each small respectively. The unknown parameter vector x̂ can be
baseline (SB) subset is composed of at least two SAR written as follows
acquisitions.
The differential phase ( j ) for a generic coherent xˆ T  v , a , a  …………………………………(5)
Now one can express the Equation (3) as a linear
pixel of range and azimuth coordinate (r, a), in system with M equations, corresponding to the
interferogram j, generated by combining SAR
acquisitions at times tB and tA is given in Equation (1) unwrapped phase value of  , and three unknowns as
(Berardino et al. 2002).  AM xˆ   j ……………………………………(6)
 j r , a    t B , r , a    t A , r , a  …………(1) Berardino et al. (2002) jointly estimate the DEM error
4 B and low pass component of deformation signal via the
 d t B , r , a   d t A , r , a   4  perp , j  z least square solution. One can easily linearly combine
  r sin  the DEM error term in the Equation (6).
  atm ( t B , r , a )   atm ( t A , r , a )    n j  AM , C xˆ   j ………………………………(7)
(2) where, C is an M×1 vector proportional to the DEM
where λ is the transmitted radar wavelength, error and shown in Equation (8).
 ( t B , r , a ) and  ( t A , r , a ) are the phases
 4  B perp , 1 4  B perp , M 
corresponding to the times tB and tA, whereas CT   ,......... ..., 
d ( t B , r , a ) and d ( t A , r , a ) are the radar LOS   r1 sin  1  rM sin  M 
projection of the cumulative deformation referenced (8)
to the first scene (i.e. implying  ( t 0 ) ). Now the unknown parameter vector x̂ is a
Ignoring the different error terms, the above Equation combination of low pass deformation component of
(1) can be organized in a matrix form the unwrapped signal and DEM error.
A  j   j ………………………………………(3) xˆ T  v , a ,  a ,  z  ……………………………(9)
A is an M×N matrix where A ( j , IS j )   1 if
Cumulative Phase Estimation
IS j  0 and A ( j , IE j )   1 if IE j  0 and zero The estimated DEM error term can be subtracted from
otherwise. So, A can be considered as an incidence the unwrapped phase, forming the following linear
like matrix which depends on the set of interferogram system
generated from the available dataset. The inversion of A  j    j …………………………………… (10)
the A matrix is depend upon on the number of subset where,   j   j   z .C is the phase with DEM
available from the dataset. If all the dataset belong to
a single small baseline subset one can easily inverse A error contribution subtracted. In order to reach a
matrix and the solution of Equation (3) can be physically sound solution Berardino et al. (2002)
obtained in least square sense. However, the dataset manipulate the Equation (10) in such a way to replace
belong to a single subset is exceptional in most of the the unknown parameters with the mean phase velocity
practical situation. A simple solution for inverting the between time adjacent acquisitions. Now the
Equation (3) is provided by singular value unknown parameter vector can be written as

  t 1   t N    t N  1   to unwrapping errors. For each complex
v T  v1  ,.... v N   interferogram, a best fitting linear phase ramp due to
 t 1  t0 t N  t N 1  imprecise orbit knowledge was estimated and
(11) removed. Two complex interferograms having smaller
Incorporating the new unknown parameter vector, perpendicular baseline (9 m) and larger perpendicular
shown in Equation (11), in place of Equation (1) one baseline (-230 m) are shown in Figure 3a and Figure
can get the following expression 3b respectively. Further, to facilitate the phase
IE j
unwrapping pixel thresholding criteria was used.
 t k  t k  1 v k    j ………………………(12)
k  IS j  1
Interferometric correlation coefficient or coherence
The Equation (12) can be organized in a matrix form was used as a threshold to select the coherent pixels.
and finally leads to the following expression The pixels which exhibited an estimated coherence
value larger than 0.3 in at least 30% of the
Bv    j ………………………………………(13) interferograms were selected for further processing.
The matrix B is an M×N and (j, k) element of B Total 284763 pixels were identified through
matrix is B ( j , k )  t k  1  t k for thresholding technique. The mean coherence image of
IS j  1  k  IE j and B ( j , k )  0 elsewhere. Mansadevi study area is shown in Figure 4. In order
to link all the sparse coherent patches, a Delauny
However, now the singular value decomposition is triangulation and interpolation were performed for all
applied to the matrix B. A trivial integration step is images, based on the selected pixels. Pixels occurring
needed to achieve the final solution. in low coherence areas are just filled in via an
interpolation step (implemented via a Delauny
Removal of Atmospheric Effect triangulation) starting from data located in high
The undesired atmospheric signal can be estimated as coherent pixels. The phase unwrapping operation was
follows (Lauknes et al. 2010): carried out using SNAPHU software (Chen and
i) The low pass deformation signal from Equation Zebker 2001). The phase unwrapping algorithm was
(4) is removed from the estimated deformation only applied to the pixels that exhibit an estimated
time series. coherence value above the chosen coherence
ii) Following this, the residual phase signal is threshold. After the phase unwrapping, the phase
detected as the result of first a temporal high pass values of all pixels were calibrated with respect to the
filter with respect to the time variable then a phase value of a particular pixel in an assumed stable
spatial low pass filter on the residuals. The area. In order to select a pixel in stable area a priori
estimated atmospheric phase signal is then
subtracted from the deformation phase signal and
then multiplied a factor; the final deformation
signal can be achieved.
SBAS PROCESSING RESULTS

In this section the preliminary results of SBAS
processing are discussed. All the processing including
focusing, coregistration were performed in Norut
GSAR processor (Larsen et al. 2005) and SBAS
processing was performed on ENVI/IDL on Linux
platform. The overall processing was performed by
applying a complex multi-look operation taking two
looks in range and eight looks in azimuth, produced
pixels with ground range dimensions of about 40 m in
range and 30 m in azimuth direction. A total of 42
interferograms were constructed from 19 scenes using deformation knowledge is needed. However, in this
perpendicular baseline threshold of 250 m and 4 years study a priori information about deformation was not
temporal baseline threshold. The different available. Therefore, the stable pixel was taken in an
combinations of interferograms are shown in Figure 2. area having very high coherence. The phase value of
It can be observed from Figure 2b, due to the stable pixel was taken as the mean of the phase values
threshold method being used, two other images (May located within a 3×3 neighborhood in that high
22, 1992 and December 18, 1992) with extreme coherence area After phase unwrapping, DEM errors
baselines were excluded, reducing the number of and LP deformation component were estimated as
unique images being used in the study from 19 to 17. described above. Following this operation, the
Out of 42 interferograms 20 good quality estimated LP deformation component and DEM error
interferograms were effectively used in SBAS were subtracted from each input interferogram in
algorithm. Remaining interferograms, which were order to reduce the fringe rate density. A new phase
connected with long temporal baseline, were removed unwrapping algorithm was then applied to the residual
from the dataset because those were more susceptible wrapped phase patterns. Final refined unwrapped

DInSAR phase patterns were achieved by adding back Fig. 6: Mean Velocity map Supper Imposed on Mean
the subtracted LP phase component. Figure 5a and Intensity Map of the Study Area
Figure 5b shows two unwrapped interferograms of the DISCUSSIONS
The deformation time series and mean velocity maps can be
study area corresponding to the same complex
used to highlight various types of phenomena occurring on
interferogram shown in Figure 3 (a-b). Relatively the terrain. The perfect linear trend of the deformation time
smooth unwrapped image was obtained in series is not observed for both the study areas. These can
interferogram having lower perpendicular baseline probably be attributed to the contribution of number of
value whereas, sudden (more than 2π) phase jump interferogram combinations of each SAR scene. It can be
was observed in interferogram having higher expected that if few interferograms are computed with
perpendicular baseline value. Based on good quality respect to a particular SAR scene, the SVD inversion will be
unwrapped interferograms, SBAS algorithm, less robust for this point, and the noise level will be higher.
described in section 4, was used. Final mean velocity Another reason may be due to the irregular sampling of the
was estimated by inverting the Equation (13) using data set perfect linear trend cannot be achieved.
Three points were selected from the assumed vulnerable
L2-norm based singular value decomposition method landslide prone area. In mean velocity image a strong
(Berardino et al. 2002). Twenty unwrapped topographic effect was observed may be attributed to error
interferograms were used in SBAS inversion associated with the external DEM. The deformation time
algorithm for this area. Figures 6 shows the estimated series of three selected points are shown in Figure 7 (a-c). A
mean LOS velocity (mm/year) of Mansadevi study clearly different mean velocity signal was observed between
area. The mean velocity maps are superimposed over these three points. A strong uplift signal was observed on
the corresponding grayscale SAR intensity images point 3. The uplifting throughout the Himalayan range can
and for better understanding. In the color scale of the be expected due to the collision of Indian subcontinent with
velocity image zero valued pixels are not shown Eurasia. The amount of convergence rate between India and
Asia was estimated by several researchers (Molnar 1990;
because of better visibility of the actual area of Jackson and Bilham 1994) using different techniques.
interest. Based upon velocity, displacement maps and However, convergence rate is not uniform and is absorbed
landslide point of view, areas of interest were over a large region of central Asia. Due to this horizontal
identified. These areas were further investigated by convergence the vertical uplift of Himalayan range also can
generating displacement time series of some selected be expected. Molnar (1990) suggested that the vertical
points inside these areas. velocity of rocks in the Himalaya varies both across the
range and along the arc. Therefore, it was more reasonable
to study relative LOS deformations of selected points. The
relative LOS deformation of points 1 and 2 with respect to
point 3 are shown in Figure 8 (a-b). A small amount of
negative deformation in LOS direction can be observed
from the Figure 8. Unfortunately, validation of this
deformation was not possible due to the lack of available
Fig. 2: Interferograms of small baseline in Mansadevi ground based deformation measurement data. However, the
Area field photograph (Figure 9) taken on early 2011 clearly
indicate massive landslides at position in agreement with
SBAS results. Slow surface movements relative to
neighboring area can be noticed during 1992 to 1998 and it
may be assumed that today’s landslide is a cumulative effect
of those slow surface movements.
A B
Fig. 3: Measured interferometric phase: (a) measured

phase 930226-930402, BPerp = 9 m, (b) measured
phase 970224-980209, Bperp = 230 m
0.00 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9
C
0.25
Mansadevi 0.50
0.75
1.0
Fig. 4: Mean Coherence Image of Mansadevi Area 9 9 9 9 9 9 9 9

Fig. 5: Unwrapped phase: (a) measured phase
Fig.7: Displacement time Series (A) Point 1 (B) Point 2 and
930226-930402, BPerp = 9 m, (b) measured phase (C) point 3 of Mansadevi area
970224-980209, Bperp = 230 m

will prove to be a critical tool in future landslide mitigation.
Although InSAR cannot be considered a replacement for
ground based monitoring of active landslides but it can
provide an important complementary dataset.
REFERENCES
Berardino, P., Fornaro, G., Lanari, R. and Sansosti, E.
9 9 9 9 9 97 98 9 9 9 9 9 9 98 (2002). A New Algorithm for Surface Deformation
2 Monitoring Based on Small Baseline Differential SAR
Interferograms. IEEE Transactions on Geoscience and
Fig. 8: Displacement time Series with respect to Point 3 (A) Remote Sensing, 40(11), 2375−2383.
Point 1 (B) Point 2 of Mansadevi area Casu, F., Manzo, M. and Lanari, R. (2006). A Quantitative
Assessment of the SBAS Algorithm Performance for
Fig. Surface Deformation Retrieval from DInSAR Data.
9: Remote Sensing of Environment, 102(3–4), 195−210.
Chen, C. W. and Zebker, H. A. (2001). Two-dimensional
Phase Unwrapping with use of Statistical Models for
Cost Functions in Nonlinear Optimization. Journal of
the Optical Society of America A, 18(2), 338−351.
Costantini, M. and Rosen, P. A. (1999). A Generalized
Phase Unwrapping Approach for Sparse data.
Proceedings of International Geoscience and Remote
Sensing Symposium 1999, Hamburg, Germany, 1,
267−269.
Ferretti, A., Prati, C. and Rocca, F. (2000). Nonlinear
Subsidence Rate Estimation using Permanent Scatterers
in Differential SAR Interferometry. IEEE Transactions
on Geoscience and Remote Sensing, 38 (5), 2202−2212.
Fruneau, B., Achache, J. and delacourt, C. (1996).
Observation and Modelling of the Saint-Etienne-de-
Tinee Landslide using SAR Interferometry.
Tectonophysics, 265, 181-190.
Ghiglia, D.C. and Pritt, M. D. (1998). Two-dimensional
Field picture of the selected points (taken February-2011) of Phase Unwrapping: Theory, Algorithms, and Software.
the study area New York: Wiley.
Goldstein, R.M., Zebker, H.A. and Werner, C.L. (1988).
CONCLUSIONS Satellite Radar Interferometry: Two-dimensional Phase
This study demonstrated the capability of InSAR technique Unwrapping. Radio Science, 23(4), 713−720.
to identify the relative movement of land surface in spatial Hanssen, R. (2001). Radar Interferometry: Data
level with high degree of accuracy. However, a clearly Interpretation and Error Analysis. Dordrecht,
irregular sampling of the dataset significantly affects the Netherlands: Kluwer Academic.
deformation estimation. The quality of the results can Hilley, G. E., Bürgmann, R., Ferretti, A., Novali, F. and
significantly improve using regular sampling dataset. Rocca, F. (2004). Dynamics of Slow-Moving
Further, the number of interferogram combinations also Landslides from Permanent Scatterer Analysis. Science,
plays an important role in SBAS technique. Increasing the 304 (5679), 1952−1955.
higher temporal and spatial baseline thresholds the number Hooper, A., Zebker, H., Segall, P. and Kampes, B. (2004).
of used images can be increased, which in turn decrease the A New Method for Measuring Deformation on
number of different subsets and increase the robustness of Volcanoes and other Natural Terrains using InSAR
SVD inversion technique. However, this condition also Persistent Scatterers. Geophysical Research Letters, 31,
increases the baseline decorrelation phenomenon. The L23611-L23615.
limiting factor is the ability to unwrap the interferograms, Hooper, A., Segall, P. and Zebker, H. (2007). Persistent
and still have a significant contribution from distributed Scatterer Interferometric Synthetic Aperture Radar for
targets. Further study would have to be done in order to test Crustal Deformation Analysis, with Application to
the use of larger baselines. Another important factor is the Volcán Alcedo, Galápagos. Journal of Geophysical
DEM accuracy. In this study SRTM DEM’s with spatial Research, 112, B07407-B07426.
resolution of 90 meters and height standard deviation of ± Hooper, A. (2008). A Multitemporal InSAR Method
16 meters in plane terrain were used. However, SBAS Incorporating both Persistent Scatterer and Small
method is less sensitive to the DEM accuracy, since this Baseline Approaches. Geophysical Research Letters,
method estimate the error in the topography contribution for 35, L16302-L16306.
each pixel, still 90 meters spatial resolution DEM can Jackson, M.E. and Bilham, R. (1994). 1991-1992 GPS
poorly estimate the DEM error term spatially strong Measurements Across the Nepal Himalaya, Geophysical
topography like Himalayas. Research Letters, 21(12), 1,169-1,172.
In spite of the several limitations SBAS technique provides Kampes, B.M. (2006). Radar Interferometry-Persistent
a relatively new and unprecedented technique to determine Scatterer Technique. Dordrecht, The Netherlands:
potential future landslide prone area. The direct linkage of Springer.
InSAR data to structural geology and sliding processes is a
relatively new and novel approach in landslide research and

Klees, R. and Massonnet, D. (1999). Deformation
Measurements using SAR Interferometry: Ptential and
Limitations. Geologie en Mjinbouw, 77, 161-176.
Lanari, R., Casu, F., Manzo, M., Zeni, G., Berardino, P.,
Manuta, M. and Pepe, A. (2007). An Overview of the
Small Baseline Subset Algorithm: A DInSAR
Technique for Surface Deformation Analysis. Pure and
Applied Geophysics, 164(4), 637−661.
Larsen, Y., Engen, G., Lauknes, T. R., Malnes, E. and
Høgda, K. A. (2005). A Generic Differential
Interferometric SAR Processing System, with
Applications to Land Subsidence and Snow-water
Equivalent Retrieval. Proceedings Fringe 2005
Workshop, ESA ESRIN, Frascati, Rome, (ESA SP-610).
Lauknes, T. R., Shankar, P. A., Dehls, J. F., Zebker, H. A.,
Henderson, I. H. C. and Larsen, Y. (2010). Detailed
Rockslide Mapping in Northern Norway with Small
Baseline and Persistent Scatterer Interferometric SAR
Time Series Methods. Remote Sensing of Environment,
114, 2097-2109.
McKean, J. and Roering, J. (2004). Objective Landslide
Detection and Surface Morphology Mapping using
High-resolution Airborne Laser Altimetry.
Geomorphology, 57, 331-351.
Molnar, P. (1990). A Review of the Seismicity and the
Rates of Active Underthrusting and the Deformation of
the Himalaya. Journal of Himalayan Geology, 1, 131-
154.
Nagler, T., Rott, H. and Kamelger, A. (2002). Analysis of
landslides in Alpine Areas by means of SAR
Interferometry. Proceedings of the IEEE International
Geoscience and Remote Sensing Symposium (IGARSS
2002), Toronto (Canada), 1, 198–200.
Rizzo, V. and Tesauro, M. (2000). SAR Interferometry and
Field data of Randazzo Landslide (Eastern Sicily, Italy).
Physics and Chemistry of Earth (B), 25 (9), 771-780.
Rott, H., Scheuchi, B., Siegel, A. and Grasemann, B.
(1999). Monitoring very Slow Slope Movements by
Means of SAR Interferometry: A case study from a
Mass waste above a Reservoir in the Otzal Alps,
Austria. Geophysical Research letters, 26, 1629-1632.
Schmidt, D.A. and Bürgmann, R. (20030. Time-dependent
Land Uplift and Subsidence in the Santa Clara valley,
California, from a Large InSAR Data set. Journal of
Geophysical Research, 108(B9), 2416-2428.
Singhroy, V., Mattar, K.E. and Gray, A.L. (1998). Landslide
Characterisation in Canada using Interferometric SAR
and Combined SAR and TM images. Advance Space
Research, 21, 465-476.
Squarzoni, C., Delacourt, C. and Allemand, P. (2003). Nine
years of spatial and temporal evolution of the La
Vallette landslide observed by SAR interferometry.
Engineering Geology, 68, 53-66.
Vietmeier, J., Wagner, W. and Dikau, R. (1999). Monitoring
Moderate Slope Movements (landslides) in the Southern
French Alps using Differential SAR Interferometry.
Proc. 2nd Int. Workshop on ERS SAR Interferometry
(FRINGE’99). Liege, Belgium.
Zebker, H.A., Rosen, P.A. and Hensley, S. (1997).
Atmospheric Effects in Interferometric Synthetic
Aperture Radar Surface Deformation and Topographic
Maps. Journal of Geophysical Research, 102 (B4),
7547−7563.

ISBN 978-81-921121-0-7
TIME PERIOD ANALYSIS OF LIQUID FILLED CYLINDRICAL

CONTAINER DURING SLOSHING
P. R. Maiti
Assistant Professor, Department of Civil Engineering, Institute of Technology, Banaras Hindu University, Varanasi-221005, Email:
prmaiti.civ@itbhu.ac.in
Abhinav Srivastava
M. Tech., Department of Civil Engineering, Institute of Technology, Banaras Hindu University, Varanasi-221005, Email:
abhinavsrivastava@gmail.com
ABSTRACT: Sloshing or the movement of free surface of liquid in cylindrical liquid-filled containers under external
excitation is a well studied phenomenon. Various design rules exist for liquid storage tanks to sustain the corresponding
liquid pressure due to external excitation and to take into account the necessary freeboard. A partially liquid filled tank
having a free surface wave produces a subsequent sloshing of liquid due to the rigid body acceleration of the tank. In
the present study an attempt has been made to determine the impulsive and convective time period of a cylindrical
liquid filled container under base excitation considering fluid structure interaction. For analysis of motion of liquid, the
liquid is assumed as inviscid, and motion of liquid is irrotational. Finite element analysis of sloshing in rigid container is
presented. Convective and impulsive time period for cylindrical tank is determined by using different codal provision.
Stress analysis of the tank wall is done by using ANSYS software and the results are presented for base excitation of a
cylindrical tank during sloshing.
Keywords: Sloshing, Convective time period, Impulsive time period, Rigid container.
INTRODUCTION hydrodynamic pressure on the pool and submerged

Sloshing means any motion of the free liquid surface components such as racks and fuel assemblies. The
inside its container. It is caused by any disturbance to sloshing motion can affect the stability of the free-
partially filled liquid containers. Depending on the type of standing spent fuels during earthquakes. The sloshing
disturbance and container shape, the free liquid surface characteristics in a storage pool may vary considerably
can experience different types of motion including simple depending on the size and location of the stored spent
planar, non-planar, rotational, irregular beating, fuel. There have been numerous studies done for dynamic
symmetric, asymmetric, quasi-periodic and chaotic. When behaviour of fluid containers; most of them are concerned
interacting with its elastic container, or its support with cylindrical tanks.
structure, the free liquid surface can exhibit fascinating The basic problem of liquid sloshing involves the
types of motion in the form of energy exchange between estimation of hydrodynamic pressure distribution, forces,
interacting modes. Modulated free surface occurs when moments and natural frequencies of the free-liquid
the free-liquid-surface motion interacts with the elastic surface. These parameters have a direct effect on the
support structural dynamics in the neighborhood of dynamic stability and performance of moving containers.
internal resonance conditions (Akyildiz and Unal 2005). Generally, the hydrodynamic pressure of liquids in
Under low gravity field, the surface tension is dominant moving rigid containers has two distinct components. One
and the liquid may be oriented randomly within the tank component is directly proportional to the acceleration of
depending essentially upon the wetting characteristics of the tank. This component is caused by the part of the fluid
the tank wall. moving with the same tank velocity. The second is known
Earthquake-induced as well as sea-wave-induced liquid as ‘‘convective’’ pressure and represents the free-surface-
motion in a container is of great concern in civil and ship liquid motion. Mechanical models such as mass-spring-
engineering. Extensive research has been carried out in dashpot or pendulum systems are usually used to model
the past two decades using analytical, numerical and the sloshing part. A liquid’s motion inside its container
experimental methods for evaluating sloshing responses has an infinite number of natural frequencies, but it is the
in rigid containers. However, usually the sloshing lowest few modes that are most likely to be excited by the
phenomenon is studied considering that the container is a motion of a vehicle. Most studies have therefore
rigid one and is subjected to base excitation. concentrated on investigating forced harmonic
Sloshing is the displacement of free surface of liquid in oscillations near the lowest natural frequencies, predicted
vertical liquid-filled container under external excitation. by the fluid field linear equations. However, nonlinear
The dynamic load acting over a tank structure as a result effects result in the frequency of maximum response
of the fluid motion with free surface confined inside a being slightly different from the linear natural frequency
tank. The sloshing behaviour is of important concern in and dependent on amplitude. Nonlinear effects include
the design of liquid containers. The clear understanding of amplitude jump, parametric resonance, chaotic liquid
sloshing characteristics is essential for the determination surface motion, and nonlinear sloshing mode interaction
of the required freeboard to prevent overflow of the due to the occurrence of internal resonance among the
contaminated cooling water, and for the estimation of liquid sloshing modes. The nonlinearities associated with

free-surface motion inside moving containers are different A circular cylinder of radius R is filled with
from those nonlinear water waves in ocean and canals. incompressible liquid of density  as shown in Figure 1.
Seismic analysis of liquid storage tanks requires special
considerations. These special considerations account for
Z
the hydrodynamic forces exerted by the fluid on tank
wall. Knowledge of these hydrodynamic forces is
essential in the seismic design of tanks. Evaluation of R
hydrodynamic forces requires suitable modeling and
dynamic analysis of tank liquid system, which is rather
complex. However, availability of mechanical models
(analogues) of tanks has considerably simplified the
analysis. These mechanical models, convert the tank-
liquid system into an equivalent spring mass system.
Design codes use these mechanical models to evaluate H
seismic response of tanks. While using such an approach,
various other parameters also get associated with the
X
analysis. Some of these parameters are: Pressure Y
distribution on tank wall due to lateral and vertical base
excitation, time period of tank in lateral and vertical
mode, effect of soil-structure interaction and maximum Fig. 1: Cylindrical tank filled with liquid of
sloshing wave height. Design Codes have provisions with depth H
varying degree of details to suitably evaluate these Laplace equation (1) may be written for cylindrical
parameters. Babu and Bhattacharya (1996) presented a
container
finite element numerical scheme to predict sloshing
displacements and hydrodynamic pressures in thin walled  2 1  1  2  2
liquid retaining structure due to external excitations. The
   0 (2)
r 2 r r r 2  2 z 2
displacement of the tank wall due to the developed
pressure considering coupled fluid structure interaction
Boundary condition
problem was solved. Biswal et al. (2003) evaluated the
Velocity of liquid normal to the tank wall is zero which
natural frequencies of liquid in a liquid-filled cylindrical
gives the no-penetration boundary condition at tank wall
rigid tank considering with and without baffles. The
authors used an annular plate as a baffle, which was fitted 1
to the inner periphery of a cylindrical tank. Finite element 
technique was used to discretize both the liquid and the 0 (3)
structural domain. The slosh frequencies of liquid were n
computed for different dimensions, thicknesses and Free surface boundary condition  2
positions of baffles, both rigid and flexible considering
the circumferential wave number as one. The coupled 1  2 
.  0 (4)
vibration frequencies of the tank-flexible-baffle system g t 2 z
were computed considering the effect of sloshing of Thus slosh displacement at free surface may be written as
liquid.
An attempt has been taken to study the stress distribution 1 
  (5)
on the tank wall due to base excitation of a liquid filled g t
tank. The slosh characteristic of a partially liquid filled
Pressure P ( r ,  , z , t ) in terms of sloshing displacement
cylindrical tank is studied. In this study, provisions given
in various codes on seismic analysis of tanks are may be written as
reviewed. Codes considered are: ACI 350.3, AWWA D- P 1 
110, AWWA D-115, Eurocode 8 and New Zealand
  g  (6)
 g t
National Society for Earthquake Engineering (NZSEE)
Where g is the acceleration due to gravity force.
guidelines.
Possible solution of equation (2) with the tank wall
MATHEMATICAL FORMULATIONS
boundary condition is given by
The motion of liquid is assumed as inviscid and
 
irrotational, it may be expressed by Laplace equation in
liquid domain Ω.
 (r , , z, t )    Amn (t ) cos(m )  Bmn (t )sin(m ) 
m 0 n 0
cosh[mn ( z  H )
 2  0 (1)  J mn (mn R )
Where,    ( r ,  , z , t )
cosh(mn H )
(7)

Where Amn and Bmn are time dependent can be tank called impulsive component and another one which
undergoes sloshing motion called convective component.
determined from free surface boundary condition and
J m (.) is the Bessel function of first kind of order m and Response to lateral excitation of tank
 mn J m mn
mn  is the root of r R
0 Impulsive modes refer to lateral mode of tank-liquid
R r system. Lateral seismic force on tank depends on the
The natural frequency of slosh may be written as impulsive mode time period. Time period of tank-liquid
2 g mn  H system depends on the flexibility of support also. It may
mn  tanh( mn ) (8) be noted that time period of tank –liquid system depends
R R on mass of wall as well as liquid. However, recognizing
the fact that mass of the wall is usually quite small as
Finite element model
compared to liquid mass, certain approximation made and
The finite element model is based on Galerkin weighted
mass of the wall is neglected.
residual method. The potential function  may be
approximated as field variable Expression for impulsive time period of fixed base
N circular tank is stipulated in different codes is shown in
   N i ( x, y, z )i (t ) (9) Table 1.
i 1
Where N i is the shape function. Applying divergence Table 1: Impulsive time period Ti for cylindrical container
theorem to the residual form of governing differential
Refer Code Impulsive Time Period Ti (s)
equation (1) for the liquid and minimizing the energy
function one may write
ACI 350.3 c
0.628h
 N N N N N N  Ti  E
  xi  xi i  yi  yi i  zi  zi i d Cw t
R
1 cw is a coefficient given as function of h/R
  Ni  Nii d 1   Ni nd
ˆ 2
g 1 2
(10) NZSEE 5.61 h 
The equation (10) may be written as Guidelines Ti  E
Kh
 M   [ K ]  0 (11)
K h is function of h/R and t/R
The fluid stiffness matrix [K] is formed by direct
summation of element stiffness matrix and element
stiffness matrix is given by EUROCODE h
2R
Ti 
8 Ci
Et
 Ke     R B T
B dA (12)
A Ci  0.01675(h / R)2  0.15(h / R)
B is the derivative of N and dA is the area of the element.
And mass matrix of the system is found by direct  0.46
summation of Where
1 n ρc = Mass density of tank wall
[M ]   M m (13) ρ = Mass density of liquid
g m 1 E = Modulus of elasticity of tank wall
Element mass matrix is given by t = Thickness of tank wall
 M m e   N T N d  (14) R = Radius of tank
h = Depth of liquid
Where n is the total number of element.
DIFFERENT CODES PROVISION Response to vertical excitation of tank
In mechanical model of tank liquid system i.e. spring Under the influence of vertical excitation, liquid exerts
mass system is considered in different codes to simplify axis-symmetric hydrodynamic pressure on tank wall.
the determination of convective and impulsive time Knowledge of this pressure is essential in properly
period of tank liquid system. In these mechanical models assessing the safety and strength of tank wall against
it is recognized that vibrating fluid inside the container buckling. Response to vertical excitation is mainly
has two components, one that moves in unison with the governed by the time period of fundamental breathing
mode or axis-symmetric mode of vibration of tank-liquid

system. Distribution of hydrodynamic pressure due to
vertical excitation also gets influenced by wall flexibility.
Expression for convective time period Tv given in various
code is stipulated in Table 2.
Table 2: Convective Time period (Tv) for cylindrical

container
Refer code Convective time period Tv
ACI 350.3
 
2  
Tv     h 

 t   E 
 R
NZSEE
 5.61 h  
Tv   
Guidelines
 Kv E  Fig. 3: Variation of impulsive time period for radius
Kv is a function of h/R and t/R R=4m
 From Fig. 2 and 3, it is observed that as the H/R ratio
EUROCODE 8(1  2 ) I 0 ( )   increases the time period also increases and this increase
8 Tv    h
  (t / R )(h / R ) I1 ( )  E is attributed to the fact that mass of the system also
  increases (as time period is directly proportional to mass
of the system). Moreover, it is observed that both
EUROCODE and ACI350.3 codes gave similar results.
I 0 , I1 are respectively modified Bessel
R
function of order 0 and 1 and 
2h
Where
h = depth of liquid
ρ = Mass density of liquid
E = Modulus of elasticity of tank wall
t = Average thickness of tank wall
RESULTS & DISCUSSIONS
Impulsive time period of cylindrical tank for radius R=

2m and R=4m is presented in Fig. 2 and Fig. 3. The
variation of impulsive time period (Ti) is presented with
the help of EUROCODE 8 and ACI 350.3 for different
liquid depth to radius ratio.
Fig. 4: Variation of Convective time period with H/R
ratio of the tank for radius R=2m
Fig. 2: Variation of impulsive time period with H/R

ratio for radius R=2m Fig. 5: Variation of Convective time period with H/R
ratio of the tank for radius R=4m

Convective time period for vertical mode of vibration is
presented in Fig. 4 and 5. It is observed from figure that
the variation of convective time period is linearly varied
with increasing the H/R ratio of the tank,
STRESS ANALYSIS
A cylindrical tank of radius 2m and height 4m is analyzed

under external excitation. Material properties of tank wall
are chosen for the analysis is as below. Modulus of
elasticity E=2.2 x1010 N/m2, Poisson’s ratio 0.15 and
density 24 kN/m3 .
The normal stress distribution of the tank wall is

performed by using ANSYS software and presented in
Figure 6-7.
Fig. 8: Shear Stress distribution at tank wall in X-Y
direction
Fig. 6: Stress distribution at tank wall in X-

direction
Fig. 9: Shear Stress distribution at tank wall in X-Z

direction
CONCLUSIONS
Time period analysis of liquid filled tank is important as it
decides the natural frequency of the tank liquid system. If
external excitation frequency closes to one of the natural
frequency of the tank liquid system resonance occurs.
Presence of hydrodynamic forces requires special
considerations in the seismic analysis of liquid storage
tanks under vibration. Invariably all the codes use
mechanical models (analogues) to evaluate hydrodynamic
forces. These mechanical models evaluate impulsive and
convective masses, time period of impulsive and
convective modes of vibrations, hydrodynamic pressure
distribution and sloshing wave height. It is noted that
Fig. 7: Stress distribution at tank wall in Y-direction separate mechanical models are available for tanks with
rigid and flexible walls. Rigid tank models are much
simpler than flexible tank models. As far as evaluation of
The shear stress variation of the tank wall is analyzed for impulsive and convective mass is concerned, there is no
a partially water filled tank and presented in Fig. 8 and 9. significant difference in the values obtained from rigid
and flexible tank models. Time period of tank-fluid

system, for which close form expressions are available, structure, Journal of Sound & Vibration, 281, 835-
depends on tank flexibility. Such an approach makes 847.
these American codes much simpler as compared to Ikeda, T., and Ibrahim, R. A. (2005). Nonlinear random
Eurocode 8 and NZSEE guidelines. The type of base on responses of a structure parametrically coupled with
which tank is resting influences the time period of tank. liquid sloshing in a cylindrical tank, Journal of sound
Some of the codes deal with tanks with different types of and Vibration, 284, 75-102.
base supports. NZSEE guidelines suggest consideration Jain, S. K. and Medhekar, P. S., (1993) Proposed
of higher convective mode while evaluating sloshing provisions for a seismic design of liquid storage tanks:
wave heights. All other codes consider only first sloshing Part I – Codal provisions, Journal of Structural
mode. Eurocode 8 and ACI 350.3 mention higher values Engineering, 20(3), 119-128.
of sloshing wave height. Although different codes use Liu, D. and Lin P., (2008) A numerical study of three
different explicit formulae for finding various parameters dimensional liquid sloshing in tanks, Journal of
of seismic analysis of liquid storage tank but on computational Physics 227(8), 3921-3939
evaluation the results and the nature of the graphs show Malhotra, P. K., Wenk, T., and Wieland, M., (2000)
distinct similarities thereby suggesting that all codes have Simple procedure for seismic analysis of liquid storage
similar design consideration for liquid storage tanks. tanks, Structural Engineering, IABSE, 10 (3), 197-201
NZS 3106, 1986, Code of practice for concrete structures
for the storage of liquids, Standards Association of
REFERENCES New Zealand, Wellington.
ACI 350.3, (2001) Seismic design of liquid containing Priestley, M. J. N. (1986) Seismic design of storage tanks,
concrete structures, An American Concrete Institute Recommendations of a study group of the New
Standard. Zealand National Society for Earthquake Engineering.
ACI 371 (1998) Guide for the analysis, design and Srivastava, Abhinav (2011), Study of sloshing and
construction of concrete pedestal water towers, An hydrodynamic pressure in cylindrical tank, M. Tech
American Concrete Institute Standard. Dissertation submitted to Department of Civil
Akyildiz, H and Unal, E. (2005) Experimental Engineering, IT BHU, Varanasi
investigation of pressure distribution on a rectangular Veletsos, A. S., (1984) Seismic response and design of
tank due to the liquid sloshing, Ocean Engineering liquid storage tanks, Guidelines for Seismic design of
32(11-12) 1503-1516 oil & gas pipelines system, ASCE, NY, 255-370.
API 650 (1998) Welded storage tanks for oil storage,
American Petroleum Institute Standard, Washington
D. C.
Aslam, M., Godden, W. G., Scalise, D. T. (1979)
Earthquake Sloshing in Annular and Cylindrical
Tanks, Journal of Engineering Mechanics ASME, 105,
pp.371-389
AWWA D-100, (1996) “Welded steel tanks for water
storage”, American Water Works Association,
Colorado
AWWA D-103, (1997) Factory-coated bolted steel tanks
for water storage”, American Water Works
Association, Colorado
Biswal, K. C, Bhattacharyya, S. K., Sinha, P. K. (2003)
Free Vibration Analysis of Liquid-Filled Tank with
Baffles, Journal of Sound and Vibration, 259(1),
pp.177-192
Eurocode 8, (1998), Design provisions for earthquake
resistance of structures, Part 1- General rules and Part
4 – Silos, tanks and pipelines, European committee for
Standardization, Brussels.
FEMA 368, (2000), NEHRP recommended provisions for
seismic regulations for new buildings and other
structures, Building seismic safety council, National
Institute of Building Sciences.
Babu S. S and Bhattacharyya S. K. (1996) Finite element
analysis of fluid-structure interactions effects on liquid
retaining structures due to sloshing, computer
structure, 59(6), 1165-1171
Bauer, H. F. and Chiba, M. (2005), Axis-symmetric
oscillation of a viscous liquid covered by an elastic

ISBN 978-81-921121-0-7
AVERAGE P-T ESTIMATES AND MINERAL CHEMISTRY OF THE

TWO PYROXENE BEARING GRANULITES FROM SONAPAHAR,
WEST KHASI HILLS DISTRICT, MEGHALAYA
S.B.Dwivedi
Associate Professor, Department of Civil Engineering, Institute of Technology, BHU, Varanasi-221005, Email:
sbd.civ@itbhu.ac.in
K.Theunuo
Research Scholar, Department of Civil Engineering, Institute of Technology, BHU, Varanasi-221005, Email:
ktheunuo@gmail.com
ABSTRACT: The two-pyroxene granulite of the Sonapahar belongs to Shillong-Meghalaya-Gneissic complex

and includes orthopyroxene-clinopyroxene-hornblende-plagioclase ± quartz as a common stable assemblage
during thermal peak of metamorphism. The detailed petrography and mineral chemistry on the basis of EPMA
data of the coexisting phases reveal that the basic granulite represents single events of metamorphism. The
average temperature, average pressure and average P-T condition of metamorphism suggest a thermal peak of
metamorphism at 1029 ± 620C / 7.6 ± 1.7 kbar. The calculation of P-T phase diagram in NCKFMASH system
afford seven stable invariant or intersection points at 780 ± 2490C/ 10.43 ± 1.9kbar, 917 ± 2390C /8.07 ± 3.9
kbar, 1073 ±3900C /9..39±2.5kbar,1149 ± 4720C /5.13 ± 6.2 kbar, 796 ± 2560C / 6.57 ± 5.6 kbar, 1005 ± 4150C /
9.64 ± 2.5kbar and 923±2450C /4.27±6.8kbar.
Keywords: Two-Pyroxene, Petrography, Mineral Chemistry, Phases and P-T conditions
INTRODUCTION of the Sonapahar granulites including the corundum-

Granulite represents the middle to lower crustal rock spinel-sapphirine metapelites that occur as enclaves
and play a major role to discuss the geodynamic within granite-granodiorite suit were 7500C and 5
evolution of the crust and upper mantle. Coexisting kbar. However no detailed mineralogical and
pyroxene believed to be interesting and important petrological studied have been carried out on the two
rock forming mineral which provide physico- pyroxene granulite of the Sonapahar. The main
chemical information about the mafic component of purpose of the present investigation is (a) to discuss
the regionally metamorphosed high grade crustal the petrology and mineral chemistry of the two
rock. Two pyroxene granulites occur in several pyroxene granulite and (b) to calculate the average P-
localities in India such as Southern Granulite belt, T condition of metamorphism of the two-pyroxene
Eastern Ghat granulite belt as well as in Chhotanagpur granulite and their P-T projection with stable
Granite-Gneiss complex. The occurrences of Two- intersections.
pyroxene granulites are rare in Shillong-Meghalaya
gneissic complex (SMGC). The investigated area GEOLOGICAL SETTING
consists essentially of rocks of upper amphibolite to The area around Sonapahar ( longitude 25o 40’N and
granulite facies ( Mazumdar,1976 Lal et al. 1978, latitude 91o05’E ) belongs to the Shillong-Meghalaya-
Chatterjee et al.,2007and Dwivedi and Theunuo ,2011 gneissic complex (SMGC) and consists essentially of
and references therein), flanked on the east by rocks of upper amphibolite to granulite facies (Lal et
greenschist facies rocks belonging to the Shillong al. 1978, Chatterjee et al.2007, Dwivedi and Theunuo
series of Precambrian age and to the south by ,2011) unconfomably overlain by the Shillong Group
sedimentary rocks and basic volcanics of Cretaceous of greenschist facies intra cratonic sandy and clayey
age. Chatterjee et. al. (2007) reported the mineral rocks (Ahmed,1983,Nandy,2001). We investigated
dates at Sonapahar tightly clustered at 500 Ma ( pan- the area around Sonapahar and prepared the
African) of the SMGC which corresponds with the geological map of the area (Fig.1).
880-480 Ma Rb-Sr dates of prophyritic granites
(Ghosh et al.,2005and references therein) that
predominantly intruded the east-central part of
SMGC. Chatterjee et al. (2007) also proposed that the
progressively eastward dominance of Cambro-
Ordovician ages in the SMGC indicates a Pan-African
final amalgamation of the Indian plate with the
Australo-Antarctic plate and a northward extension of
the Prydz Bay suture through the SMGC, with the
west boundary of the suture possibly located between
the Garo Golapara, 50 km WSW of Sonapahar. Lal et
al. (1978) established that the metamorphic conditions

Fig.1 Geological map of the area around Sonapahar of
SMGC.
The basement rocks of the area around Sonapahar
include (i) cordierite-bearing granulites, (ii) basic
granulites (iii) quartz –sillimanite schist and (iv)
granite gneisses. The Mineral assemblages suggest
regional metamorphism in granulite facies condition
and the rocks are affected by doubly plunging E-W
trending folds (F1) superposed by northerly striking
later folds (F2) formed during the D1 and D2 phases of
deformation respectively. Time relations between the
two phases of deformation and metamorphic
crystallization, as revealed by Si/Se relationships of the (b)
porphyroblastic minerals, with the matrix foliation,
indicates that the regional metamorphism which
initiated with the D1 and finally outlasted the D2 Fig.2 (b) BSE image of coexisting minerals for the
deformation, represents a single events of EPMA analyses..
metamorphism including both prograde and
retrograde metamorphic sequences . Four thick bands Weak foliation is sometimes present which is defined
of two-pyroxene granulites are exposed towards the by parallel orientation of hornblende, biotite and on
south of the Sonapahar along east and west of the the odd occasion pyroxene and plagioclase. It appears
Village Nogmaweit (Fig.1). These are dark colored that during the regional metamorphism
rocks and contain orthopyroxene-clinopyroxene- recrystallization or annealing of the oriented fabric
hornblende and plagioclase as major constituents. has taken place to varying degree, as commonly seen
in the different granulite facies terrains. The two-
PETROGRAPHY AND MINERAL CHEMISTRY pyroxene granulite of the area around Sonapahar
Petrography contains the following minerals: - Orthopyroxene is
Two-pyroxene granulite of the investigated area is feebly to strongly pleochroic and occurs as
dark grey to dark greenish grey in color and is subidioblastic grains. It shows inclusion of
medium to coarse–grained. It has granulitic texture hornblende and quartz (Fig.2a) which provide the
/fabric with granular mosaics of the pyroxenes, evidence of prograde reaction such as Hbl + Qz =
hornblende and feldspar (Fig.2a &b). Opx + Cpx + Plag + H2O (i). Orthopyroxene is also
partly rimmed by hornblende and suggesting the
evidence of reversal of reaction (i). Clinopyroxene
occurs as subidioblastic prism (Fig.2a &b) and
contain inclusion of hornblende and plagioclase which
also provide the evidence of reaction given above.
Some of the crystals of orthopyroxene show
exsolution patches of colorless clinopyroxene and
vice versa has also been noted. Hornblende is present
in intermittent amounts and occurs as armored relic
within the pyroxene (Fig.2a&b). Hornblende is
commonly present in significant amount (a) as rims
around pyroxene, (b) as coarse poikiloblast and (c) as
subidioblastic prisms in textural equilibrium with the
(a) pyroxene. Plagioclase An-content ranges from (An 40
to An55) and occurs as granular aggregates or coarse
Fig.2(a) Textural relations in the two pyroxene lath-shaped subidioblastic prophyroblast Myrmekitic
granulite of the area around Sonapahar intergrowth of plagioclase and the pyroxene are seen
photomicrograph of coexisting mineral phases under in the grain contact of plagioclase and pyroxene. In a
plane polarized light Mineral abbreviations are Opx: few sections the polysynthetic twin lamellae are
orthopyroxene; Cpx: clinopyroxene; Hbl: slightly deformed indicating post-crystalline
hornblende; Plag: plagioclase and Qz; quartz deformation. Quartz occurs as xenoblasts in
association with plagioclase, and also along the
itergranular space between prisms of pyroxene. In
Fig.2a the quartz grains occurs as inclusion within the
pyroxene along with hornblende and providing the
evidence of prograde reaction (i).

Mineral Chemistry the hornblende varies between 0.187 and 0.209 p.f.u
Detailed petrographic studies of the rocks of the area and wt% of TiO2 varies from 1.66 to 1.83wt%. The ca
have revealed the presence of a number of minerals content of the hornblende varies from 1.828 to 1.893
given above. The coexisting mineral phases have been p.f.u. suggesting the presence of clacic amphibole.
analyzed by the EPMA and BSE image of the mineral The tschermakite and edenite composition of the
phase is shown in Fig.2b. hormblende of the investigated area are plotted after
Raase et al, .1986 (Fig.4).
ANALYTICAL METHOD
Chemical composition of minerals were determined
by CAMECA SX51 EPMA installed at Indian
Institute of Technology, Kharagpur with the
accelerating voltage 15kV and a beam current of 15
nA. X-Ray Lines used for analyses are K for Si, Al,
Na, Mg, Fe, Mn, Cr, Zn, Ba, P, K, Ca and Ti with
standard zedite, orthoclase, Al2O3, MgO, Fe2O3, Zns,
BaSO4 Apatite, wollastonite and rutile.
Pyroxenes The microprobe analyses and structural
formula of the pyroxenes are calculated and the
pyroxenes are plotted in a triangular end member
CasiO3-MgSiO3 –Fe2+SiO3 diagram (Fig.3). Fig.4 Plot of tschermakite against edenite components
expressed as 0.5[AlVI-(Na +K)A] and (Na + K)A atom
per formula unit respectively of calcic amphibole of
the study area.
The plot clearly suggest that the all the hornblende of
the study area lies in the field of granulite facies
(Raase et al.,1986) and indicates that these hornblende
have crystallized near the thermal peak of
metamorphism under granulite facies metamorphism.
Plagioclase: The microprobe and structural formula
of plagioclase shows that Ca/ (Ca +Na +K) ratio of
plagioclase from the study area are 0.49 to 0.51 (i.e.
andesine to labradorite). The Fe is present as Fe3+ as
minor constituent in plagioclase it may be due to Al3+
substitution or may be due to extremely fine inclusion
Fig.3 Plot of EPMA data of pyroxene shown in part of opaque in plagioclase.
of triangular diagram CaSiO3-MgSiO3-FeSiO3. The
clinopyroxene coexisting with orthopyroxene is PRESSURE-TEMPERATURE (P-T)
joined by a line. CONDITION OF METAMORPHISM
The orthopyroxene plot lies between enstatite and The P-T condition of metamorphism of two pyroxene
ferrosilite join at En53-57 and contain up to 0.024 to granulites has been deliberated through
0.028 Ca per formula unit based on 6 oxygen. The orthopyroxene-clinopyroxene solves and exchanges
AlVI of the orthopyroxene varies between 0.01 to conventional geothermobarmeters as well as
0.031 p.f.u. while AlIV varies between 0.008 and internally consistent dataset (Thermocalc ver. 3.1).
0.023.The XMg ranges between 0.56 to 0.57 and The temperature estimates at 7 kbar by conventional
corresponds to hyhersthene. The hypersthene from geothermometers are given in Table.1
basic granulites of the Sonapahar contains 0.59 to
0.99 wt% Al2O3 which is significantly lower from the Table.1 Temperature estimates of the two-pyroxene
basic granulites of other terrains. The coexisting granulite from solves and exchange conventional
clinopyroxene plot lies in the field of salite (Fig.3). geothermometer at assume pressure
The XMg of clinopyroxene ranges between 0.72 to Opx-Cpx Temp in 0C Temp in
0.75. The clinopyroxenes have higher XMg and Al-
0
content in comparisons to orthopyroxene. Geothermometers (Core) C (Rim)
Hornblende: The analyse of hornblende shows
significant amount of Fe2O3 wt% which varies from Wood& Banno 833 766
1.63 to 3.07Wt%. The AlIV and AlVI content of (1973)
hornblende varies from 1.374 to 1.493 and 0.01 to
0.031 per formula unit at 23–oxygen basis Wells (1977) 867 769
respectively. The XMg of hornblende varies from 0.56
to 0.57 which does not show the significant change of Powell (1978) 843 796
the XMg values of the hornblende. The Ti content of Average 848±17 777 ±16

The temperature estimate of coexisting 10) 5hed + 5mgts + tr + 5q = 5en + fact + 5an 9.8 4.10 13.98
orthopyroxene-clinopyroxene from the basic
granulites of the area around Sonapahar vary in the 11) 5hed + 3mgts + ts + 5q = 3en + fact + 5an 6.0 4.36 13.64
range of 833 to 8670C for core and 766 to 7960C for
rim at 7 kbar. The average pressure (Pav), average 12) 4cats + en + ts + ab = mgts + parg + 4an 6.9 5.60 12.83
temperature (Tav) and average P-T conditions of
metamorphism (P -T av) are estimated through
average temperature calculation, average pressure P av sd fit
calculation and average P-T calculation of
Thermocalc v.3.1 with EPMA data of phases Average Pressure in Kbar 8.45 1.52 1.08
involving orthopyroxene , clinopyroxene,
hornblende and plagioclase of domain 1/1.the
calculations are given in table2. Single end member diagnostic information of (P-Tav) {for x(H2O) = 0.3}
Table.2 The average temperature and average End P in Sd Temp Sd(T) Core Fit
pressure and average P-T condition estimate through member kbar (P) in 0C
internally consistent dataset (THERMOCALC v3.1)
where mineral abbreviations are di= diopside, di 7.64 1.66 1032 66 0.219 0.91
hed=hedenbergite, cats=Ca-Tschermak pyroxene,
hed 7.65 1.67 1030 63 0.249 0.92
en=enstatite, fs=ferrosilite, mgts= Mg-Tschermak
pyroxene tr=tremolite ,fact=ferroactinolite, ts= cats 6.15 2.21 1005 66 0.394 0.79
tschermakite, parg=pargasite, an=anorthite, ab=albite,
q=quartz and H2O=water fluid. en 7.53 1.68 1048 79 0.073 0.90
Reactions used to calculate average temperature(Tav) {for x(H2O) = 0.3} fs 7.74 1.66 1033 62 0.238 0.86
mgts 7.71 1.73 1031 64 0.285 0.91

Independent set of reaction Tem Sd( lnK
p in T) tr 8.15 1.72 1069 72 0.326 0.77
0
C
fact 7.61 1.66 1022 63 0.233 0.83
1) 2tr = 4di + 3en + 2q + 2H2O 63
959.1 0.612 ts 7.75 1.66 1026 62 0.225 0.86
2) 2di + fs = 2hed + en 481 - parg 8.34 1.85 1005 68 0.001 0.84

917.5 0.780
an 7.60 1.78 1029 62 0.237 0.91
3) 3cats + 2tr = 7di + 3mgts + 2q + 2H2 O 905.3 139 -
0.889 ab 7.68 1.67 1028 62 0.219 0.91
4) 4mgts + 2tr + 2q = 7en + 4an + 2H2 O 972.2 q 7.64 1.66 1029 62 0.232 0.92
121 12.68
8 H2O 7.64 1.66 1029 62 0.232 0.92
5) 5di + fact = 5hed + tr 610.0 1.108 Average 7.6 1.7 1029 62 0.232 0.92
493
6) hed + 2fs + ts = di + 2mgts + fact 1166. 677 -

6 3.009
The calculation of the Tav, Pav has been performed
on the basis of concerning end member reactions
7) 3tr + parg = 7di + cats + 6en + ab + 4H2 O 996.3 59 -
(Table.2). The obtained average temperature
1.373 (independent set of reactions 1 to 7) and average
pressure (independent set of reactions 8 to12) are
T av sd fit 1021± 510C and 8.45 ± 1.52 kbar respectively. The
Average Temp(0C) 1021 51 1.02
average P-T condition (P-T av) is estimated on the
Reactions used to calculate Average(Pav) pressure{for x(H2O) = 1.0}
source of end member diagnostic information of
pyroxene, hornblende and plagioclase linking
Independent set of reaction P Sd lnK diopside, hedenbergite, Ca-Tschermak pyroxene,
(Kba (P) enstatite, ferrosilite, Mg-Tschermak pyroxene
r) tremolite, ferroactinolite, tschermakite, pargasite,
anorthite, albite, quartz and water fluid. The estimated
8) cats + q = an 10.4 1.83 2.519
average P-T condition of metamorphism is 1029 ±
9) di + mgts + q = en + an 8.9 3.97 3.019 620C / 7.6 ± 1.7 kbar (table.3) which is within the
limit of ultrahigh-temperature (UHT) of regional

metamorphism in which crustal rocks are subjected to given in Fig.5 and helpful to know the P-T history of
temperature of 900-11000C at moderate pressure of 7- the rock at given mineralogical data (Yang & Powell,
13 kbar (Harley,1998). The P-T phase diagram has 2006)
been also calculated for the two-pyroxene granulite
from the Thermocalc ver3.1 of the same domain CONCLUSION
involving all above mentioned end member phases in The two pyroxene granulite of the area around
NCKFMASH system and the P-T projection is given Sonapahar is located towards the south of the
in Fig.5. Fifty-four (54) reactions and 22 intersections Sonapahar along east and west of the Village
have been obtained out of which only seven Nogmaweit and contains the mineral phases
intersections are stable at definite P-T conditions orthopyroxene-clinopyroxene-hornblende-plagioclase
(Fig.5). The intersection or invariant point (1) ± quartz. The pyroxene appears in the basic granulite
involving phases di, hed, cats, fs,,mgts ,an,q located at through the prograde reaction Horblende + quartz=
780±2490C /10.43±1.9kbar from which four orthopyroxene + clinopyroxene + plagioclase + vapor.
univariant reactions are diverging, point(2) involving The reversal of this reaction is also observed which
phases di, hed, en, fs, mgts, an, q at 917±2390C result in the formation of hornblende wrapping the
/8.07±3.9 kbar from which five univariant reactions orthopyroxene. Textural relations between the two
are diverging, Point(3) involving phases hed, cats, fs, phases of deformation and metamorphic
mgts, fact ,ts, an, q at 1073±3900C /9..39 ±2.5 kbar crystallization represent single events of
from which 3 univariant reactions are diverging, metamorphism including both prograde and
point(4) involving phases hed en, fs, mgts, fact ,ts, an, retrograde metamorphic sequences. The mineral
q at 1149±4720C /5.13±6.2 kbar from which five chemistry of the pyroxene, hornblende and
univariant rections are diverging, point (5) involving plagioclase and BSE image provide the evidence of
phases di, hed, cats, fsmgts, ts ,parg, an, ab at occurrence of the two-pyroxene granulite and also
796±2560C /6.57±5.6 kbar, point(6) involving phases show the variation of cations within the granulite
cats, fs, mgts, fact ,ts, an, H2O at 1005±4150C facies rocks. The average temperature, average
/9.64±2.5 kbar and finally invariant point(7) involving pressure and average P-T conditions of
phases di, hed, an, fs ,mgts ,ts ,parg, an, ab at metamorphism are 1021± 510C, 8.45 ± 1.52 kbar and
923±2450C /4.27±6.8 kbar (Fig.5) 1029 ± 620C / 7.6 ± 1.7 kbar respectively. The P-T
condition indicate the thermal peak of the
metamorphism of two-pyroxene granulite at 1029 ±
620C / 7.6 ± 1.7 kbar which support the idea of
ultrahigh-temperature of metamorphism. The seven
stable intersections or invariant points have been
observed in NCKFMAS system with phases involving
di, hed, cats, en, fs, mgts, tr, fact, ts, parg, an ab and q,
H2O as excess phases at 780±2490C /10.43±1.9 kbar,
917±2390C /8.07±3.9 kbar, 1073±3900C / 9..39 ±2.5
kbar, 1149±4720C /5.13±6.2 kbar, 796±2560C
/6.57±5.6 kbar, 1005±4150C /9.64±2.5 kbar and
923±2450C /4.27±6.8 kbar.
ACKNOWLEDGEMENT
We are grateful to Prof.R.K.Lal for his valuable
suggestion to give the framework of the manuscript.
This work has been possible through a DST (DCS)
sponsored project (ESS/16/304/2005). We are also
thankful to Head, Department of Civil Engineering,
IT, BHU for providing the infrastructure. K.. Theunuo
would like to thank UGC for financial support as JRF.
Fig.5 P-T projection in NCKFMASH system plot REFERENCES

showing the seven stable intersections. Where mineral
abbreviatins are di= diopside, hed=hedenbergite, Ahmed, M. (1983) Depositional environment of the
cats=Ca-Tschermak pyroxene, en=enstatite, basal conglomerate of the Barapani Formation,
fs=ferrosilite, mgts= Mg-Tschermak pyroxene Shillong Group, KhasiHills, Meghalaya. Quart. J.
tr=tremolite ,fact=ferroactinolite, ts= tschermakite, Geol. Miner. Metal. Soc. India v. 55 pp. 62–68.
parg=pargasite, an=anorthite, ab=albite, q=quartz and Chatterjee, N., Mazumdar,A.C.,Bhattacharya,A.,
H2O=water fluid. Saikia,R.R. (2007) Mesoproterozoic granulites of
Pressure, temperature, parameters of the minerals at the Shillong-Meghalaya Plateau: evidence of
the invariant points in the NCKFMASH system is westward continuation of the Prydz Bay Pan-

african suture into northeastern India. Precamb. Mazumdar, S.K.(1976) A summary of the
Res. v.152.pp.1-26. Precambrian Geology of Khasi Hills, Meghalaya.
Dwivedi, S.B and Theunuo, K. (2011). Two pyroxene Geol. Ind. Misc. Publ.v.23 (2) . pp. 311–334.
mafic granulites from Patharkhang, Shillong- Nandy, D.R. ( 2001) Geodynamics of the
Meghalaya Gneissic Complex. Curr. Science v.100 Northeastern India and theAdjoining Region. Acb.
pp.100-105. Publ, 209 p.
Ghosh, S., Fallick, A.E., Paul, D.K., Potts, P.J. ( Powell, R.(1978) Thermodynamics of pyroxene
2005) Geochemistryand origin of Neoproterozoic geotherms. Royal Soc.(London) Phil. Trans..
granitoids of Meghalaya, Northeast India: v.288A. pp.457-469.
Implications for linkage with amalgamation of Raase,P., Raith.,M., Ackermand,D. and Lal, R.K.
GondwanaSupercontinent. Gond. Res. v.8. pp. (1986) Progressive metamorphism of mafic rocks
421–432. from greenschist to granulite facies in the
Harley, S.L.( 1998) On the occurrence and Dharwar Craton of South India. J.Geol.
characterization of ultrahigh temperature crustal V.94.pp.261-282.
metamorphism. In: Wells,P.R.A. (1977) Pyroxene thermometry in simple
Treloar,P.J.,O.Brien,P.J.(Eds.), What drives and complex system.Contrib.Mineral.Petrol. v.62.
metamorphism and metamorphic reactions. Geol. pp.129-139.
Soc. London, Special publication v.138 pp81-107. Wood,B..J. and Banno,S. (1973) Garnet-Opx and
Holland,T.J.B.,Powell,R.(1998) An internally – Opx-Cpx relationship in simple and complex
consistent thermodynamic dataset for phases of systems. Contrib. Mineral.Petrol. v.42. pp.109-
pertological interest. J.Met.Geol. v.16. pp 309-344. 124.
Lal, R.K., Ackermand, D., Seifert, F., Haldar, S.K.. Yang.J.J. and Powell,R. (2006) Calculated phase
(1978) Chemographicrelationships in sapphirine- relatins in the system Na2O-CaO-K2O-FeO-MgO-
bearing rocks from Sonapahar,Assam, India. Al2O3 -SiO2 -H2O with application to UHP
Contrib. Miner. Petrol. V. 67. pp 169–187. eclogites and whiteschists. J.Petrol.v.47 pp.2047-
2071.

ISBN 978-81-921121-0-7
CRACK SENSING CAPABILITIES OF SMART CANTILEVER

STRUCTURES
Nilanjan Mallik
Senior Assistant Professor, Mechanical Engineering, Institute of Technology, Banaras Hindu University,
Varanasi, Pin-221005, UP, India, e-mail: get_nilu@yahoo.com
ABSTRACT: A theoretical analysis is presented to predict crack sensing capabilities of smart sensors on
cantilever type beam structures under static loading conditions. Smart sensors are made of piezoelectric
materials and their composites. A single sensor is bonded on the top surface of a cantilever beam structure. The
sensor output is modeled with respect to strain developed in the structure due to mechanical point loading. The
effect of crack location, crack size, sensor location, sensor size and sensor properties are studied on the sensor
output.
Keywords: SHM; Structural Health Monitoring; Piezoelectric Sensors; Smart Structures; Smart Sensing;
Intelligent Structures; Cantilever Beam Structures.
INTRODUCTION [7] examined metallic materials coated with

Beams are common structures used in machines and piezoelectric coating. Zhang et al [4] and Yin et al [5]
civil structures to carry and transfer high loads. examined the damage detection in composites coated
Cracks are the main cause of failure for such with PVDF films. Several researchers [8, 9] presented
structures. Sudden failure during high load operation impedance based method for structural health
can be disastrous due to the presence of crack. Thus monitoring. Park et. al. [8] introduced a damage
early crack detection is a must for structural detection methodology based on impedance change
components. At present real challenges arising for the with high frequency excitation. Several researchers
control, maintenance and retrofitting of existing [10-13] investigated structural damage detection using
structures and machineries concern the diagnostic distributed sensors. Okafor et al [9] collected the
identification of damages. Non-destructive techniques modal data through built-in piezosensors. They
(NDT) are of great interest for this purpose. detected delaminations by analyzing the data through
Conventional methods of NDT include visual a neural network. Fukunaga et al [10] proposed a two-
inspection, radiography i. e. X-ray technique, thermal stage damage identification scheme using a limited
analysis and ultrasonic testing. However, these number of piezoelectric sensors. Castillo et al [11]
methods require overhauling and too long time for focused on the use of small embedded or surface
inspecting every part of structures and are not bonded piezoelectric patches for monitoring the
practical for real time or online health monitoring. integrity. Schulz et al [14] reported a method based on
Moreover, in the case of conventional NDT the transmittance function. It required deployment of
techniques, additional hardware is required for multiple sensors termed as structural neural network
excitation of structures and collection of vibration (SNS). Several researchers [15-19] also demonstrated
data. Of late effective properties of piezoelectric use of carbon nanotubes as strain sensors of
composites has been obtained by Mallik et. Al. [1]. structures. Several researchers [20-23] also
Smart sensors made of materials like piezoelectric demonstrated crack detection technique by using
materials, piezoelectric composites, rheological fluids piezoelectric patches on beam structures. Detailed
i.e. magnetorheological and electrorheological fluids, parametric study of crack detection by detecting strain
shape memory alloys, carbon nanotubes etc. exploits with the help of smart sensors on beam like structures
damage detection online. The piezoelectric materials has not yet been explored much. This provides the
are one of the most common smart materials used in main motivation for the present study. The present
structural health monitoring (SHM). These sensors study theoretically demonstrates a strain-based
also offer the capability of actuation. Crawley et. al. approach to determine crack in a cantilever beam like
[2] demonstrated that the combined features of structure using piezoelectric composite sensors. A
sensing and actuation of piezoelectric materials can be series connected sensor system is considered to
used to diagnose structures and generate corrective develop the mathematical modeling. Detailed
measures to provide safety and increase life. Such parametric study is illustrated for a single sensor case
structures are called intelligent structures when they and demonstrated its effectiveness for crack sensing.
are coupled with piezoelectric sensors and/or
actuators. Egashira and Shinya [3], Zhang et al [4], MODELING
Yin et al [5], Shiozawa et al [6] and Rees et al [7] Sensor Modeling
reported strain based method for structural health Figure 1 shows n sensors bonded on the top surface
monitoring. Egashira and Shinya [3] and Rees et al of a cantilever beam structure and connected in series

to a measuring device. Each sensor has electrodes at sensor. Again, electric displacement ( D ) of a
its top and bottom surfaces as shown in Fig. 2. It is piezoelectric sensor is related with piezoelectric stress
assumed that the bonding between sensor patch and coefficient ( e ) through constitutive equation as
beam structure is perfect. The voltage current
relationship of the series connected sensor circuit can D  e   E (5)
be given as
where,  and E are mechanical strain and electric
V1  V2  V3  .......  Vn  IR  0 (1) field, respectively. Electric displacement is defined as
the electric charge per unit cross section thus, electric
In concise form Eq. (1) can be written as charge ( Q ) of a piezoelectric sensor can be obtained
from Eq. (5) as
p
P
hp Q  eA   AE (6)
Using Eq. (4) and the relationship between electric

h V
field and voltage, E  , Eq. (6) can be rewritten as
hp
L w
Q  eA   CV (7)
Fig. 1 Cantilever beam with crack mounted with
piezoelectric sensor patches
Current (time rate of charge) through sensor can be
obtained by differentiating Eq. (7) w.r.t. time. Thus,
  eA 
IQ 
  CV (8)
The relationship between voltage and current of a

piezoelectric sensor can thus be written by rewriting
Eq. (8) as
  I  eA 
V  (9)
Fig. 2 A piezoelectric sensor patch C C
n For n sensors connected in series Eq. (9) becomes
 Vi  IR  0 (2)
i 1 n
nI eA n
 V i  C  
C i 1
where, Vi is the voltage across the i th sensor, R is i 1
the resistance of the measuring device and I is the (10)
current in the circuit. Differentiating both sides of Eq.
(2) w.r.t time, following relationship is obtained Substituting Eq. (10) into Eq. (3) current equation for
the sensor is obtained as
n
 V i  IR  0 (3)
I 
nI eA n
i 1  
RC RC i 1
(11)
 and I are the time rate of change of Taking Laplace transform of Eq. (11) following
where, V
equation in Laplace domain is obtained
voltage and current, respectively. For a piezoelectric
sensor capacitance ( C ) is related to dielectric
constant (  ) as nI (s) eAs n
sI(s) 
RC
  i ( s )
RC i 1
(12)
A
C (4)
hp Thus current through the sensors connected in series
is given by
where, A is the cross sectional area of the electrode
surface of the sensor and h p is the thickness of the

eh p nh p
n where, a1  and b 1 
. The transfer
eAs  RA
I(s )  
RCs  n i 1
i (s ) (13)
function relating sensor output to strain input in case of
a single sensor mounted on the structure and connected
to a measuring device is given by
And voltage across series connected sensors ( V ) can
be obtained from the relation V  IR as
sin gle V sin gle (s ) eARh p s a s
n G (s )    1 (20)
eARs  (s ) RAs  h p s  b 3
V (s )   i (s)
RCs  n i 1
(14)
eh p hp
where, a1  and b 3  . For step input
Also voltage across i th sensor can be obtained from  RA
the relation Vi  IR si as 
with magnitude as strain  (s )  and
s
n
eAR sis n
Vi (s)   i (s)
RCs  n i 1
(15) n  i sum
 i (s)  i1s 
s
, the time responses of the
i 1
sensor for series connection and for single sensor
where, R si is the resistance of the i th sensor. mounted on the structure are respectively given by
Substituting Eq. (4) into Eq. (14) and (15), voltage
across series connected sensors and across i th sensor v series (t ) sum a1e  b1t (21)
can be obtained respectively as
sin gle
eARh p s n v (t )  a 1e  b3t (22)
V (s ) 
RAs  nh p
  i (s)
i 1
(16) The exponential frequency ( a ) or the initial rate of
change of response, the time constant ( T ), rise time
( Tr ) and settling time ( Ts ) for these two cases are
eAR si h p s n
Vi (s )   i (s) respectively given by
RAs  nh p i 1
(17) neh p 2 sum
| a series | a1b 1 sum  ,
R 2 A
The voltage strain relationship for a single sensor sin gle eh p 2 

|a | a1b 3  ,
mounted on the structure and connected to a R 2 A
measuring device can be obtained from Eq. 16) by
substituting n  1 as 1 R 2 A
| T series |  ,
series
|a | neh p 2 sum
eARh p s
V sin gle (s )    (s) sin gle 1 R 2 A
RAs  h p |T |  ,
| a sin gle | eh p 2 
(18)
2 .2 2.2R 2 A
For series connection the transfer function of the Trseries   ,
sensor assembly is given by | a series | neh p 2 sum
2.2 2.2 R 2 A
Trsin gle   ,
eARh p s | a sin gle | eh p 2 
V(s) a 1s
G series (s )    (19)
n RAs  nh p s  b1 4 4R 2 A
  i (s ) Tsseries   ,
i 1
| a series | neh p 2 sum
4 4R  2 A
Tssin gle   . The steady state error
| a sin gle | eh p 2 

for the two cases are given as v series (  ) sum and
-7
x 10
14
Bending strain w/o crack
v sin gle ()  . The d. c. gain or open loop gain for 12 Strain due to crack
Total strain
the two cases are given as g series  eAR sum and 10
Strain at sensor location

sin gle
g  eAR  . The frequency response of the 8
sensor when the structure is strained for the two cases
6
are respectively given by
4
2
sum a 1
v series ( j)   A 1  j B 1 (23) 0
b 1  j
-2
0 0.2 0.4 0.6 0.8 1
Non-dimensional parameter for sensor location
sin gle a1 Fig. 3 Strain at sensor location with non-dimensional

v ( j)   A 4  jB 4 (24) sensor location parameter
b 3  j
-7
x 10
14
Bending strain w/o crack
neRAh p 2 sum 12 Strain due to crack
where, A1  , Total strain
n 2h p 2   2 R 2  2 A 2 10
Strain at sensor location

2 2
eR A h p sum 8
B1  , Sensor to the Sensor to the
2
n 2h p   2 R 2  2 A 2 6 left of Crack right of Crack
eRAh p 2 
4
A4  , 2
h p 2   2 R 2 2 A 2
0
eR 2 A 2 h p 
B4  . The magnitude and phase -2
hp 2  2R 2 2 A 2
-0.4 -0.2 0 0.2 0.4 0.6 0.8 1
Non-dimensional parameter for relative position of sensor and crack
of frequency response for the two cases are given by
eh p AR sum Fig. 4 Strain at sensor location with non-dimensional
M series  , parameter of relative location of sensor and crack
n 2h p 2   2  2 A 2 R 2
A non-dimensional parameter for sensor location is
sin gle
eh p AR  series  AR
M  ,   , defined which is ratio of distance of sensor from fixed
h p 2  2 2 A 2R 2 nh p end to lotal length of beam. Another non-dimensional
parameter is defined as ratio of distance between
 AR sensor and crack to total length of the beam. It is
 sin gle  .
hp found that sensor placed to the left of crack or sensor
placed between fixed end and crack is much more
sensitive than when it is placed to the right of the
crack. Frequency response parameters are also
NUMERICAL RESULTS
obtained which are not shown here.
Numerical results are obtained for a single crack
mounted on a cantilever beam subjected to static load.
CONCLUSIONS
Smart sensor used is a piezoelectric composite whose
The present work demonstrates capability of a single
material properties are obtained from Ref. [1]. Both
sensor to predict location and size of crack in a
time response and frequency response of sensor
cantilever beam structure. The results show that a
output is obtained.
single sensor can locate a crack and predict the size of
crack with very low margins of response. However,
better results can be obtained by using multiple
sensors results of which are not shown in this work.
ACKNOWLEDGEMENTS
The author acknowledges the funding of Department
of Science and Technology (DST), India vide sanction

order no. SR/FTP/ETA-086/2009 dated 25th February, damage identification using piezoelectric sensors.
2010 under Fast Track Scheme. International Journal of Solids and Structures,
34, (393–418).
REFERENCES Castillo D., Pardo de Vera C. and Guemes J. (1999).
Mallik N. and Ray M. C. (2003). Effective Properties Structural integrity analysis with piezoelectric
of Piezoelectric Fiber-Reinforced Composites. patches. Key Engineering Materials, 167, (91–
AIAA Journal, Vol. 41, No. 4, (704-710). 101).
Crawley E. and Lius J. (1987). Use of piezoelectric Kirikera G. R., Shinde V., Schulz M. J., Ghosal A.,
actuators as elements of intelligent structures. Sundaresan M. and Allemang R. (2007). Damage
American Institute of Aeronautics and localization in composite and metallic structures
Astronautics Journal, 24, (1373–1385). using a structural neural system and simulated
Egashira M. and Shinya N. (1993). Local strain acoustic emissions. Mechanical Systems and
sensing using piezoelectric polymer. Journal of Signal Processing, 21, (280-297).
Intelligent Material System and Structure, 4, Inpil K., Schulz M. J., Kim J. H., Shhanov V. and Shi
(558–560). D. (2006). A carbon nanotube strain sensor for
Zhang H., Galea S., Chiu W. and Lam Y. (1993). An structural health monitoring. Smart Materials and
investigation of thin PVDF films as fluctuating- Structures, 15, (737- 748).
strain-measuring and damage monitoring devices. Xiaojun W. and Chung D. D. L. (1995). Short-carbon
Smart Materials and Structures, 2, (208–216). fiber reinforced epoxy as a piezoresistive strain
Yin L., Wang X. and Shen Y. (1996). Damage sensor. Smart materials and structures, 4, (363-
monitoring in composite laminates by 367).
piezoelectric films. Computers and Structures, Zhang W., Sakalkar V. and Koratkar N. (2007). In-
59, (623–630). situ health monitoring and repair in composites
Shiozawa D., Kubo S. and Sakagami T. (2002). using carbon nanotube additives. Applied Physics
Passive electric potential CT method using Letters 91, (102-133).
piezoelectric material for crack identification. 4th Watkins A. N., Ingram J. L., Jordan J. D., Wincheski
International Conference on Inverse Problems R. A., Smits J. M. and Williams P. A. (2004).
in Engineering (Rio de Janeiro, Brazil). Single wall carbon nanotube- based structural
Rees D., Chiu W. and Jones R. (1992). A numerical health monitoring sensing materials. NSTI Conf –
study of crack monitoring in patched structures Nanotech, Vol 3, (149).
using a piezoelectric sensor. Smart Materials and Mallik N. et. Al. (2009). Study on carbon nanotube
Structures, 1, (202–205). spun thread as piezoresistive sensor element.
Park G., Sohn H. and Farrar C. R. (2003). Overview Advanced Materials Research Journal, vol. 67,
of piezoelectric impedance based health (155-160).
monitoring and path forward. The shock and Cacciola P., Impollonia N. and Muscolino G. (2003).
Vibration Digest, 35(6), (451-463). Crack detection and location in a damaged beam
Zagrai A. N. and Giurgiutiu V. (2001). Electro- vibrating under white noise. Computers and
mechanical impedance method for crack Structures, 81, (1773–1782).
detection in thin wall structures. 3rd International Owolabi G. M., Swanidas A. S. J. and Seshadri R.
Workshop of Structural Health Monitoring, (2003). Crack detection in beams using changes
September 12-14, 2001, Stanford University, (1- in frequencies and amplitudes of frequency
10). response function. Journal of Sound and
Schulz M., Pai P. and Inman D. (1999). Health Vibration, 265, (1–22).
monitoring and active control of composite Suresh S., Omkar1 S. N., Ganguli R. and Mani V.
structures using piezoceramic patches. (2004). Identification of crack location and depth
Composites B, 30, (713–725). in a cantilever beam using a modular neural
Okafor A., Chandrashekhara K. and Jiang Y. (1996). network approach. Smart Materials and
Delamination prediction in composite beams with Structures, 13, (907–915).
built-in piezoelectric devices using modal Sumant P. S. and Maiti S. K. (2006). Crack detection
analysis and neural network. Smart Materials and in a beam using PZT sensors. Smart Materials
Structures, 5, (338–547). and Structures, 15, (695–703).
Fukunaga H., Hu N. and Chang F. (2002). Structural

ISBN 978-81-921121-0-7
ESTIMATION OF GUST EFFECTIVE FACTOR USING IS 875

(PART 3) – 1987 AND MODIFICATIONS PROPOSED BY PREM
KRISHNA (2002)
Deepak Kumar
Senior Engineer, Tata Consulting Engineers Limited, Mumbai, 400083, Email deepakk@tce.co.in
S. Mandal
Associate Professor, Department of Civil Engineering, Institute of Technology, BHU, Varanasi, 221005, Email:
smandal.civ@itbhu.ac.in
ABSTRACT: There are various approaches to estimate the equivalent Static Wind Load on structures.
They are primarily of three types, viz., Displacement based Gust Loading Factor (DGLF), which is
traditionally adopted in the various national codes, Moment based Gust Loading Factor (MGLF), and Shear
based Gust Loading Factor (SGLF). The turbulent random flow field characterized by proper (divergent
infinite types) boundary conditions will give always varying different values of wind pressure as well as
force on any typical building. Thus the design value prescribed for any particular building will be function
of many varying parameters mostly influenced by terrain category, other buildings interfering/ shielding the
turbulence characteristics and flow pattern influenced by boundary conditions [Ojha et al. (2001)]. IS 875
(Part 3) – 1987 is based on DGLF, whereas Prem Krishna (2002) presented more versatile methods based
on MGLF, which considers the wind structure interaction in better way [Holmes (2002); Kim et al. (2005),
Deepak et al. (2007)]. This is a recently published document based on recent research publications carried
out globally. The results from these two approaches produce a different set of values for Gust Effective
Factor as well as ESWL.
INTRODUCTION (2007)]. However, wind turbulence modeling

In addition to the effect of mode combinations, aspects should be taken care while finally
six basic parameters, viz., gf (peak factor), r prescribing a set of comprehensive mathematical
(roughness factor), B (Background factor), β equations [Ojha et al. (2001), Deepak et al.
(damping coefficient), U (wind speed) (2007), Mandal et al. (2001), Mandal et al.
significantly affect estimation of equivalent static (2005)] for gust effective factor.
wind load (ESWL). The resonant ESWL
(RESWL) can be expressed in terms of the GOVERNING EQUATIONS
inertial load in each structural mode, which The IS 875 (part3) – 1987 procedure is available
depends on the mass distribution and mode in detail in the published code. In this section,
shape. The background ESWL (BESWL) only the modified procedure as proposed by
depends on the external wind load characteristics Prem Krishna (2002) is outlined.
(Davenport, 1985). The present code IS 875 (Part The Equivalent Static Wind Load (ESWL) or the
3) – 1987 for calculation of wind loads is based wind force (F) is given
on displacement based gust loading factor taking F = Cf Ae Pd (1)
the fundamental mode or first mode of vibration
into consideration. This approach often gives where, Cf = Force Coefficient, Ae = Effective
overestimated background response as well as Area, Pd = Design wind pressure, which is given
higher ESWL. The effect of height, terrain by (Prem Krishna, 2002)
category and materials of construction (steel or Pd = Pz Kd Ka (2)
concrete) on background and resonant 2
where, Pz = 0.6 V z , and Vz is the velocity of
components as well as Gust Effective Factor, (ii)
Relative magnitude of GEF from both the codes wind at height Z which is given by Kd = wind
and their variation along the height are directionally factor, Ka = area averaging factor,
and design wind speed is given by a modified
presented. Zhou and Kareem (2003) have
compared the DGLF and MGLF methods and expression, where cyclone factor (k4) is
concluded that MGLF determines that ESWL in introduced as
more realistic way that the Traditional DGLF. It
Vz = Vb k1 k2 k3 k4 (3)
has been also noted in recent researches that
higher modes of vibration may result into locally
higher values of wind loads [Huang & Chen

N= Reduced frequency, expressed as
where k1 = risk factor, k2 = terrain category f o Lh 1   g v I h 
factor, k3 = topography factor, k4 = cyclone N= (9)
factor (this factor has been introduced newly in 0.65Vh
the proposed draft to take care of cyclonic wind Vh = design 3-sec gust wind speed at height h
in coastal area). The Gust Factor is expressed as β = ratio of structural damping to critical
damping of a structure, as given in Table 1.
hs g R2 SE
1  2 I h g v2 Bs 
 The expression for gust factor as per IS 875 (part
G (4) 3)-1987 is given by.
1 2gv I z
SE
G  1  g f r B 1    
2
(10)
where, Ih = turbulence intensity of the wind, hs = 
average roof of a structure above the ground, gv This expression is valid for all load effects (Such
= peak factor for upwind velocity fluctuations as BM, SF and top displacement) and this
which shall be taken as 3.5, Bs = background expression is specifically applied to the slender
Factor, which is measure of the slowly varying and flexible structure.
background component of the fluctuating It can be appreciated that the turbulence
response, caused by low frequency wind speed intensity is maximum in terrain category as it is
variations, given as follows the category having numerous high closely
1 spaced obstructions. For across wind response,
Bs  (5) Prem Krishna (2002) has given equations, which
1
36h  s  2
 64bsh2 0.5
was not provided in the IS 875 (Part3) – 1987.
2 Lh
s = height of the level at which wind effects are EXAMPLE STRUCTURE
calculated for a structure, bsh = average breadth A typical framing plan for the example buildings
of the structure between heights s and h, Lh = has been shown in Fig. 1. This same RC
measure of the integral turbulence length scale at commercial building located in Mumbai coast
0.25 has been analysed by Bhandari et al. (2006).
h The detailed results and discussion are presented
height h = 100    , gR = peak factor for
in Deepak (2008). The study has been done by
 10 
varying aspect ratio b/a as 0.5, 1.0, 1.5, 2.0 as
resonant response (one hour period) given by
shown in Fig. 1b. In this Figure, the dimension
‘b’ perpendicular to the direction of wind flow
gR  2 log e 3600 f o  (6) has been varied from 12 m to 48 m, and the other
where, fo = first mode natural frequency of dimension ‘a’, along the flow direction has been
vibration of structure in the along wind direction kept constant at 24 m. Floor to floor height
in hertz, h = total height of a structure, S = size considered here is 3 m.
reduction factor given as follows All the studies in this paper have been done in
respect of the middle frame shown by hatched
lines. The basic wind speed adopted in the
1 present paper is 44 m/s (as per Fig. 1 of IS 875
S
 4 f o h1  g v I h    4 f o boh 1  g v I h   (part 3) – 1987) for Mumbai.
1   1  
 0.65Vh  1  gv I h  RESULTS AND DISCUSSION
(7) For the above exanple building, several
boh = average breadth of the structure between parametric studies on gust effective factors were
heights 0 and h., Ih = turbulence intensity, carried out varying plan aspect ratio, total height
E = (π/4) times the spectrum of turbulence in the of building, terrain category of building where
approaching wind stream, given as they are located, steel building vs concrete
N building (Deepak, 2008) as per data and
E (8) guidelines given in IS 875 (Part 3) – 1987, Prem
1  70 N 2 5/ 6
Krishna (2002) & AS 1170.2 – 1989. The salient
where,

results from the above studies are presented and values (noted at the tip of each building) are
discussed here. gradually decreasing (unlike present code) from
Variation of BESWL and RESWL TC1 through TC4.
Fig. 2 shows a typical variation of background
and resonant factor along the heights of a 32 ESWL values from IS 875 (Part 3), Australian
storeyed steel framed building using the methods AS 1170.2 -1989 & Prem Krishna (2002) are
of Prem Krishna (2002) and present code. Back aslso compared in Fig. 7. Several similar graphs
ground factor is much higher than resonant are generated in the present study for different
component for this building. Also, the values of aspect ratio of building and varying total height.
these parameters are higher in case of Prem On the basis of these Figures, it is concluded that
Krishna (2002) and increases with height unlike the ESWL values calculated as per IS 875 (part
the IS 875 (part 3) values, which are constant 3) are most conservative.
along the height.
CONCLUSIONS
Effect of plan aspect ratio Based on the present study on regarding
Fig. 3 presents the resonant and background variation of GEF, it can concluded that
components B and R calculated as per Prem 1. As total height of a building increases gust
Krishna (2002) with varying plan aspect ratio. factor value at the tip of the building (i.e.
The width of building along the wind flow is
maximum GEF for that building) increases.
kept fixed at 24 m and width normal to flow is 2. Using present code, GEF remains constant
increased from 12 m to 48 m. It can be seen that throughout height of the building for any
resonant factor is maximum for a square particular terrain category.
building, thus it will result into highest GEF
3. Using present code, values GEF (which is
value. Eimani-Kalehsar et al. (2006) also
constant throughout the height) obtained for
concluded that buildings square in plan are
varying total height of the building shows a
subjected to more gust induced forces, and
slightly decreasing trend for all categories.
across wind forces in such cases may be more 4. Buildings square in plan provides maximum
than long wind forces. value of background response.
5. GEF value is higher in case of steel building
Effect of height of building when compared with similar reinforced
It is seen that in 60 storey building resonant
concrete which is due to low damping
response is greater than background response,
coefficient.
contrary to 22 storey RC building. In other
words, for other parameters remaining identical,
rigidity of structural system decreases The recent research trend in gust effective factor
exponentially with height (Vihwal and Mandal, may look into modeling of fatigue effect of tall
2004). Also resonant response is directly structures, shielding/ interference effect, across
increasing with total height (resulting from wind GEF for buildings symmetric on plan.
higher flexibility of the structure) of a building
leading to higher value of resonant component
compared to background response. Fig. 4 shows REFERENCES
the variation of GEF for terrain category 1 with AS/NZ1170.2 (2002) Australian/New Zealand
varying height. As expected, GEF is maximum Standard, Structural design actions, Part 2:
for the tallest building. The study of GEF values wind actions, Standards Australia & Standards
for steel framed vs RC framed buildings showed New Zealand.
that the steel frame building have higher GEF Bhandari, N. M., Kumar, K., and Krishna, P.
values. This is so mainly due to the higher (2006). “IS: 875(Part3): Wind Loads on
damping coefficient (β) of RC structures than Buildings and Structures -Proposed Draft &
that of steel structures. Commentary.” IITK-GSDMA-Wind02-V5.0,
Dept. of Civil Engg., IIT-Roorkee, Roorkee.
Davenport, A. G. (1985). “The representation of
Effect of terrain category the dynamics effects of turbulent wind by
In the present code, GEF values are gradually equivalent static effects of turbulent wind by
increasing when terrain category changes from equivalent static wind loads.” Engineering
TC1 through TC4 (refer Figs. 5 & 6) according Structures, 16, 287-292.
to Prem Krishna (2002) & IS 875 (Part 3) – 1987
respectively. Using Prem Krishna (2002), GEF

Deepak Kumar (2008). Estimation of equivalent 2004 Seoul Conference”, Oct 10-13, Seoul,
static wind load on tall structures. M. Tech. Korea, II, 1080-1087.
Thesis, Dept. of Civil Engineering, IT BHU, Mandal, S., Ojha, C. S. P., and Bhargava, P.
Varanasi, India. (2001). “Significance of correlation coefficient
Deepak Kumar, Mandal, S. and Kumar, V. in evaluating Reynolds Stresses.” J. Wind
(2007). “Some Aspects in Estimation of Engineering, Japanese Association for Wind
Equivalent Static Wind Load on Tall Engineering, 89(10), 317-320.
Buildings.” Proc. Fourth National Conference Mandal, S., Ojha, C. S. P., and Bhargava, P.
on Wind Engineering, SERC Chennai, 30 Oct. (2005). “Wind Turbulence modeling at near
– 1st Nov., 373 – 380. wall zone using k- model: A review.” J.
Eimani-Kalehsar, H., Kumar, K., and Godbole, Wind & Engineering, Indian Society for Wind
P. N. (2006). “Acrosswind loading of tall Engineering, Vol. 2, 52-59.
rectangular buildings.” J. Wind & Engineering, Ojha, C. S. P., Mandal, S., and Bhargava, P.
ISWE, Vol. 3, 32-37. (2001). “Aspects of Inlet boundary prescription
Holmes, J. D. (2002). “Effective Static Load for a turbulent flow field.” J. Hydraulic
Distribution in Wind Engineering.” Journal of Engineering, ASCE, 127(8), 694-700.
Wind Engineering and Industrial aerodynamic, Prem Krishna (2002) “Proposed Modification to
90, 91-109. IS:875 (Part 3)-1987.” Distributed during
Huang, G., and Chen, X. (2007). “Wind load National Conference on Wind Engineering, IIT
effects and equivalent Static wind loads of tall Roorkee.
building based on Synchronous pressure Vihwal, V. and Mandal, S. (2004). “A
measurements.” Engineering Structures, 29, Parametric Study on frame Shear Wall
2641-2653. Interaction under Wind Load.”, Proceedings of
IS: 875 (part 3)-1987. Indian standard code of Second National Conference on Wind
practice for design loads (other than Engineering (NCWE- 04), Nagpur, India.
earthquake) for buildings and structures, part Zhou Y., Ming G., and Xiang H. (1999). “Along
3-wind loads. Bureau of Indian Standards, New wind static equivalent wind loads and response
Delhi, 1989. of tall buildings. Part I: Unfavorable
Kim, D. Y., Kim, J. Y., Lee, M. H., & Kim, S. distributions of static equivalent wind loads.”
D., (2004). “A study on the Evaluation Journal of Wind Engineering and Industrial
Methods of Equivalent Design Wind Loads for Aerodynamics, 79,135-150.
tall buildings.” Proceeding of the CTBUH
(a) b = 12 m, a = 24 m (b) b = 24 m, a = 24 m
(c) b = 36 m, a = 24 m (d) b = 48 m, a = 24 m
Fig. 1a : Plan dimensions of the building

analyzed
Fig. 2 : Background and Resonant responses for

32 SSB
Fig. 1b : Framing plan of the example building

70
B a c kg ro u n d
60
R eson an t
50
40
Height (m)
12B
30 12R
24B
24R
20
36B
36R
10 48B
48R
0
0 1 2 3 4 5 6 7 8 9 10
F a cto r
Fig. 6 : GEF for RC building as per IS 875 (Part

Fig. 3 : Resonant & background responses for 22 3)-1987
SRCB (Prem Krishna, 2002)
T e r r a i n C a te g o r y 1
1 .0 5
1 .0 0
12 S to re y
Gust Factor
17 S to re y
22 S to re y
0 .9 5
27 S to re y
32 S to re y
37 S to re y
42 S to re y
0 .9 0
60 S to re y
0 .8 5
0 20 40 60 80 100 120 14 0 160 180
Fig. 7 : ESWL from IS 875, Australian code
H e ig h t ( m )
AS 1170.2 -1989 & Prem Krishna (2002)
Fig. 4 : GEF for RC building in TC1
Fig. 5 : GEF in different Terrain Category for a

60 SRCB (Prem Krishna, 2002)

ISBN 978-81-921121-0-7
ANALYSIS OF SKEW SLAB DECK USING ANALYTICAL METHODS

Rajesh Kumar
Associate Professor, Department of Civil Engg., I.T. BHU, Varanasi
rkumar.civ@itbhu.ac.in
Veerendra Kumar
Professor, Department of Civil Engg., I.T. BHU, Varanasi
Bikram Kesharee Patra
M.Tech. Student, Department of Civil Engg., I.T. BHU, Varanasi
ABSTRACT: Bridges and culverts form important parts of a rail or road or any other type of communication
network and the major part of the project of the cost of the project goes into the construction of these structures. In
order to cater to high speeds and more safety requirements of the traffic, modern highways are to be straight as far as
possible and this has required the provision of increasing number of skew bridges. In brief the study includes the
behavior of skew slab bridges with respect to support reaction and deflection of the slab under standard IRC 70R
wheeled loading. The modeling of the slab is done by using both Finite element method (FEM) and Grillage method
and results are compared.
Key words: Bridge, Skew, Concrete slabs, Finite element method, Grillage analogy method, Highway, slab culvert.
INTRODUCTION
Considerable amount of research work has been
reported in the literature on the analysis of skew
bridges under different load combinations. However,
less attention has been paid for exploration of
suitable analysis technique for highly skewed decks.
Since bridges are an important structural element
today, efficient methodology to accurately predict the
behavior of skew bridges assume importance. Fig.1 Different components of skew bridge
Bridges and culverts form important parts of a rail or
road or any other type of communication network REVIEW OF PREVIOUS WORK
and the major part of the project of the cost of the A lot of research works have already been done on
project goes into the construction of these structures. the behavior of skew slabs [Victor (1991)
The ideal bridge crossing of any obstacle is a square Raina(1990)]. Kennedy (1983) presented an
crossing, which ensures minimum span, deck area experimental study on the structural response of
and support pier lengths, with attendant economies reinforced concrete skew bridges of (a) orthogonally,
and it also the easiest structure to design and detail. (b) non orthogonally shaped waffle slab construction
In order to cater to high speeds and more safety and (c) of solid slab construction .The response is
requirements of the traffic, modern highways are to examined under elastic and ultimate load condition.
be straight as far as possible and this has required the Bakht (1988) analysed some skew bridges as right
provision of increasing number of skew bridges. The bridges for angle of skew less than 20°, It is shown
inclination of the centre line of traffic to the normal that the errors in analyzing skew slab on girder
to the centre line of the river in case of a river bridge bridges as right bridges are not characterized by the
or other corresponding obstruction is called the skew angle of skew but by two dimensionless parameters,
angle. For bridges in which the plan form is a which depend upon the angle of skew, the spacing
parallelogram is shown in Fig. 1, the angle obtained and span of girders, and their flexural rigidities
by subtracting the acute angle of the parallelogram relative to the flexural rigidity of the deck slab.
from 90º is termed the skew angle of the bridge. Bishara and Elmir (1990) presented a three
dimensional finite element algorithm for computing
internal forces in intermediate and end cross frames

of multi beam composite steel bridges. Bishara and etc. which are of permanent stationary nature. These
Elmir (1990) have shown the interaction between dead loads are customarily considered to be borne by
cross frames and girders of bridges. Girder moments the longitudinal grid members giving rise to
in continuous skew composite bridges have been distributed loads on them. This distributed load is
found by Ebeiodo and Kennedy (1996). The idealized into equivalent nodal loads. If the dead load
developed algorithm was applied to four medium is uniformly distributed load but its centre is non-
span simply supported multi girder tangent four lane coincident with the longitudinal grid line then it is
bridges, With x-type intermediate cross frames and substituted by a vertical udl together with a torsional
end cross frames of the warren truss with vertical udl. The self-weight of cross-beams is given taking
type. The methods which are listed above have not the total weight of all the cross-beams per span and
been consistently explored for high skew angle and equally dividing it in the form of distributed loads to
for different aspect ratio of the slab. Therefore, a various longitudinal members of the grillage. The
consistent and an efficient method are required to dead weight of railings, kerbs, footpaths etc. is
know the behavior of skew deck slab under dead load lumped on the edge longitudinal grid lines. For the
and live load. In present study, Grillage Analogy and analysis using STAAD-V8i, in each case the slab is
Finite Element Method have been chosen for analysis discritized in grid of interconnected beams in case of
and compared to each other. Grillage analogy and grid of interconnected
connected plates in Finite element analysis. For the
CASE STUDY sake of convenience in analysis and also for
A study on the behavior of Skew slab culverts under comparison of results the grid pattern are kept same
standard IRC-70R wheeled loading is presented and
in both methods. For analysis purpose both dead and
it is based on the analytical modeling of slab culverts
by Grillage Analogy and Finite element method. live load considered simultaneously.
Effects of support reactions, deflection for different
spans and skew angle have been studied. The analysis GRID PATTERNS USED FOR MODELING
provides the useful information about the variation of The following grid patterns as shown in figures
support reactions, deflection, with respect to change (Figures 2 to 3) are used in the modeling for different
in skew angle and aspect ratio. The analysis is done slabs with different skew angle and different aspect
for slabs with different spans and skew angles using
ratio. The selection of a particular grid pattern for
STAAD-v8i software.
The clear roadway between the krebs is taken as 7.5 modeling mainly depends upon the availability of
m for important roads to suit two lane traffic. Krebs right and skew region and convergence of the results.
of minimum width 500 mm is desirable. Hence for all
the analysis, clear carriage way distance is kept
constant at 8.5m. To study the effect of aspect ratio
on the behavior of deck slab, different cases have
been studied by taking right span as 3m, 4.5m, 6m,
and 7.5m with skew angle varying from 0-75°. Here
the right span is kept constant but the skew span is
varying. Here the study is carried out for, two lane
slab bridge for 3m,4.5m,6m and7.5m right spans for
0°,15°,30°,45°,60°,75° skew angles. Grillage
Analogy and Finite Element Method have been used Fig. 2 Grid pattern for skew angle 30°, 45°, 60°,
for the analysis of the deck. 75° (With higher right region and lesser skew
region)
APPLICATION OF LOADS ON DECK
The loads consisting of dead load, live load (IRC
70R) and impact load acting on the bridge
superstructure are appropriately distributed to the
nodes of the grillage and FEM. The dead load
comprises of self-weight, wearing coat, parapet, kerb

reaction varies from 40% to 51% i.e. decreases near
acute corner and increase near obtuse angle corner.
Then for 45°-60° skew, the reaction near acute angle
corner starts negative which increase with increase in
skew angle and aspect ratio and there is heavy
reaction built up at the node near to obtuse angle
corner. As soon as the skew angle reaches 75° there
is a heavy negative reaction built up near the obtuse
angle corner due high amount of torsion and the joint
becomes unstable. The variation in the results of
FEM and grillage method is 20% to 30%. This is due
Fig. 3 Grid pattern for skew angle 30°, 45°, 60°, to the different approaches of analysis in both the
75° (With lesser right region and higher skew methods.
region)
SUPPORT CONDITION AND SLAB USED FOR

ANALYSIS
To simulate the actual field condition of slab
abutment joint with neoprene rubber bearing, it is
assumed that one edge of the slab has been supported
on hinge and another edge on the roller support. Thus
the deck slab is analyzed as simply supported on the
two edges. There are four different types of slab used
in the analysis procedure which are shown in table 1,
notation like F.P1, G.P1. and F.P115,G.P115 ,
F.P230 the first suffix indicates the Method used i.e. Fig. 4 Reaction curve for slab, having skew 0°-75°,
‘F’ for FEM and ‘G’ for Grillage, and the middle one due to edge loading (Combined Grillage & F.E.M)
indicates P1,P2 indicates the plate size, and last
Reaction at supports due to central loading:
suffix indicates angle of skew. For right slab third
For right slab the reaction found under the wheel is
suffix is employed.
highest and reduces gradually towards both sides of
Table 1 Four different type of slab
centre. For slabs with skew 15° to 30° (Fig.5), the
Right Cross Slab Notation used maximum and minimum support reaction varies from
section (m) thickness(mm) in Graph 40% to 51% i.e. decreases near acute corner and
8.5x3.0 400 P1 increase near obtuse angle corner. Then for 45° to
8.5x4.5 500 P2 60°, the reaction near acute angle corner starts
8.5x6.0 600 P3 negative which increase with increase in skew angle
and aspect ratio and there is heavy reaction built up at
8.5x7.5 700 P4 the node near to obtuse angle corner. As soon as the
skew angle reaches 75° there is a heavy negative
reaction built up near the obtuse angle corner due
high amount of torsion and the joint becomes
Reaction at the Supports due to Edge Loading: unstable. Again the variation in the results of FEM
Reaction at the supports has been found for different and grillage method is 20% to 30%. This is due to
skew angles using both the methods and plotted in the different approaches of analysis in both the
the figures (4). For right slab the reaction found methods
under the wheel is highest and reduces gradually
towards acute angled corner. For Slabs with skew
15°-30°, the maximum and minimum support

Fig. 7 Deflection curve for slab, having skew 0°-
Fig. 5 Reaction curve for slab, having skew 0°-75°, 75°, due to central loading(Combined Grillage &
due to central loading (Combined Grillage & F.E.M)
F.E.M)
Conclusions
Out Plane Deflection due to Edge and Central
1. In analysis of skew slab bridge both
Loading:
grillage and finite element method can
For skew angle 0-15° (Fig. 6, for edge loading), the
be effectively applied. Both methods
maximum deflection point remains at mid location.
are easy to use and versatile.
Maximum deflection is increasing with increase of
2. The results of both methods are
skew angle for different aspect ratio. It has also been
matching at a particular grid
observed that maximum deflection point shifting
spacing(generally 8-12 grid spacing).
away from the mid location. Out of plane deflection
Appropriate modelling of the slabs
for central loading has also been shown in the fig.7.
must be choosen based on aspect ratio
The nature of the graphs is almost same as due edge
and skew angle.
loading. The deflection value in grillage method is
3. Skew effect on support reaction and on
about 20-30% lower than FEM.
deflection is less upto skew angle 30°.
4. The slab becomes unstable at and above
60° under central and edge loading.
Also the varition in deflection and in
support reaction is very high.
5. The behaviour of skew slab is also
governed by aspect ratio, at lower
aspect ratio the skew effect is more as
compared to higher aspect ratio at a
higher skew angle.
6. The skew effect should be taken into
acccount properly for analysis and
Fig. 6 Deflection curve for slab, having skew 0°- design for skew angle ranging from
75°, due to edge loading (Combined Grillage & 30°to 60°.
F.E.M)

REFERENCES BAKHT , B . (1988), “ANALYSIS OF SOME SKEW
VICTOR , D . J ., “ESSENTIALS OF BRIDGE BRIDGES AS RIGHT BRIDGES”, JOURNAL OF

STRUCTURAL ENGINEERING , VOL. 114, NO .10,
ENGINEERING ”, OXFORD & IBH PUBLISHING CO .
PP.2307-2322.
PVT . LTD ., 1991.
BISHARA, A. G ., ELMIR , W. E.(1990), “INTERACTION
RAINA V. K., “CONCRETE BRIDGE PRACTICES”, TMH
BETWEEN CROSS FRAMES AND GIRDERS
PUBLISHING COMPANY LIMITED , 1990.
”, JOURNAL OF STRUCTURAL ENGINEERING, VOL
KENNEDY , J . B (1983), “ORIENTATION OF RIBS IN 116, NO.5, PP.1319-1333
WAFFLE –SLAB SKEW BRIDGES” JOURNAL OF EBEIDO , T ., AND KENNEDY , J . B . (1996), “GIRDER
STRUCTURAL ENGINEERING , VOL. MOMENTS IN CONTINUOUS SKEW COMPOSITE
109, NO.3, PP.811-816 BRIDGES”, JOURNAL OF BRIDGE ENGINEERIN G ,

VOL.1, NO .1, PP 37-45.

ISBN 978-81-921121-0-7
REMOVAL MECHANISM OF Cr (VI) AND Fe (III) METAL IONS FROM

EXPANSIVE SOIL BY ELECTROKINETIC EXTRACTION
Ch. Ramakrishna
Department of Environmental studies, Gitam University, Visakhapatnam -530045
B. S. Nagendra Prakash
Department of Civil Engineering, University Visvesvaraya College of Engineering, Bangalore University Bangalore-560056.
P.V. Sivapullaiah
Professor, Department of Civil Engineering, Indian Institute of Science, Bangalore-560012, Email: siva@civil.iisc.ernet.in
ABSTRACT: Laboratory studies are conducted to understand mechanisms of removal of different heavy metals in
expansive soils containing montmorillonite clay minerals by electrokinetic process while passing water, acetic acid and
EDTA solution with varying voltage. The permeability of soil, which is low, increases in the presence of contaminants.
The permeability of soil increases further due to osmotic permeability. Only Chromium and Iron are removed from
Black Cotton Soil by electrokinetic processes. Their removal rate of chromium is higher by flushing the soil with water.
Electro kinetics assists the removal of Iron by enhanced flow through the soil as well as by desorption of Iron from soil.
INTRODUCTION Black Cotton Soil

Contamination of subsurface and ground water Black Cotton Soil used in this study was obtained from
subsequently occurs due to leachate migration from Davanagere, Karnataka state, India. The soil was
landfills, industrial activities and other sources. Soils collected by open excavation from a depth of one meter
often get contaminated with heavy metals due to improper from the natural ground. The soil was dried and passed
disposal of sludges and waste water, from rinsing through the IS sieve size of 425 microns. The soil so
operations at electro plating and metal finishing plants. obtained has clay content of 35%. The clay content
Decontamination of fine-grained soils from ionic consisted predominantly of Montmorillonite mineral. The
pollutants by conventional methods is not only costly but cation exchange capacity of the black cotton soil was
also mostly ineffective because of their low hydraulic about 29.0 meq/100gms. Compaction characteristics
conductivity, low biodiversity and strong adsorption on required for remoulding soil are also determined.
their surface (Li & Li, 2000). The demand for innovative Properties of Soil are summarized in Table 1.
and cost effective in situ remediation stimulated the effort
to employ conduction phenomena in soils under an Table 1 Properties of Soils used
electrical field to remove chemical species from soils Property B C Soil
(Sellers, 1998). This method called Electrokinetic
Specific Gravity 2.74
Remediation increases flow through soils and desorbs
ions from soil surfaces (Acar & Alshawabkeh, 1993). The Liquid Limit (%) 56
method seems to have met with varying degree of success Plastic Limit (%) 23
depending on contaminants from soils containing Shrinkage Limit (%) 10.34
different clay minerals. Several factors such as CEC of
soil, adsorption capacity of contaminant ion ectc., Clay Content (%) 35
responsible for this. The ions, which are sorbed strongly Max Dry Density (kN/m3) 15.1
on the clay particles, are not removed by enhanced flow Optimum Moisture Content (%) 27.76
alone but by the capacity of applied voltage to desorb and
Cation Exchange Capacity,
make them mobile. Also the ions which form stable 29
(meq/100g)
hydroxides are not removed unless the pH is efficiently
controlled. The rate of neutralisation of alkalinity by
acetic acid depends on the rate of passing the solution. Chemicals Used
The removal of ions from soil by electro kinetic The chemicals, Potassium dichromate (K2Cr2O7), ,
extraction method is by i) enhanced flow of water by Ferric chloride (FeCl3) used to prepare contaminated soils
electrolysis of water and ii) desorption of metal ions form with Chromium, and Ferric ions, were obtained from
the surface of the clay particles by ionic migration standard suppliers.
(Shapiro & Probstein 1993; Pamukcu & White, 1992). It
is proposed to examine the relative importance of these
processes for their removal. The above aspects for soil ELECTRO-KINETIC APPARATUS
containing heavy metals viz., chromium, nickel, An apparatus designed, fabricated and assembled for
cadmium, lead, zinc and iron as individual contaminating carrying out the studies. The Electro-kinetic cell was
ions are studied. fabricated using nylon material for the body, which is a
EXPERIMENTAL INVESTIGATIONS non-conductor of electricity, corrosive resistant, not

affected by acid or alkali and can withstand a pressure of bring it to 90% of maximum dry density on dry of
100 kPa. The electrokinetic cell consists of two end caps optimum on the compaction curve. The soil is divided
and a specimen cylinder made up of nylon. The test into three equal parts by weight and then each part is
sample was 80mm in diameter and 300 mm long. Ten emplaced into the specimen cylinder and compacted one
electrical measurement nodes have been installed on the by one using a screw jack to ensure uniform compaction
specimen cylinder at 25 mm intervals so that the electrical for the entire specimen.
voltage distribution along the sample can be monitored After compaction, the perforated electrodes were
continuously during the test. The end caps house the covered with the filter papers and the end caps were
graphite plate electrodes, inflow and outflow tubings. closed. Fluid in-let and out let tubes was connected to the
cell. TA low constant hydraulic head has been applied
using the self-compensating mercury control device and
the sample saturated.
Sequential Extraction of Contaminants

The trends of removal of contaminants under different
processes considered are:
Flushing with water

When water flows under hydraulic gradient, soluble
contaminant is removed along with hydraulic flow. Since
hydraulic gradient is applied from anode to cathode
leachate containing soluble contaminant is collected at the
cathode.
Removal by enhanced flow due to electro-osmosis and

ion migration
To enhance the flow of water and to make the
contaminant desorb from the surface of clay, DC voltage
is applied across the soil. The cumulative volume of flow
and the concentration of contaminants in the leachate are
monitored. The increase in pH due to electrolysis of water
is also monitored and plotted.
By catholytic neutralisation and removal of

contaminants
The increase in pH at the cathode due to electrolysis of
Fig 1 Electro Kinetic cell
water precipitates the metallic cationic contaminant at
cathode leading to reduction in flow and removal of
As porous graphite is extremely expensive and fabrication
contaminants from the soil. Acetic acid is passed along
of porous graphite electrodes is very labour intensive,
with water to neutralize the alkalinity. The enhanced or
normal grade graphite plates were used in this study.
continued removal of contaminant after passing acetic
Holes of 1mm dia were drilled through the graphite plate
acid is monitored (Rodsand et. al 1995).
electrode to facilitate water transport during electro-
osmosis. Details of electrokinetic cell are shown in Fig.1.
Electro-osmosis while passing acetic acid to enhance
The required electrical circuit has been designed,
flow and removal
connected and assembled. The circuit is designed to
To further enhance the extraction of the contaminant
facilitate future automation of the experiment by means of
electrical potential has been applied across the specimen
computerized data acquisition system. Provision has been
while passing acetic acid under 1Volt/cm and 2 Volts/cm.
made to pass water through the soil compacted in the cell.
The hydraulic pressure of inlet is controlled
Use of complexing agent to enhance removal
independently by “Self Compensating Mercury Control
It is known that EDTA, which is a good chelating agent
"device (Page & Page, 2002).
and can form complexes with many metallic ions under
suitable pH condition. Thus 0.03 M solution of EDTA
solution was passed to desorb ions from clay surface by
Sample Preparation
its ability to form complexes and carry along with
The oven dried soils were mixed with water containing
hydraulic flow (Ugaz et.al 1992).
specific known amount of chemical contaminants
Electro-osmosis while passing EDTA solution to
thoroughly to ensure homogeneous distribution of
enhance flow and removal
contaminants and kept in a polythene bag and placed in a
To further enhance the extraction of the contaminant, use
humid desiccator overnight to achieve uniform moisture
of electrical potential across the specimen while passing
content. The soil is then compacted in the cylinder to

EDTA solution has also been adopted under 1 Volt/cm
and 2 Volts/cm.
Electrokinetic flow Flow through Black Cotton Soil without any

In electro kinetic flow, the flow is driven by both by Contaminant
hydraulic gradient and electric gradient occurs. Attempt
has been made to separate and calculate the flow due to The rate of flow of water through Black Cotton Soil
electric gradient for different types of clays. without any contaminant is about 0.7 ml/hr.
Electro-osmotic fluid volume flow rate is described by an Considerable increase in the rate of flow occurs in on
equation analogous to Darcy’s law application of voltage of 30V. The rate of flow through
soil is tripled to 2.1 ml/hr. The increased flow is
Q = ke Ie A Eq. (1) predominantly due to osmotic flow induced by ion
migration. For Black Cotton Soil with high cation
where Q = increased fluid volume rate on application of exchange capacity the effect of ion migration is
voltage (m3/s); considerable and hence the increased flow. Increasing
Ke = coefficient of electro-osmotic conductivity the voltage to 60V, however, decreases the rate of flow
(m2/ V-s); to 0.9 ml/hr, as seen from Table 2 and Fig. 2.
Ie = electric field strength (V/m);
A = total cross sectional area perpendicular to the Table 2 Hydraulic Permeability and Osmotic
direction of fluid flow. Permeability of Black Cotton soil with water
Osmotic Permeability, Ko
RESULTS AND DISCUSSION Hydraulic cm2/V-s.
Contaminant Conductivity,
To understand the rate of removal of contaminants it is kh cm/s. 30 V 60V
necessary to know the
Variation of Flow through Soil with and without NIL 1.24E-07 7.69E-06 2.18E-06
Contaminants during Different Phases of electrokinetic
extraction as brought out earlier.
Cr(VI) 2.41E-07 -4.60E-06 5.08E-05
VARIATION OF FLOW THROUGH BLACK
COTTON SOIL Fe(III) 3.30E-07 -2.50E-05 2.30E-04
The importance of rate of flow of flow by electro kinetic
extraction has been very clearly brought out. The flow
through soils varies considerably depending on many
factors like the clay mineral present, particle size
distribution, porosity, soil fabric and the nature of pore
fluid etc. As brought in previous chapters, the rate of flow
to any specific fluid can vary depending on the type of
contaminant present. How the rate of flow through Black
Cotton Soil varies with different contaminants and fluids
has been studied in this chapter. For the sake of
comparing the effect of passing fluids it is necessary to
know the rate of flow rate to water not only for Black
Cotton Soil without contaminants but also for soil with
different contaminants (Yeung,1994). As explained the
fluids that are employed in this other than water are acetic Fig 2 Variation of effluent flow with cumulative time in
acid and EDTA solutions. Then for each of these fluids Black Cotton soil (without pollutant)
the effect of employing voltage on the rate of flow has
been studied.
The effect of applied voltage on flow rate while The decrease in flow is due to increased pH as seen in
passing water is monitored. Since acetic acid is normally Fig. 3. To reduce the alkalinity at the cathode and
used to remove the effect of precipitation, the effect of accelerate the flow, acetic acid is passed through the soil.
applied voltage is studied while passing acetic acid. With As seen from Table 2 and Fig. 2, Flow through the soil
a view to enhance decontamination by passing EDTA to acetic acid, which is almost stopped slowly, observed
different levels of voltage across the soil has been applied. at rate of about 0.8 ml/hr on application of 30V. With
The effect of voltage while passing EDTA has also been application of voltage (60V) the rate of flow decreases to
investigated 0.5 ml/hr. To enhance the removal of contaminants
The above studies are conducted on Black EDTA solution is passed through the soil. Even though
Cotton Soil with Chromium, Nickel, Cadmium, Lead, passing of EDTA solution is not expected to remove any
Zinc and Iron as individual contaminating ions contaminants with soil alone, to enable comparisons
later while conducting experiments with the soil

containing contaminants EDTA is passed with and Fig 4 Variation of effluent (Cr) pH with cumulative time
without current. The rate of flow, which reduces to 0.2 in Black Cotton soil
ml/hr. to EDTA solution, increases on application of
voltage to 1.2 ml/hr. This possibly is the counteraction of other ions over that
of Chromium. The increase in pH as seen in Fig. 5 after
applying 60 V cannot precipitate Chromium (VI) and
hence cannot decrease the flow. For the sake of
uniformity acetic acid is passed through the soil to reduce
the pH at the cathode. It is to be observed from Table 2
and Fig. 4 that passing of acetic acid without voltage the
flow rate is considerably reduced to about 2 ml/hr.
Application of voltage decreases the rate of flow. The rate
of flow to EDTA solution without current is less at about
0.3 ml/hr. which, increases on application of voltage.
Fig 3 Variation of effluent pH cumulative in Black Cotton

soil (without pollutant)
Flow through Black Cotton Soil with Chromium ions

To maintain uniformity, hydraulic flow is maintained
towards the cathode in the Black Cotton Soil containing
Cr (VI) ions also. But ion migration under the influence
of voltage is towards the anode. Though not very Fig 5 Variation of effluent (Cr) flow with cumulative time
efficiently, Cr (VI) may move along with water in the in Black Cotton soil
direction of cathode. The effect of these processes namely
ionic movement and other flows being in the opposite Flow through Black Cotton Soil with Iron ions
direction only the overall flow rate and osmotic flow have Presence of Iron ions has increased the rate of flow in soil
been considered. to about 2 ml/hr, as seen Table 2. Application of voltage
The rate of flow of water through Black Cotton of 30V increases the rate of flow considerably to 13 ml/hr
Soil containing Chromium is higher at 1.5 ml/hr. under inspite of increase in pH (Fig. 6). Thus for Black Cotton
application of constant hydraulic pressure of 100 kPa. It is Soil, application of voltage is more effective to enhance
well known that in the presence of higher valency ions, the flow than replacement of exchangeable ions. However
cation exchange takes place Leading to suppression of with increase in voltage (60V) there is a reduction in the
diffused double layer of clay particles. This in turn flow rate though pH is almost unaffected as seen from the
increases the rate of flow through the soil. Application of Table 2 and Fig. 7.
voltage (30V) decreases the flow rate. This can be
attributed to opposite directions of flow of water due to
the influences of water hydraulic head and Chromium ion
migration. It has been earlier brought that Chromium ions
in Oxidation State VI are stable as anion and hence move
towards the anode. The decrease in flow on application of
voltage indicates that the ion still carries hydrated water
with it. Increase in voltage (60V) increases the rate of
flow considerably, as seen from Table 2 and Fig 4.
Fig 6 Variation of effluent (Fe) flow with cumulative time

in Black Cotton soil
Passing of acetic acid solution increases the rate of flow

to 12.7 ml/hr. This indicates that possible dissolution of
Iron precipitated as hydroxide during passing current

earlier. As will be seen later Iron is removed by passing 2.2 X 10-6 cm2/V-s to water; 5.2 X 10-6 cm2/V-s to acetic
acetic acid. Application of voltage of 30 V reduces the acid solution and 1.1 X 10-5 cm2/V-s to EDTA solution
rate of flow probably due to precipitation of Iron. With (Table 3). Thus application of higher voltage is increases
increase in voltage there is significant increase in flow to flow to all fluids. While the increase in flow is not
33.5 ml/hr. Passing EDTA solution reduces the rate of proportional to increase in voltage for water it is almost
flow without or with voltage without changing the pH as proportional to acetic acid and is more than proportional
seen from the Table 2. to EDTA solution.
Table 3 Hydraulic Permeability and Osmotic

Permeability of Black Cotton soil with EDTA solution
Hydraulic Osmotic Permeability,

Contaminant Conductivity, Ko cm2/V-s.
kh cm/s. 30 V 60V
NIL 5.24E-06 1.05E-05
3.77E-08
Cr(VI) 2.08E-06 1.42E-05
5.22E-08
Fe(III) 1.12E-06 -1.20E-05 -2.20E-05
Fig 7 Variation of mixed effluent (Fe) pH with Table 4 Hydraulic Permeability and Osmotic
cumulative time in Black Cotton soil Permeability of Black Cotton soil with acetic acid
Variation of Hydraulic Permeability and Osmotic Osmotic Permeability,
Permeability of Black Cotton Soil without Hydraulic
Contaminant Conductivity, Ko cm2/V-s.
Contaminant 30 V 60V
The hydraulic conductivity of Black Cotton Soil without kh cm/s.
contaminant is in the range of 1.2 X 10-7 cm/s to water, NIL 0 4.61E-06 5.20E-06
and 3.8 X 10-8 cm/s to EDTA solution. No flow through
the soil could be observed for long time while acetic acid Cr(VI) 3.25E-07 -2.30E-06 -1.70E-05
solution was allowed to enter from the reservoir under a
pressure 100 kPa after stopping passing of water under Fe(III) 2.11E-07 -2.50E-05 8.23E-03
60V. Thus the hydraulic conductivity of Black Cotton
Soil to both acetic acid solution and EDTA solution is
significantly lower than to water. As explained earlier, the
initial condition of the soil while passing water, acetic Variation of Hydraulic Permeability and Osmotic
acid and EDTA solutions is not the same. Acetic acid Permeability of Black Cotton Soil with Chromium
solution is passed after applying voltage and possible The hydraulic conductivity of Black Cotton Soil with
precipitation of hydroxides. EDTA solution is passed after Chromium as contaminant to water is 2.4 X 10-7 cm/s 3.3
passing of acetic acid without and with application of X 10-7 cm/s to acetic acid solution and 5.2 X 10-8 cm/s to
voltage across the soil. Thus before passing EDTA EDTA solution. Thus the hydraulic conductivity of the
solution, the hydroxide formed must have precipitated. soil to water and acetic acid is higher than to EDTA
But as seen from pH of the effluent, often the pH is high solution. It has been observed earlier for soil without
sometimes due to electrolysis during application of contaminants that the flow to water is higher than to
current even while passing acetic acid. While considering acetic acid and EDTA solutions. The effect of reduced
the effects on hydraulic conductivity due to the nature of amount of Chromium by the time fluid is changed might
the fluid, one has to bear in mind these differences in have a predominant effect. It will be observed later that a
initial soil condition (Yeung et. al 1996). large percent of Chromium is removed with passing of
On application of voltage, the flow through the water and acetic acid solution and hence the amount of
soil under the same hydraulic head has increased while Chromium present while passing EDTA solution is less.
passing any fluid. The electro osmotic permeability of The flow through the soil on application of voltage while
Black Cotton Soil at (30V) is about 7.7 X 10-6 cm2/V-s to passing water and acetic acid solution has actually
water; 4.6 X 10-6 cm2/V-s to acetic acid solution and 5.2 reduced. Consequently, the electro osmotic permeability
X 10-6 cm2/V-s to EDTA solution (Table 3 & 4). The has become negative. This may be due to migration of
osmotic flow rates are 1.3 ml/hr, 0.8 ml/hr and 0.9 ml/hr water along with Cr (VI) towards anode rather than
respectively (Table 2). Thus the osmotic flow in Black cathode. With increased voltage to 60V, the osmotic flow
Cotton Soil is much higher than hydraulic flow. The to water was 5.1 X 10-5 cm2/V-s. With increased voltage
hydraulic flow rates as seen earlier were 0.7 ml/hr, 0 and to 60V, the osmotic flow to acetic has become negative.
0.2 ml/hr to water, acetic acid and EDTA solutions. The Thus the flow which is less at 30V in water has increased
osmotic permeability of soil under application 60V was on increasing voltage but unable to increase in acetic acid

solution on increasing voltage. Application of voltage of water under a constant hydraulic head was monitored and
30V while passing of EDTA solution has increased the the results are presented in Fig. 8. It can be seen from the
rate of flow and the osmotic permeability is 2.1 X 10-6 figure that about 18 % of Chromium VI present in soil is
cm2/V-s. Though osmotic flow is observed at 60V, it is removed by passing 0.15 pore volumes of water. The rate
not proportional to applied voltage. of removal has not enhanced on application of 30V of DC
current. About 12% of Chromium is removed by passing
Variation of Hydraulic Permeability and Osmotic another 0.15 pore volumes of water. It has been noted
Permeability of Black Cotton Soil with Iron earlier that the rate of flow in Black Cotton Soil
The hydraulic conductivity of Black Cotton Soil with Iron containing Chromium has enhanced on application of
as contaminant is in about 3.3 X 10-7 cm/s with water, 2.1 voltage. Thus the rate of removal with respect to time is
X 10-6 cm/s with acetic acid solution and 1.1 X 10-6 cm/s enhanced. At higher voltage the removal of Chromium is
with EDTA solution. Thus the hydraulic conductivity of not observed. Passing acetic acid solution without or with
Black Cotton Soil has increased with the acetic acid and voltage is not able to revive the removal of Chromium
EDTA solutions. (Reddy & Parupudi ,1997). Passing EDTA solution
Application of voltage (30V or 60V) increases without or with voltage is also unable to remove
the flow of water through the soil. The osmotic Chromium even at a slower rate.
permeability values are 6 X 10-5 cm2/V-s with 30 volts
and 8.2 X 10-5 cm2/V-s with 60 volts. Thus the osmotic
permeability increases with increase in voltage. The
osmotic permeability in the soil with Iron to acetic acid
solution is observed only with the application of 60V
only, which is about 2.3 X 10-4 cm2/V-s, which is very
high. This may be due to dissolution of Iron hydroxide
while passing acetic acid. As will be seen in next section
Iron is removed from the soil by passing acetic acid.
However with EDTA solution there is no osmotic
permeability
ELECTRO-KINETIC EXTRACTION OF METAL

IONS FROM BLACK COTTON SOIL
It has been shown that the efficiency of removal of ions

from soils by electro kinetic extraction method makes use Fig 8 Chromium removed with cumulative pore volume
of from Black Cotton soil
i) enhanced rate of flow of water due to electro
osmosis, The removal trends of Chromium from Black Cotton Soil
ii) desorption of metal ions from the surface of the shows that it is best by soil flushing. Application of lower
clay particles by ionic migration, and Voltage can remove Chromium at a faster rate.
iii) use of enhancing agents such as EDTA to
complex with contaminating ion and remove
them, Electro Kinetic Extraction of Iron
varies greatly depending on contaminant ions and applied The only success of electro kinetic extraction method for
voltage etc. The effects of ion migration and osmosis on the removal of heavy metals from Black Cotton Soil,
the rate of flow through Black Cotton Soil with different other than Chromium, is Iron. The Iron removed with
contaminants have already been presented. The effect of cumulative pore volume of water after continuous
precipitation of ions as hydroxide and the role of acetic washing with water passed under a constant hydraulic
acid to neutralize the alkalinity on the flow rate have been head was monitored and shown in the initial portion of
brought out. To efficiently apply this method to remove Fig 9. Only a small amount of Iron of about 5% is flushed
ions from soils, it is necessary to know the trends of out without application of voltage by passing about 0.5
removal by passing water with application of current pore volumes of water. Passing about 0.5 pore volume
under a given hydraulic head and the role of other fluids water under 30V could not remove any Iron. But about
to enhance the rate of removal (Reddy & Chintamreddy, 20% of Iron could be extracted by passing of about 0.5
1999). pore volume of water under 60V. This indicates that the
ionic mobility of Iron is less. By passing 0.5 pore volume
Electro Kinetic Extraction of Chromium VI of acetic acid another 8% of Iron could be removed. Thus
The efficiency of the Electro kinetic extraction method to the rate of removal of Iron is not high with acetic acid.
remove Chromium VI from Black Cotton Soil containing With 0.5 pore volume of acetic acid solution another 5 %
known amount of Chromium VI has been tested. of Iron is removed under application of 30V. Thus again
Initially, Chromium removed with number of cumulative the rate of removal is not good. Increased voltage has not
pore volumes of water while continuous passing with further effected any removal. Passing EDTA solution is
able to remove Iron from Black Cotton Soil even without

application of voltage. Application of Voltage has not contaminated soils under reducing environments,
enhanced the rate of removal while passing EDTA Waste Management, Vol.19, pp.269-282.
solution as seen from Fig 9. Reddy, K.R. and Parupudi, U.S. (1997) Removal of
Chromium, Nickel and Cadmium from Clays by In-
Situ Electrokinetic Remediation, Journal of Soil
Contamination, Vol.64, pp.391-407.
Rodsand, T., Acar, Y.B. and Breedveld, G. (1995)
Electro-kinetic extraction of lead from spiked
Norwegian marine clay, Characterization,
containment, remediation, and performance in
environmental geo-techniques, Geotechnical Special
Publication No.46, ASCE, pp.1518 - 1534.
Shapiro, A.P. and Probstein, R.F. (1993) Removal of
contaminants from saturated clay by electro-
osomosis, Environmental Science and Technology,
Vol.27, pp.283-291.
Ugaz, A., Puppala, S., Gale, R.J. and Acar, Y.B.
(1992) Complicating features of electro-kinetic
Fig 9 Iron removed with cumulative pore volume from remediation of soils and slurries: saturation effects
Black Cotton soil and the role of the cathode electrolysis, Department
of Chemistry and Civil Engineering, Louisiana State
CONCLUSIONS University, 31p.
Yeung, A.T., Hsu, C. and Menon, R.M. (1996)
Based on the results presented in this paper, the following EDTA-enhanced electro-kinetic extraction of lead,
conclusions are drawn:. Journal Geotechnical Engineering, ASCE, Vol.122,
No.8, pp.666-673.
2. Osmotic permeability of soil to water on application of Yeung, A. (1994) Effects of electro-kinetic coupling
30V of voltage is observed with all the contaminants on the measurement of hydraulic conductivity,
other than Chromium. Hydraulic Conductivity and Waste Contaminant
The osmotic permeability with contaminants vary in the Transport in Soil, ASTM STP 1142, Eds. Daniel, D.
order E. and Trautwein. P., pp.569-585.
Zn 2+  Fe 2+  Cd 2+  Pb2+  Ni2+
4. Both Chromium and Iron are removed from Black
Cotton Soil. Their removal rate is higher by flushing the
soil with water.
5. Electro kinetics assists the removal of Iron by enhanced
flow through the soil as well as by desorption of Iron
from soil.
REFERENCES
Acar, Y.B. and Alshawabkeh, A.N. (1993) Principles

of electro-kinetic remediation, Environmental
Science and Technology, Vol.27, pp.2638-2647.
Sellers, K. (1998) Fundamentals of hazardous waste
site remediation, Lewis Publishers, Boca Raton.
Li, Y.L. and Li, R.L. (2000) Enhancement of
electrokinetic extraction from lead- spiked soils,
Journal of Environmental Engineering, ASCE,
Vol.126, pp.849-857.
Page, M.M. and Page, C.L. (2002)
Electroremediation of contaminated soils, Journal of
Environmental Engineering, ASCE, Vol.128, No.3,
pp. 208-219.
Pamukcu, S. and White, J.K. (1992) Electro-kinetic
removal of selected heavy metals from soil, Journal
of Hazardous Materials, Vol.55, pp.203-220.
Reddy, K.R. and Chinthamreddy, S. (1999)
Electrokinetic remediation of heavy metal-

ISBN 978-81-921121-0-7
APPLICATION OF ISOTOPIC TECHNIQUES IN HYDROLOGY OF

MOUNTAINOUS REGION
S. P. Rai
Scientist E1, National Institute of Hydrology, Roorkee-247667, Email:spr@nih.ernet.in
Bhishn Kumar
Scientist F and Head, National Institute of Hydrology, Roorkee-247667, Email:bk@nih.ernet.in
Y.S. Rawat
S.R.F., National Institute of Hydrology, Roorkee- 247 667, Email: yadhvir2007@rediffmail.com
P. Purushothaman
Research Associate, National Institute of Hydrology, Roorkee- 247 667, Email: purus26@gmail.com
ABSTRACT: A large part of earth surface is covered by mountains and are rich in water sources. There is a lack of
detailed information about the hydrology of mountains due to the inaccessible and difficult terrain. The application of
conventional methods cannot be implemented. The usage of advanced techniques, which are simple and easier to use, will
help in understanding the hydrology of mountainous region. Isotopic tracer methods is one such can be used in the
mountainous and terrains where usage of conventional method is difficult. In this article we present case studied of some
of the isotopic studies which were used to address various hydrological issues in the Himalayas range. The article gives a
picture on the usage of isotopic techniques in discharge measurement, hydrograph calculation, identification of leakage
and seepage in dams and reservoirs, estimation of water balance in lakes and sedimentation rate of lakes.
INTORDUCTION Therefore, an advance technology such as isotopic

Mountainous region occupies more than 25% of the techniques play major role to explore the hydrological
earth surface (Kapos et al. 2000). The mountains are fact of river catchment, lakes, spring and reservoir
rich is water sources and provide plenty of water to the located in mountainous area. Thus isotopic technique is
society. Their significance, however, is due not only to the viable alternative method is to understand the
the impact of the area they occupy but also to the fact hydrological processes of mountainous region. This
that they are an essential constituent of regional and paper presents few case studies carried out by National
supra regional climate systems. Changes in temperature Institute of Hydrology, Roorkee to solve the field
and precipitation have in recent years led to the retreat problems using isotopic techniques.
of glaciers. Climatic changes do not only affect glaciers
or the nival zone; a change in climatic conditions also STUDIES CARRIED OUT USING ISOTOPE
has an impact on the entire hydrological and biogenous TECHNIQUES
system of mountainous environments (Hermann and Discharge Measurement
Schroder 2004). Most of the major rivers in the world Two streams Lachen Chu and Lachung Chu meet at
originate in the mountains. The Himalayan range act as Chungthang (at the height of 1600 metres) forms the
the source of some of the major river system in the Teesta river. From the point of confluence onwards the
world such as: the Indus, the Ganges and the resulting stream is known as Teesta which is a perennial
Brahmaputra. These three river system receive river with substantial flow even in lean season. It is the
significant contribution from snow and glacier melt major river of Sikkim that flows North-South along the
water. The mountains also host the springs, lakes which state. River Teesta is an ideal source of hydropower
also provide fresh water and play very important role in generation because of enormous fall of the order of
regional and local hydrological cycle. In spite of the 3600 metres over a river stretch of 175 kilometers.
above said importance very less hydrological Although data for high flow flood discharge are very
information of the mountainous river is available. important, the low flow observations are also equally
Extremely difficult terrain, the harsh environment, important for the estimation of firm power draft for
sparse settlements, and the need for a special design of hydel power projects. The float method is mostly used
instruments are major problems in the understanding of for the measurement of discharge by Sikkim
mountain hydrologyproperly and closely. Similarly, Investigation Division (SID) of Central Water
naturally occurring lakes in the Himalayan region are Commission (CWC). Hence the chances of erroneous
being used as a source of water for drinking, irrigation, results of discharge are very high. SID of CWC
pisciculture and recreation. However, the lakes area and therefore proposed National Institute of Hydrology
volume is reducing at alarming rate and water quality (NIH), Roorkee for taking up the work of flow
also degrading rapidly. Likewise, springs are drying up measurement of Teesta river using the radio-isotopes.
in the Himalayan region. Therefore, there is an urgent The study was carried out by Bhabha Atomic Research
need for hydrological studies of lakes/reservoir and Centre (BARC), Mumbai and NIH, Roorkee jointly.
springs etc in mountainous region. Four sites (Chungthang, Sanklang, Dikchu, Sirwani)
Due to unavailability of hydrological data of have been selected in the river Teesta to install the
inaccessible and difficult mountainous terrain of India, microhydels. The trace method was used to calculate the
the application of conventional technique is limited. discharge of the river at these locations using tritium

(3H) and bromine (82Br) and inactive Iodide isotopes. comparison to the pre-event runoff, the mixing ratios of
The tracer was injected at a constant rate using a the two components during that event can be estimated
constant head device and a peristaltic pump. The re- by hydraulic continuity and simple mass balance
producibility of tracer injection rate was better than equations.
0.5% in repeated measurements. In chungthang site 82Br To evaluate the runoff hydrograph, the following
tracer was used to check the mixing length. This was balance equations are used:
followed with 3H and inactive iodide tracers. The Qt = QS + Q0 and tQt = SQS + 0Q0 (1)
tritium and iodide tracer methods were used to xalculate where Qt denotes the total runoff, QS (Q0) the surface
the discharge in the selected locations. In sankalong runoff (base flow) component, and t is derived from the
only tritium tracer was used. In chungthang, dikchu and individual isotope (i) and run-off (Qi) values measured
sirwani both tritium as well as iodide tracers were used. during the event, by the relationship:
The study shows that the isotopic tracer methods
 i Qi
calculate discharge more precisely than the conventional t  (2)
float method. Q i
The hydrograph of Bhagirathi River at Gaumukh site
Table 1 Comparative details of the discharge of the four comprises of multiple peaks (Fig. 1). The variation in
sites in m3/s. discharge occurs due to variations in climatic conditions
Dilution Method which affects the contribution of different components
Site By Float Method
Tritium Iodide to river discharge. To separate out the rainfall
Chungthang 35 28.5 28.2 contribution in Bhagirathi river, the isotopic
Sankalang 70 44.3 composition of river water and observed rainfall were
Dikchu 85 55.2 54 studied. The 18O of river was found isotopically
Sirwani 110 98.0 different during pre and post event of rainfall. The pre-
event river 18O represents the combined signature of
Hydrograph Separation snow and glacier melt while post event river represents
The flow characteristics of head water streams are of mixed signatures of rain, snow and glacier components.
much interest to the hydrologists as they give insight to The significant variation in 18O of river due to south-
the basic hydrologic processes involved. Stable isotopes west monsoon rains was observed occurred during July
are useful in understanding the processes by aiding in and September 2005 (Fig. 1). The isotopic composition
hydrograph separation. Pearce et al. (1986) suggested of pre-event river (-14‰ to -16‰) was found sharply
three main components in lower order stream generation depleted to -20‰ due to much depleted rain of -30.3‰
process in head water catchments. viz. partial overland The proportion of the two components in total discharge
flow, saturation over land flow and subsurface flow. can be separated out using two components model. The
However, in most of the studies, investigators use two water mass balance equation can be written as:
component mixing models. The terms new water and
old water, which are relevant in head water catchment Qt  Qsm  Qr (3)
studies, are generally used to denote the two main Where Q is discharge component and subscripts t, sm
components in the stream flow process. The old water is and r represent total river flow, snow/ice melt and rain
defined as the water that existed in the catchment prior contribution, respectively. Similarly, the isotopic mass
to a given precipitation event or snowmelt, and includes balance equation can be written as
groundwater, soil water and surface detentions. The new  t Qt   sm Qsm   r Qr (4)
water is defined as the water that is added to the stream
as a consequence of a given precipitation or snow melt. Where   [( R sample / R std )  1] * 10 3 ‰
Sklash et al. (1976) and Pearce et al. (1986) reported
By substituting Qr  Qt  Q sm and rearranging
that in humid, temperate climates the old water is the
dominant component. The study on the role of Equation (4), one gets
groundwater in storm runoff by Sklash and Farvolden Qsm  Qt ( t   r ) /( sm   r ) (5)
(1979) revealed that groundwater plays a more Using the Equation (3), the runoff component can be
significant role in the storm and snowmelt runoff separated out.
processes.
The contributions of the surface-runoff to the total 300 0
runoff in small catchment areas can be determined by 250 20
Discharge (m /Sec)
Rainfall
the so-called hydrograph separation, for which 40
Rainfall (mm)
200 Discharge
3
60
hydrochemical and especially isotope data are usefully 150
Rain Contribution
80
employed. The approach is often related to two- 100 100
component separation of total (stream) hydrographs into 50 120
surface/rainfall contribution and subsurface 0
140
(groundwater)/ snowmelt for individual storm events. If

5/1/05
5/15/05
5/29/05
6/12/05
6/26/05
7/10/05
7/24/05
8/7/05
8/21/05
9/4/05
9/18/05
10/2/05
10/16/05
the solute (conservative chemical constituent or isotope)

concentrations of the components are significantly Fig. 1 Hydrograph of river Bhagirathi River at
different during an individual storm event in Gaumukh during ablation period of the year 2005

seepage of the drainage galleries using environmental
In order to separate out rainfall generated runoff isotopes.
component, different events of precipitation were The D and 18O relationship of reservoir and leakage
selected during the years 2004 and 2005. In the month water has been studied on seasonal basis in order to
of August 2004, large variation in isotopic composition identify the source of leakage. Fig. 2 represents the
of river was recorded at the time of rainfall. The relationship between D and 18O for reservoir and
maximum rainfall contribution to the tune of 40% was leakage water for the monsoon and post monsoon
computed for the rain occurred in the month of July, up months. The seasonal variation in isotopic composition
to 30% in case of rainfall occurred in the third week of reveals that isotopic values fall under different
September while up to 40% in case of rainfall occurred categories in different seasons. Reservoir surface water
during the last week of September. The rain is isotopically enriched during summer months and gets
contribution was in the order of 14% to 15% of the total gradually depleted in monsoon period and maximum
river discharge for each storm. the rainfall contribution depletion is observed during the winter months. Based
of all the rain events occurred during ablation period of on isotopic composition, seepage and reservoir water
the year 2005 was computed, which comes out to be an fall under two groups (Fig. 2). During monsoon months
aggregate of 3% of the total discharge of river. the isotopic values (D and 18O) of reservoir and
leakage water fall in the same region. This clearly
IDENTIFYING LEAKAGES FROM DAMS AND indicates that the source of leakage water in the two
RESERVOIRS galleries is reservoir during monsoon months. During
This simple tool is of great practical importance for this period reservoir level rises above 770m.When the
studies focusing on leakages from dams and surface reservoir level raises above 770m a similar variation in
reservoirs. Springs and leakages occurring downstream isotopic composition of reservoir and leakage water is
of dams always present a potential thread for these big observed during the monsoon period. However, a varied
engineering constructions. In this context the question is isotopic composition during summer months is observed
asked whether the given leakage (spring) carries water when the reservoir level falls below 770m. This
from the reservoir behind the dam or whether it is indicates that the fracture zones responsible for leakage
simply a manifestation of the local hydrological system. lie above 770m in the reservoir. During the winter
Identification of the first possibility often means very months, segregation of isotopic values (D and 18O) of
costly engineering measures to detect the leaking part of the reservoir and leakage sites start and isotopic values
the construction. (D and 18O) of reservoir and leakage water falls under
Source Identification of Seepage in Drainage different groups. This is due to the change in isotopic
Galleries of Tehri Dam composition of seepage water as reservoir level declines
During the filling of reservoir, quantum of seepage after the month of October. During the summer months
water through various drainage galleries is being isotopic composition becomes entirely different. Thus,
regularly observed in Tehri Dam. However, in two the D and 18O relationship of reservoir and seepage
drainage galleries (i.e., AGR3 and AGIR) relatively water confirms the connectivity of D5, D6, D7, D10 and
high seepage discharge has been observed during the
D11 water with the reservoir. The isotopic values of D
monsoon period at few locations. Therefore, it becomes
and 18O of D9 always remains in another group which
necessary to identify the source of leakage in the two
clearly reveals that leakage source for D9 is not
drainage galleries of the reservoir. Thus, this study was
connected with the reservoir.
initiated with the objective to identify the source of

δ 18O(‰) Isotope Mass Balance Method
-12 -11 -10 -9 -8 -7
Isotope mass balance of a lake may be written as
-50
Reservoir -55
April-June-08
-60
V=(PPI+SiSI+DiDI +gSSI)-(EEO+SoSO+LSSO) (6)
δD(‰)
D9 & D10 -65
D11 -70
Equation (6) can be rearranged to get sub-surface terms:
-75
-80 gSSI -LSSO=(EEO+SoSO" LV)-(PPI+SiSI+DiDI) (7)
-85
HP Reservior D9 D10 D11 Linear (PPT) Where, SSI, SSO, EO, SO, PI, SI, DI and V, g, Go, E,
So, P, Si, Di and L are the corresponding isotopic
δ 18O(‰)
values. Rearranging equation (7) we get equation (8).
-12 -11 -10 -9 -8 -7 The equation (8) used to determine the sub-surface
-50
Leakage & Reservoir -55
outflow component of the lake, which in turn is used to
D9 -60 estimate groundwater inflow to the lake by the
δD(‰)
-65 following relation given in equation (9).
-70
-75 [G ( SO+EO  V - SI - DI - PI )- (L SO+ E EO   LV - P PI - D DI -  S SI )]
August-September-08 -80 SSO=
(  L - G )
-85
HP Reservior D9 D10 D11 Linear (PPT) (8)
SSI = [(EO + SO -V)-(PI + DI + SI)]+ SSO (9)
δ 18O(‰)
-12 -11 -10 -9 -8 -7 This method of estimation does not require prior
-50
October-November-08 Reservoir -55
estimation of the outflow from the lake through springs
D9 & HP -60 and pumping wells. Isotope mass balance has been
δD(‰)
Seepage -65 attempted for the period between February, 1994 and
-70
-75
February 1995. Since in the month of February, the lake
-80 remains well mixed and homogeneous, it eliminates the
-85 stratification effects on the calculation. The results
HP Reservior D9 D10 D11 Linear (PPT)
indicate that sub-surface components are dominant over
δ 18O(‰) other components. The SSI and SSO account for 51% and
-12 -11 -10 -9 -8 -7 56% of total inflow and total outflow respectively.
-50 The results presented in table 2 shows that the estimates
Leakage
D9 & HP -55 of sub-surface inflow to the lake and outflow from the
-60
Reservoir lake, obtained through isotopic and chemical balance,
δD(‰)
-65
-70 compare well with those obtained through conventional
-75 water balance method.
December-08-Feburary-09 -80
-85
HP Reservior D9 D10 D11 Linear (PPT)
Table 2 SSI and SSO data estimated by isotopic,
chemical and conventional mass balance methods.
Fig. 2 Seasonal variation of D vs 18O in Groundwater, Method of
Reservoir surface water and leakage galleries of AGR-3 18O (‰) Chloride Conventional
estimation
AGIR SSI SSO SSI SSO SSI SSO
Volume
WATER BALANCE STUDY OF LAKE (x 103 m3) 2269 2618 2777 3140 2234 2416
Nainital Lake located in the Kumaun Lesser Himalaya % to total
is the main drinking water source to the people living in inflow or
and around the lake basin. Isotopic investigations have outflow 51 56 55 59 50 54
been carried out in order to understand the surface water Lake WRT -
and groundwater interaction and to throw some light on Year * 1.93 1.77 1.92
the present sedimentation status to estimate the expected
* Lake water retention time has been calculated,
useful life of the lake. The investigations were aimed to
assuming mean depth of the lake as 18.52 m. Time
aid the Lake Development Authority in adopting
difference between the dates of sampling considered
appropriate management strategy. Increase in
was 380 days. Appropriate corrections have been made
population pressure has resulted in the deterioration of
to calculate the total inflow in 365 days.
the environment of Lake Nainital. The conventional
methods, either using discharge-sediment concentration
Estimation of Sedimentation Rate
or periodic bathymetric surveys usually provide
Radiometric Dating Method
adequate information required for computation of lake
The sedimentation rates have been estimated at different
sedimentation rates and water balance of the lake.
locations in the lake employing Cesium-137 (Cs-137)
and Lead-210 (Pb-210) radiometric dating techniques of
sediments. Radiometric dating techniques have been

proved as reliable tools for estimation of lake basin from 0.260.04 to 1.060.10 cm/yr. The results
sedimentation rate and are being used the world over. show that Rate of sedimentation is comparatively higher
Although several radioisotopes are useful in in Hazaratbal sub basin of Dal Lake. The Mansar lake
geochronological studies of lake sediments, 210Pb and represents the condition of the Siwalik Himalayas of
137
Cs find wider application. 210Pb dating technique was Jammu and Kashmir lakes where sedimentation rate
initiated by Goldberg (1963) and the technique was varied from 0.14  0.03 to 0.37 0.03 cm/y. The
established for dating of lake sediments by minimum rate of sedimentation is in central and deepest
Krishnaswamy et al. (1971). Since then numerous part of the lake and maximum closer to the debauching
studies on the use of 210Pb and 137Cs both in the research point of an inflow stream very similar to the Kumaun
and application fronts of lake sediment dating have been lakes.
reported (Krishnaswamy and Lal 1978, Ritchie and
McHenry 1985, Walling and He 1993). REFERENCES
Sedimentation rate of the Nainital, Bhimtal, Naukchiatal
Goldberg, E. D. (1963). Geochronology with 210Pb In:
and Sattal in Uttaranchal, Dal and Mansar in Jammu Radioactive dating. International Atomic Energy
were determined using the radiometric dating Agency, Vienna, Austria
techniques 137Cs and 210Pb. Activities of 137Cs and 210Pb Krishnaswamy, S. and Lal, D. (1978). Radionuclide
in sediment samples were measured at National Institute
limnochronology. In: A. Lerman (Ed.) Lakes:
of Hydrology, Roorkee.
Chemistry, Geology, Physics, Springer-verlag.
The sedimentation rate in the Nainital Lake is higher
Krishnaswamy, S., Lal, D., Martin, J. M. and Meybeck,
(1.350.05 cm/yr) in the intermediate portions located M. (1971). Geochronology of lake sediments. Earth
just adjacent to the bank zones while comparatively Planetary and Science Letters, 11, 407-414.
moderate (0.640.18 cm/yr) in steeper bank zones. The Pearce, A. J., Stewart, M. K. and Sklash, M. G. (1986).
deeper portions away from the bank of the lake, receive Storm runoff generation in humid headwater
sediment at lower rate (0.480.04 cm/yr). In Bhimtal catchments, I, Where does the water come from?
Lake, highest rate of sedimentation (1.500.03 cm/yr) at Water Resources Research, 22, 1263-1272.
the debouching point of the major inflow drain and Ritchie, J. C. and McHenry, J. R. (1985). A comparison
minimum (0.430.04 cm/yr) in deepest and central part of three methods for measuring recent rates of
of the lake. Similarly, the highest rate of sedimentation sediment accumulation. Water Resources Bulletin,
of 0.950.04 cm/y is observed in the eastern part of the 21, 99-103.
Naukuchiatal lake closer to the debauching point of a Sklash, M. G. and Farvolden, R.N. (1979). The role of
major inflow stream. In the northern part of the lake, the groundwater in stream runoff. Journal of Hydrology,
rate of sedimentation is about 0.59 cm/y and at the 43, 45-65.
central portion the rate of sedimentation is lowest (0.38 Sklash, M. G., Farvolden, R. N. and Fritz, P. (1976). A
cm/y). Therefore, it appears that the sedimentation conceptual model of watershed response to rainfall,
pattern within all the Kumaun lakes is more or less developed through the use of oxygen-18 as a natural
similar. tracer. Canadian Journal of Earth Sciences, 13, 271-
The current sedimentation rate different sub-basin of 283.
Dal lake determined by 137Cs dating technique varies Walling, D. E. and He, Q. (1993). Towards improved
between 0.140.03 to 1.600.13 cm/yr. In the Hazratbal interpretation of 137Cs profiles in lake sediments.
basin varies 0.400.05 to 1.600.13 cm/yr, in Bod-Dal In: J. Mamanus and Duck R.W. (Eds.)
basin from 0.390.05 to 0.610.06 cm/yr, Gagribal Geomorphology and Sedimentology, Wiley.
basin from0.140.03 to 0.220.05 cm/yr and in Nagin

ISBN 978-81-921121-0-7
GROUND WATER ZONES AROUND JHANSI TOWN USING SATELLITE

REMOTE SENSING DATA IN THE GRANITIC TERRAIN OF
BUNDELKHAND
S.P. Singh
Department of Geology, Bundelkhand University, Jhansi-284128, spsinghbu@rediffmail.com
Yamini Singh
Geologist, MSPL, Badota Road, Hospet, Karnataka,
ABSTRACT: The geologically area around Jhansi town represents a rocky terrain of several varieties of granitoids and
quartz reef. These rocks have received ductile to brittle conditions deformations which are mainly responsible for the E-
W, NE-SW and NW-SE trending shear zones, fractures and joints into these granitoids. The NW-SE trending regional
/crustal scale reactivated fault passing from the Jhansi area mainly responsible for the development of present
topography. This tectonism is also responsible for the thin to thick soil cover on the fractured grantoids in the different
parts of Jhansi and causes the development of braded channels and river shifting of Betwa river and drainage system.
The integrated approaches based on the pattern of fractures and shear zones, joints, soil, drainage system vegetations
and forest demarcates the low, moderate and high potential areas for the subsurface water conditions around the Jhansi
town.
than 60% area around the Jhansi town is occupied by
INTRODUCTION merely fractured and sheared granites, fractured quartz
Groundwater is the primary source of drinking water and reefs and thin soil cover.
perhaps is the safest and most reliable resource for fresh The comparison of satellite imaginary of months of
water in the country. More than 80% of the Indian November and March reveals that most of the major
population living in small town and rural areas still reservoir constructed around Jhansi town becomes dry
depend on this water. Only 3% of the earths freshwater is in summer season and check dams becomes cultivated
located in the streams, lakes, reservoirs and rest 97% of land for Rabi crops. Due to rapid increase of demand of
freshwater is available to underground. Groundwater is a water for man and animal and vegetation owing to
natural stored water which occurs in the several industrialization, urbanization modernization and expand
formations of saturated/weathered rocks, sand, gravel and of agricultural land around Jhansi. It has been realized
soil and some times in the fracture zones. Unlike surface that people should also depend on several mitigations and
water, the groundwater does not show much flow in a find out the possible solution for the sources of water
series of lake and river. The availability and quality of potential around the Jhansi town. The present study is
groundwater varies widely across the India. Generally, aimed to analysis and evaluates the groundwater potential
the well yield ranges from less than 3 gallon per minute in zones around Jhansi with the help of remote sensing and
western and central parts of country to thousands of geological data.
gallons per minute well in aquifers beneath and adjacent
to major rivers of India. In Bundelkhand region, the GEOMORPHOLOGY
ground water usually available between 32 feet and 65 The Bundelkhand craton is delineated from Indo -
feet depth in the rocky terrains where the chances of Gangetic plain by NW-SE trending Yamuna fault in the
aquifer conditions are very limited. The ground water in north and separated from southern India shield by E-W
Bundelkhand is usually meager and some places are trending Son Narmada faults. The western part of craton
excessively hard where the presence of dissolved salts of is separated from NE-SW trending Aravalli Mountains by
calcium, magnesium, manganese and iron are enriched. Great Boundary fault (Singh et al. 2008). Thus
The population of Jhansi city is rapidly increasing due to Bundelkhand terrain has developed a triangular shaped
rapid migration of people from its rural areas owing to Precambrian land mass in the northern part of Indian
premium centre for education, transportation, industries shield where several faults on regional /crustal scale are
and business centre in Bundelkhand since last two passes (Fig 1) and controlling the geomorphology of
decades. The Jhansi is the biggest town in Bundelkhand entire Bundelkhand terrain (Singh and Basu 2011).
which is situated in the northern part of the granite terrain Geomorphologically, the northern part of craton has low
where four major multipurpose reservoirs were elevation compare to southern part. As a result of this all
constructed before the independence. However the the major and moderate type rivers of Bundelkhand
pressure on the requirement of water to common people terrain are originated in south and followed the NE
of Jhansi has been increasing in the multitudes and the direction along the trend quartz reefs. The major rivers of
present condition is inadequate to fulfill the requirement Bundelkhand ultimately become parallel in Zone –I near
of demand. The water crisis around Jhansi town has the Yamuna fault before their mergence into the river
initiated in the end of last century mainly due to the Yamuna (Fig.1).
myopic exploration of groundwater to fulfill the need of The Bundelkhand terrain has been divided into four
drinking and multifarious domestic requirements. More geomorphic zones based on the geological and

geomorphological and some other factors like soil cover, fault lineament passes from the central part of the
drainage, linear fabrics, hydrogeological constraints along Bundelkhand that demarcates the rocky terrain dominated
with input from the computation of remote sensing data area in the south and peripheral of Ganga basin alluvium
(Fig.1). Zone I of this terrain is characterized by thick deposits in the north while thePanna-Gwalior lineament
alluvium deposits with rare occurrences of hard rock’s passes from Jhansi that demarcates the boundary of rocky
like granite, quartz reef etc. The area is dominated by terrain to the terrain of thin soil cover in the massif part.
agricultural flat land topography. Variation in land use
and land cover, thickness of forest and shrubs and ravine GEOLOGY AND ROCK TYPE
type topology has been observed along the tectonically The area around Jhansi town is mainly dominated by
active zones. Zone II is characterized by mainly the rocky massive hard and compact, pinkish grey coloured coarse
terrain with thin soil cover. The zone III consists the grained granite that is also called Jhansi granite (Jhigaran
massive granite of the central part. The NE-SW, E-W and 1958, Sharma 1982). The western part of the area is
NW-SE linear fabrics have been noted from the massif covered by Pahuj water shade, a tributary of Kali-Sindh of
part but E-W and NE-SW linear fabrics were not found to Chambal basin while the eastern part is the part of Betwa
basin (Fig.1). The Jhansi town is situated in the central
part of study area and is bounded by two prominent quartz
reefs trending in NE-SW and N70E-S70E directions (Fig
2). The northeast part of the study area is characterized by
recent sedimentary covers where signatures of
paleochannels and meanders of river Betwa are still well
preserved (Fig 2).
continue in the Proterozoic rocks (Bijawar, Gwalior and

Vindhyan).The zone IV is dominated by thin rocky soil
and plateau type mountains.
Fig 1: The map shows the major river basin of
Bundelkhand (Chambal, Betwa and Ken) and their
relation with Gangetic basin. The NW-SE trending Fig 2: Geological map of the study area. The porphyritic
Yammuna fault is major dislocation identified from pinkish grey granite is the major rock type around Jhansi
satellite imagery deleneates the Varanasi alluvium (VA) town.
and Banda alluvium (BA). The Kabrai fault, Panna Since more than 60% of the proposed study belongs to
Gwalior Fault, Shahgarh Babina fault, Narhat fault rock terrain therefore, a detailed geological mapping
parallel to yamunna fault showing normal step fault. have been carried out to establish the relationship
Zone-I : Dominated by thick alluvium Zone-II: Thin between ground water conditions, and to locate recharge
alluvium and granite rock, Zone-III: Granite and rarely areas with geological structures and rock types. The
alluvium, Zone-IV: Gneissis, with thick alluvium, Zone- geological map of Jhansi town ( Fig.2) points that terrain
V: Sandstone.Terrain have been demarcated by Remote mainly comprises hornblende- biotite- bearing coarse
sensing Data of IRS-ID. The study area Jhansi is part of grained granite, biotite granite, leuco granite pink granite,
Zone-II, shown in the square. grey granite, quartz reef, dolerite dyke. These rocks have
The geomorphologic studies based on different techniques been mapped and their outcrop pattern were further
of remote sensing indicate them as normal step faults that correlated with structural data vis a vis satellite imagery
are related to recent movements and stress development in of IRS-ID.
the peripheral part of Gangetic plain and massif part. Pink granite is medium to coarse - grained, usually
Three such faults/lineaments have been identified (Fig 1) massive and devoid of any major structure. The rock is
that are mainly responsible for major geomorphologic and very hard and compact. The orthoclase, microcline and
topographical changes in Bundelkhand terrain. The NW- quartz are the major mineral constituent. Beside these,
SE trending Kabrai-Jigni- Behat fault/lineament, Panna– magnetite and flakes of biotite and occasionally needles
Gwalior fault/lineaments and Shahgarh-Babina of hornblende minerals also found to be embedded in the
faults/lineaments were analyzed from the satellite data, massive rock. The pink granite is mainly exposed in the
which are more or less, parallel to each other. Jigni- Behat southern and northern parts of the Jhansi town. The small

patches of pink granites have been found as intrusive surface zones and surrounding soils are characterized by
around Jhansi town near Konchha bhawar, Parichha and red colour texture. The width of individual out crop varies
many other places. These granitoids are intrusive into from 50 to 100m. There are four basic dykes in the area
grey granite as well as biotite granite and hornblende- and each trend in NW-SE direction. The contact of basic
biotite granite in Jhansi. The fractures and joints were dykes with quartz reef is one of the favorable conditions
very common in pink granite, exposed at Barwasagar. for ground water.
Gray granite is widely exposed to the southern part of
Jhansi mainly around Pratappura, north of Bijauli, GEOLOGICAL STRUCTURES
Konchha Bhanwar, Bhagwantpur and Raksha villages The polyphase deformations for the rocks of Bundelkhand
(Fig 2). Small patches of gray granitoids are found around have been suggested by many workers (Singh et al 2005,
south west of Ambabai. They are usually medium to Basu 2007). Three phases of deformation have been
coarse grained, hard and very compact. In the local identified for the granitoids of central and northern part of
language this rock is called “Teliya pathar”. Patches of Bundelkhand (Singh et al. 2007). The E-W trending
gneisses and migmatite are also noticed with this granite. vertical shear zone has been considered as oldest
The E-W shearing is prominent in the gray granites. The deformation while NW- SE trending shears and faults are
north part of gray granite sometimes consist enclaves of youngest phase of development (Singh and Bhattacharya
gneisses. The biotite-hornblende granite is common 2010). The study suggests three E-W trending shear zones
intrusion in the southern and northern part of grey from Jhansi viz. (1) E-W trending vertical shear zone
granites respectively. passes through Ambabai – Karari to Isagar to Palar, in
Hornblende –biotite granite (Jhansi granite) is most the north of Jhansi town (2) Bijauli – Orachha Barwa
dominating granite in the area. It is pink gray to brownish sagar shear zone passes to the north of pink granitoid
grey in colour, very hard, compact and massive in nature. in the south of Jhansi and (3). The Rajapur-Raksha shear
Due to subsequent deformations, the lineation and which occurs in the central part, passes through Jhansi
fracture has been observed at several places. The coarse town to Konchha Bhawar to Baghat (Fig 3 ).
grained hornblende and porphyritic feldspar is the These E-W shear zones are few meter to 100 meter in
character of this rock. Besides this, clustered of width and individual crop never exceed more than 300
magnetite, epidote and biotite are also present. This meter in length usually, they are vertical sheet like
granite is mainly exposed in the central part of study mylonite and sometimes fine grained granite observed
area and is also called Jhansi granite (Jhingaran 1958). along Raksha- Rajapur – Pichhor shear zone. The NE-
The plagioclase, quartz, orthoclase, hornblende and SW trending shear zone is also prominent in the study
biotite are the major mineral composition. area which is confined mainly at Konchha bhawar,
Biotite granite is medium grained light grey to light pink Shivgarh, and Simardha and Bijauli villages (Fig 3). The
in colour. Biotite is usually small in size, streaky in NE-SW trending shear are found to be parallel to quartz
texture and defines the flow direction of magma. The reef near Badol and Karguwan (Baghat) villages. The
massive biotite granite expose mainly near village Bijauli, NE -SW shears are ductile to brittle ductile in nature and
Ambabai and Simardha villages. The small hillocks of displaced the E-W trending sheared granitoids at many
biotite granite are also exposed at Jhansi fort, Pichor and places. The NW-SW trending shears and fractures are
Garhmau areas. The Joint and fractures are frquent very prominent in the east of Jhansi town and also near
Quartz reef is elongated, linear and detached hillocks, Bijauli and Barawa sagar (Fig 3). The patches of NW-SE
characterized by hog back structures and teeth like hill trending shear zone also noted in SE of Raksha and
tops of Bundelkhand massif. These hillocks are elevated Shivagarh villages. These shear and fault zones are
hundred too few hundred feet from its surrounding and sometimes more or less parallel to basic dykes. The NW-
consists very course crystals of quartz and feldspar. SE shears displace the quartz reef hillocks, biotite
Surface of quartz reef is usually highly weathered and granite and E-W trending shear zone at many places.
fractured. The boulders and gravels of quartz reef are These are brittle to brittle in nature, solidifications and
frequent on both side of the slope. The quartz reef is presence of pegmatite’s and quartz veins are very
found in detached hillocks, trending in NE-SW direction. common several faults trending NW-SE have been noted
These hillocks are some time displaced and traversed by which has displaced the quartz reef from meter to
basic dykes or shear zones which has been used as the 500/600 meters. Besides this N 150 W and N 150 E
linear fabrics for the identification of underground water trending lineament, fault and fine granite have been
locations. There are four major quartz reefs has been identified to the west and east of Jhansi town (Fig 3.).
mapped in the study area. The Jhansi town is bounded of These decibed shear zones are also responsible for
two NE-SW quartz reefs. The reef passes through changes in the geomorphology and land use and land
Lahargird to Palar is infact a water divider for two major cover of the study area. The Jhansi town is bounded by
basins viz. Betwa basin and Chambal basin in the study two quartz reefs viz. N700E – S700W trending Palar -
area. Lahargird quartz reef and Bijauli- Garhmau quartz reef.
Dolerite dyke is usually exposes as smaller outcrop of The N-S trending two fault causes the uplift of the area
boulders of dark grey to black coloured rock that around east of Jhansi city. The continuous reactivation of
commonly occurs at low land areas. However, hillock of E-W shear zone and Panna–Gwalior fault causes the
dolerite has been also found at Bijauli. The weathered braded channels in Betwa river and subsequent change

in the elevation due to upliftment of block of Jhansi FOREST AND AGRICULTURAL LAND
causes the river meandering. Not only this, the presence The general trend of slope in the investigated area is
of shifted paleochannel around Barata (east of Jhansi towards northeast; however, the area has been divided
town Fig 2) may be due to change in river course of into three parts on the basis of NE-SW trending quartz
Betwa may also be considered the part of reactivation reef, which occurs as moderated to highly elevated hills.
of these fault in Bundelkhand craton (Singh and Basu Thick vegetations including shrub have been found to
2011). develop along these quartz reefs. The dense forestation
has been encountered along the river Betwa (Fig. 4). It is
noted that dense forests are more frequent on pink granite
terrain while least on grey-granite terrain. The shrubs are
U. P. mostly around Jhansi town and are confined on the
hornblende- biotite bearing pinkish grey granite (Jhansi
granite). Thick soil has been found along old fracture
zones, especially along E-W trending shears. At the north
east of Jhansi and north of upstream of Betwa at
Baragaon, a wide fertile land has been encountered. The
M. P. land has been possibly developed due to recent
sedimentation owing to signature of change in river
course and meandering of Betwa. The paleochannels and
menders are well preserved in satellite imagery. The small
Fig 3: Lineament map of the study area around Jhansi patches of agricultural land has also encountered around
town based on visual interpretation of IRS-ID data. Barawasgar. The small patches of agriculture land on
local scale have been found throughout the study area.
DRAINAGE SYSTEM
Betwa, Dhasan, Ken, Pahunj and Sindh are the major
rivers of Bundelkhand, which are originated at south and
proceed to northward and finally merge into river
Yamuna or its tributaries. These rivers have followed the
various geological constraints and geological structures
on local and regional scale (Fig.1). The Betwa River is the
most powerful river of Bundelkhand that consists various
tributaries. The major part of investigated area lies in the
Chambal basin though river Betwa passes from Orachha
to Parichha in east. The river Pahunj, tributary of Sindh is
originated about 45km SW of Jhansi and followed the
north ward upto simardha village and then its turns to
eastward along shear and further turns to west and finally
merges in the Sindh at very close to confluence of
Chambal and Yamuna (Fig 4). The changes in the
direction from east to west for the Pahuj river to tne east Fig 4: The map shows the forest, agriculture land and
of Jhansi is related toNw-SE tectonic structure. water bodies/ reservoirs with drainage system in the study
Drainage and slope elevation study around Jhansi area.
suggests that Jhansi town is situated at higher elevation
than its surrounding. The changes in direction of streams SURFACE WATER AND RAINFALL
have been noticed at many places which are related to The Bundelkhand region comes under the arid type
local geology (Fig 4). All the drainage start from the climate where more than 8 months are characterized by
central part and has followed the structure of the area. The bright sun and hot temperature. The average rainfall is at
urban part of jhansi is bounded by two river viz. Betwa in Jhansi is about 835 mm that encounters only for 70 days
the east and Pahuj in the western part. Besides these, there two months (from 25th June to 25 th Sept). Most of the
are several small water channels and nalas ( Fig. 3 and 4). rainfall (more than 80%) has been covered under these
Kumhara and Mathhrapura are two major nalas to the two months only. Since more than 80% area of Jhansi
west of Betwa that run from west to east and controlled and adjoining part consists igneous plutonic acidic rock
by the E-W trending shear zone. Among all the Angora or rocky terrain with thin soil cover, therefore heavy rain
nala is the most prominent nala which originated from fall causes the rapid runoff and fast erosion. On the other
heart of Jhansi town and fallows the NE trend. The hand the temperature in Bundelkhand is high, more than
drainage pattern around Jhansi is mostly in the dendrite 30ºC in eight months of year and low humidity through
pattern. However, some of them develop in the linear all season that causes high rate of evaporation. Thus the
pattern. The lineament and drainage pattern around Jhansi adverse geological conditions for water retentions viz.
clearly indicate their structural and geological controlled. granite rocky terrain and harsh climatic conditions cause
the poor chances for sustaining the conditions of aquifer
in this region. The people of this region has realized since

the beginning that construction of reservoir for the
storage of rain water, is the only solution for the
survival of man , animals and environment in this area.
Therefore, different size of storage type reservoirs were
constructed time to time as per there requirement. There
are six major reservoirs for storing the surface water with
in the 15 km circle of Jhansi town (Fig 4). Three major
dams viz. Parichha, Sukua Dukwa and Pahuj have been
constructed for minimize the water scarcity. Besides
these, various check dams of as small reservoir covering
few sq. meters were also constructed to store the water for
agricultural works. The geological study point that some
of these check dams are loaded on the recharge zone and Fig 5: Map showing the possible moderate to high
dam at Parichha is very close to the recharge zone of occurrences of groundwater potential zones around Jhansi
Jhansi. with the different parameters discussed in the text. Thick
line represents the quartz reefs.
SUB SURFACE WATER The slope and topography study reveals that general slope
The ground water condition in entire rocky terrain of of study area corresponds to the general slope of
Bundelkhand is more or less perched water conditions. Bundelkhand terrain with few exceptions. The southern
The ground water usually accumulates in weathered zone part of Betwa has characters of braided channels. The
or sometime in the fractured zone of granites. The stored river Jamini diverted from the confluence and develops a
ground water has been presently explored from the depth very enlarged and elongated area of confluence with
of 45 to 60 feet around Jhansi. The high LMP areas Betwa at south of Orachha (Fig 4). The presence of
mostly confined around the Barata, Miyanpur and braided structure and development of various NE-SW
Baragon areas (Fig 3.). The underground water is also trending shear and fault suggests that this part have been
pumped from the depth below 100 feet at some places in active (Singh and Basu 2011). This view is also supported
Barata basin. The shallow water conditions have been by the presence of various paleochanels and signatures
recorded from few areas of Jhansi, Baragaon and Barawa meanders of river Betwa to the north of Orachha (Fig 2).
sagar. The geological and field study suggest that Along the thick sheared and fractured zones, the crust
moderate to good potential zones of ground water are rocks promoted soil formation and increases the
mainly confined around Chopra, north of Pichore, vegetation and forestation. The presence of E-W trending
Baragaon, Barwa sagar and also nearby area of Laxmital Kumharra and Masthhapura nala along E-W fault is an
and Nayatal. The Bijauli, Raksa, Bijaur areas have been evidence of the reactivation of Blocks (Fig 3). Series of
considered as low to poor for underground. The water NW-SE trending fracture zones parallel to Panna-Gwalior
level rises in month of July and August and gradually fault have been identified from study area. It is the fact
goes down ward from month of October and by the end in that possibilities to explore the favorable ground water
month of April most of the wells and hand pumps condition in granite terrain are always a difficult task. The
becomes dry due to myopic exploration of ground water. presence of fractures, joints and some faults and some
time shears zones from the granitic terrain of Jhansi has
DISCUSSION AND CONCLUSION encouraged the possibility of ground water in the
Jhansi town of Bundelkhand region is bounded by NE- secondary structural zone in some extent around Jhansi
SW and ENE – WSW trending two important quartz reefs Town (Fig 5). The thin to thick soil cover and vegetations
which are further traversed by the crustal scale NW-SE have been noted all around the Jhansi town. Such fracture
trending Panna-Gwalior fault. Both the structures were zone promoted the surface water percolation through
developed after the emplacement of the granitoids and are fractures and joints of granites and perhaps becomes the
encountered from all the rocks. Therefore, these structures potential zone for storing the ground water in the
and small faults associated with them cause’s fractures weathered zone of granite and some times in the
and crushing into granitoids. As a result series of shears, secondary fracture zone.
fractures and linear fabrics developed in the different The study suggests that eastern part of the study have
rocks of Jhansi. Such fractures and associated joints are highly fractured rocks with thick soil cover. The moderate
very prominent at the contact zones of pink granite and to dense vegetation and thick soil cover and dominance of
grey granite possibly due to geological parameters. The forest has been also encountered from these areas
peripheral parts of biotite-granites are also found to be compared to the other parts of Jhansi indicate the
more sheared and fractured than core part. The core parts presence of high soil moisture content and subsurface
of pink granite, biotite granite, pinkish-grey granite and water. The geophysical exploration and water pumping
gray granite contain the poorly developed joints than rim station around Jhansi point the presence of perched
part. All these features and joints suggest the development water conditions at depth of 25 to 45 feet around Laximi
of secondary fracture zones in the granitoids of Jhansi and tal, 25 to 60 feet around Barwa sagar and 30 to 100m
their density vary from place to place. around Barata and Garhmau areas. Low to moderate LMP
of ground water has been also identified around Bijauli,

Pahuj Tal and Ambabi areas. Thus on basis of geological,
structural, geomorphology and evaluation of Remote
sensing data, the ground water potential zones around
Jhansi town has been broadly categorized into (1) Good
(2) Moderate (3) Poor. The eastern and north eastern part
and some part of south of Jhansi has been considered in
the first category while western and south western are
considered under the third category and rest of area is
under the second category. The construction of major
dams and their linkage with smaller dams on one side and
strengthening the recharge system into the thick soil cover
area may minimize the water scarcity for continuous
myopic exploitation of water.
REFERENCES
Basu, A.K. (1986). Geology of parts of the Bundelkhand
Granite Massif, Central India. Rec. Geol. Sur Ind., 117
Basu,A.K.(2007). Role of the Bundelkhand Granite
Massif and the Son-Narmada megafault in Precambrian
crustal evolution and tectonism in Central and Western
India. J. Geol. Society India. 70: 745-770.
Jhingran, A.G. (1958). The problem of Bundelkhand Granites
and Gneisses. Presidential Address, Section Geology and
Geography. 45th Ind. Sc. Cong., Madras.
Sharma,R.P, (1982). Lithostratigraphy, structure and
petrology of the Bundelkhand group. In. Valdia, K.S.,
Bhatia, S.B. & Gaur K. Eds. Geology of Vindhyanchal,
30-46
Singh and Basu (2011). Giant Ground Fissures from the
South of Ganga Basin: Evidences of Neotectonism in
Northern Part of Bundelkhand Craton. Proceeding
Neotectonism in edited Dhruv Sen Singh and N.L.
Chhabra in Geological process and climatic changes.
Macmillan Publisher New Delhi pp235-250
Singh S.P. Hemraj, Shukla, R.S. Shrivastva, S.K. and
NambiarK.V. (2008). Granite collapse Breccias in
Precambrian Terrainof Bundelkhand Craton. Jour.
Economic Geology and mineral resource management
Vol 5, 36-52.
Singh, M. M., Singh, S. P. and Srivastava, G. S. (2005).
Crustal evolution in Bundelkhand massif Central India.
Abstract, International conf on PCGT, Bundelkhand
University, Jhansi, India. 242-243.
Singh S.P., Singh M.M., Srivastava G.S, and Basu A.K.
(2007). Crustal evolution in Bundelkhand area, Central
India. Jour. Himalayan Geol., 28, pp.79-101.
Singh S.P. and Bhattacharya A.R. 2010. Signatures of
Archaean E–W Crustal–Scale Shears in the Bundelkhand
Massif, Central India: An Example of Vertical Ductile
Shearing. Journal Earth Science India, Vol. 3 (IV),
October, 2010, pp. 217-225.

ISBN 978-81-921121-0-7
INFLUENCE OF CRUSHABILITY AND MIGRATION OF MOISTURE ON

CBR CHARACTERISTICS OF POND ASH
M.V.S.Sreedhar, Research Scholar & Asst. Professor, Osmania University, email: mvs_sreedhar@yahoo.com
G.Venkatappa Rao, Former Professor, IIT Delhi & Hon. Professor, Osmania University, email: gvrao.19@google.com
R.Ramesh Reddy, Professor, Osmania University, Hyderabad-500007, email: rameshreddy@hotmail.com
ABSTRACT: The pond ash being a non-plastic cohesion-less medium with relatively better CBR value, it has high
potential for being used as an overlay material on soft sub-grades of the pavements. On the other hand the CBR value is
considered to be an important parameter in the design of pavement structures. It is therefore necessary to understand the
CBR characteristics of pond ash more systematically. It was noticed that, in general, there is significant difference in the
CBR value of the same pond ash sample when it was tested both on top and bottom faces, which happened to be
marginal in respect of natural inorganic soils. When the difference is considerable, it invites attention as to which value
is to be adopted in the design. In this research, efforts are made to understand the mechanism in a greater detail. Based
on the investigations it is concluded that, there is considerable amount of crushing of coarse and medium sand sized
pond ash particles when compacted leading to a differential dry densities at the top and bottom faces which is
responsible for the difference in CBR values. This study suggests adoption of CBR value of the face that remains at the
top during compaction in the design as it resembles the field condition appropriately.
INTRODUCTION
One of the environment related challenges faced by METHODOLOGY
mankind in the present century is utilization of the Coal The methodology includes collection of Pond Ash and its
Ashes. While fly ash with pozzolanic property is finding characterization. Conducting CBR tests in soaked and un-
considerable utilization, there is utmost necessity to soaked conditions on both CB and CT faces. To
increase the bulk utilization of Pond Ash. understand the difference, after the CBR test is carried out
on both faces, the sample is carefully divided in to top and
Pond Ash being a non-plastic cohesionless medium has bottom halves and tests are performed to determine the
great potential to be used as an overlay on soft soil sub- moisture content, dry density, gradation of the soil sample,
grades. The California Bearing Ratio (CBR) value of the separately for the two halves. The effect of compaction
pond ash overlay will govern the pavement design. In this energy on crushability of pond ash is also investigated.
context, a phenomenal considerable difference in the CBR
value of the pond ash tested on top and bottom faces is Characterisation of pond ash
identified against the marginal difference in inorganic The Pond Ash was collected from National Thermal
soils. As per clause 4.3.3.2 of the IS:2720 (Part 16)-1979, Power station, Ramagundam in Andhra Pradesh. The Pond
after compaction, the sample is inverted due to which the Ash was collected from a near the discharge point that is
bottom face during compaction (denoted at CB) will about 3m from the bank, at a depth of 2m below the then
become the top face. Many a times, the penetration test is ash level in April-2010. The index and engineering
conducted on the CB face only and its validity is not properties of the Pond Ash used in this project are as given
checked for the reverse end of the sample, as suggested in in Table 1. The photograph of the Pond Ash is shown in
clause 5.2.1. of the said code. Though it may not make Fig. 1.
much difference in normal inorganic soils, it is bringing
huge difference in respect of pond ash. This reduces the Table 1 Geotechnical characteristics of Pond Ash
factor of safety and thus performance of the pavement. It Parameter Value units
is therefore necessary to investigate the mechanism
involved. Efforts are made in this direction in this project. Specific Gravity 1.97
Particle Size distribution
Gravel Size particles 7.20 %
REVIEW OF LITERATURE Sand Size particles 80.60 %
Several studies are carried out on the CBR characteristics Silt size particles 12.20 %
of natural soils and it was concluded that, marginal Consistency Non-Plastic
difference was noticed between the CBR values of bottom Coefficient of uniformity, Cu 4.65
and top faces during compaction, named CB, CT Coefficient of curvature, Cc 0.84
respectively. It was reported that, the CBR for CB face IS Heavy Compaction test results
was marginally higher than that of CT face. Though, Maximum Dry Density 1.17 g/cc
studies on CBR characteristics of pond ash were carried Optimum Moisture content 28.90 %
out, this aspect was not thoroughly investigated. Classification as per IS:1498-1970 SP

Test set up
The experimental test set up is shown in Fig 1. The CBR
tests are conducted strictly as per IS:2720 (Part 16)-1979.
A surcharge of 5.0 kg is used to simulated the overburden
due to upper pavement layers.
Fig. 3 CBR plot for four day soaked condition
OBSERVATIONS
Based on the CBR plots shown in fig. 2-3, the CBR value
corresponding to 5.00mm penetration, which happened to
be higher even after repetition, duly applying correction
Fig. 1 The test set up
for initial upward concave curvature, as shown in table-2.
These CBR values appears to be un-realistic. But even
after three repetitions with three different proving rings,
RESULTS the results are established almost same. This is attributed
About 32 sets of CBR tests are performed on the pond ash. to the greater interlocking between the pond ash particles.
The difference between the CBR value of CB, CT faces
was noticed. One set of such typical penetration curves are Table -2. Difference in CBR value
shown in fig. 2-3. CBR value (%) for Un-soaked Soaked
CB face 50.20 49.75
CT face 35.90 36.20
Percentage 39.83 37.43
Difference of CB
face over that of CT
face
As it can be seen from table-2, the difference is very huge.

In field conditions, the pond ash is compacted and the
CBR of compacted top face influences its behaviour.
Accordingly, the true CBR value should have been that of
CT face. However, if CBR value of CT face is not
determined and that of CB face is adopted, it leads to
under performance of the pavement.
After having understood that, the difference is huge, it is

necessary to investigate the reasons behind this difference.
In line with this, the corresponding samples are placed on
the dummies that are placed in a clean and dry tray and
pushed gently downwards, such that the sample is ejected
Fig. 2 CBR plot for unsoaked condition out and the spilled out ash is collected in to the tray. When
the ejection is up to half thickness of the sample, the
surface is trimmed. Knowing the weight of one half, the
bulk density is computed. Similarly for the remaining half.
From the so collected pond ash, samples for water content
and particle size distribution are considered. The results
are as shown below. The samples consists of (-)4.75mm
size particles only.

Table -3 Difference in dry density and moisture
content values
Un-soaked Soaked
Overall moulding bulk 1.27 1.49
desnity (g/cc)
Overall Moisture content 29.62 33.97
after the test (%)
Overall moulding dry 1.123 1.119
density (g/cc)
On CB face
Moisture content 30.335 37.115
Dry density 1.151 1.148
On CT face
Moisture content 28.905 30.82
Dry density 1.095 1.09
The variation in particle size is as shown in fig. 4-5. Fig. 6 Effect of compaction energy on crushability of
pond ash
Fig. 4 Variation in particle size distribution

of un-soaked CBR sample
Fig. 7 Effect of compaction energy on crushability of
medium sand size pond ash particles
Fig. 5 Variation in particle size distribution Fig. 8 Magnified (373x) view of Pond Ash particles
of soaked CBR sample (Curtesy: A.Bhoominathan et.al.)

Based on the results, the following observations can be 5. Subject to feasibility, the pond ash possessing
made relatively higher CBR value and being non-
plastic has great potential to be an overlay on soft
1 From fig. 4-5, it is evident that, the coarse and sub-grades.
medium sand size particles in un-compacted pond
ash are being crushed significantly and thus
increasing the fraction of fine sand sized and silt ACKNOWLEDGEMENTS
sized particles which in turn may be contributing to The first author acknowledge the support provided under
relatively better dry density. TEQIP phase-I and Research Infrastructure fund of UGC.
2. This phenomena of crushing is more pronounced in

CB face than in CT face for the reason that, the REFERENCES
pond ash in bottom most layer has the opportunity 1. Bhoominathan, A and S.Hari (2002), Liquefaction
of receiving more compaction energy than the strength of fly ash reinforced with randomly
upper most layer. This is evidenced by fig. 6-7. distributed fibers, Soil Dynamics and Earthquake
Engineering, 22 (2002) 1027–1033.
3. Further, the globule like structure of the coarse 2. Datta M (1998) Engineering properties of coal ash,
pond ash as shown in fig. 8, when crushed results in Proceedings of Indian geotechnical conference, Delhi,
a fraction with high angularity. This not only helps 2, pp 41–46
in modifying the gradation to well graded state and 3. Datta M, Singh A, Kaniraj R (1996) Spatial variation
thus dandifying the medium, but also helps in of ash characteristics in an ash pond, Narosa
imparting significant structural and frictional Publishers, Delhi, pp 111–119.
contribution to its shear strength. This may be the 4. Goutham Kumar Pothal & G.V.Rao (2008),Model
reason why the CBR of CB face is significantly studies on Geosynthetic reinforced Pond Ash,
higher than that of CT face. JGE,Vol. 13, Bund. H
5. IS:2720 (Part-16)-1979, Laboratory Determination of
4. As seen from the table, the moisture content of the CBR, (first revision), Bureau of Indian Standards.
CB face is slightly higher than the OMC and that on 6. Jakka RS, Datta M, Ramana GV (2010) Shear
the CT face is lower than OMC. This may be due to Behaviour of Loose and Compacted Pond Ash,
the fact that, the moisture held inside the cavity of Geotech Geol Eng (2010) 28:763–778
coarser pond ash particles may have flown out 7. Kim B, Prezzi M, Salgado R (2005) Geotechnical
when it is crushed and under the impact of blows, it properties of fly and bottom ash mixtures for use in
migrated downwards in to the CB face. This is highway embankments, J Geotech Geoenviron Eng-
justified by the fact that, in the compaction test, ASCE 131(7):914–924
when moisture is added more than OMC, bleeding 8. Pandian NS, Sridharan A, Chittibabu G (2001) Shear
of water is observed from the base plate. This strengthof coal ashes for geotechnical applications.
increase in moisture content has apparently no In: Proceedingsof the Indian geotechnical conference,
influence on the CBR value as the medium is non- Indore, India, pp 466–469
plastic. 9. Sridharan A, Pandian NS, Rajasekhar C (1996)
Geotechnical characterization of pond ash, Narosa
Publishers, Delhi, pp. 97–110.
CONCLUSIONS
Based on the investigations carried out in this project, the
following conclusions are made
1 The pond ash consisting of clinker like particles
with intra cavities, is subjected to high degree of
crushability which amounts to variation in
density levels within a compacted sample.
2 There is considerable variation with in the CBR
value of a pond ash sample when tested on
compaction bottom face and compaction top face.
3. Though migration of water downwards is
noticed, it has no significant influence on CBR
value of the sample as the pond ash is non-
plastic.
4. It is therefore advisable to consider the CBR
value of compaction top face in the design as it
resembles the field condition effectively.

ISBN 978-81-921121-0-7
PERFORMANCE HISTORY OF WARM MIX ASPHALT

Rajan Choudhary
Assistant Professor, IIT Guwahati, Guwahati – 781039, Email: rajandce@iitg.ernet.in
Eleena Gao
M. Tech Student, IIT Guwahati, Guwahati – 781039, Email:elena@iitg.ernet.in;eleena.gao@gmail.com
Ashok Julaganti
PhD Student, IIT Guwahati, Guwahati-781039, Email: j.ashok@iitg.ernet.in
ABSTRACT: With immense rise in the human activity, the composition of the atmosphere started changing and as a
result of this authorities in various countries are creating legislation and incentives to reduce global warming. The
increase in environmental awareness and strict regulations on emissions compelled the Hot Mix Asphalt (HMA)
industry also to reduce emissions during HMA production. Warm mix Asphalt (WMA) is an asphalt mix which is
mixed and compacted at lower temperatures compared to that of HMA. Lowering of about 20 to 55 °C in mixing and
compaction temperatures can be achieved with warm mix asphalt. The potential benefits such as improved compaction,
extended haul distances, reduced energy consumptions and emissions etc. have created an interest of researchers in the
WMA. WMA is been gaining popularity in the recent years. A number of technologies for producing WMA have been
developed and many test sections have also been constructed in some countries. Based on the data collected from these
test sections, the performance of WMA mixes is found comparable with HMA. This paper describes the data of some
test sections of WMA constructed in different countries.
INTRODUCTION warming, the Kyoto protocol was signed in 1992 to

Construction sector is the second largest employer in reduce the emission of greenhouse gases. These had
India after agriculture. Road and infrastructure influenced the road building and production industry too.
construction is wheel of Indian economic growth. Road Asphalt industry is trying to lower its emissions by
transport completely dominates the transport sector in reducing the mixing and compaction temperatures of the
India with a share of 4.6 percent in India’s Gross asphalt mixtures without affecting the properties of the
Domestic Product (GDP) in 2007-2008 as per the data mix. The technologies to decrease the energy
released by the Central Statistical Organisation (CSO). In consumption and emissions associated with conventional
India 95% of road network is occupied by bituminous* hot mix asphalt production have created an interest of
pavement only (https://globaljournals.org). Hot Mix researchers in the Warm Mix Asphalt (WMA).
Asphalt (HMA) mainly used as a paving material in India
consists of aggregate and asphalt binder which are heated Warm Mix Asphalt (WMA) refers to technologies that
and mixed together. HMA is generally produced by batch allow for a significant reduction of mixing and a
or drum mix plants at a discharge temperature of 138oC to compaction temperature of asphalt mixes. These
160oC. During construction, the temperature must be high technologies tend to reduce the viscosity of the asphalt
enough to heat the aggregate and bitumen for proper and provide complete aggregate coating at lower
coating and to achieve desired workability. Due to the temperatures. WMA is generally produced at
need of high temperature for the production of HMA, cost temperatures 20 to 55 °C lower than typical HMA.
of the energy is significantly large because high demand Reduced mix production and paving temperatures
of fuel is needed for heating the aggregate and asphalt. decreases the energy needed to produce HMA, reduces
Even higher temperatures are used for HMA mixtures emissions and odours from plants, and provide better
containing polymer-modified asphalt binders and crumb working environment at the plant and paving site.
rubber modified asphalt binders. Moreover, emissions Manufacturers and material suppliers have reported that
from the HMA plants itself are very high. The the cost of energy can be reduced by 30% and also carbon
Environmental Protection Agency (EPA) estimates that, dioxide emissions by 30% (You and Goh, 2008). The
the production of 2 lakh tons of asphalt mix in a drum decrease in emissions from the asphalt production plant
mix asphalt plant in a year would emit about 13 tons of facilitates the asphalt plant to be located in and around the
carbon monoxide (CO), 5 tons of volatile organic non-attainment areas. Lower production temperatures for
compounds (VOC), 0.4 tons of sulfur oxides, 2.9 tons of asphalt paving mixtures will decrease the aging of the
nitrogen oxides and about 0.65 tons of total Hazardous asphalt binder associated with HMA production. This
Air Pollutants (HAP) during that period (US EPA Report, decrease in aging can improve thermal and fatigue
2000). cracking resistance.
Today’s versions of WMA are the brainchild of the
With the rise of human activity, the composition of the Bitumen Forum of Germany in 1997 (Zettler, 2006).
atmosphere started to change. The need of conservation European interest in reducing the emission of green house
of non- renewable resources and energy, environmental gases was mainly a result of the Kyoto agreement.
pollution and poor working environment are becoming Europeans started using these technologies to lower
the global issues. With increasing concerns of global mixing/paving temperatures, and the results are also

promising (Barthe and Von Devivere, 2003). Warm mix than the HMA, as indicated by decreased stiffening. It
asphalt was introduced to the United States (U.S) when was found that there was no difference in performance
the National Asphalt Pavement Association (NAPA) grade between the HMA and WMA produced using the
sponsored an industry scanning tour to Europe for the Evotherm emulsion.
asphalt paving contractors in 2002. In U.S., these
technologies were tried as an attempt to see the potential WMA with the addition of Aspha-min, a synthetic zeolite
benefits (Kuennen, 2004). successfully incorporates with a higher percentage of
RAP materials than HMA does. Aspha-min was added to
PERFORMANCE HISTORY OF WARM MIX a Superpave mixture containing 20% RAP during a
ASPHALT demonstration project in Orlando, Florida. It was
Warm Mix Asphalt (WMA) is gaining popularity in the designed for 3 to 10 million (equivalent single axle loads
recent years. Recent interest in WMA has lead to several (ESALs) with an Ndesign of 75 gyrations. The addition of
countries hosting WMA field demonstrations. Most of Aspha-min was able to reduce the production and
these projects are evaluating different aspects of this new compaction temperatures by 36°C, while yielding the
technology. Many researchers and users are interested in same in-place density. According to Hurley and Prowell
the long term performance of WMA. The result obtained (2005a), no distress was observed in the section. It was
from some WMA trial section shows equal or better also found that there was no difference in moisture
performance than HMA. However, there is still a damage resistance between the WMA and the control
significant challenge ahead to move WMA from trial mix.
projects to main stream pavement products. Some of the
field performances of different projects of Warm Mix Alabama Department of Transportation in 2007 hosted a
Asphalt are described below: warm mix asphalt field demonstration in Tarrant City.
This field demonstration was evaluated by the National
The first pavements incorporating WMA additives were Center for Asphalt Technology (NCAT). The mixes used
constructed in Europe in 1997. In 2007, the FHWA in the study consist of Reclaimed Asphalt Pavement
International Scanning Tour team collected laboratory and (RAP) and the WMA additive used was EvothermTM
short-term field performance data from some of the oldest Dispersed Asphalt Technology (DAT). Two mixes, WMA
WMA pavements in Europe. The scan team viewed six and HMA, were produced for demonstration. Based on
WAM-Foam pavement sections in Norway with average the study, they concluded that the WMA was more
daily traffic ranging from 3,500 to 25,000 vehicles. susceptible to rutting than the HMA but show acceptable
Performance of the sections has been mixed. Where poor rut depths for low volume roads. They also concluded that
performance occurred, it was attributed to factors other the WMA was more moisture susceptible and less stiff
than the use of WAM-Foam. WAM-Foam sections than HMA. The creep compliance testing showed that the
appeared to perform similarly to previous HMA overlay. WMA is less resistant to load-induced damage than the
Also, good performance was achieved for a section with HMA. The site revisits indicated that both the HMA and
15 percent RAP. WMA showed segregation. After one year, it was found
that the indirect tensile strengths of the WMA and HMA
Eurovia constructed test sections with WMA mixes using were similar. It was observed that the bond between
Aspha-Min (Zeolite) and HMA control sections in pavement lifts is not affected by the use of WMA.
Europe. Barthel et al. (2004) reported that 3 years after
construction of the first WMA test section using zeolite The Nebraska Department of Roads (NDOR) initiated the
measurements indicated no significant changes in the WMA field trial in 2007 using different amounts of a
surface characteristics. wax-type WMA additive called Sasobit. In 2008, NDOR
paved four trial sections, installing two WMA pavements
In the year 2006, three WMA trial sections were (Evotherm WMA and Advera zeolite WMA) and their
constructed in various locations in Virginia. Two WMA control HMA sections in Antelope County, Nebraska. The
sections were constructed using Sasobit and the third field performance data indicated that, for the two-year
section was produced using Evotherm ET. The sections public service after placement the rut depth and
were evaluated over a period of 2 years to assess the International Roughness Index (IRI) of both the WMA
initial performance of the WMA and compare it with that and HMA sections were insignificant, compared to the
of hot-mix asphalt (HMA) control sections constructed at typical failure criteria. Both WMA and HMA trial
the same time. Visual inspections and coring were sections showed similar good performance without raising
performed during the initial construction. It was observed any major concerns.
that there were no significant distresses in either the
WMA or HMA sections during the first 2 years in service. Washington State Department of Transportation
The air-void contents of the core indicated that the air (WSDOT) in 2008 rehabilitated a pavement on Interstate
void contents for the WMA and HMA were not 90 between the Columbia River and George town.
significantly different in each trial. Performance grading Rehabilitation was done by grinding the existing
of the recovered binder suggested that the WMA pavement to a depth of 0.25 feet and inlaying with the
produced using Sasobit aged at a slightly reduced rate same depth of HMA or WMA. Both the mix design for

the HMA and WMA were same except the addition of period. One of the WMA sections showed good rutting
Sasobit in the WMA. The RAP was recovered from the performance even though traffic was returned to the
grinding of 0.25 feet of the existing pavement. The results sections less than 2 hours after paving commenced.
indicated that density with WMA is same as HMA even at However, laboratory tests indicated an increased potential
lower compaction temperatures. It was also reported that for moisture damage with the WMA mixtures.
the stiffening affect of Sasobit was similar to stiffening
observed on other projects. In late 2009, it was estimated that over 1 million tons of
WMA had been placed and another 1 million tons were
Prowell et al. (2007) reported findings from accelerated under contract in Texas. The Texas Department of
loading tests on WMA sections containing Evotherm at Transportation has used so much WMA that it no longer
the NCAT Test Track. The Test Track was constructed in tracks quantities and now routinely allows the use of
Opelika, Alabama near Auburn University in 2000. This WMA as a contractor option. The Double Barrel Green
test track is a 2.7 km oval track consisting of 45 different additive was used to produce the WMA. The mix
flexible pavement sections. Evotherm was placed on three production temperature of 132°C was about 28ºC lower
test sections (each section 61 m in length) for than that used with conventional HMA and resulted in 18
rehabilitation of NCAT Test Track. Fig 1 shows a section percent savings in burner fuel costs. Placement
view of the layers. Sections N1 and N2 were milled to a temperatures ranged between 105°C and 122°C. Results
depth of 127 mm to complete the structural rehabilitation from this project were also positive.
of those sections. Two lifts of 19.0-mm Nominal
Maximum Aggregate Size (NMAS) WMA were placed in From the studies discussed in this paper the main
Sections N1 and N2. The binder for the 9.5-mm nominal advantages found of using WMA are:
maximum aggregate size surface mix was varied for each  Low production temperature will lower fuel
section. An HMA control containing a PG 67-22 binder usage and decrease emissions. This would also
(similar to AC-30) was placed in section N2. The binder result in improved working conditions at plant
for section N1 was the Evotherm emulsion produced and paving site.
using the same PG 67-22 base asphalt as used in Section  Producing WMA at HMA temperatures will
N2 plus 3 percent, by weight of binder residue, latex. All permit an extension in the hauling time and
mixtures contained a blend of granite, limestone, and compaction. This also extends the paving season.
coarse sand. The surface mix for section E9 was the  With lower mixing temperature, the short term
Evotherm emulsion produced using the same PG 67-22 aging of the binder will be significantly reduced.
base asphalt as used in Section N2.  Use of WMA is beneficial for stiff mixes, RAP
mixes and low-temperature paving.
 WMA mixes have comparable performance with
HMA mixes.
WARM MIX ASPHALT IN INDIA

Due to the potential benefits of WMA technologies, many
researchers in India are interested to study on these
technologies on Indian perspectives. Few trial sections
have been constructed in some parts of India. In one of
the field trial section, Evotherm warm asphalt technology
was used on a 400m stretch of three lanes on National
Highway – 10 near Rohtak, Haryana. It was found that the
mixing and compaction temperatures were reduced by
about 30°C relative to the conventional hot-mix asphalt.
Fig 1: Section View of Evotherm Layers at NCAT Test WMA were able to compact to a density equivalent to the
Track (Prowell et al., 2007). conventional hot-mix asphalt at lower production and lay-
down temperatures. At another trial conducted near
Before opening the sections to traffic, three cores were Chandanagar, West Bengal, Evotherm warm mix
taken from sections N2 and E9 and four cores from produced at 120oC was used and no fumes were observed,
section N1. It was reported that in-place densities of the thus giving a more pleasant work environment with a
WMA surface layers were equal to or better than the reduced impact on surrounding communities.
corresponding HMA test sections, even though
compaction temperatures were reduced by 8 to 42°C. The Despite the promising benefits, the industry and many
Asphalt Pavement Analyzer (APA) results indicated researchers have been concerned to implement WMA
similar performance for the WMA and HMA surface techniques into actual practice. Moisture susceptibility
mixtures which contained PG 67-22 base asphalt. The two has been a primary concern for some WMA approaches.
WMA sections and corresponding HMA section This is because lower temperatures in the process of
demonstrated excellent field performance in terms of mixing and compaction could result in incomplete drying
rutting after the application of 515,333 ESALs in a 43-day

of the aggregate, compromising the bond between asphalt Kuennen, T., “Warm Mixes are a Hot Topic,” Better
and aggregate. Roads, James Informational Media, Inc., Des
Plaines, Illinois, June 2004.
CONCLUSIONS Kvasnak, A. and Prowell, B., “Warm Mix Asphalt
Warm mix asphalt is a promising approach to production Field Study: Final Report,” National Center for
and placement of paving materials. The technologies to Asphalt Technology, Report 10-XX, May 2010
decrease the energy consumption and emissions Newcomb, D., “An Introduction to Warm Mix
associated with conventional hot mix asphalt production Asphalt,” National Asphalt Pavement
have made an interest in the warm mix asphalt. The Association, Lanham, Maryland, 2006.
WMA revolution is sweeping across the world. Research Prowell, Brian D. and Hurley, Graham C., “Warm-
work worldwide is evidently demonstrating that WMA Mix Asphalt: Best Practices,” Quality
systems are providing significant benefits with regards to Improvement Series 125 (QIP-125), National
the environment, in facilitating paving practices and, with Asphalt Pavement Association, December 2007.
regards to field performance. However, there is still a Prowell, B., Hurley, G., and Crews, E., “Field
significant challenge ahead to move WMA from trial Performance of Warm Mix Asphalt at the NCAT
projects to main stream pavement products. Test Track”, Presented at the 86th Annual
Meeting of the Transportation Research Board,
REFERENCES Washington, DC, January 2007.
Barthel, W., Marchand, J., and Von Devivere, M., United States Environmental Protection Agency,
“Warm Asphalt Mixes by Adding a Synthetic (2000), “Hot Mix Asphalt Plants – Emissions
Zeolite.” 3rd Eurasphalt and Eurobitume Assessment Report”, Emissions Monitoring and
Congress, Vienna, 2004. Analysis Division, Office of Air Quality
Button, J.W., C. Estakhri, and A. Wimsatt, “A Planning and Standards, United States
Synthesis of Warm-Mix Asphalt,” Report Environmental Protection Agency, Research
FHWA/TX-07/0-5597-1, Texas Transportation Triangle Park, NC, December 2000.
Institute, Texas A&M University, College You, Z. And Goh Shu W., “Laboratory Evaluation
Station, Texas, 2007. of Warm Mix Asphalt: A Preliminary Study”,
D Angelo, J.,“Warm Mix Asphalt European International Journal of Pavement Research and
Practice.” Report No. FHWA-PL-08- 007, Technology, Vol.1 No.1, January 2008
Washington DC, February 2008. Zettler, R., “Warm Mix Stands Up to Its Trials,”
Diefenderfer, S.D., and Hearon, A.J., “Performance Better Roads, James Informational Media, Inc.,
of Virginia’s Warm Mix Asphalt Trial Sections Des Plaines, Illinois, February 2006.
”, VTRC 10-R17, Virginia Transportation
Research Council, Charlottesville, Virginia,
February 2010
Hurley, Graham C. and Prowell, Brian D.,
“Evaluation of Aspha-Min Zeolite for Use in
Warm Mix Asphalt”, National Center for
Asphalt Technology, NCAT Report 05-04, June
2005.
“Evaluation of Sasobit for Use in Warm Mix
Asphalt”, National Center for Asphalt
Technology, NCAT Report 05-06, June 2005.
“Evaluation of Evotherm for Use in Warm Mix
Asphalt”, National Center for Asphalt
Technology, NCAT Report 06-02, June 2006.
Hurley, G. and Prowell, B., “Evaluation of Potential
Processes for Use in Warm Asphalt Mixes”,
Journal of the Association of Asphalt Paving
Technologists, Vol. 75, 2006.
Khandhal, P.S., “Warm Mix Asphalt Technologies:
An Overview,” Journal of the Indian Road
Congress, Paper No. 561, Vol. 71-2, 2010.
Kim, Y.R., Zhang, J., and Ban, H., “Implementation
of Warm-Mix Asphalt Mixtures in Nebraska
Pavements”, MPM-02, University of Nebraska-
Lincoln, Lincoln, November 2010.

ISBN 978-81-921121-0-7
DETERIORATION OF GROUND QUALITY DUE TO OPENCAST

COAL MINES AND OTHER RELATED INDUSTRIES IN SINGRAULI
AREA
Raj K. Sharma, Department of Mining Engineering, IT-BHU, Varanasi
C.S. Singh, Department of Mining Engineering, IT, BHU-Varanasi.
ABSTRACT: Opencast mining and related industries have significant impact on the ground water quality. The
total suspended solids is a common pollutant in the most of the streams after mixing with the mine and different
industries effluents. The concentration of toxic heavy metals in concentration due to industrial activities is much
higher than the permissible limits in the study area This contamination is much dangerous for the human health
and our environment also.
INTRODUCTION around the mining areas. It starts from the mineral

During the process of opencast mining, a variety of exploration stage when forests are cleared, roads
rock types with different composition are exposed are built and drilling is commenced. It continues
to atmospheric condition and undergo accelerated
through the exploration, beneficiation and the
weathering. The different mineralogical component
of the rocks chemically decomposes and under utilization stages. Mining operation has made rapid
varying conditions would produce alkanity or strides in the last fifty years and brought economic
acidity. The development of natural resources is a gains of the country, but side-by-side also led to the
compulsion for any progress- oriented society. The pollution of the environment including
mining activity provides raw materials for all the contamination of the water sources, rendering
metallurgical industries and help to generate harmful effects of human health, plant life,
revenue by giving export items etc. Increase in
microorganisms, flora and fauna in aquatic life.
mining activities is therefore, stepping stone in the
development process of any nation (Ghosh, 1993). Thus stepped up mining activities for
Mining is second after agriculture as the world’s industrialization and consequent pollution may be
oldest and the most important industry. Mining treated as the two sides of the same coin. In broad
industry has a long history in India. Even during prospective “Pollution means such contamination
Vedic times, a great importance has been attached of water or such alteration of the physical,
to this industry as a source of revenue (Rao, 1993). chemical, biological properties of water or such
The independence of primitive societies upon
mined products is illustrated by the nomenclature discharge of domestic sewage or industrial
of different epochs like stone age and iron age, a effluents or of any other liquid, gaseous or solid
sequence that also reflects the increasing substances into the water as major is likely to
complexity of relationship of society with mining create a nuisance or render such water harmful or
activities (Desai, 1993) injurious to public health or safety, or to domestic,
commercial, industrial, agriculture or other
legitimate uses, or to life or health of animals or
plant or of aquatic organisms. (Govt. of India 1974)
Keeping above facts in mind the present study has

been conducted in Singrauli coal field area which is
one of the biggest opencast coal mining complex in
the world to determine impact of opencast coal
mining on surface and ground water quality.
ENVIRONMENTAL PROFILE OF
SINGRAULI (U.P.) AREA
It is well known that mineral resources Singrauli- The problem area, as far as identified
development results in environmental degradation. industries are concerned, lies in the state of Uttar
The mining activities lead down to serve Pradesh. On the east, the area has its boundaries
with the Bihar State, while it shares its southern
contamination of all the three meso-environmental,
border with M.P.State. The area comprises a part
i.e., atmosphere, hydrosphere, and lithosphere of erstwhile district of Mirzapur in south east of

Uttar Pradesh. It encompasses a vast geographical the area and rank next to alluvial formations, in
area extending from Obra in the north to Anpara in terms of groundwater potential. In general, the
district Sonebhadra and further south to Bina depth of water table ranges between 4 and 6 m.
collieries. On geographical consideration, the area below ground level (BGL). Rainfall is the main
lies between 200o 10′ – 240 30′ North latitude and source of groundwater recharge, apart from seepage
8200 47′ – 8300 15′ East longitude. Although the from the surface water bodies. The general
area is characterized by an uneven topography, the hydraulic gradient prevailing in the area is towards
average altitude in the area is 270-280 m. above River Sone or locally towards the subsidiary
MSL. However, it ranges between 262 m to 474 m surface water bodies. More or less, the flow of
above MSL. The area witnesses a semiarid climate groundwater is in conformity to the surface
with annual rainfall ranging between 760-1300 topography.
mm. The area is characterized by proven natural
resources of coal, limestone and building stone. SOURCES OF POLLUTION IN SINGRAULI
Predominantly hard-rock topography and limited AREA
water resources in the area have ultimately resulted As a point source of groundwater pollution in
into the situation wherein 40 percent of the total Singrauli problem area, cement industries, stone
geographical area lies under forest cover, while crushers and thermal power plants are insignificant,
only 20-30% of the total area is under cultivation. however during the coal mining operation, the
From the water resource point of view, the area has water may be contaminated by coal slurry, and
river Sone as a major surface water body on the other metallic and non-metallic proportions, is
northern fringe. Besides, there are five medium liable to disturb the aquatic ecosystem of the
sized streams namely Belan, Karamnasa, surrounding areas, if enough precautions towards
Chandraprabha, Rihand and Kanhar, which, swell treatment of the effluent and reclamation activities
considerably during the rainy season and shrink to are not taken. The discharge of effluent from
near insignificant flow during summer. These aluminum industries and chemical industries act as
streams are mainly flowing in NE -SW direction. a potential source of pollution of sub-surface water
For raw water requirement, the area more or less on account of mercury, fluoride, titanium, iron,
depends on Rihand reservoir across river Renu, or aluminum and pesticides. Besides, the problem
on river Son and its subsidiaries besides tapping associated with disposal of red mud, mercury
sub-surface water resources. sludge and such other refuse also carry a significant
potential of surface and groundwater pollution. One
GEOLOGY & GROUNDWATER important factor worth mentioning is the presence
The area is largely covered by hard rock formations of Rihand reservoir, a fresh water resource for the
belonging to Vindhyan and Precambrian groups, entire area, which has become a waste disposal site.
which have comparatively low groundwater Besides ensuring the availability, almost stagnant
potential in general. The groundwater utilisation water body of the reservoir, with very high
and development in the district is rather poor. hydraulic gradient, pose a major threat in
Groundwater in alluvial area generally occurs significantly increasing the salinity of the top soil
under unconfined and semi confined conditions. In as well as groundwater in semi confined,
deeper aquifers, groundwater occurs under highly unconfined and perched aquifer conditions. In the
confined conditions. From the point of view of areas where critical safe balance between draft and
groundwater occurrences, the lithological units recharge is not maintained (groundwater mining),
within the area are broadly classified into five chances of reservoir water contaminating the
groups, comprising - Granites - Gneisses; Phyllites- nearby groundwater cannot be ruled out.
Shales; Limestones; Quartizite – Sandstone and
alluvium. Groundwater occurs in the pore spaces of Major industries in Singrauli Area
the unconfined alluvial / weathered (in hard rock
terrain) material. The near surface sediments S.no. Name of Industry Major environmental issue(s)
comprise fine to medium sands with inter-layered
pockets of clayey material. This top un-confined
aquifer supports large number of dug wells and 1. HINDALCO Air and water pollution due to
Industries, Renukoot fluoride and disposal of red mud
other shallow groundwater structures. In hard rock
2. Kanoria Chemical Water pollution due to
formations like granites, phyllites and quartzites Industries, Renukoot pesticides, chlorine and mercury
etc. groundwater occurrence is limited to weathered in wastewater and sludge
zones and plains of foliation or weak fracture 3. Hi-Tech Carbon, Air dispersion of carbon ;oil &
zones. The depth to water table ranges from 2 to 24 Renukoot grease in wastewater
m. On the whole, these formations have moderate 4. Northern Coal Fields Mine water discharge, oil &
Ltd., Site, Bina grease from workshop, Air
to low groundwater potential with an exception in pollution
case of quartizites and sandstone which 5. Northern Coal Fields Mine water discharge, oil &
occasionally form good aquifers (unconfined) in Ltd., Site , Kakri grease from workshop, Air

pollution GLOBAL FRESH WATER
6. Northern Coal Fields Mine water discharge, oil & DISTRIBUTION(TODD, 1970)
Ltd., Site , Khadia grease from workshop, Air
pollution
7. Northern Coal Fields Mine water discharge, oil & Location Volume % of total
Ltd., grease from workshop, Air (x102 m3) volume on
Site : Dudhichua pollution - Globe
8. Rihand Super Air Pollution; Ash Pond
Thermal Power Plant Overflow
Fresh water lakes 125 0.009
(NTPC) Rivers 1.25 0.0001
9. Renusagar Power Co. Air pollution; Disposal of fly – Soil moisture 67 0.005
Anpara ash Ground water 8350 0.61
(Captive unit for
HINDALCO) (Average Depth
10. U.P.State Cement - Currently started of 4000m)
Corporation, Churk Ice caps & 29200 2.14
11. U.P.State Cement - Currently started Glaciers
Corporation, Dala
Total on land 37743.25 2.7641
12. U.P.State Electricity Air Pollution;Ash Pond
Corporation, Obra Overflow area
13. Anpara Thermal Air Pollution; Ash Pond Atmosphere 13 0.001
Power Station Overflow Total in all 37756.25 2.7651
14. Singrauli Super Air Pollution Ash Pond locations
Thermal Power Plant Overflow
15. Stone Crushers – Air Pollution & water pollution
cluster of appx 150 MATERIALS AND METHOD
nos. The ground water samples have been collected
from in and around Singrauli area during summer,
WATER QUALITY AND ITS rainy season and winter 2010. Samples are
DETERIORATION collected in clean and clear plastic bottles having
Water quality is defined in terms of many air tight caps. The measurements of pH, TDS,
parameters, which can be classified broadly as Salinity, Dissolve oxygen, are taken with the help
of water analysis instrument kit. Other parameters
physical, chemical, bacteriological and
like Carbonate, Bicarbonate, Chloride, Sulphate,
radiological. These parameters must first be Nitrate, Calcium, Magnesium, Sodium, Potassium,
analysed to identify their right source and later, iron are determined in the laboratory.
need to be continuously monitored at the source at
the end of the production and supply line. Settling RESULT
of water quality standards and regulations alone The mining and other industrial activities have
cannot help to achieve the water quality, but what affected the water quality which may impact the
is required is a limited time regulations within physiochemical characteristics of soil and the
which the changes envisaged must be implemented. plants cellular structures & functions particularly in
If the world’s water supply were compared to one vicinity of the emission sources. The herbs with
litre of fresh water would make about 31.05 medicinal values are also being destroyed due to
mililitres or about 3.1% of the total volume which water pollution. The results of water analysis have
is readily accessible fresh water. It makes less than been summarised as given below.
one drop (Miller, 1998). Human beings already put
The physiochemical quality of ground water in
approximately 54% of all accessible surface water
Singrauli coal field area:
to use, which is expected to increase to 70% by
2025 (Postel Daily & Enrich,1996). The total Para Away from mines Near mining areas
quantity of fresh water on earth can satisfy all the meter related area
needs of human population, if it would have been s
evenly distributed and was easily accessible. Out of Mg/l Sum Rai Wi Sum Rai Wi
world’s fresh water resource about 75% is held mer ny nte mer ny nter
within ice sheets and glaciers, nearly 24.5% in r
ground water and the remaining small quantity PH 7.31 ≈ 7.3 6.37 ≈6.8 6.09
7.86 4 - 6- -
occurs in rivers (0.03%), in lakes (0.3%) and in the 8.36 8.94 7.92
atmosphere (0.06%). (Parivesh News Letter, 1995). TDS 186. ≈15 153 102. ≈11 110.
0 9.0 .20 64- 2.61 82-
908. - 161
16 1512 2.16

.16 of trace inorganics from mine dumps to surface
Carbo 12.0 ≈12. 01. 23.0 ≈23. 24.0 water sources.
nate 1 0 12
Bicarb 121. ≈12 108 423. ≈42 492. REFERENCES
onate 01 2.08 .70 12 8.13 10 Central Pollution Control Board Report 1996, pp
Chlori 46.0 ≈48. 35. 176. ≈18 191. 261-264
de 0 0 1 7.12 20
Desai M (1993) Environmental impact of mining in
Sulph 4.8 ≈4.9 11. 231. ≈27 282.
the Himalayas- a study on Sikkim Himalayas in
ate 5 70 2.12 02
Nitrat 2.9 ≈3.0 2.7 117. ≈10 109. Envirnmental issue in mineral resources
e 0 2.70 60 development.
Calciu 25.7 ≈26. 20. 64.2 ≈74. 82.3 Govt of India(1974) The Water (Prevention and
m 2 12 25 20 6 Control of Pollution) Act.
Magn 16.8 ≈16. 12. 69.2 ≈65. 64.2 Ghosh (1993) Damage to natural sources in jharia
esium 9 01 01 21 5 coal field through half of the century up to
Sodiu 21.0 ≈21. 17. 117. ≈15 165.
1974. In environmental issues in mineral
m 9 79 89 0 8.23 21
resources development.
Potass 2.91 ≈2.6 2.1 36.4 ≈34. 30.0
ium 9 7 0 02 8 Miller GT (1998) Living the environment 10th
Iron 0.19 ≈0.2 0.1 6.09 ≈6.9 8.09 Edition Wardsworth publishers, Belmont
0 6 2 Calofornia.
Total 138. ≈12 108 81.0 ≈82. 84.9 Postel Daily & Enrich (1996) Human appropriation
hardne 25 9.16 .96 - 23- 3- of renewable fresh water science Vol 271, pp
ss 364. 423. 453. 785-788
91 89 29
Privesh news letter (1985) Assessment and
It is evident from the water analysis that pH of
development study of river basin. Editorial
water away from mining area is varies from neutral
central pollution control board. Vol 1 pp4
to basic whereas in mining areas it varies from
Todd DK (1970) The water Encyclopedia “Water
acidic to basic. Similarly, TDS & Carbonate,
information centre post Washington, NewYork.
Bicarbonate, Chloride, Sulphate, Nitrate, Calcium,
Magnesium, Sodium, Potassium, Iron, is much less
in water samples collected from away areas as
compared to near mining areas.
CONCLUSION
It is observed that the quality of water in the stream
flowing through the mining regions is quite good
before confluence with the effluent, but after
mixing the mine water stream in the downstream
side have been found affected by pH and total
suspended solids. The ground water in the mining
areas is having higher concentration of Carbonate,
Bicarbonate, Chloride, Sulphate, Nitrate, Calcium,
Magnesium, Sodium, Potassium, Iron, as compared
to ground water away from the mining areas. In the
present study the mining activity significantly
changes the water quality of the area.
The mining and related industry professionals need
to urgently find the ways of carrying out the
operations with the minimum possible adverse
impact on the environment. If such damage remains
inevitable steps should be taken to reclaim the land
and afforest the areas as early as possible after
mining is over to avoid the erosion of soil, silting,
or nearby reservoir and to arrest faster movement

ISBN 978-81-921121-0-7
STABLE ISOTOPIC CHARACTERISATION OF GROUNDWATERS IN

BIST- DOAB REGION, PUNJAB
P. Purushothaman
Research Associate, H.I.D., National Institute of Hydrology, Roorkee- 247 667, Email: purus26@gmail.com
M.S. Rao
Scientist ‘E1’, H.I.D., National Institute of Hydrology, Roorkee- 247 667, Email: somesh@nih.ernet.in
B. Kumar
Scientist ‘F’ & Head, H.I.D., National Institute of Hydrology, Roorkee- 247 667, Email: bk@nih.ernet.in
Y.S. Rawat
S.R.F., H.I.D., National Institute of Hydrology, Roorkee- 247 667, Email: yadhvir2007@rediffmail.com
Gopal Krishan
Res. Scientist, H.I.D., National Institute of Hydrology, Roorkee- 247 667, Email: dr.gopalkrishan@gmail.com
ABSTRACT: A study was conducted to characterize groundwaters (shallow and deep) in the Bist- Doab region of
Punjab using stable isotopes. The Bist- Doab region is very complex with varied geomorphology and drainage
system and hence is necessary to characterize the groundwater for better understanding of the recharge sources.
Samples for analysis were collected from the shallow and deep groundwater from the Bist- Doab region. The
isotopic composition δ18O of shallow and deep groundwater varied from -4.13‰ to -8.93‰ and -26.03‰ to –
61.11‰ (shallow) and -5.72‰ to -12.12‰ and -38.06‰ to -84.07‰ (deep aquifer) in 18O and D respectively.
The study shows that shallow groundwater gets recharged from precipitation at northwestern and central part of the
study area. The locations adjoining the Bist- doab canal (from R. Satluj) and R. Beas & Kandi canal (R. Beas) show
similar composition showing significant contributions from these sources to the groundwater recharge. The Deep
groundwater show isotopic composition similar to that of the precipitation.
Key words: Stable Isotope, Groundwater, Isotopic characterisation, Bist- Doab, Punjab
INTRODUCTION in the region has gone up to 300% with almost 80% of

As an alternative to a purely hydrodynamic approach, the region showing the stage exceeding 100 % (CGWB
geochemical and isotopic investigations of aquifer 2007). The high amount of over exploitation of
systems have often proven to be quite successful in groundwater in this region has raised an alarming
determining groundwater recharge and flow paths situation for the identification of groundwater recharge
(Allison and Hughes 1978, Chandrasekharan et al. sources in the region. The present study is aims to
1988, Edmunds and Gaye 1994, Adar and Leibundgut characterize groundwater by stable isotopes for
1995, Van Warmerdam et al. 1995, Singh et al. 1996, understanding the source of recharge.
Sukhija et al. 1996, Kaplan et al. 1997, Hunkeler et al.
2004, Schmidt et al. 2004). Isotope measurements STUDY AREA
complement hydrogeologic observations by providing The Bist- Doab is a triangular region covering an area of
direct evidence for groundwater sources and transport 9060 sq.km lies between 30° 51′ and 30° 04′ N latitude
rates. This has been demonstrated for many simple and 74° 57′ and 76° 40′ E longitude (Fig. 1). It
groundwater flow systems (e.g., Stute et al. 1992, comprises of the districts Jalandhar, Kapurthala,
Ekwurzel et al. 1994). Nawanshahar, and Hoshiarpur of Punjab State. The
The Bist- Doab region is an interfluve region between region is bounded by Siwalik’s (360- 900m amsl) all
R. Sutlaj and Beas (Doab: local word for interfluves) along its NE-SE portion, the river Beas in the north and
occupying northwest region of Punjab. It is west direction and the river Satluj in SE and southern
experiencing high amount of groundwater depletion boundary. The two rivers coalesce at Harike. The area is
despite input from two perennial rivers. Due to drained by the river Satluj and Beas, with dense
increasing agricultural activity, water demand in the drainage density at the NE strip bordering Siwaliks and
region is increasing day by day putting more pressure moderate to low in the rest of the area with sub-parallel
on groundwater causing groundwater mining and and sub-dendritic patterns. The region experiences an
exploitation. The irrigation activity in this region is average of 715 mm annual rainfall with a maximum at
done through canal systems and groundwater. But there Siwaliks (Kandi) (Chopra and Sharma 1993).
has been decrease of over 35% in the use of canal
system in the state for irrigation purpose with increase
in usage of groundwater (Tiwana et al. 2007). Due to
these intense usage of groundwater the groundwater in
all the blocks of the districts in the Bist- Doab region
has been experiencing high groundwater draft and have
resulted in overexploitation with high groundwater stage

(a)
INDIA
Hoshiarpur
Kapurthala
Jalandhar
Nawanshahr
Harike
PUNJAB
R. Satluj Ropar
Fig. 1 Location map of the study region
Geological Setting
Geomorphologically the region is divided into three
zones: the Siwaliks, Kandi watershed, and flood plain
area (Bowen 1985). The area, part of Indo- Gangetic (b)
plain, receives huge pile of sediments of Pleistocenece
to recent age due to erosion in the mountains and
deposited by the rivers Satluj, Beas and other streams
draining the area. The relevant lithologies and
sequences are as follows:
- Surface deposits (Recent).
- Sirowal sediments and occasional gravels to the
northeast with Kandi coarse clastics including red clay
beds to the southwest (Holocene).
- Boulder beds with interbedded clays (Pleistocene).
SAMPLING AND ANALYSIS

The water samples were collected randomly from the
entire study area to avoid any biasness in the analysis.
The groundwater from the hand pump and peizometers Fig. 2 Sampling sites in Bist- Doab region a: Shallow
(shallow aquifer (<30m)) and tube well and peizometers groundwater, b: Deep groundwater
(deep aquifer (>80m)) were collected during year 2009
and 2010 (Fig. 2). The isotopic analyses (δ18O and δD) Results and Discussion
of collected water samples were done by standard The stable isotopes of oxygen and hydrogen are
equilibration method in which water samples are generally considered to be transported conservatively in
equilibrated with CO2 and H2 (Epstein and Mayeda shallow aquifer settings, i.e. low temperature (<80°C),
1953, Brenninkmeijer and Morrison 1987). The samples prevalence of high water–rock ratios and an absence of
were analysed using a Continuous Flow Isotpe Ratio significant evaporation (Kim et al. 2003). The isotopic
Mass Spectrrometer (CF-IRMS) to measure oxygen composition of groundwater in both shallow and deep
(18O/16O) and Dual Inlet Isotope Ratio Mass aquifer shows much variation with the values ranging
Spectrometer (DI-IRMS) to measure D/H ratio and from -4.13‰ to -8.93‰ and -26.03‰ to –61.11‰
computed the δ18O and δD using a triple point (shallow) and -5.72‰ to -12.12‰ and -38.06‰ to -
calibration equation with Vienna standard mean ocean 84.07‰ (deep aquifer) in 18O and D respectively. The
water (V-SMOW), Greenland ice sheet precipitation spatial distribution (Fig. 3) of stable isotopic
(GISP) and Standard light Antarctic precipitation
composition (18O) of shallow and deep groundwater in
(SLAP) standards. The results were expressed by
the study area shows enriched character in northwestern
convention as parts per thousand deviations from the V-
region of the study area compared to that of the other
SMOW the calculation is as follows:
parts of the region. The stable isotopic composition of
sample= (Rsample-RV-SMOW/RV-SMOW)x1000‰ deep groundwater shows that the groundwater in the
Where, R is the ratio of D/H or 18O/16O in sampled study region flows from Kandi belt to towards the
water (Rsample) or in VSAMOW (RSMOW). The confluence of Rivers at Harike.
reproducibility of measurements was better than ±0.1‰
for δ18O and better than ±1‰ for δD. Isotopic Analyses
of samples were done at the National Institute of
Hydrology, Roorkee, India.

Shallow GW δ18O(‰)
-15 -10 -5 0 5
40
δD = 8.13 x δ 18 O + 7.63; R² = 0.96 (Precipitation)
20
δD = 7.05 x δ18 O + 0.51; R² = 0.94 (R. Satluj)
δD= 7.44 x δ18 O+ 7.53; R² = 0.93 (R. Beas) 0
-20
δD(‰)
-40
-60
Linear (R. Satluj)
-80
Linear (R. Beas)
-100
Linear (Precipitation)
-120
Fig. 4 Regression line of the 18O- D bi plot of
Precipitation; R. Satluj and River Beas in the Bist- Doab
region
The 18O- D bi plot (Fig. 5a) of shallow groundwater

shows lesser slope and intercept (eq. 4) than that of
precipitation, RWL Beas and RWL Satluj. The low
-9 -8.5 -8 -7.5 -7 -6.5 -6 -5.5 -5 -4.5 (‰)
intercept of shallow groundwater (-0.91) shows
Deep GW significant evaporation effect of groundwater during
infiltration. The shallow groundwater near the
tributaries of Bist- Doab canal that is draining from R.
Satluj in the Jalandhar, Nawanshahr and parts of
kapurthala districts show enriched isotopic composition
than that of R. Satluj and depleted than isotopic
composition of precipitation shows possible recharge of
groundwater from these tributaries (canal) which have
suffered evaporation due to agricultural activities. The
groundwater at locations near R. Beas, and Kandi canal
draining from R. Beas (pong dam) at upper reach of the
study area shows isotopic composition lying between
the LMWL and RWL Beas line shows possible recharge
from the R. Beas. The other locations at the study area
especially locations at the north western part and central
part of the study area show isotopic composition similar
to that of precipitation showing significant recharge
-8.5 -8 -7.5 -7 -6.5 -6 -5.5 (‰)
through precipitation. The locations at the north western
18 part of the study area show high evaporation effect due
Fig. 3 Spatial Distribution of  O in groundwaters in to the presence of salt encrustation in these part. The
Bist- Doab region. isotopic composition of deep groundwater show lesser
slope than that of LMWL but higher slope than the river
To characterise the isotopic composition of groundwater lines (RWL Beas and Satluj) with lower intercept than
and understand recharge source the local meteoric water the precipitation and R. Beas (eq. 5). The slope and
line and river water lines (for both R. Satluj and R. intercept of the deep groundwater show possible
Beas) were made using the isotopic composition of recharge through precipitation (Fig. 5b). A detailed
precipitation in the study area and river water samples. isotopic study of the shallow and deep groundwater will
The slope of regression equation (eq. 1) derived for the help to understand major source of groundwater that is
precipitation in the study area is above IMWL- North contributing to the recharge of these aquifers.
(D = 8.15 x 18O + 9.55) and less than GMWL (D =
8.17 x 18O + 11.27) and change in intercept indicates Shallow GW: D = 6.83 x 18O - 0.91; R2 = 0.90 (eq. 4)
change of source or the rain out process. The regression Deep GW: D = 7.97 x 18O + 7.1; R2 = 0.96 (eq. 5)
lines derived for the river water shows that both the
rivers show evaporation effect with R. Satluj showing
higher evaporation effect than R. Beas (eq. 2 and Eq.3
respectively; Fig. 4).
LMWL: D = 8.13 x 18O + 7.63; R2 = 0.95 (eq. 1)

RWL Satluj: D = 7.05 x 18O + 0.51; R2 = 0.92 (eq. 2)
RWL Beas: D = 7.44 x 18O + 7.53; R2 = 0.93 (eq. 3)

δ18O (‰) (a) recharge to an unconfined aquifer. Australian
-10 -8 -6 -4 -2 0 Journal of Soil Research, 16, 181-195.
0 Bowen, R. (1985). Hydrogeology of the Bist Doab and
δD = 6.83 x δ18O - 0.91; R² = 0.90 (GW Shallow)
δD= 7.44 x δ 18O + 7.53 (R. Beas) -10 Adjacent Areas, Punjab, India. Nordic Hydrology,
δD = 7.05 x 18 O + 0.51 (R. Satluj)
-20 16, 33-44
δD= 8.13 x δ 18 O + 7.63 (LMWL) Brenninkmeijer, C.A.M. and Morrison, P.D. (1987). An
-30
δD (‰)
automated system for isotopic equilibration of CO2
-40
GW(Shallow) and H2O for 18O analysis of water. Chemical
Linear (Beas) -50
Geology, 66, 21-26.
Linear (LMWL) -60
Linear (GW(Shallow))
CGWB. (2007). Groundwater Information Booklet,
Linear (Satluj)
-70 Punjab. Central Ground Water Board North Western
-80 Region, Chandigarh.
δ18O (‰) (b) Chandrasekharan, H., Navada, S.V., Jain, S.K., Rao,
-10 -8 -6 -4 -2 0 S.M. and Singh, Y.P. (1988). Studies on natural
0 recharge to the groundwater by isotope techniques in
δD = 7.97 x δ18 O + 7.1;R² = 0.96 (GW-Deep)
δD = 7.44 x δ 18 O + 7.53 (R. Beas) -10 arid western Rajasthan, India. In: Estimation of
δD = 7.05 x 18O + 0.51(R. Satluj)
-20 Natural Groundwater Recharge, NATO ASI Ser.,
δD = 8.13 x δ 18 O + 7.63 (LMWL)
-30 Ser. C, vol. 222, edited by I. Simmer, 205-221.
Chopra, R. and Sharma, P.K. (1993). Landform analysis
δD (‰)
GW(Deep) -40
and ground water potential in the Bist Doab area,
Linear (Beas) -50
Linear (LMWL)
Punjab, India. International Journal of Remote
-60 Sensing, 14, 3221-3229
Linear (GW(Deep))
Linear (Satluj)
-70 Edmunds, W.M. and Gaye, C.B. (1994). Estimating the
-80 spatial variability of groundwater recharge in the
Fig. 5 18O- D bi plot of a) shallow and b) deep Sahel using chloride, Journal of Hydrology, 156, 47-
groundwater in the Bist- Doab region. 59.
Ekwurzel, B., et al. (1994). Dating of shallow
CONLCUSION groundwater: Comparison of the transient tracers
The groundwater in the Bist- Doab region is over- 3H/3He, chlorofluorocarbons, and 85Kr. Water
exploited due to increase in usage of groundwater for Resources Research, 30, 1693-1708.
agricultural and industrial activities. The stable isotopic Epstein S. and Mayeda T. (1953). Variations of the
18
composition of oxygen and hydrogen is used to O/16O ratio in natural waters. Geochimica et
characterize and understand the recharge source of Cosmochimica Acta, 4, 213-224.
groundwater in Bist- Doab region. The study shows that Hunkeler, D., Chollet, N., Pittet, X., Aravena, R.,
the shallow groundwater in northwestern and central Cherry, J.A. and Parker, B.L. (2004). Effect of source
part of the region is significantly recharged from variability and transport processes on carbon isotope
precipitation. The Bist- Doab canal drained from R. ratios of TCE and PCE in two sandy aquifers.
Satluj recharges few places where the canal network is Journal of Contaminant Hydrology, 74, 265-282
present i.e. in the Nawanshahr and Jalandhar districts. Kaplan, I. R., Galperin, Y., Lu, S.T. and Lee, R.P.
The locations which are near R. Beas and the Kandi (1997). Forensic environmental geochemistry:
canal (at upper reach of the study area) gets recharged Differentiation of fuel-types, their sources and
from R. Beas. The northwestern part of the study region release time. Organic Geochemistry, 27, 289-317.
show significant evaporation effect during infiltration Kim, Y., Lee, K.-S., Koh, D.-C., Lee, D.-H., Lee, S.-G.,
process. The deep groundwater in the study regions Park, W.-B., Koh, G.-W. and Woo, N.-C. (2003).
shows similar character to that of precipitation Hydrogeochemical and isotopic evidence of
indicating possible recharge from the precipitation. groundwater salinization in a coastal aquifer: a case
study in Jeju volcanic island, Korea. Journal of
ACKNOWLEDGEMENT Hydrology, 270, 282-294
The present work is part of the research project funded Schmidt, T.C., Zwank, L., Elsner, M., Berg, M.,
by the World Bank under Purpose Driven Study of HP Meckenstock, R.U. and Haderlein, S.B. (2004).
II, Ministry of Water Resources, Government of India, Compound-specific stable isotope analysis of organic
(GoI), initiated by the Hydrological Investigations contaminants in natural environments: A critical
Division of NIH, Roorkee, India. Hence, the financial review of the state of the art, prospects, and future
support from the World Bank is acknowledged. challenges. Analytical Bioanalytical Chemistry, 378,
926-931
REFERENCES Singh, D., Singh, Y.P., Bairwa, S.P., Porwal, C.P. and
Adar, E.M. and Leibundgut, C. (1995). Application of Mathur, K.M. (1996). Isotopic and hydrochemical
Tracers in Arid Zone Hydrology. IAHS Publ. 232, study of groundwater in Shahgarh Bulge area of
IAHS Press, Wallingford, UK, pp. 452. Jaisalmer. Journal of Applied Hydrology, 9, 84-87.
Allison, G.A. and Hughes, M.W. (1978). The use of Stute, M., Schlosser, P., Clark, J.F. and Broecker, W.S.
environmental chloride and tritium to estimate total (1992). Paleotemperatures in the southwestern

United States derived from noble gases in
groundwater. Science, 256, 1000-1003.
Sukhija, B.S., Nagabhushanam, P. and Reddy, D.V.
(1996). Groundwater recharge in semi-arid regions of
India; an overview of results obtained using tracers.
Hydrogeology Journal, 4, 50-71.
Tiwana, N.S., Jeerat, N., Ladhar, S.S., Singh, G., Paul,
R., Dua, D.K. and Parwana, H.K. (2007). State of
Environment: Punjab- 2007. Punjab state council for
science and technology, pp- 243.
Van Warmerdam, E.M., Frape, S.K., Aravena, R.,
Drimmie, R.J., Flatt, H. and Cherry J.A. (1995).
Stable chlorine and carbon isotope measurements of
selected organic solvents. Applied Geochemistry, 10,
547–552

ISBN 978-81-921121-0-7
LINEAR AND NON LINEAR ANALYSIS OF SQUARE PLATE

Ashish Johri
M.Tech. Student, Department of Civil Engg., IT-BHU, Varanasi
Rajesh Kumar
Associate Professor, Department of Civil Engg., IT-BHU, Varanasi
Veerendra Kumar
Professor, Department of Civil Engg., IT-BHU, Varanasi
ABSTRACT: The behaviour of four elements namely shell 63, shell 93, shell 143 and shell 181 (of ANSYS
software) has been studied in this paper and their convergence pattern has been found out separately for simply
supported and fixed square plates for different combinations of edge lengths, thickness and load. Shell 93 and shell
63 perform closest to classical result for simply supported and fixed square plates respectively. These two elements
are used for comparing results of linear and non linear analysis of both simply supported and fixed square plates.
The conclusion thus obtained is used to establish a relationship between central deflection (classical) and variation
between linear and non linear results for a square plate.
INTRODUCTION 8
In the linear analysis, the displacements and strains
central deflection (mm)

developed in the structure are small. That is, the 6
geometry of the structure assumed remains 4
unchanged during the loading process and linear classical
strain approximations can be used. However, the 2 linear
geometry of the structure changes continuously 0
during the loading process, and this fact is taken into non linear
account in the geometrically nonlinear analysis 0 10 20
[Timoshenko and Woinowsky (1959)]. Finite
thickness (mm)
element method is quite important for analysis of thin
plates [Krishnamoorthy (2005)]. Large deflection
analysis of plates using finite element method and
other approximate methods has been done by many Fig. 1: Central Deflections of Simply Supported
researchers [Savithri and Vardan (1993), Rao et Square Plate (0.4mx0.4m) for different plate
al.(1993), Luo (2000) and Little (1987)]. Simply thickness (a=0.4m, q = 10 KN/ , E = 2.07*
supported and square plates have been taken for . ∗ ∗
N/ , µ = 0.3, = , Element used =
analysis in this study.
shell 93, Mesh density = 128x128)
SIMPLY SUPPORTED SQUARE PLATE
Simply supported square plate under a uniformly 40
central deflection
distributed load is analyzed for varying plate 30

thickness using classical, linear and non linear
approach. Linear and non linear analysis have been 20 classical
(mm)
done using ANSYS (Finite Element based) software. 10

linear
Three edge lengths namely 0.4 m, 2 m and 4 m are 0
used. In following examples “a” is edge length, “q” is 0 50 non linear
uniformly distributed load, “E” is Young’s modulus
of elasticity, “µ” is Poisson’s ratio, is thickness (mm)
maximum out plane displacement and “D” is flexural
rigidity of plate. The linear and non linear analysis is
done using the software ANSYS 10.0
Fig. 2: Central Deflections of Simply Supported
Square Plate (2.0mx2.0m) for different plate
thickness (a=2.0m, q = 10 KN/ , E = 2.07*
. ∗ ∗

100

80 80
60
60 classical
40
40 classical 20 linear
20 linear 0 non linear
0 20 40
0 non linear
0 50 100 thickness (mm)
thickness (mm)
Fig. 5: Central Deflections of fixed Square Plate
for different plate (2.0mx2.0m) thickness (a=2.0m,
Fig. 3: Central Deflections of Simply Supported q = 10 KN/ , E = 2.07* N/ , µ = 0.3,
Square Plate (4.0mx4.0m) for different plate . ∗ ∗
= , Element used = shell 63, Mesh
thickness (a=4.0m, q = 10 KN/ , E = 2.07*
. ∗ ∗
density = 128x128)
FIXED SQUARE PLATE 200

Fixed square plate under a uniformly distributed load central deflection (mm) 150
is analyzed for varying plate thickness using
classical, linear and non linear approach. Three edge 100 classical
lengths namely 0.4m, 2m and 4m are used. The linear 50
and non linear analysis is done using the software linear
ANSYS 10.0. 0 non linear
20 0 50 100
15 thickness (mm)
10 classical
Fig. 6: Central Deflections of fixed Square Plate
5 linear for different plate (4.0mx4.0m) thickness (a=4.0m,
0 non linear q = 10 KN/ , E = 2.07* N/ , µ = 0.3,
. ∗ ∗
0 10 20 = , Element used = shell 63, Mesh
density = 128x128)
thickness (mm)
80
linear and non linear
% variation between
simply
Fig. 4: Central Deflections of fixed Square Plate 60 supported
for different plate (0.4mx0.4m)thickness (a=0.4m, square
result
q = 10 KN/ , E = 2.07* N/ , µ = 0.3, 40

plate
. ∗ ∗
= , Element used = shell 63, Mesh 20 fixed
density = 128x128) square
0 plate
0 100 200 300 400
central deflection (classical) as % of…
Fig 7: Graph showing relation between central

deflection (classical) and variation between linear
and non linear results for a square plate.

RESULTS AND DISCUSSIONS REFERENCES
A relation between central deflection (classical) and Timoshenko, S.P. and Woinowsky-Krieger, S.,
variation between linear and non linear results is ‘Theory of plates and Shells’, New York:
obtained using above results and plotted on a graph McGraw-Hill, 1959.
for both simply supported and fixed boundary Krishnamoorthy, C.S., ‘Finite Element Analysis’,
conditions separately. This graph is shown in Fig. 7 2nd ed. New Delhi: McGraw-Hill, 2005.
and it is true for any uniformly distributed load, any Savithri, S. and Vardan, T.K., (1993), ‘Large
edge length and any thickness of the square plate. In deflection analysis of laminated composite
both the plates variation between linear and non plates’, International Journal of Non Linear
linear results increases with decrease of thickness of Mechanics, 28(1), 1-12.
plate. This is may be due to the higher value of Rao, D.V., Sheikh , A.H. and Mukhopadhyay,(1993),
membrane stress in thin plate as compare to thick ‘A finite element large displacement analysis of
plate. stiffened plates’, Computers and Structures,
47(6), 987-993.
CONCLUSIONS Luo, A.C.J.,(2000), ‘An approximate theory for
Out of different elements shell 93 and Shell 63 are geometrically non linear thin plates, International
suitable for simply supported and fixed plate Journal of Solids and Structures’, 37(51), 7655-
respectively. The maximum difference between the 7670.
results of linear and non linear analysis is 66.9% for Little, G.H., (1987), ‘Efficient large deflection
simply supported square plate and 55.5% for fixed analysis of rectangular orthotropic plates by direct
square plate. In both the plates variation between energy minimization’, Computer and Structures,
linear and non linear results increases with decrease 26, 871-884.
of thickness of plate.

ISBN 978-81-921121-0-7
LOCAL STRATEGIC PLANNING PROCESSES AND SUSTAINAB-

ILITY TRANSITIONS IN INFRASTRUCTURE SECTORS
Rajesh Singh
Research Scholar, Department of Mechanical Engineering, IT-BHU, Varanasi (U.P.) INDIA
rajeshsingh_sms@rediffmail.com
P.K.S. Dikshit
Professor , Department of Civil Engineering, IT-BHU, Varanasi (U.P.) INDIA
pk_sd@yahoo.com
ABSTRACT: Sustainability transitions have repeatedly been demanded in various economic sectors to confront
global environmental problems. The present paper analyses the role of strategic planning processes for
improving the prospects of achieving sustainable transitions in infrastructure sectors. Current planning
approaches in these sectors tend to perpetuate the established socio-technical configurations by neglecting
context uncertainties, by ignoring radical system alternatives and by focusing on narrow value considerations. In
order to improve the prospects for sustainable transitions, more reflexive and discursive strategic planning
methods are needed. We present the 'Regional Infrastructure Foresight' methodology as a specific procedural
proposal and analyze results from its application to a case study in the Sanitation Sector Uttar Pradesh.
INTRODUCTION increasing sustainability pressures in most regions of

Local Planning Processes and Sustainability the world. Urban water management has long been
Transitions in Infrastructure Sectors considered a largely non-controversial economic
Interdependencies (Markard and Truffer 2006). sector. Therefore, a strong socio-technical regime
Regarding sustainability issues, however, developed, which is prevalent in most industrialized
infrastructure sectors take a central position in modern nations in the world today. The material basis of the
production and consumption (Konrad et al. 2008). sanitation regime follows a civil engineering logic
The long life span of its material components consisting of concrete sewerage pipes and centralized
(transmission lines, underground piping systems, waste water treatment plants. The key technological
centralized treatment and generation facilities, etc.) components (generation and treatment plants) exhibit
puts strategic planning processes on a central stage strong economics of size, which leads to a strong
when analyzing possibilities for sustainable sector drive for centralization and compulsory connection of
reforms. Strategic decisions taken at the beginning of captured users. Given the long material life times,
a life cycle of key components determine the institutional conditions have to remain stable over
characteristics of the systems for much extended time long periods. As a consequence, these sectors were
periods (usually 25-50 years). The way these strategic largely organized as public services run by public
planning processes are currently carried out tends to authorities (Dominguez et al. 2009). In the 1960s and
narrow the evaluation space away from potentially 1970s, when this socio-technical regime was
more sustainable alternatives: it tends to implemented countrywide, there were constant growth
systematically downplay context uncertainties, tends rates in service demand, the technological paradigm
to ignore radical system alternatives and tends to remained largely unchallenged and the goals on which
focus on a narrow subset of possible value system performance was assessed remained
considerations (Dominguez et al. 2010, Dyner and uncontested (elimination of nutrients, un-
Larsen 2001). As a consequence, radically different discriminated with uncertainties related to future
and potentially more sustainable alternatives will be context conditions. Therefore, we understand
systematically undervalued. Infrastructure sectors sustainability transitions less in a substantive way
therefore have a tendency to be locked-in to (predicting how a sustainable sector structure will
unsustainable states for long periods. The present ultimately look like) but rather propose a procedural
paper aims to elaborate on how strategic planning approach that allows us to reflect on path-
should be designed to enhance the prospects for dependencies and by this improve the implementation
sustainable transitions in infrastructure sectors. The potential for radical system alternatives with better
proposed approach integrates insights from the sustainability performance. Considering these
sustainability transitions literature with lessons from requirements, the 'regional infrastructure foresight
environmental decision-making and strategic (RIF)' method has recently been put forward as a
planning. Essentially, it argues for a more reflexive specific approach to sustainability orientated strategic
and participatory planning approach that allows planning (Stormer et al. 2009). Stormer and Truffer
considering a broader range of context conditions, 2009). It involves local decision-makers as well as
options and value positions. We demonstrate the broader stakeholder representatives in a highly
virtues of such an approach by applying it to cases structured elaboration and discourse process.
from the sanitation sector, a domain that is exposed to Participants explore a broad range of possible future

system configurations and evaluate the performance studies. One of the earliest and most prominent
of these options for a diverse set of regional context examples is Hughes (1987, 1993), who provided an
scenarios. Evaluation is based on an encompassing list analysis of the social and technological determinants
of sustainability goals and is carried out relative to of the growth and diffusion of the electricity system in
(assumed) preference profiles of key future the US. He coined the concept of Large Technical
stakeholder groups. In particular, RIF supports Systems (LTS), which are characterized by a strong
investments in the capabilities of local decision- interrelationship between their material and
makers (see also Harmaakorpi and Uotila 2006) that institutional structures. More recently, research on
form a precondition for more encompassing socio-technical transitions was developed based on
sustainable transformation processes. historical case studies of specific sectors (Rip and
Kemp 1998, Konnola et al. 2006). Coherent socio-
The paper is structured as follows. In the next section technical configurations have been epitomized as
we outline the theoretical background of sustainability socio-technical regimes. Over extended time periods,
transitions and strategic planning on which the RIF specific regimes may dominate the provision of
method builds. We then introduce the methodology of certain services. These stable periods are interspersed
the RIF approach. The following section presents the by phases of transition that often are associated with
major results of a case study run in the Swiss substantial leaps in performance (Krausmann et al.
sanitation sector, where the advantages of the RIF 2008, Elzen et al. 2004). Such transitions are often
method compared with a conventional planning triggered by mounting pressures from context
approach are well illustrated. We conclude by coming conditions ('socio-technical landscape forces'; see
back to the question of how strategic planning can Geels and Schot 2007) to which the established
enhance the prospects for sustainability transitions in regime has problems to adapt. Additionally,
infrastructure sectors. In particular, we discuss how established regimes may be challenged by upcoming
this local planning process should be embedded in alternatives ('technological niches') that can take
more encompassing multi-level governance structures advantage by better fitting the upcoming context
aiming at sustainability transitions. pressures (Hoogma et al. 2002). The potential for
sustainability transitions may therefore be assessed by
SUSTAINABILITY TRANSITIONS AND analyzing
STRATEGIC PLANNING IN
INFRASTRUCTURE Local Planning Processes and Sustainability
As expounded above, the urban water management Transitions in Infrastructure Sectors
sector (in line with many other infrastructure sectors) The interplay between dominant sector structures,
exhibits strong path dependencies. Strong material changing landscape pressures and the prospects of
and institutional interdependencies have been built up, technological niches (see Geels 2002). Compared
in order to guarantee a proper working of these with a conventional market-based account of sector
systems. Conceptually these configurations may be transformations, socio-technical systems emphasize
understood as socio-technical systems in which the need to develop matching configurations between
material and institutional characteristics are tightly technology and institutions and therefore allow us to
intertwined in order to establish 'configurations that account for resistance to change and path dependency
work' (Rip and Kemp 1998). As a consequence, more explicitly.
incremental innovations are mostly preferred to The analysis of socio-technical regime changes has
radical system reconfigurations (Markard and Truffer been undertaken with a view on policy implications.
2006). Leaps in environmental performance may only Originally, 'constructive' approaches to technology
be achieved Alternative socio-technical configurations assessment have been developed (Rip et al 1995,
can establish their own matching structures. If these Schot 1999, 2001). Here, the role of early learning
new configurations gain dominance in a sector, we and decision-making processes in technology
may call it a socio-technical transition. If the new development was emphasized, as well as the need to
system shows a substantially higher performance with encompass experimentation and social learning to
regard to sustainability indicators, these are termed improve success for sustainability transitions
'sustainability transitions' (Geels 2006b). Below we (Hoogma et al. 2002). More recently, the approach of
present some core tenets of the sustainability transition management has been formulated, which
transitions literature. In a second step, we will - based aims to establish a long-term policy perspective on
on the recent literature on environmental decision- sustainability transitions consisting of the
making and strategic planning - elaborate on how development of broadly shared visions on sustainable
strategic planning could enhance the potential for futures, experimentation with radical system
sustainability transitions in infrastructure sectors. alternatives and involvement of stakeholders in
participatory decision contexts (Voss et al. 2009,
Infrastructures as Socio-technical Regimes Loorbach 2007). By its very origins, transition
Infrastructure-based sectors have long received management had a strong technology policy
considerable attention from scholars of innovation orientation. More recently, the decision-making

processes of firms and other societal actors have been for substantially different technological system
incorporated by drawing on foresight and sustainable configurations with a grossly enhanced performance
assessment approaches (e.g. Truffer et al. 2008, with regard to sustainability indicators (Daigger
Eriksson and Weber 2008, Smith et al. 2005). 2007). Third, evaluation criteria for assessing system
alternatives have become more diversified. For
Strategic Planning for Sustainable Urban Water instance, economic and institutional performance
Infrastructures characteristics have gained increased public attention
Water infrastructures in general and sanitation in in the course of recent liberalization and privatization
particular have been the object of analysis in a debates worldwide. An important driver for this
number of empirical transition studies. Geels (2005, development was a reduced financial capacity of
2006a), for instance, has carried out a multi-level many local communities and nation states, which have
analysis of historical transformations of urban water questioned the established overcapacity paradigm for
management in the Netherlands. The role of handling uncertainties. The increasing focus on
technological niches for the sustainable economic efficiency has in turn also raised concerns
transformation of this sector has been analyzed by about environmental and social goals of infrastructure
Hegger et al. (2007), among others, and Hiessl et al. management in many parts of the world (Dominguez
(2003) have carried out detailed sustainability et al. 2009).
assessments for alternative socio-technical system In order to respond to these new challenges, strategic
configurations in urban water management. In the planning has to explicitly address the increased
transitions terminology, the sanitation sector may be uncertainties in context conditions, options and
described as a specific socio-technical regime, which evaluation criteria. Decision-makers would have to
is characterized by sewage systems connected to respond proactively to emerging opportunities. An
centralized wastewater treatment plants, a incremental improvement of existing planning
professional culture of civil engineering, practices will probably not be sufficient (Dominguez
organizational structures which are (often) highly et al. forthcoming). A number of recent proposals for
localized and thus spatially fragmented, strong state environmental decision-making may be instructive to
regulation or even state-run service organizations, and find better suited procedures.
an almost total invisibility for end-users (here actually Thereby, shared orientation is the outcome of strategic
waste-water generators). From these characteristics, planning rather than mere optimization within a
we may deduce that strong path dependencies exist predefined choice set. Decision-makers have, in
due to resistance to technical, institutional and user- particular, to develop strategic capabilities to better
related reforms of the predominant regime. Planning confront ambivalent choices and manage open-ended
and decision procedures have explicitly been tailored development processes rather than to make optimal
to maintain and perpetuate these regime one-short choices (Harmaakorpi and Uotila 2006).
characteristics. Urban water treatment systems are The implementation will then follow some variant of
locally planned, decision-making is dominated by adaptive management involving experimental action
technical experts, and evaluation of alternatives is and reflexive monitoring and adaptation (see Gregory
based on rather narrow ranges of alternatives and et al. 2006). An increasing number of such approaches
value considerations (dominguez et al 2010). to strategic planning for sustainable infrastructure
Therefore, radically new (and potentially more development have been formulated recently: sectoral
sustainable) alternatives have difficulty in gaining a foresight for sustainable infrastructure transformations
foothold in these contexts and there are well- have been reported in Germany (Truffer et al. 2008)
developed barriers set up by incumbent actors to and more specifically strategic planning in individual
exclude potential rivals (Hegger et al. 2007). sectors has been carried out such as in urban water
In the past few years, however, criticism has increased management (Dominguez et al. 2009, Stormer et al.
to challenge the long-term optimality of this 2009), Lienert et al. 2006), electricity (Dyner and
established socio-technical regime (Daigger 2007, Larsen 2001), transportation (Bartholomew 2007,
Wilderer 2004, Larsen and Gujer 1997), based on Zegras et al. 2004), waste management (Scholz et al.
three lines of argument. First, increasingly uncertain 2009, Shmelev and Powell 2006) and regional
context conditions, or landscape forces under which planning (Gamboa 2006).
these systems have to operate, call for more flexible Summarizing, strategic planning aiming to support
system configurations: population dynamics, sustainability transitions should more explicitly deal
shrinking regions, industrial transformations, new with uncertainties and consider the interrelated
contaminants and pollutants, risk of terrorism and character of socio-technical systems. As a
reduced capacities of local communities to run the consequence, decision-making processes should be
established institutional structures (OECD 2007). participative (respecting different knowledge
Second, new treatment alternatives based on perspectives and value positions), open to
membrane and telecommunication technology (socio- experimentation and learning (to enhance prospects
technical niches) exhibit fundamentally different for alternative system configurations), flexible
economies-of-scale characteristics that would allow (relative to changing context conditions) and reflexive

(with regard to value considerations of different Dubendorf, a suburban region in the Zurich area).
stakeholder groups and the long-term orientation of a Each of the three planning processes ran over 9-12
sustainable sector future). Given the strong localized months and involved about 25-30 stakeholders. The
nature of water infrastructures such a planning processes were organized as follows. A core team
approach has to be implementable at the community consisting of 4-7 local decision-makers was formed
level. We maintain that if these conditions are together with (about 3-5) members from the research
fulfilled, innovations with a higher performance on team. This team analyzed the local system
the overall set of sustainability dimensions will have a characteristics and prepared inputs for the stakeholder
better chance of being implemented and as a workshops in the course of 7-8 half-day workshops.
consequence sustainability transitions will become The core team identified key context factors
more likely. potentially impacting the future of their sanitation
system and assessed the scope of uncertainties of their
The Regional Infrastructure Foresight Approach future development. Compared with the conventional
Based on this analysis, we provide a methodological planning approach, the RIF method represents a rather
approach to strategic decision-making, the RIF radical break with standard procedures. It increases
process. Based on the above analysis, RIF is designed transparency, reflexivity and communication
to fulfill the following goals. (1) To consider possible substantially. The process will also deliver a different
context conditions in the time horizon of 25-50 years. set of outputs: the processes will not deliver an
In particular, it will address uncertainties explicitly undisputed set of optimal solutions; rather, it will
and prevent decision-makers from closing down the provide a broader argumentative structure, which
range of possible developments prematurely. Future allows us to explore possible courses of development,
context conditions are integrated into context elaborate arguments in favor and against a departure
scenarios using a discursive and deliberative from the established regime structures, and identify
approach. This allows us to elucidate expectations of potential conflicts lines and sustainability deficits (e.g.
the involved stakeholders and to create a shared as in Gamboa 2006). The outcome of the process will
understanding of different development paths (Holden thus not present 'the' sustainable solution but rather a
2008, Truffer et al. 2008). (2) The choice set of socio- mapping of the many trade-offs that have to be
technical system alternatives will be broadened to confronted to develop 'more sustainable' sector
include radical system alternatives. Risks and structures (more extensively, see Stormer et al. 2009).
opportunities associated with these options will be And finally, RIF does not supplant conventional
assessed considering the requirements following from planning; rather, it sets an encompassing framework,
the different context scenarios. In particular, which enables decision-makers to run subsequent
established system boundaries are questioned both planning processes in a more reflexive way.
thematically (e.g. in order to consider options of
integrated infrastructure service management) and Results of the RIF process in the Klettgau Region
spatially (e.g. in order to consider catchment-level To illustrate the potential contribution of the RIF
integration of urban water management). (3) approach to ease sustainability transitions, we present
Decision-making will involve a broad range of the results of one case in more detail. In the Klettgau
perspectives and vale positions and thus tap into region, a conventional strategic planning process had
broader resources of knowledge and value positions. been completed shortly before the start of the RIF
Different value positions will be analyzed by process. It therefore represents a very good case to
systematically considering the different sustainability discuss the specific performance of the RIF method
dimensions (e.g. by applying a systematic compared with more conventional approaches.
sustainability value tree) and by involving a broad set The Klettgau region is located in the canton
of influential and affected stakeholders. Stakeholder Schaffhausen in the northernmost part of Switzerland,
involvement is organized in three levels of intensity in bordering Germany. The sanitation system in the
participation: the lion's share of the analysis will be Klettgau encompasses 10 communities, nine of which
carried out by a 'core team' of delegated local are organized in an association of convenience
decision-makers (in the project they were running a wastewater treatment plant in Hallau, a
complemented by members of the project team). For community located downstream of the river
key steps of the analysis, workshops with additional Halbbach, which subsequently crosses the German
stakeholder representatives have to be undertaken. border. Sewerage is owned and operated by each
The third level of involvement relates to the formal individual community. The Klettgau region is
decision-making institution in the region. They representative of peripheral, rural regions in
commission the strategic planning processes and Switzerland with a strong proportion of agriculture
receive the recommendations elaborated during the (here mostly viniculture) and only little industry. It is
RIF process. In the corresponding research project, somewhat distant from the local urban centre
the procedure was run in three Swiss regions: Schaffhausen. The wastewater treatment plant
Klettgau, a rural region in northern Switzerland; (WWTP) in Hallau was built in 1976 and is now in
Kiesenthal, a peri-urban region in the Berne area; and urgent need of renewal. The decision body of the

organization commissioned a first strategic planning scenarios. The scenarios may be characterized based
study to the local planning engineer. He on a combination of high versus low growth and
recommended two options: either refurbishing the strong versus weak role of environmental demands
existing WWTP or build a new structure in its placed on the sanitation system. The two growth
immediate vicinity. Local policy-makers were scenarios have been labeled 'Blooming Growth' when
dissatisfied with these proposals. They complained combined with high levels of importance citizens put
that these options merely represented an incremental on healthy environmental conditions and 'Critical
optimization of the solution they had inherited. Given Growth' when increasing environmental problems are
that they had to live for another 30-40 years with the combined with a generally low levels of
new solution they wanted to have 'real strategic environmental consciousness. The two economic
alternatives'. This pre-project history illustrates nicely downturn scenarios were labeled 'Strong
how the conventional approach to planning produced Globalization', describing a world with high pressures
a strongly reduced set of options and how on environmental resources but a low level of public
uncertainties in the context conditions were not finances, and 'Diet Region' for a continuation and
addressed in an adequate form. reinforcement of the peripheral status of the Klettgau
When the project team issued a call for participation region (implying a very high share of elderly people
among the Swiss wastewater organizations, Klettgau with little financial resources).
representatives showed keen interest and gained Socio-technical system alternatives were elaborated in
approval for the first pilot study to run and test the the core team between the first and the second
method. The core team encompassed a member of the stakeholder workshops. To this end, basic
local decision body (and at the same time a member technological and organizational characteristics of
of the board of one of the communities), the operator future sanitation systems were identified, which were
of the WWTP, a (newly commissioned) planning then combined into four coherent system alternatives.
engineer and the official from the canton responsible These alternatives, which might be interpreted as
for regulating sanitation in Schaffhausen. A contract alternative future regime configurations, had not been
was set up between the project team and planning selected because they ere considered a priori as the
commission to carry out the RIF process. The process best options. Rather, they represent cornerstones of a
ran from November 2006 to September 2007. The continuum of feasible alternative systems. The first
core and project teams met for eight sessions of 3-4 h option aimed at a thematic expansion of the current
each. Two workshops were held with 16 organizational competence range by encompassing
representatives (including the core group) from several infrastructures in the region (multi-utility). In
different stakeholder groups in March and June 2007. a first step, this would include the management and
The final recommendation was presented to the maintenance of the sewerage in order to better run the
planning commission and the participants of the WWTP. Later, water provision or waste management
workshop in September 2007. Since then, the results could be added to the activity portfolio of the
of the RIF process have been used as a basis for more organization. A second option (WWTP light) foresaw
detailed planning activities. Overall, the process ran a radical cost reduction of the WWTP organization.
very smoothly and participants were highly To this end, industry would be compelled to treat its
committed. After completion, they expressed high waste water separately, and new investments into the
levels of satisfaction with the process and the results. WWTP would be kept at an absolute minimum. In the
long run, operation of the WWTP could be outsourced
Scenarios, Options and Goal Trees to some private company or one of the neighboring
The extended core team chose a time frame of 25-30 sanitation organizations. The third option
years (i.e. about the life time of a newly built (decentralization) would prepare for a radically
WWTP). For that period, major challenges were decentralized structure of wastewater treatment in the
identified that could impact the Klettgau region and region. Building on household scale, WWTPs would
its sanitation infrastructure. Furthermore, about 10 allow investments in sewerage to be reduced and, in
interviews with experts in the region were carried out the long run, render the existing plant obsolete. This
and the results analyzed. About 15 context factors option depends on the timely and cost-competitive
were derived that might have a substantial impact on availability of decentralized treatment technology as
the suitability of any chosen option. These factors well as the guarantee that these plants may be
encompassed population structure, the kind and operated and controlled safely. The fourth option
number of industry workplaces, impacts from global (decommissioning) describes the removal of the
change, connection to the highway system, etc. The current WWTP and delegation of the service to one of
different factors were evaluated with regard to the larger neighboring organizations, near
importance and uncertainty. From this, nine were Schaffhausen or even downstream to one of the larger
selected as inputs into the first scenario process, German communities.
where the list was discussed, and it was As a third element, an encompassing sustainability
complemented by additional factors; finally, five value tree was elaborated along which the options
context factors were chosen to construct four could be assessed. Starting from the three core values

of sustainable development (environmental, economic be elaborated and evaluated in more details, especially
and social goal) and a fourth dimension relating to the regarding feasibility, cost, time of realization, etc. In
political autonomy of the community, more specific parallel, the 'multi-utility' option should be specified,
goals were elaborated upon and corresponding as well. The 'WWTP light' option was considered to
evaluation criteria were derived. Context scenarios, be associated with too many risks and would quickly
strategic options and the value tree represented the lead to open conflict between the communities and
inputs to the second workshop in which the industry (especially if growth conditions in the region
assessment of the options in the different scenarios proved to be meager). The 'decentralization' option
was carried out. was initially judged as utterly unrealistic by the core
team. At the end of the process, however, positive
Evaluation of Options and Ranking of Sustainable aspects of this option appeared under several
Alternatives scenarios. Given the relatively high pressure for
This elaboration of arguments with regard to the action in the region and the considerable uncertainties
different preference orders (of future citizens, industry associated with the future development of
and a more balanced consideration of the decentralized technologies, this option was dismissed
sustainability value tree) and with regard to a wide as a fully fledged alternative. However, in the context
range of context conditions enabled participants to of the 'multi-utility' option, a caveat was formulated
assess the four alternatives in detail. Figure I gives a that dependent on the maturing of these technologies;
synthetic representation of these assessments by implementing services for decentralized wastewater
posting each alternative in each scenario in a two- appliances should be actively considered.
dimensional diagram. They y-aix gives the average
ranking attributed by citizens and industry (an The Contribution of RIFs to Sustainability
indicator of the future social preference for the Transitions
respective option) and the x-axis indicates the likely How far can we now claim that RIF procedure will be
conflict potential that a specific option would entail supportive for sustainability transitions in the
under the respective scenario (measured by the sanitation sector? From the Klettgau case, we may
difference between the assessments of citizens and conclude that the initially postulated characteristics of
industry). Each polygon connects the assessments a more sustainability orientated strategic planning
attributes to one option, the corners being defined by approach were fulfilled: a broad set of stakeholders
the different scenarios. Size and position of each assessed different system configurations considering a
polygon may be interpreted as a measure of the wide spectrum of future regional context conditions.
dependence of the assessment relative to context Evaluations took into account a systematic
conditions and stakeholder perspectives: the larger the sustainability value tree and rankings considered
surface of the polygon, the higher the context potential preference profiles of relevant future
sensitivity, and the more its location is in the right stakeholder groups. Emphasis was less on ultimate
corner, the higher disagreements between ranking scores but rather on trade-offs between
stakeholders (for more detail see Stormer et al. 2009). different goals and different preferences. We may
When comparing the different polygons, the diagram therefore assume that the reported case is well suited
shows that option 'decommissioning' generally to demonstrate the contribution of the RIF method to
outranks the others. However, under the 'Critical sustainability transitions.
Growth' and 'Blooming Growth' scenarios, the multi- The effectiveness of the RIF exercise can be
utility option would also fare quite well. In the 'Diet determined by comparing the result with
Region' and 'Strong Globalization' scenarios this recommendations from a conventional planning
option would not be well suited. The 'decentralized' exercise conducted shortly before the RIF project
and 'WWTP light' option reveal a more problematic started: the strategies proposed in the conventional
pattern, the latter because it was perceived as not planning approach represented a typical case of
responding to most of the challenges identified in the incremental improvement of the prevailing socio-
scenarios and the former because of many technical regime. Context uncertainties were not
uncertainties related to the timely availability of these explicitly addressed. For example, the development
radical system alternatives. path of the Klettgau region was assumed to continue
After the second workshop, the sum of all discussion as previously. Options were to consider only minor
transcripts and assessments was analyzed by the core variations of the current layout of the WWTP,
team and synthesized into recommendations for the respecting the current spatial boundaries of the
official planning commission of the Klettgau region. organization. Finally, value considerations were
The report was presented to the commission and the mostly reduced to investment cost and partially to
workshop participants in a final meeting in September political feasibility.
2007. The recommendations foresee that the The broadening of the decision space resulted in
'decommissioning' option, which implies a connection substantially different strategic recommendations: the
with other larger WWTPs, should be given priority. recommendation of the conventional planning
As a next step, conditions for its implementation will exercise is close to the 'WWTP light' option

developed in the RIF process. This option received, Also, with regard to the promotion of technological
however, very low scores in the RIF exercise because niches, decentralization was a potentially radical but
it did not respond to most challenges identified in the still immature system system alternative and was
scenarios. Instead, the 'decommissioning' option was therefore only considered as an option for the future.
favored by all stakeholders and under most future However, when compared with the initial attitudes of
context conditions. In the conventional strategic decision-makers, decentralized technologies were in
planning process, this option was not mentioned as a the end evaluated far more favorably. We may
legitimate course of action, essentially because it therefore conclude that the RIF process did not
extended the system boundaries beyond the national directly lead to a radical system transformation but
borders to Germany. The RIF process produced a prepared decision-makers to consider a broader range
radically different result. It opened up the highly of options and arguments and by this weakened the
entrenched modes of thinking of the regime actors, path dependencies of the dominant regime
furthered a balanced consideration of long-term considerably.
implications of the different options and their likely How could a sustainability transition in the Swiss
conflict potential, and increased willingness to sanitation sector look like? Running a few isolated
consider radically different system configurations. All strategic planning processes at the local level would
these points were also explicitly acknowledged by the not be sufficient to create enough momentum for
local decision-makers in the final feedback sustainability transitions. These local decision
questionnaires. processes would have to be embedded within broader
We therefore consider that our approach was effective concepts of sustainability-orientations. These local
in reaching our stated procedural goals. Still, the decision processes would have to be embedded within
preferred option should not be interpreted as the broader concepts of sustainability-orientated reforms
single best (or ultimately sustainable) solution for the at the regional and national level (Turffer 2008). This
region. The different system alternatives were judged was taken into account in the RIF processes by
based on rather general and qualitative assessments involving cantonal policy-makers. In the final project
and many impacts were associated with high degrees evaluation, they expressed a strong interest in scaling-
of uncertainty. For instance, costs of the different up these experiences to the cantonal level (the state
system alternatives were not assessed in detail. level in Switzerland), so that lessons could be
Additionally, although representatives of the transferred from one region to another. If several
neighboring German communities participated in the contons developed similar strategies this would
workshops, no feasibility study for the ultimately also create a basis for sector reform for
'decommissioning' option had been carried out. The federal government to discuss and implement
final recommendation of RIF team therefore radically new system configurations in sanitation. In
comprised a series of more specific feasibility studies this way, a multi-level governance structure could
and cost estimates. Based on these results, preference emerge to promote bottom-up learning processes that
rankings of the options could change considerably. are able to overcome path dependencies of established
However, as was mentioned by most of the socio-technical regimes. This would radically reverse
participants, the broadening of the analysis brought a the logic of the governance structure initially
more reflexive consideration of system alternative and established to build up the Swiss sanitation sector. In
future conditions and so decision-makers felt better the1960s and the 1990s, implementation of the (in
prepared to respond to different situations. This was those times) 'new' socio-technical regime was
nicely corroborated in the final presentation of the prescribed by national law (elaborated by expert
report to the planning commission, where the committees) and subsidized top-down from the
participants argued strongly in favor of decentralized confederation to the cantons. This delegated the
technologies that had been very critical at the outset implementation to individual communities. As in
of the process. recent years, federal subsidies and support have
With regard to broader concerns of sustainability decreased substantially, and there is little room left to
transitions for the whole sanitation sector in steer the system in a top-down manner. A coordinated
Switzerland (or even beyond), even if the RIF process set of local strategic planning processes could thus
does provide supportive conditions for radical system create momentum for reforming the whole sector
alternatives, it is not sufficient to lead to radical from the bottom up and open up the decision-making
transformations of the sectoral regime. The preferred process to newly emerging technological options. In
regime. The preferred option, namely removing the that sense, methods such as RIF could become an
regional WWTP and connection the sewers to a larger essential element of a new type of adaptive
plant in Germany, would not challenge the core of the management in infrastructure sectors orientated at
existing socio-technical regime but rather reinforce it. supporting sustainability transitions.

REFERENCES
Bartholomew K. 2007. Land use-transportation Geels FW, 2005. Co-evolution of technology and
scenario planning: promise and reality. society: the transition in water supply and personal
Transportation, 34:397-412. hygiene in the Netherlands (1850-1930) - a case
Berkhout F, Angel D, Wieczorek Aj. 2009. study in multi-level perspective, Technology in
Sustainability transitions in developing Asia: are Society 27:363-397.
alternative development pathways likely? Gregory R, Failing L, Higgins P. 2006. Adaptive
technological Forecasting Q Social Change 76: management and environmental decision making :
215-217. a case study application to water use planning.
Daigger G. 2007. Wastwater management in the 21st Ecological Economics 58:434-447.
century. Journal of Environmental Engineering Harmaakorpi V, Uotila T. 2006, Building regional
133 (7): 671-680. visionary capability. Futures research in resource-
Dominguez D, Truffer B, Gujer W. 2010. Tackling based regional development. Technological
uncertainty in infrastructure sectors through Forecasting and Social Change 73:778-792.
strategic planning. The contribution of discursive Hegger DLT, Van Vliet J, Van Vliet BJM. 2007.
approaches in the urban water sector. Water Niche management and its contribution to regime
Policy. change: the case of innovation in sanitation.
Dominguez D, Worch H, Markard J, Truffer B, Gujer Technology Analysis and Strategic Management
W. 2009. Closing the capability gap: strategic 19:729-746.
planning for the infrastructure sector. California Hiessl HEA. 2003, Alternativen kommunaler
Management Review 51:30-50. Wasservesorgung and Abwasserentsorgunj-
Dominguez D, Gujer W. 2006. Evolution of a AKWA 2100. Schrifterreihe des Frauenhofer-
wastewater treatment plant challenges traditional Holden M. 2008. Social learning in planning:
design concepts. Water Research 40:1389-1396. seattle's
Dyner I, Larsen ER. 2001, From planning to strategy Holden M. 2008, Social learning in planning :
in the electricity industry. Energy Policy 29: 1145- Seattle's sustainable development codebooks,
1154. Progress in Planning 69: 1-40.
Elzen B, Geels FW, Green K. (eds) 2004, System Hoogma R, Kemp R, Schot J, Truffer B. 2002.
Innovation and the Transition to sustainability: Experimenting for Sustainable Transport. The
Theory, Evidence and Policy. Edgar Elgar, approach of Strategic Niche Management. E&F
Cheltenham. Spon Publisher (Routledge), London.
Eriksson EA, Weber KM. 2008, Adaptive foresight: Hughes TP, 1993. Networks of power: electrification
navigating the complex landscape of policy in western society, 1880-1930. John Hopkins
strategies. Technological Forecasting and Social University Press, Baltimore.
Change 75: 462-482. Hughes TP. 1987. The evolution of large
Failing L, Gregory R, Harstone M. 2007. Integrating technological systems. In: W. Bijker, T.P. Hughes
science and local knowledge in environmental risk and T. Pinch (eds), The Social Construction of
management: a decision-focused approach, Technological Systems. Cambridge/MA, pp. 51-
Ecological Economics 64:47-60. 82.
Frame B, Brown J. 2008, Developing post-normal Jaffe AB, Newell RG, Stavins RNN. 2005. A tale fo
technologies for sustainability. Ecological two market failures: technology and
Economics 65:225-241. environmental policy. Ecological Economics
Funtowiez S, Ravetz JR, 1993, Science for the post- 54:164-174.
normal age, Futures 25:739-759. Kemp R, Rotmans J. 2004, The management of the
Gamboa G. 2006, Social multi-criteria evaluation of co-evolution of technical, environmental and
different development scenarios of the Aysen social systems, In: Weber KM, Hemmelskamp J
region, Chil. Ecological Economics 59: 157-170. (eds), Towards Environmental Innovation
Geels FW. 2002, Technological transitions as Systems, Springer, Heidelberg; 33-56.
evolutionary reconfiguration processes: a multi- Kemp R, Loorbach D. 2006, Transition management:
level perspective and a case-study. Research a reflexive governance approach. In: Vo B J,
Policy 31:1257-1274. Bauknecht D, Kemp (eds), Reflexive Governance
Geels FW, Hekkert MP, Jacobsson S. 2008, The for Sustainable Development. Edward Elgar,
dynamics of sustainable innovation journeys. Cheltenham, pp, 103-130.
Technology Analysis and Strategic Management Konnola T, Unruh GC, Carrillo-Hermosilla J. 2006.
20(5): 521-536. Prospective voluntary agreements for escaping
Geels FW, Schot J. 2007, Typology of sociotechnical techno-institutional lock-in. Ecological Economics
transition pathways, Research Policy 36: 399-417. 57:239-252.
Geels FW, 2006b. System innovations and transitions Konrad K. Truffer B, VoBJ. 2008. Multi-regime
towards sustainability: challenges for innovation dynamics in the analysis of sectoral transformation
theory, 14.

potentials: evidence from German utility sector. Stormer E, Truffer B, Dominguez D, Gujer W, Herlyn
Journal of Clearner Production 16: 1190-1202. A, Hiessl H, Kastenholz H, Klinke A, Markard J,
Krausmann F, Schandl H, Sieferle RP. 2008, Socio- Maurer M, Ruef A. 2009. The exporatory analysis
ecological regime transitions in Austria and the of trade-offs in strategic planning. Lessons from
United Kingdom. Ecological Economics 65:187- regional infrastructure foresight. Technological
201. Forecasting and Social Change 76:1150-1162.
Larsen TA, Gujer W. 1997. The concept of Stormer E, Truffer B. 2009. Strategic decision making
sustainable urban water management. Water in infrastructure sectors. Participatory Foresight
Science and Technology 35:3-10. and strategic planning for sustainable sanitation.
Lienert J, Monstadt J, Truffer B. 2006. Future Geographica Helvetica 64:73-80.
scenarios for a sustainable water sector-A case Swart RJ, Rasking P, Robinson J. 2004. The problem
study from Switzerland, Environmental Science of the future: sustainability science and scenario
and Technology 40:436-442. analysis. Global Environmental Change-Human
Loorbach D, 2007, Transition management. New and Policy Dimensions 14: 137-146
mode of governance for sustainable development. Tippett, J, Handley JF, Ravetz J. 2007, Meeting the
International Books, Utrecht. challenges of sustainable development-A
Markard J, Truffer B. 2008. Technological innovation Conceptual appraisal of a new methodology for
systems and the multi-level perspective: towards participatory ecological planning, Progress in
an integrated framework, Research Policy 37:596- Planning 67:9-98.
615. Truffer B, Voss J, Konrad K. 2008. Mapping
Markard J, Truffer B. 2006, Innovation processes in expectations for system tranformatinos: lessons
large technical systems: market liberazation as a from sustainability foresight in German utility
driver for radical change? Research Policy 35: sectors. Technological Forecasting and Social
609-625. Change 75: 1360-1372.
Meppem T, Bourke S. 1999. Different ways of Truffer B 2008. Society, technology, and region:
knowing: a communicative turn sustainability. contributions from the social study of technology
Ecological Economics 30:389-404. to economic geography. Environment and
OECD, 2007, Infrastructure to 2030 (Volume 2): Planning A 40: 966-985.
Mapping Policy Electricity, Water and Transport. Tukker A, Charter M, Vezzoli C, Sto E, Andersen
OECD Publishing, Paris. MM. 2008, System Innovation for Sustainability I,
Raskin P. 2008. World lines: a framework for Perspectives on radical changes to sustainable
exploring global pathways. Ecological Economics consumption and production. Greenleaf
65:461-470. Publishing Ltd.: Sheffield (UK); 470 pages.
Rip A, Kemp R, 1998, Technological change. In: S. Van den bergh JCJM. 2007. Evolutionary thinking in
Rayner and E.L., Malone (eds), Human choice and environmental economics. Evolutionary
climate change-Resources and technology. Economics 17: 521-549.
Battelle Press, Columbus. Printer Publisher Ltd.: Voss J, Smith A, Grin J, 2009. Designing long-term
London, New York; 327-399. policy: rethinking transition management, Policy
Rip A, Misa T, Schot J. (eds), 1995. Managing Sciences 42: 275-302.
Technology in Society: The Approach of Wilderer PA. 2004. Some thoughts about future
Constructive Technology Assessment, London. perspectives of water and wastewater
Scholz RW, Spoerri A, Lang DJ, 2009. Problem management. Water Science and Technology
structuring for transitions: the case of Swiss waste 49:35-37.
management. Futures 41:171-181. Zegras C, Sussman J, Conklin C. 2004. Scenario
Schot J, 1999, Constructive Technology Assessment planning for strategic regional transportation
Comes of Age. The birth of a new politics of planning. Journal of Urban Planning and
technology. Development-Asce 130:2-13.
Schot J, 2001. Towards new forms of participatory
technology development. Technology Analysis &
Strategic Management 13: 39-52.
Schot J, Rip A. 1997. The Past and Future of
constructive Technology Assessment.
Technological Forecasting and Social Change
54(2-3): 251-268. Shmelev SE, Powell JR, 2006,
Ecological-economic modelling for strategic
regional waste management systems Ecological
Economics 59:115-130.
Smith A, Stirling A, Berkhout F. 2005. The
governance of sustainable socio-technical
transitions. Research Policy 34: 1491-1510.

ISBN 978-81-921121-0-7
ENVIRONMENTAL CAPACITY CONCEPT IN NOISE POLLUTION

ANALYSIS OF A ROADWAY
Kanakabandi Shalini
Ph.D. Research Scholar, Transportation Engineering Group (TEG), Department of Civil Engineering,Institute of
Technology, Banaras Hindu University, Varanasi
Brind Kumar
Assistant Professor, Department of Civil Engineering, Institute of Technology,Banaras Hindu University, Varanasi Email:
kumar_brind.civ@itbhu.ac.in
INTRODUCTION may be stated that, in contemporary times roads

Road traffic noise is established as a highly should not only fulfil the engineering, safety and
fluctuating source of noise. Noise pollution has economy requirements, but also the stipulations for
already reached alarming proportions along busy environmental safeguard. Conducting
highway corridors and most of the metropolitan Environmental Impact Assessment [IRC:104
cities of our country. It is steadily increasing its (1988)] is order of the day for road projects in
domain and engulfing many more areas under its developed nations. Such countries have the
umbrella. Noise pollution due to traffic flow is fast predominance of segregated traffic flow.
becoming a major source of annoyance to people. Unfortunately, there seems to be a lack of enough
Being an integral component of the ecosystem, its know-how with regards to mixed traffic flow in
containment within the specified limits prescribed order to describe road traffic noise on plains and
by the environment monitoring agencies is of vital hills (alignment is predominantly on gradient and
importance for maintenance of the ecological curvature). Consideration of noise pollution at
balance. In fact, for ensuring the homeostasis of the planning stage of highway projects is essential, but
ecosystem, none of the environmental there is lack of building blocks for noise analysis
factors/parameters can be allowed to escalate for mixed traffic which may simplify the
beyond control [Odum and Barrett (2006)].
Apart from the plain areas, the hilly areas are also formulation of prediction methodologies.
gradually beginning to witness the surge of Krishnaveni et al. (1994) mentions that, there is a
population and industrial activities. People living need to evolve an acceptable categorisation of
therein were unaccustomed to any form of vehicles and develop simplified predictor of noise
pollution. Now, they are suddenly exposed to the identical to Passenger Car Unit for the context of
ills of air, water and noise pollution leading to mixed traffic flow. Such a measure should help to
various forms of physiological and psychological simulate the traffic with vehicles of different noise
damages to the individuals as is reported from emissions. Towards this goal, the consideration of
many recent studies. So it may be stated that, environmental capacity in noise pollution analysis
factors similar to plain areas are available on hills of a roadway is of prime importance.
for generating noise pollution. Additionally, the
vehicles are required to perform under prohibitive STATEMENT OF THE ENVIRONMENTAL
geometric conditions generating more noise. CAPACITY
Leonovich and Kashevskaja (1993) have reported Different classes of vehicles use the same right-of-
that modification of the road scheme (slope, way without segregation on roads where mixed
curvature etc.) has a vital impact on noise traffic flow is prevalent. This unrestricted mixing
contamination. Also, the noise emission is irregular of different vehicle types forms a heterogeneous
through the whole slope in the road scheme. The traffic flow, which is required to be converted to a
longitudinal gradient of the highway was stated to single arguable unit from the point of view of noise
pollution analysis [Srivastava (1994)]. In the age of
be linearly proportional to the Leq under fixed road environmental alertness, two types of highway
traffic conditions. Wilson (1989) states that, on a capacity are defined – conventional capacity
level road the noise level varies slightly with (related to movement and parking of vehicles) and
increase in payload because of the increase in the environmental capacity (aims to restrain the
engine noise to overcome the rolling resistance. A traffic volume to maintain environmental
major change in noise level may be expected due to standards). Environmental capacity for restraining
increased payload on an uphill grade where an traffic noise pollution may be defined as the
increased torque is required to counteract the environmental threshold which will ensure that
increased component of the vehicular weight. traffic does not dominate so much as to cause noise
Planning of a environmentally competent road pollution. Environmental threshold is the weighted
alignment is a highly demanding job. Therefore it

traffic flow ( Qw ) or the Equivalent Passenger Car Rickshaws(R):representing non-motorised
rickshaw-cycles
Noise Equivalence ( EPCNE / hour ) which will Cycles(Cy):representing non-motorised bi-cycles
cause noise level to exceed the acceptable limits. Others(O):representing non-motorised hand drawn
This implies that environmental capacity will be tri-cycles and other forms of hand drawn vehicles
possible for many permutations and combinations
of traffic volume composition. The environmental The weighted traffic flow may be represented as:
threshold will always ride over all such possible Qw  NC  aN 2W  bN 3W  cN J  dNTr  eN LCV  fN B  gNT  hN R  iN Cy  jN o
combinations of traffic volume composition as a
(1)
limiting value. Thus, a convenient model for
where, Nc, N2-W, N3-W, NJ, NTr, NLCV, NB, NT, NR,
management of traffic noise could be arrived at by
NCy, and NO = Number of Cars, 2-Wheelers, 3-
using this concept.
Wheelers, Jeeps, Tractors, LCV’s, Buses, Trucks,
The environmental threshold may be alternately
Rickshaws, Cycles and Others vehicle types
described by the weighted traffic flow or the
passing the point of observation, respectively.
equivalent passenger car noise equivalence, which
a, b, c, d, e, f, g, h, i, and j = PCNE factors of the
seeks to direct the whole discussion of restraining
respective vehicle types.
traffic volume for noise containment using
passenger car as the standard unit for all possible
PCNE factor
vehicle types.
Weighted traffic flow may be defined as the total The noise intensity ( I ) generated by a vehicle is
number of passenger cars (representing the related to the sound pressure level ( L ) in dB(A) by
equivalence of various vehicle types) that traverses the relationship:
a given section of the roadway in an hourly I  I 010( L /10) (2)
duration so that the environmental threshold of
noise is not violated. It may be obtained by where I o = reference noise intensity.
multiplying the number of all participating vehicle For a car producing noise level, LC in dB(A), the
types in the traffic stream with their corresponding
Passenger Car Noise Equivalence (PCNE) factors intensity ( I C ) is given by,
and summing them up.
I C  I 010 ( LC /10) (3)
PCNE is defined as that number of passenger cars
which will generate same intensity of noise along
the road-side to that of a particular category of For any particular vehicle type producing noise
vehicle under same operating and traffic flow level, Lx in dB(A), the intensity ( I x ) is given by,
conditions. Therefore PCNE factor becomes a
simple ratio of noise intensity generated by a class
of particular vehicle type to that of a class of I x  I 010( Lx /10) (4)
passenger cars, both traversing under similar By definition, PCNE factor is the intensity of noise
operating and traffic flow conditions. generated by those many passenger cars which
becomes equal to one unit of a particular vehicle
MATHEMATICAL DERIVATIONS AND type. Alternately, it is the total noise intensity due
PAST STUDIES to n number of cars which is equal to the noise
intensity due to one particular vehicle type under
Weighted flow similar operating and traffic flow conditions. We
The vehicle types on roads representing a typical may state that,
mixed flow condition may be classified as follows: I x  n  IC (5)
2-wheelers (2W): representing motorised 2-
wheelers where, n = PCNE factor of any particular
3-wheelers (3W): representing all forms of vehicle type whose characteristic noise level is Lx
commercial 3-wheelers (passenger and freight) in dB(A).
Cars (C): representing all forms of passenger cars
(private and commercial) or, Io 10Lx /10  n  I o10 LC /10
Jeeps (J): representing all forms of jeeps (private So, n  10( Lx  LC )/10 (6)
and commercial)
Tractors (Tr): representing all forms of tractors Past studies
(laden and unladen) Joyce et al. (1975), has evaluated the conversion
LCV’s (LCV): representing all forms of LCV’s factors for some vehicle types in U.K. which is
(laden and unladen) Buses (B): representing all summarised below:
forms of passenger buses (laden and unladen)
Trucks (T): representing all forms of commercial Car = 1
trucks (laden and unladen) Bus = 7

Motorcycle = 2 the percentage of heavy vehicles in the traffic
Heavy Commercial Vehicle = 15 stream is of vital importance towards the evaluation
Light Commercial Vehicle = 2 of noise pollution [Kumar (2000)].
Gangil (1979) has worked out the following
conversion factors for SH-45 at Roorkee in 1979: CONCLUSIONS
Car = 1 Environmental capacity concept should be applied
Bus = 7 for quantification of noise pollution due to
Rickshaw = 0.025 operation of traffic. Environmental capacity for
Scooter = 1.5 restraining traffic noise pollution may be defined as
Truck = 15 the environmental threshold which will ensure that
Cycle = 0.04 traffic does not dominate so much as to cause noise
Motorcycle = 5 pollution. Environmental threshold is the weighted
Tonga = 0.4
traffic flow ( Qw ) or the Equivalent Passenger Car
Kumar (1997) has obtained the following PCNE
factors for Delhi in 1997: Noise Equivalence ( EPCNE / hour ) which will
Car = 1 cause noise level to exceed the acceptable limits.
3-Wheeler = 7.08 Weighted traffic flow may be defined as the total
Truck = 2.00 number of passenger cars that traverses a given
2-Wheeler = 1.26 section of the roadway in an hourly duration so that
Bus = 7.08 the environmental threshold of noise is not
Rickshaw = 0.08 violated. It may be obtained by multiplying the
number of all participating vehicle types in the
METHODOLOGY FOR EVALUATING PCNE traffic stream with their corresponding Passenger
FACTORS Car Noise Equivalence (PCNE) factors and
Firstly, the categorisation of vehicles pertinent to summing them up.
the geographical area under study should be made PCNE is defined as that number of passenger cars
such that it is representative of most vehicle types. which will generate same intensity of noise along
In this, the engine-less vehicles (non-motorised and the road-side to that of a particular category of
hand drawn) should also be included since they vehicle under same operating and traffic flow
have the effect of reducing the traffic speed, conditions.
thereby, higher levels of vehicular noise. The event PCNE may be obtained by mapping the event of
of passage of a singular vehicle at the point of single vehicle passage of every category of vehicle
observation is recorded with respect to its sound and applying statistical analysis to obtain L50 .
level, traversing speed, roadway geometrics and
road surface characteristics. The point of
REFERENCES
observation should be such that it is at a fixed
Odum, E.P. and Barrett, G.W., “Fundamentals of
distance from pavement edge and is devoid of any
Ecology”, Thomson Brooks/Cole, Thomson
absorbent material between source and receiver.
Business Information India Pvt. Ltd., ISBN: 81-
The roadway geometrics considered may be
315-0020-9, Fifth Edition, 2006.
gradient, degree of curve etc. Presence of any hill-
Leonovich, I., and Kashevskaja, E., “Technical
side wall and box type deep open-trench cutting
Solutions on Environmental Control from Traffic
along the right-of-way, which may reflect the
Noise and Their Effects”, Proceedings, The East-
sound signals may also be observed and
West European Road Conference, Warsaw, Poland,
considered. The type of road surfacing available
Vol. 2, pp. 363-365, September 22-24, 1993.
which may affect the noise profile due to jerking
Wilson, C.E., “Highway and Other Surface
and bouncing of vehicles should also be
Transportation Noise: Prediction and Control”,
considered. Large number of single vehicles under
Noise Control – Measurement, Analysis and
each vehicular category may be observed to
Control of Sound and Vibration, Harper and Row
evaluate PCNE factors.
Publishers, New York, pp. 368-393, 1989.
From the noise level data obtained, the cumulative
IRC: 104-1988, “Guidelines for Environmental
frequency distribution curve for each vehicle type
Impact Assessment of Highway Projects”, The
is developed and the 50 percentile time exceeding
Indian Roads Congress, Jamnagar House, New
noise level ( L50 ) is obtained. The PCNE factors Delhi, 1989.
are then computed using Eq. (6). Krishnaveni, Mrs. P.I., and Balagopal, Dr. T.S.
For every operating condition, it may be expected Prabhu, “A Review of Traffic Noise Studies
that the PCNE factors will increase with the Related to Environmental Planning”, Indian
increasing axle load capacity of vehicles. Since the Highways, Vol. 22, No. 2, pp. 27-35, February
noise polluting potential of the vehicles is more, 1994.
their PCNE factors will also be higher. It has Srivastava, J.B., “Traffic Flow and Environmental
already been established in many past studies that Impact Analysis of a Highway Corridor”, Ph.D.

Thesis (Unpublished), Department of Civil
Engineering, University of Roorkee, Roorkee,
April 1994.
Joyce, F.E., Williams, H.E., and Johnson, D.M.,
“The Environmental Effects of Urban Road Traffic,
2. Evaluating Alternative Transport Proposals”,
Traffic Engineering and Control, Vol. 16, No. 4,
pp. 198-171, April 1975.
Gangil, N.L., “Relationship between Vehicular Noise
and Stream Flow Parameters”, M.E. Dissertation,
Department of Civil Engineering, University of
Roorkee, Roorkee, 1979.
Kumar, V., “Analysis of Urban Traffic Noise”, M.E.
Dissertation, Department of Civil Engineering,
University of Roorkee, Roorkee, 1997.
Kumar, B., “Noise Pollution Analysis for a Hill Road
Corridor (NH-52A in Arunachal Pradesh)”, Ph.D.
Thesis (Unpublished), Department of Civil
Engineering, M.M.M. Engineering College, DDU
Gorakhpur University, Gorakhpur, January 2000.

ISBN 978-81-921121-0-7
ESTIMATION OF SCOUR DEPTH AROUND BRIDGE PIERS BY

USING HEC-RAS
Dilip Kumar
Assistant Professor, Civil Engg. Dept., G.B.Pant Engg. College, Pauri, Uttarakhand (India); jhadilip@sify.com
Hira Lal Yadav
Assistant Professor, Civil Engg. Dept., G.B.Pant Engg. College, Pauri, Uttarakhand (India); hiralalyd@gmail.com
Sushil Kumar Himanshu
PG Student, Deptt. of Hydrology,IIT Roorkee, Uttarakhand (India).
ABSTRACT: Scour is the formation of a scour hole around the bridge pier due to erosion of soil by
flowing water which is responsible for the bridge failure without warning. The depth of scour below the
riverbed level around the bridge piers in alluvial stream varies depending on the depth of flow, shape of
pier, size of the pier, angle of attack of flow and sediment characteristics. 86% of the 577,000 bridges in
the national bridge registry are built over waterways in which more than 26,000 of these bridges are
reported to be scoured critical. Excessive scour affects the stability of the bridge foundation by the
removal of the bed material. The foundation of bridge piers have to be taken deeper and to provide
sufficient anchorage length required for stability of the foundation. Experimental study was performed to
understand sensitive parameters influencing the determination of scour depth. The scour evaluation for a
proposed bridge site highly depends upon the accuracy and analysis of the hydrologic and hydraulic data.
In this Paper, one-dimensional hydraulic model such as HEC-RAS (Corps of Engineers) model is used
for developing the water surface profiles for the experimental data around bridge pier. Estimation of
scour depth during the design is able to reduce considerable cost in the construction of bridge
foundations. Therefore, there is a need of the estimation of scour for innovative, effective and
economical countermeasures to be considered for the design of new bridges and for replacing existing
bridges.
Key words: - Scour, Bridge pier, Scour Estimation, HEC-RAS
INTRODUCTION the shear stress over the stream bed. Therefore,

Scour at a bridge pier is the formation of a hole it is important to estimate the probable scour
around the pier due to erosion of soil by depth so that bridge foundations can be
flowing water. It means lowering of the designed to support the structural load safely
riverbed level by water erosions such that there below the probable scour depth.
is a tendency to expose the foundations of a The present paper is concerned with the
bridge. Such a process reduces confinement of numerical simulations for the local scour
foundation and the pier, there by affecting its around a bridge pier. The objective of this paper
load carrying capacity. The depth of scour is to verify scouring depth around bridge piers
below the riverbed level around the bridge by applying the computer software package
elements in alluvial stream can vary depending HEC-RAS and empirical equations.
on the depth of flow, shape of pier, size of the
pier, angle of attack and sediment METHODOLOGY
characteristics. The scour evaluation for a
proposed bridge site highly depends upon the River Flow Modeling
accuracy and analysis of the hydrologic and In order to determine the scour depth near the
hydraulic data. In this study, one-dimensional bridge structure, the flow profile near the
hydraulic model such as HEC-RAS (Corps of structure is very important. Since exact
Engineers) model is commonly used for measurements of hydraulic parameters are not
developing the water surface profiles for each measured, river flow modeling will provide the
reach of the river. Scouring of bed material required details for the determination of scour
from around bridge foundation is a common depth near the bridge structure. Two types of
cause of bridge failure. Reduction of the flow modeling, namely steady and unsteady
area causes an increase in the average velocity modeling are possible, which are explained in
for the same discharge, as well as increase in the following sub sections.

Steady Flow Modeling unsteady flow model the principle of
Steady flow consists of flow regime, boundary conservation of mass (continuity) and the
conditions and peak discharge information. principle of conservation of momentum
Flow regime consists of sub-critical, super- (momentum) equations are used.
critical, and mixed flow. In subcritical flow the
boundary conditions are necessary at the Determination of Scour Depth
downstream end of the river, for supercritical Numerous equations have been proposed to
flow the boundary conditions are necessary at predict scour at piers, mostly developed from
the upstream of the river, and for mixed regime laboratory data. Some of the commonly used
condition the boundary conditions are necessary scour depth equations can be summarized as
at the upstream and downstream of the river. follows:
Boundary conditions are required to establish ds = 1.3b0.7y0.3 (1)
the water surface at the ends of the river system. Where ds is the maximum predicted local scour
Discharge and energy equation is required to depth, b is the width of the pier and y is
compute the water surface profile. For the depth of flow.
computing the water surface profiles the
upstream and downstream discharge data of the Description of HEC-RAS
river is necessary. Water surface profiles are HEC-RAS is a computer program designed by
computed from one cross-section to the next the U.S. Army Corps of Engineers. This
cross-section by solving the energy equation program incorporates the standard step method
with iterative procedure called standard step for computing the water surface profiles. HEC-
method (Chow, 1988). Whenever the water RAS is an integrated system of software,
surface passes through critical depth, the energy designed for interactive use in a multi-tasking
equation is not applicable. The energy equation environment. HEC-RAS is designed to perform
is applicable for gradually varied flow one-dimensional hydraulic calculations for a
situations, and the transition from subcritical to full network of natural and constructed
super critical or supercritical to sub-critical flow channels. This software allows performing one-
condition. In HEC-RAS the momentum dimensional steady flow and unsteady flow
equation can be applied where the occurrence calculations. The function of the HEC-RAS
of a hydraulic jump, low flow hydraulics at program is to determine water surface
bridge, and stream junctions (HEC 18 Manual, elevations at all locations of interest. The data
2001). Momentum equation is derived from the needed to perform these computations are
Newton’s second law of motion. In this separated into geometric data and steady flow
modeling, flow is assumed to be steady because data (boundary conditions).
time-dependent terms are not included in the
energy equation. Flow is assumed to be Details of the Study Area
gradually varied because energy equation is This study is carried out in the laboratory. The
based on the hydrostatic pressure distribution at experiments were carried out in a 12 m long,
the cross-section. When the flow is rapidly 0.52 m wide, and 0.45 m deep flume. The
varied, the program switches to the momentum flume is provided with centrifugal pump to
equation. Flow is assumed to be one recirculate the water from under ground tank to
dimensional because total energy head is same feed the flume with the required flow discharge.
for all points in a cross-section. Upstream the flume, a rectangular weir is
installed to measure the flow discharge. Figure
Unsteady Flow Modeling (1) shows the overview of the experimental set
In unsteady flow some of the boundary up. The bottom of the flume was covered with
conditions like flow hydrograph, stage 0.14 m sand stratum with d50=0.47mm,
hydrograph, stage and flow hydrograph, and g=2.19. The flow discharge was .06m3/s, the
rating curve are used. Upstream end of the range of the flow depths were between (6.0 to
river system is modeled by using flow 8.0) cm. The control of water depth during
hydrograph, stage hydrograph, and flow and experimental work was achieved by using a tail
stage hydrograph as boundary conditions, and gate located downstream the flume. The water
downstream ends of the river is modelled by depths upstream the piers were measured by the
using rating curve, normal depth, stage point gauge of the flume. The data obtained is
hydrograph, flow hydrograph, and stage and used in HEC-RAS model to sudy the sour
flow hydrograph as boundary conditions. In depth.

Fig. 1 Schematic representation of the experimental flume.
1)- Centrifugal pump 9)- Pier model

2)- First head tank 10)- Sand trap
3)- Second head tank 11)- Tail gate
4)- Flow measuring weir 12)- Collecting tank
5)- Approach basin 13)- Graduated tank
6)- Inlet screen 14)- Laboratory sump
7)- Sand bed 15)- Side walk
8)- Wood bracing 16)- Rails for point gague
RESULT AND DISCUSSION effect of obstructions due to the presence of bridge
The river flow modeling has been carried out using has been incorporated at the relevant location. The
HEC-RAS. Steady flow component in HEC-RAS flow profile was plotted as shown in the figure
is capable of modeling sub-critical, super-critical, given below.
and mixed flow regime water surface profiles. The
1
Wn n o Wo
G
0 Ln Lo
Z / Wo
-1
a) Wn/Wo=0.5
-2 (G/Lo) =1.0
(G/Lo) =2.0
(G/Lo) =3.0
(G/Lo) =4.0
-3
-30 -20 -10 0 10 20 30 40 50 60 70 80
X / Wo
Wn n o Wo ….
G
Lo
n
G Wn ….
Ln

1
Wn n Wo
G
0 Ln Lo
Z / Wo
-1
-2 (G/Lo) =1.0
(G/Lo) =2.0
(G/Lo) =3.0
(G/Lo) =4.0
-3
-30 -20 -10 0 10 20 30 40 50 60 70 80
X / Wo
Fig. 2 Scour profile along (C ) of the pier, (Ln/Lo=1), (F=0.254).
0.8
0.7
Observed values of (So / y).
0.6
0.5
0.4
0.3
0.2
0.1
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Predicted values of (S o / y).
Fig.3 Relationship between observed values of (So/y) and predicted ones
CONCLUSIONS gives total scour below HFL in flood plain sandy

In the present study total scour depth is analyzed material, and also used for estimation of the scour
using HEC-RAS model. Scour depth values depth at abutments and guide banks. Scour depth
obtained from the empirical equations and HEC- estimated from this formula is independent of
RAS. Different formulas for estimating total scour hydraulic parameters and dimensions of the
depth have been checked for their over or under bridge. It was concluded that local scour depth
prediction. Lacey’s method is commonly used in reduces as pier inclination angle increases, critical
India for estimation of scour depth around bridge velocity and scour depth increases with particle
piers placed in alluvial rivers, recommended by size and bed roughness respectively.
the Indian Road Congress and Indian Railways.
This method is basically empirical and essentially

REFERENCES
Bateman, A., and Torres, A., "Experimental study on Local Scour at Accompanied Pier Systems". RCEM -River,
Coastal and Estuarine Morphodynamics, UPC, (2003).
Chiew, Y. M.,: " Scour Protection at Bridge Piers", Journal of Hydraulic Engineering, ASCE, Vol. 118, no. 9,
(September, 1992), pp. 1260-1269.
Jones, J.S., and Sheppard, D. M., " Local Scour at Complex Pier Geometries" Water Resources, Joint Conference on
Water Resource Engineering and Water Resources Planning and Management, (2000).
Kumar,Dilip, " Dam break analysis-using HEC-RAS." Lambert Publication. Faculty of Eng. G.B.Pant Engg.
College,Pauri, (2011), pp. 61-70.
Melville, B. W., and Raudkivi, A.J., "Effects of Foundation Geometry on Bridge Pier Scour" Journal of Hydraulic
Engineering, ASCE, Vol. 122 , Issue 4 , (April , 1996), pp. 203-209.
Melville, B.W., and Sutherland , A.J., : "Design Method for Local Scour at Bridge Piers", Journal of Hydraulic
Engineering, ASCE, Vol.114, No.10 , (October, 1988), pp.1210-1226.
Parola, C.A., "Effect of Rectangular Foundation Geometry on Local pier Scour" Journal of Hydraulic Engineering,
ASCE, Vol.122, Issue 1, (January, 1996), pp. 35-40
Yahaya, A.S., and Ghani, A.A., "Acomparative Study of Models for Predicting Local Bridge Pier Scour Depth ",
[River Engineering And Urban Drainage Research Center] (REDAC), Malaysia, (2005).
NOTATION
The following symbols are used in this technical note:
d50 = mean particle size diameter,
F = Froude number,
G = separate distance length between piers,
Ln = length of the new pier,
Lo = length of the old pier,
So = old pier maximum scour depth,
Wn = new pier width,
Wo= old pier width,
g = geometric standard deviation of particles
Y=Depth of approach flow

ISBN 978-81-921121-0-7
COMPARATIVE STUDY OF BUILDING WITH COMPOSITE AND

RCC STRUCTURE
A.K. Chitkeshwar
P.G. Student (M.Tech- Structure), Y.C.C.E., Nagpur -441110, Email: anupchitkeshwar@rediffmail.com
V.G. Meshram
Associate professor, Dept of Civil Engineering,Y.C.C.E., Nagpur -441110, Email: meshramvg@gmail.com
ABSTRACT: Steel-concrete composite members are an interesting option for structural designers, but the
reliability of design procedures both in the case of gravity and seismic loads is in continuous development.
composite steel concrete design provide a significant economy through reduced materials and faster
construction, such system make use of each type of member in the most efficient manner to maximize the
structural and economic benefits. Except in certain structures such as towers, portal frames, industrial structures
covered with trusses and purlins, cranes, etc., steel work exist with reinforced concrete. In many places, the
properties of reinforced concrete (strong in compression, greater rigidity) can be advantageously combined with
those of structural steel. Such composite construction includes (a) concrete-encased steel column (b) concrete-
filled steel column (c) concrete-encased steel beam and (d) steel beam supporting concrete slabs. In contrast to
ordinary structural steel design, which considers only the strength of steel, composite design assumes that the
steel and concrete works together in resisting the loads.
In this paper the comparative study of composite and RCC building is done. The building selected is G+10
made up of composite and RCC. The building is analysis by using Response spectra method in Staad Pro
software. The comparison is done in the parameters like axial force, moments, lateral deflection, story drift and
size of foundation. From this analysis it was observed that the moments on the columns are more than RCC
column but due to the more flexural stiffness the size of composite column is not increase, the lateral
displacement of composite building is more as compare to RCC building and from this result the story drift is
comes more than the RCC building, dead weight of structure is found to be much lesser than (16 to 20%) that of
RCC building due to that the size of foundation is reduced as compare to RCC building.
Key words: composite structure, concrete-encased steel column, steel beam supporting concrete slabs, flexural
stiffness.
INTRODUCTION are increasingly designing composite and mixed

The most important and most frequently encountered building systems of structural steel and reinforced
combination of construction materials is that of steel concrete to produce more efficient structures when
and concrete, with applications in multistory compared to designs using either material alone.
commercial buildings and factories as well as in It should be noted that the
bridges. The basic principal underlying composite combination of concrete cores, steel frames and
construction is that certain materials may be used composite floor construction has become the standard
more effectively in certain types of stressed construction method for multistory commercial
conditions; thus, the combination of a material strong buildings. The structural steel/reinforced concrete
in compression with one strong in tension makes a frame is the standard system for tall building, the
very economical union for its use in structures. With main reason for this preference is that the sections and
basic materials such as concrete, steel, masonry members are best suited to resist repeated earthquake
materials, wood, and timber, a number of successful loadings which requires high amount of resistance and
combinations can be used. These essentially different ductility.
materials are completely compatible and A further more important consideration is
complementary to each other; they have almost the that the use of rolled steel sections, profiled metal
same thermal expansion; they have an ideal decking and/or prefabricated composite members
combination of strengths with the concrete efficient in speeds up execution. And also composite floor
compression and steel in tension; concrete also gives construction is highly competitive if spans are
corrosion protection and thermal insulation to the increased to 12 to 15m or even 20m. There is, of
steel at the elevated temperatures and additionally can course, a demand for larger column-free spans in
restrain slender steel sections from local or lateral buildings to facilitate open planning or greater
buckling. flexibility.
In multistory building, structural steel work
is typically used together with concrete; for example,
steel beams with concrete floors slabs. The same
applies to the road bridges, where concrete decks are
normally preferred. It is fact, however, that engineers

FUNDAMENTALS AND OVERVIEW OF plates, suitable brackets, vertical gusset plates, shear
COMPOSITE ACTION IN STRUCTURES:- heads or other structural means.
A typical form of composite construction consists of a COMPSITE VS. NONCOMPOSITE ACTION:-

slab connected to a series of parallel steel members. 1. Concrete filled steel tubes and cased rolled shapes
The structural system is therefore essentially a series possess higher strength than the reinforced concrete
of interconnected T-beams with wide, thin concrete column of comparable size. (Research as reported by
flanges, In such system, the flange width may not be Furlong (1968), Gander and Jacobson (1967), and
fully effective in resisting compression due to “shear Stevens (1965)).
lag” this phenomenon will takes into account by the 2. Encasing steel sections in reinforced concrete is
well known approach “effective width” of composite particularly beneficial for earthquake resistant
structures. design.(Okamoto(1973)).
Instead of an insitu concrete slab, pre-cast 3. From the point of structural advantage
concrete floors or decks are also to be used. Careful i. Depth of steel beam is reduced to support a given
detailing and construction practice are needed to load.
ensure adequate containment for connectors. ii. Composite columns helps to reduce the effective
COMPOSITE COLUMN:- slenderness ratio of a steel column, thus
Three different types of composite columns are increasing its buckling load. Furthermore; the
principally in use concrete encasement also carries its share of loading
 Concrete encased steel columns (a) according to the actual behavior of composite
 Concrete filled steel tubes (b&d) columns under load.
 Rolled section columns partly encased in iii. For a given load, a reduction in dead loads and
concrete (c) construction depth reduces.
iv. Composite construction is amenable to “fast-
track” construction because of using rolled steel and
pre-fabricated components, rather than cast-in-situ
concrete.
v. Encased steel beam sections have improved fire
resistance and corrosion.
The following assumptions are normally
utilized for the formulation of the basic composite
plane frame analysis (Vallenilla, 1987). It is
emphasized that nonlinear analysis techniques, such
as those used to evaluate pure steel frame and
(a) (b) (c) (d) reinforced concrete frame behavior, can be applied to
composite frames as well.
Fig 1.1 Typical composite columns 1. The building is a rectangular grid pattern. Story
In calculating the strength of such height may vary arbitrarily, as may the bay widths in
columns full composite interaction without any each direction. However, it is assumed that the
slip at the steel-concrete-interface, geometrical and columns extend continuously from base to top and the
physical non-linearity of the different materials should girders from side to side.
be observed. It is only possible, however, to meet 2. The members are straight, prismatic and symmetric
these requirements by using comprehensive numerical about the plane of the frame.
methods of analysis and computer software. The 3. Loads are applied in the plane of the frame, and
assumed complete interaction enables definition of only at the end of the elements.
section properties, stiffness and slenderness ratio for 4. Only in plane deformations are considered.
whole inhomogeneous cross-section this information 5.Only half of the concrete cross section remains
is necessary to determine the load carrying resistance, uncracked after the composite columns are
including slenderness or P-delta effects. Eurocode4 loaded.
gives simplified design methods for practical use. 6. The steel has a linearly elastic-perfectly plastic
Instead of more precise buckling curves, Eurocode4 stress-strain relationship.
has adopted the European buckling curves a, b and c 7. Slip does not occur between steel and concrete.
which was originally established for bare steel 8. Lateral Torsional buckling does not occur.
columns.
The complete interaction must be SALIENT FEATURES OF THE EURO CODE 4
ensured by means of mechanical connections. The Euro codes are the European standards for the
connections have to be provided at least at the column structural design. Euro codes will come in a number
ends and where loads or forces are acting. They of parts, covering a range of applications.
should be distributed over the whole cross-section. Table 3.1 Structural Euro codes currently under
Such connectors can be headed studs, top and bottom preparation

1) Floor load
Euro Subject a) Assume slab thickness 130 mm
code Load = 0.130 x 1 x 25 =3.250
1 Common unified rules for different types of KN/m²
construction and materials b) Floor finish = 1 KN/m²
2 Design of concrete structures
Total floor load = 4.125
3 Design of steel structures
KN/m²
4 Design of composite steel and concrete structures 2) Wall load (Assume wall thickness =150
5 Design of timber structures mm)
6 Design of masonry structures a) Wall load = (3-0.6) x 0.15 x19.5 =7.02
KN/m
7 Geotechnical Design
b) Plaster = 3 x 2 x 0.025 x20 = 3 KN/m
8 Design of structures for earthquake Resistance Total wall load =10 KN/ m
9 Design of Aluminum Structures 3) Parapet load -
It is foreseen that Eurocode 4 will consists of three a) Wall load = 1.2 x 0.15 x19.5 =3.5 KN/ m
parts: b) Plaster =1.2 x 2 x 0.025 x20 =1.2 KN/ m
Total Parapet load = 6 KN/ m (considered)
Part 1 - General Rules and Rules for building
4) Balcony wall load - 5 KN/ m
Part 2 - Bridges
5) Staircase load - 20 KN/ m
Part 10 - Fire Resistance
B) Live load:- 2 KN/m2 from IS 875-1987( part 2)
Part 1 of Eurocode 4 gives a general basis for the
C) Wind load:-
design of composite slabs, beams, columns and
Wind load is taken as per IS 875-1987 (Part 3).
frames detailed rules which are mainly applicable to
For Nagpur region basic wind speed is 39m/s and the
ordinary buildings subjected to predominant static
loading, except that subjects mainly for bridges values of wind pressure at various heights are
calculated for assigning wind load to STAAD pro.
(fatigue, Prestressing and precambering of beams)
The basic wind speed is given by:-
have been excluded.
Vz = VbK1 K2 K3
3.1.1: Safety Concept
Where,
All Eurocodes are written in a limit state design
Vz = Designed wind speed at any height z in
format. The limit state conditions are expressed as
m/s.
those of the ultimate limit state, thus:
K1 = Probability factor
K2 = Terrain height and structure Size factor
Sd Rd K3 = Topography factor
Wind speed is calculated as:-
Where: S = “solicitation” (action or action effect) Vz = VbK1 K2 K3
R = resistance Vb = 39 m/s, Class = B
d = design K1 = 1 (IS 875-1987 part 3)
This means that the action effects shall be less than or K3 = 1 (IS 875-1987 part 3)
equal to the resistance capacity. The action effect K2= as given in table below (IS 875-1987
result from structural analysis with the loads and the part 3)
respective partial safety factors according to Eurocode Table 3.1 : Wind pressure values Pz for different
2, 3 and 4. heights
PROBLEM DEFINITION AND LOADING
CONSIDERATIONS Height(in K2 Vz(m/s)= Pz(N/m²)
For comparison of composite column and M) value VbK1 K2 =0.6 Vz²
R.C.C. column a G+10 storied building is considered. K3
It is analyzed using STAAD pro software.
3.1 Description of the structure:- 0 1.03 40.17 968.17
The architectural plan of building is as 10 1.03 40.17 968.17
shown in figure 4.2. The plan dimensions of the 15 1.07 41.73 1044.83
building are 26.2 m X 15.00 m (X x Z). It consists of 20 1.10 42.9 1104.24
2 bhk flats at each floor and one staircase. The height 30 1.13 44.07 1165.29
of the building is 34.2 m. 34.2 1.132 44.14 1169.00
3.2 Loadings considered on structure:- 50 1.18 46.02 1270.7
The basic loadings on both R.C.C. and
composite type of structures are kept same. The
following are the basic loading considered on the
building. Dead load and live loads have been taken
according to IS 875-1987 (Part 1, and 2 respectively)
A) Dead loads:-

Wind pressure Pz(KN/m²) vs Height(m)
2) Material properties:-
Wind pressure K N /m²

1.4
1.2
1
Grade of Concrete: - M 30
0.8
0.6
Wind Steel: - Fe 415
Density of concrete: - 25000 N/m3
pressure
0.4 Pz(KN/m²)
0.2
0
0 10 15 20 30 34.2 50
Damping for R.C.C. structure:-5%
Height(m) Damping for steel structure:-3%
Graph 3.1:- Graph of wind pressure(KN/m²) vs. 3) Analysis of R.C.C. building:-

story height (m) for building After assigning the above loads cases and
D) Earthquake load – material properties in STAAD pro, analysis of
The earthquake loading is as per IS building is done.
1893-2002. The building is considered to be in zone
3. The damping for R.C.C. structure is 5% and for
steel structure is 3%. Design horizontal seismic 4) Design data of R.C.C. columns:-
coefficient is calculated for assigning earthquake load The force based on 3D analysis for
to STAAD pro. various column members i.e. axial forces in x
From IS 1893-2002 direction and moments in y and z directions are
Z = 0.16 (Zone -3) taken from the extract of STAAD pro for design
I = 1 (Importance factor) calculation.
R = 5 (Response reduction factor)
Design horizontal seismic coefficient is given as- Composite column structure :-
Z I Sa The analysis of composite structure is
Ah  done as 3D framed structure same as that of R.C.C.
2 R g model by using STAAD pro. In STAAD pro
In X- direction: - software there is no provision for composite
For rocky or hard soil sections. For this purpose the composite sections
0.09h are converted in equivalent steel section. The
T section is then assigned as prismatic general and the
d analysis is done.
0.09  34.2
T = 0.79
15
Sa/g = 1/T =1/0.79 =1.26
0.16 1
Ah    1.26 = 0.207
2 5
Similarly in Z-direction
0.09  34.2
T =0.6
26.2
Sa/g = 1/T =1/0.60 =1.67
0.16 1
Ah    1.67 =0.0267
2 5
R.C.C. column structure:-
The analysis for R.C.C. structure is done as a 3D-
framed structure using STAAD Pro. Beams and
columns are modeled as beam elements. The slab is
modeled as beam element to carry the live load as
distributed pressure load. The material used for the
whole structure is reinforced concrete. The design
follows IS 456 -2000 code.
There are many analysis procedures for
earthquake forces coming on the building. In analysis,
the loading cases considered are according to IS 1893
-2002. Fig 4.1 :- Architectural plan of complete
building

Center of gravity of the structure is located at:-
1) Analysis of composite structure:- x = 12.24 m, y = 13.05 m , z = 7.54 m
After assigning the above loading cases and total self weight = 14088.43 kN
material properties in STAAD pro, analysis of
building is done. 3)Table 4.2:- Max nodal displacements for R.C.C.
column building (summery)
2) Design data for composite columns:-

The data based on 3D analysis for various
columns members in composite frame i.e. axial forces
in x-direction and moments in y and z -directions are
taken from the extract of the STAAD pro.
From above max nodal displacements for node 1229

is 25.40 mm
4) Table4.3:Max support reactions for R.C.C.

column building (summery)
Figure 3.2: - 3-D view of building (STAAD pro)
ANALYSIS AND DESIGN OF R.C.C. BUILDING From above max support reactions for node 987 is
3856 kN
Analysis and design results of STAAD pro for
R.C.C. column building:- 5) Table 4.4:– Eleventh floor storey drift for
Analysis of building is done using R.C.C. column building
STAAD pro for loads and load combinations given in
chapter 4. Then design of columns is been carried out
manually for the worst loading conditions as per the
STAAD pro output extract.
After the analysis and design of the
structure following results are obtained as extract of
STAAD pro output.
Table 4.1:- Frequency and time period for R.C.C.

building
Mode Frequency(cycles/sec) Period(sec)

1 0.463 2.15861
2 0.565 1.77066
3 0.692 1.44449
2) Self weight R.C.C. building :-

6) Analysis results for column group(5) 3)Table 5.2:- Max nodal displacements for
Table 4.5: Axial load, shear force and composite column building (summery)
bending moment for R.C.C. columns group
(5)
From above max nodal displacements for node 1229

is 24.866 mm
4)Table 5.3:-Max support reactions for composite

column building( summery)
From above max support reactions for node 987 is

7) Analysis results of R.C.C. column no 16,25- 3522.2 KN
Table 4.6:- Output extract from STAAD pro
for R.C.C. column no 16,25 5) Table 5.4:- Eleventh floor storey drift for
composite column building
5.0Analysis and design results of STAAD pro 6)Analysis results for column group(5):-
for composite column building Table 5.5 : Axial load, shear force and bending
The properties of composite columns moment for composite columns group(5)
designed in chapter-6 are input to STAAD pro
investigate the changed behavior of the structure.
Loads and load combination in chapter 4 are input to
STAAD pro. The analysis and design results are
tabulated to compare with R.C.C. column building.
After the analysis and design of the structure
following results are obtained as extract of STAAD
pro output.
1) Table 5.1:- Frequency and time period of
composite building
Mode Frequency(cycles/sec) Period
1 0.639 1.56573
2 0.723 1.38315
3 0.896 1.11569
2) Self weight of composite building:- 7) Analysis results for composite column no 16,25:-
Center of gravity of the structure is located at: Table 5.6:- Output extract from STAAD pro for
x = 12.24 m, y = 14.03 m, z = 7.54 m composite column no 16,25
Total self weight = 8457.034 KN

1) A. Braconi1, W. Salvatore1, R Tremblay2 , and,
O.Bursi3, (Mar.2002), “Behavior and modeling of
partial-strength beam to column composite joints for
seismic applications”, Journal of structural engg, vol.
120, ASCE, ISSN 0733 -9445-O697
2) C.C. Weng, S.I. Yen, (June 2002), “Comparisons
RESULTS AND DISSCUSSION of concrete-encased composite column strength
The analysis and design results of both the types of provisions of ACI code and AISC specification”,
building, composite and R.C.C. column buildings are Journal of structural engg, vol 127, no 23, ASCE,
being given in chapter. The comparison of columns ISSN 0733 -9445-1190
on basis of axial force, shear force, bending moment, 3)C.C. Weng, l S.I yen, 2 and C.C.Chen2, (Oct.2001),
frequency, storey drift and time period is carried out “Shear strength of concrete- encased composite
as fallow. structural members”, Journal of structural engg,.vol.
Table 6.1:- Comparison of composite column and 127, no 1, ASCE ,ISSN 0733 -9445-1190.
R.C.C. column buildings 4)Euro code 3, “Design of composite steel and
concrete structures,” European committee for
standardization committee European de normalization
europaisches committee fur normung”.
5) Euro code 4 , “Design of composite steel and
concrete structure,” European committee for
standardization committee European de normalization
europaisches committee fur normung”.
6)F. Chung Beng, PhD DIC & R. Narayanan , MSc
(Engg.) PhD, FICC, “Composite column design to
Euro code 4,”
7)IS: 456(2000), “Indian Standard Code of Practice
Table 6.2:- Comparison of composite column and
for Plan and Reinforcement concrete (Fourth
R.C.C. column no 16,25
Revisions)”, Bureau of Indian Standards (BIS), New
Delhi.
8) IS 800(1984), “Indian Standards Code of Practice
for General Construction in Steel (First Revisions) ”,
Bureau of Indian Standards (BIS), New Delhi.
9) IS 875(1987-Part 1), “code of practice for design
CONCLUSION loads (other than earthquake) for buildings and
structures ,Dead loads,” Bureau of Indian standards
Analysis and design results of G+10 storied
(BIS), New Delhi.
building with composite columns and R.C.C. columns 10) IS 875(1987-Part2), “code of practice for live
are given in chapters 5, 6 and 7. The comparison of loads,” Bureau of Indian Standards (BIS), New Delhi.
results of composite column building and R.C.C.
11)IS 2062(1999) , “Steel for General structural
column building shows that:-
purposes specification,” Bureau of Indian Standards
1) The frequency of the composite column building
(BIS), New Delhi.
is increased by 30% to 35% than that of the
12) IS11384 (1985), “Code of Practice for Design of
R.C.C. column building.
Composite Structure,” Bureau of Indian Standards
2) Storey drift of composite columns reduce by 15%
(BIS), New Delhi.
to 30% than that of R.C.C. columns. 13)R.P. Johnson, MA, MICE, “Composite Structure
3) In composite columns, axial loads reduce by 3% of Steel and Concrete,” (Vol-1,” Blackwell, scientific
to 5% whereas shear force and bending moments
publication, UK 1987
increase by 30% to 50% and 30% to 65%
14)R.P. Johnson, MA, MICE, “Composite Structure
respectively than that R.C.C. columns.
of Steel and Concrete”(Vol-II” Blackwell, scientific
4) Cross-section dimension of composite columns
publication, UK, 1987
are lesser than that of R.C.C. columns by 5% to
15)Dr. R. Narayanna’s ,“Teaching resources material
25%.
for “structural steel Design”, volume 1, 2, & 3,”
5) Load and moment carrying capacity are increased INSDAG publication prepared by IIT, Chennai, Anna
by 5% to 20% and 30% to 70% respectively than University & Structure Engineering Research Centre,
that of R.C.C. columns.
Madras under Dr. R. Narayanna’s Leadership.
6) Good fire resistance and corrosion protection are
16)Thierry chicoine; 1 Robert trmblay2; Bruno
achieved due to increased concrete cover in
Massicotte3; James M. Ricies4; and Le-Wu Lu5
composite columns than that of R.C.C. columns.
,(Mar.2002) “Behavior and Strength of partially
Encased Composite Columns With built up sections ”,
REFERENCES
Journal of structural engg,vol 126, n0 9, ASCE.

ISBN 978-81-921121-0-7
SHEAR STRENGTH OF ORGANIC SOIL TREATED WITH CEMENT

GROUT AND MODIFIED WITH ADDITIVES
SinaKazemian
Dept.. of Civil Engineering, Bojnourd Branch, Islamic Azad University, Bojnourd, Iran, Corresponding Author Email:
sina.kazemian@gmail.com
Arun Prasad
Dept. of Civil Engineering, Banaras Hindu University, Varanasi-221005, India, Email: prasada@bhu.ac.in
ABSTRACT: The varieties of additives used in grouting are many including cement. Calcium chloride is also
added as an accelerator in a grout. It decreases the setting time of grout and also has a minor effect on fresh
grout properties. Kaolinite or some other fine mineral soils are used as filler in the grout to reduce the cost of
grout. In this study, the effect of calcium chloride and kaolinite on shear strength of soil, treated by cement grout
was investigated. Samples were prepared by mixing specific amounts of cement, water, and organic soil with
different ratios of calcium chloride and kaolinite. Unconfined compressive strength (UCS) of the treated soil
was evaluated after curing for 3 and 30 days. The results showed that kaolinite has positive effect on the shear
strength of treated soil. By increasing the kaolinite content in the grout, the shear strength of samples increased.
On the other hand, calcium chloride seemed to have different effects on the shear strength of samples.
INTRODUCTION compressive strength or finally (even more

Soil improvement methods are classified to following frequently) to reduce conductivity and interconnected
principles: (i) consolidation (e.g. prefabricated porosity in an aquifer (Moseley and Kirsch, 2004).
vertical drains & surcharge, vacuum consolidation, Gleize et al. (2003) stated that pozzolanic materials
stone columns), (ii) chemical modification (e.g. deep and fillers (kaolinite and sand) can be added to the
soil mixing, jet grouting, injection grouting), (iii) chemical grouts as these materials change the
densification (e.g. vibro-compaction, dynamic microstructure of mortars and consequently modify
compaction, compaction grouting), and (iv) some of its properties.
reinforcement (e.g. stone columns, geosynthetic Chemical stabilization is an effective method to
reinforcement) (Raju, 2009). improve the soil properties by mixing additives to
The demand for improving and stabilizing land for soils. Today, commercially available chemical grouts
different purposes is expected to increase in the future cover a wide range of materials, properties and costs,
and the best way to fulfill it is by using DMM. The and give the grouter the opportunity to select a grout
main advantage of these methods is long term for specific job requirements. These reagents bring an
increase in strength, especially for some of the almost instant setting time and produce very low
binders used. Pozzolanic reaction can continue for penetrability type gel that are unsuitable for
months or even years after mixing, resulting in the permeation treatments; but in the soils with high
increase in strength of cement stabilized clay with the ground water level, it can be effective as the water
increase in curing time (Bergado, 1996; Roslan and causes the grout to be diluted (Rawlings et al., 2000).
Shahidul, 2008). The function of the accelerator is to control gel time
Majority of the companies which are working in these and impart strength to the gel. The effect of an
fields agree that the DMM can be divided into three accelerator is important at temperatures below 37 °C
common techniques (Andromalos et al., 2000; Keller, and increases in importance as the temperature
2005; Raito, 2006; SDFEC, 2007; Kazemian and decreases. An excessive amount of accelerators may
Huat, 2009a): (i) Shallow Soil Mixing (SSM), which result in undesirable flocculation or formation of local
uses a single mechanical mixing auger located at the hardening. This causes variations in both, the gel and
end of the drilling tool (Kelly bar), (ii) Cement Deep setting times that would tend to plug injection
Mixing System (CDM), which utilizes a series of equipment or restrict penetration, resulting in poorly
overlapping augers and mechanical mixing shafts, grouted area. Therefore, a retarder should be added in
(iii) Injection and grouting methods which can be the mixture for delaying the setting time and
considered a type of soil mixing. The Federal formation of gelation (USACE, 1995). Calcium
Highway administration has suggested that these chloride has been used in concrete since 1885 and
techniques can be classified based on (Elias et al., finds application mainly in cold weather, when it
1998): (i) method of additive injection (i.e. wet or dry allows the strength gain to approach that of concrete
injection), and (ii) method by which additive is mixed cured under normal curing temperatures (Rixom and
(i.e. rotary/ mechanical energy or by high pressure Mailvaganam 1986). Calcium chloride has the ability
jet). to accelerate cement hydration and reduce set time by
Grouting or injection generally is used to fill voids in as much as two thirds. Research has shown that the
the ground (fissures and porous structures) with the important contributions of cold weather protection
aim to increase resistance against deformation, to and early strength of concrete, calcium chloride
supply cohesion, shear–strength and uniaxial provides other benefits as well: (i) improves

workability, (ii) improves strength of air-entrained Table3 Grout formulae with their notations and group
concrete, (iii) reduces bleeding, these advantages numbers
cause to produce better quality concrete faster.
Concrete acceleration with calcium chloride greatly
Grout Grout Formula Grout Grout Formula
facilitates completing jobs as quickly and No. (%) No. (%)
economically as possible (The Dow Chemical
Company, 2006). K Ca Ce K Ca Ce
This study was aiming to describe the effects of (%) (%) (%) (%) (%) (%)
different ratios of calcium chloride and kaolinite on 20 0.0 20 20 0.0 30
cementation and pozzolanic reactions process of
cement grout in the soil, characterized by unconfined 20 0.1 20 20 0.1 30
compressive strength (UCS). 20 0.25 20 20 0.25 30
1 2
MATERIALS AND METHODS 20 0.5 20 20 0.5 30
Materials 20 1.0 20 20 1.0 30

Cement used in this study (as binders) was sourced 20 2.5 20 20 2.5 30
from Anuza Enterprise Company, Malaysia. Calcium
chloride anhydrous powder (CaCl2) and kaolinite 20 5.0 20 20 5.0 30
were used as a reactor/accelerator and filler 20 10 20 20 10 30
respectively. The kaolinite [Al2Si2O5(OH)4] structure
is made up of silicate sheets (Si2O5) bonded to 0 1.0 20 0 1.0 30
aluminum oxide/hydroxide layers [Al2(OH) 4] called
10 1.0 20 10 1.0 30
gibbsite layers. Tables1 and 2 show the chemical 3 4
compositions of calcium chloride and kaolinite 20 1.0 20 20 1.0 30
respectively.
30 1.0 20 30 1.0 30
Table1 Chemical composition of calcium chloride Note:K- Kaolinite, Ca- Calcium chloride and Ce- Cement
Constituent (%) After mixing thoroughly, the confection was

Minimum Assay 96% transferred to a cylindrical container of the same
SiO4 0.02 height and inner diameter as that used in the UCS test
(38mm in diameter and 76mm high). Curing of the
Ca(OH)2 0.04 samples was accomplished by keeping the container
Mg+2 0.6 with the confection submerged in distilled water in a
Alkalis (sulfate) 0.6 large tray for 3 and 30 days ata constant temperature
(22±2ºC). For repeatability and reproducibility of the
tests, six samples were prepared for each grout
Table 2 Physico-chemical composition of kaolinite
formula as mentioned before; and the variation in the
shear strength of the six samples, during any curing
Constituent (%) period (3 and 30 days) was not more than 5%. The
SiO2 45.8 values of the shear strength reported here are the
Al2O3 39.55 average values.
Fe2O3 0.57 The combinations of the chemicals used in preparing
the grout are arranged according to the increasing
CaO 0.41
component in the confection for easy readability and
MgO 0.14
are shown in Table 3. UCS tests were performed
according to BS 1377-7-7(1990) on the samples at the
Experimental Methods end of 3 and 90 days of curing.
For preparing the samples to investigate the effect of
calcium chloride and kaolinite on shear strength of RESULTS AND DISCUSSION
cement grout, different ratio of cement was first The influence of different compounds (calcium
thoroughly homogenized with the water in the ratio of chloride and kaolinite) on cement grout has been
1:1 by a household mixer and then the grout were investigated and the results are presented in the
mixed with organic soil (750g) and the desired following paragraphs.
amount of kaolinite, and calcium chloride were added
to it (Table 3).

Influence of Calcium Chloride on Shear Strength confection and thereby leading to the re-stabilization
of Cement Grout of the colloidal fraction and deflocculating of the
The influence of calcium chloride was studied by large size particles.
preparing samples with different concentrations of ACI Committee 212 (1963) and Transportation
calcium chloride (0, 0.25, 0.5, 1, 2.5, 5 and 10% by Research Board (1990) emphasized that, according
weight of soil) as mentioned in grout Nos. 1 and 2. laboratory tests,the maximum increases in
The effect of calcium chloride on the shear strength compressive strength of concrete resulting from the
and moisture content of the samples after 3 and 30 use of 2% of calcium chloride by weight of cement
days of curing are presented in Fig. 2 for grout ranges from 400 to 1,000 psi (2.8 to 6.9 MPa) from 1
Nos. 1. through 7 days, curing at 70° F (21° C) and long-term
strength is usually unaffected and is sometimes
reduced, especially at high temperatures.
Influence of Kaolinite on Shear Strength of

Cement Grout
The effect of kaolinite on samples was investigated
by preparing different samples based on grout Nos. 3
and 4 (Table 3) and the results are presented in Fig. 3.
Fig. 2 Influence of different calcium chloride

concentrations for grout No. 1
It was observed that (Fig. 2) by increasing the

calcium chloride concentration in grout No. 1 from 0
to 1%, the shear strength of the samples increased
from 208 to 264kPa after three days of curing.
Furthermore, by increasing the calcium chloride up to
10%, the trend of the shear strength showed a Fig. 3 Influence of different kaolinite concentrations
reversal, that is, the shear strength decreased from for grout No. 3
264 to 138kPa. Similarly, the shear strength of the
samples varied with an increase in curing time from 3 The grout Nos. 3 and 4 consisted of different amounts
to 30 days for grout No. 1. It increased from 248 to of kaolinite (0, 5, 10, 20, and 30% by weight of soil),
278kPa by increasing calcium chloride from 0 to 1%, while the amount of other additives were kept
then the trend reversed and the shear strength constant. Firstly, the shear strength of the samples
decreased from 278 to 149kPa when the with grout Nos. 3 and 4, after three days of curing,
concentration of calcium chloride increased from 1 to was observed to increase with an increase in the
10%, and the trend was similar for grout No 2 also. kaolinite content. Furthermore, with an increase in the
Kazemian et al. (2009b) and Huat et al. (2011) have curing time, the shear strength increased as well. For
reported that when water comes in contact with grout No. 3, the shear strength increased from 238 to
cement, three phenomena take place; (i) cement reacts 275kPa and from 254 to 284kPa, respectively, after 3
with water, this is called hydration, (ii) there are and 30 days of curing (Fig. 3). The trends of variation
pozzolanic reactions between Ca(OH)2 from burnt in shear strength for the samples with grout No. 4
cement and pozzolanic minerals in the soil, and (iii) were similar to grout No. 3.
there is ion exchange between calcium ions (from The above findings agree well with the fact that (as
cement and additives) with ions present in the mentioned earlier), cement initiates a chemical
kaolinite, which leads to an improvement in the reaction with water called hydration. When tricalcium
strength of the mixture. First step (when the calcium silicate (C3S) and dicalcium silicate (C2S) are mixed
chloride concentration in less than 1%), the with water, calcium ions are quickly released into the
confection particles present in grout tend to go solution with the formation of hydroxide ions. When
towards the particles with a zero net charge and will the concentration of calcium and hydroxide ions
not repel each other. This is the reason for an increase reaches a certain threshold value, calcium hydroxides
in shear strength the samples. Conversely, due to the crystallize out of solutions and this finally leads to the
addition of extra calcium cations, the charge balances production of calcium silicate hydrate (C–S–H) and
in the confection, giving a positive charge to the thus, an increase in the shear strength of the treated

samples. The pozzolanic reaction takes place with the Gleize, P.J., Muller, A., and Roman, H.R.
mineral parts of the soil; since there is kaoilinite in (2003).Microstructural investigation of a silica
the mixture, it contains aluminous and siliceous fume-cement-lime mortar.Cement & Concrete
minerals. Some of these possess pozzolanic Composites, 25: 171-175.
properties, i.e. under certain conditions they can react Huat, B.B.K., Kazemian, S., Prasad, A., and
with Ca(OH)2 to form calcium aluminate silicate Barghchi, M. (2011). A study of the engineering
hydrate (C–A–S–H), which leads to an increase in the behavior of peat stabilized by DMM: Lab model
shear strength together (Janz and Johansson, 2001; and FE analysis. Scientific Research and Essays
Huat et al., 2011). As mentioned earlier, the third J. 6(1): 196-204.
reaction in this confection is ion exchange between Janz, M., and Johansson, S. (2002).The function of
calcium ions (from cement and calcium chloride) different binding agents in deep stabilization.
with ions present in the kaolinite which cause to Report 9, Swedish Deep Stabilization Research
increase the shear strength by increasing the kaolinite Center, Linköping, Sweden, 14-47.
and this was also evident in the confections (Figs. 2 Kazemian, S., Prasad, A., Huat, B.B.K., Ali, T.A.M.,
and 3). and Aznieta, F.N. (2010).Effect of cement,
sodium silicate, kaolinite and water on the
CONCLUSIONS viscosity of the grout.Scientific Research and
This study was carried out to investigate the effects of Essays J. 5(22): 3434-3442.
the calcium chloride and kaolinite on shear strength Kazemian, S., and Huat, B.B.K. (2009a).Assessment
of treated soil. It was observed that, by increasing and comparison of grouting and injection
calcium chloride within 1% by the weight of soil, the methods in geotechnical engineering.European
shear strength of grout increased, after that it Journal of Scientific Research. 27(2): 234-247.
decreased due to an increased positive charge on the Kazemian, S., Asadi, A., Prasad, A., and Huat,
surface of particles, thereby leading to the re- B.B.K. (2009b).Settlement Problems in Peat Due
stabilization of the particles and deflocculating of the to Their High Compressibility and Possible
large size particles. The shear strength increased by Solution Using Cement Columns. In Shen-en C,
increasing the kaolinite in the grout due to increasing de Leon AD, Dolhon AM, Drerup MJ, Parfitt K
of pozzolanic reactions and ion exchange between (eds) Forensic engineering 2009: pathology of
calcium ions (from cement and additives) with ions the built environment, Proceedings of the 5th
present in the kaolinite and mineral part of soil. Congress on Forensic Engineering, ASCE,
November, Washington, 3: 255-265.
REFERENCES Keller Company. (2005). Retrieved from Keller:
ACI Committee 212.(1963). Admixtures for Http://www.kellergrundbau.com.
concrete.ACI Journal Proceedings 60(11):1481- Moseley, M.P., andKirsch, K. (2004). Ground
524. Improvement.2nd ed. UK, Taylor & Francis, UK.
Transportation Research Board. (1990). Admixtures Raito Company. (2006). Retrieved from Raito:
and ground slag for concrete. Transportation Http://www.raito.co.jp/english/
Research Circular no. 365, Washington: Raju, V.R. (2009). Ground improvement principles
Transportation Research Board Index, 50pp. and applications in Asia. In Leung CF et al. (eds)
Andromalos, K.B., Hegazy, Y.A., and Jasperse, B.H. Ground Improvement Technologies and Case
(2000).Stabilization of soft soils by soil Histories, Singapore, Geotechnical Society of
mixing.Proceedings of the Soft Ground Singapore (GeoSS), 43-66.
Technology Conferenc, Noorwijkerout, Ramachandran, V.S. (1984). Accelerators. In
Netherland. Ramachandran VS, Park Ridge NJ (eds)
Bergado, D.T., Anderson, L.R., Miura, N., and Concrete admixtures handbook: Properties,
Balasubramaniam, A.S. (1996).Soft Ground science, and technology, Noyes Publications.
Improvement in Lowland and Other Rawlings, C.G., Hellawell, E.E., and Kilkenny, W.M.
Environments.New York, USA: ASCE Press, (2000).Grouting for Ground Engineering. UK,
123-183. CIRIA Press.
British Standard Institution, BS.(1990). Methods of Rixom, M.R., and Mailvaganam, N.P.
Test for Soils for Civil Engineering Purposes. (1986).Chemical Admixtures for Concrete.
London, UK: British Standard Institution. Cambridge, England: The University Press.
Construction Industry Res. and Info. Assoc.(CIRIA) Roslan, H., and Shahidul, I. (2008).properties of
(2000). Grouting for Grouting Engineering. stabilized peat by soil-cement column method.
London, UK: CRIRIA Press, pp 12-57. The Electronic Journal of Geotechnical
Elias, V., Welsh, J., Warren, J., and Lukas, R. Engineering (EJGE Journal), 13( J).
(1988).Ground improvement technical Shideler, J.J. (1952). Calcium chloride in
summaries.Department of Transportation, concrete.Journal of American Concrete Institute,
Federal Highway Administration, Publication 537-559.
No. FHWA–SA–98–086, 13-48.

Soilmec Drilling and Foundation Equipment http://www.glchloride.com/brochure/Brochure%
Company (SDFEC). (2007). Retrieved from 20%20Calcium%20chloride%20in%20Portland
SDFEC: Http://www .soil mec.it/ %20cement%20concrete.pdf
The Dow Chemical Company, (2006). Calcium US Army Corps of Engineers. (USACE) (1995).
Chloride in Portland Cement Concrete. Chemical grouting.US Army Corps, Washington
Retrieved: DC, USA.

ISBN 978-81-921121-0-7

POTENTIAL EVAPOTRANSPIRATION ASSESSMENT BY USING MWSWAT
MODELING ON SABARMATI RIVER
Shaikh M Zuned
P G Student Water Resources Management Department,
L D College of Engineering, Ahmedabad Gujarat-380015, India. E-mail: zunedshaikh@gmail.com
P P Lodha
Associate Professor, Water Resources Management Department, L D College of Engineering, Ahmedabad, Gujarat-380015, India. E-mail:
pplodha@gmail.com
ABSTRACT: Potential Evapotranspiration is a phenomenon representing the hydrologic response of a watershed to

precipitation. Hydrologic response of the watershed to precipitation is studied in order to estimate the Potential
Evapotranspiration using remotely sensed data through GIS. The present paper is focused on Sabarmati River which is passing
through states of Rajasthan and Gujarat and meets Gulf of Cambay. The study area is covering Gujarat Portion up to Vasna
barrage which is situated downstream of Ahmedabad city. Daily rainfall and maximum and minimum temperature data of eight
years (1998-2005) are used in the study. “SWAT” (Soil and Water Assessment Tool), a distributed hydrological model, is a
public domain model actively supported by the USDA has been applied for this research work. The main components of SWAT
include weather, surface runoff, return flow, percolation, Evapotranspiration, transmission losses, pond & reservoir storage,
crop growth & irrigation, groundwater flow, reach routing, nutrient & pesticide loading, and water transfer. The study adopts a
Map-Window (GIS software) interfaced SWAT based distributed model for the study of watershed to estimate the potential
Evapotranspiration. The soil data, land use data, DEM, and drainage map of study area are used in the study a geographical
information system (GIS) is used for generating drainage networks, land cover/land use data and soil series. The SWAT model
is made to run and results are measured by applying appropriate Potential Evapotranspiration methods.
INTRODUCTION spatially and temporally variable. Advances in distributed

Water is a precious natural resource, vital for sustaining all parameter hydrologic modeling and its integration with
life on the earth. Less than 3% is fresh water and less than is Geographical Information Systems (GIS) have led to the
accessible to mankind. It is however not uniformly distributed development of powerful tools like;
and space. Water is needed for irrigation for growing food, 1. Agricultural Non-point Source Pollution model
drinking needs of people and to maintain the natural (AGNPS).
ecosystem. Water as a resource requires special attention due 2. Chemicals, Runoff and Erosion from Agricultural
to its multiple benefits and the problems created by its Management Systems (CREAMS).
excesses, storages and deterioration of quality. It needs 3. Groundwater LOADING effects of Agricultural
development, i.e. storage for surface water and pumping from Management Systems (GLEAMS).
groundwater. 4. Hydrologic Simulation Program FORTRAN (HSPF).
A number of advanced techniques have been developed on 5. Simulator for Water Resources in Rural Basins
various aspects of water conservation and quality preservation (SWRRB).
by researchers. Also, there are traditional approaches 6. Soil and Water Assessment Tools (SWAT).
available and are being used very effectively in various parts SWAT (Soil and Water Assessment Tool) was chosen as the
of the world. Some of these techniques are; first model for this effort, as it is free, open source, actively
1. Systems Approach to Hydrological Problems and supported by model developers, well documented, and has
Mathematical Modeling. been tested around the world. It has a substantial and growing
2. Geographical Information Systems and Remote number of users in the developing world, as shown by the
Sensing in Water Resources Management. papers presented at its biennial conference. The SWAT
3. Water Pollution and environmental considerations. Interface was written as a plug-in for MapWindow. It is
4. Watershed Management, Education, Artificial similar in scope to the ArcView interface to SWAT, which is
Recharge and Rain Water Harvesting. probably the way most existing SWAT users set up the model.
Catchment hydrology is difficult to observe, conceptualize LITERATURE REVIEW
and model because it includes interactions among several PET is the rate at which Evapotranspiration would occur
nonlinear processes, under environmental conditions that are from a large area uniformly covered with growing vegetation

that has access to an unlimited supply of soil water and that To study MapWindow SWAT and apply it to simulate
was not exposed to advection or heat storage effects. Evapotranspiration for Sabarmati river.
MWSWAT uses one of the following three methods to To assess PET process using appropriate method with
estimate Evapotranspiration. watershed modeling.
Scope of the study
PENMAN-MONTEITH METHOD As described earlier, analytical methods are too laborious and
The Penman-Monteith equation combines components that time consuming. They also require complex equations to be
account for energy needed to sustain evaporation, the strength solved. Hence there is a need of advance techniques to replace
of the mechanism required to remove the water vapor and them. The present study is an attempt of analyzing PET using
aerodynamic and surface resistance terms. The Penman- MWSWAT. In the present study PET and various other
Monteith equation is: parameters are considered. For this various Landcover change
 (Hnet  G)   air  cp[e0 z  ez]/ ra and Temperature increase scenarios are generated using
E  SWAT (Soil and Water Assessment Tool) Model. Then using
    (1  rc / ra)
MWSWAT model PET, surface runoff, discharge, soil
erosion etc are estimated and analysed.
where λ is the latent heat flux density (MJ m-2 d-1), E
Methodology
is the depth rate evaporation (mm d-1), ∆ is the slope of the
The SWAT model requires a number of GIS data layers as
saturation vapor pressure-temperature curve, de/dT (kPaC-1),
input. These include subbasin map, land use, soils, and digital
Hnet is the net radiation (MJ m-2 d-1), G is the heat flux
elevation model (DEM) and weather station layers. All of the
density to the ground (MJ m-2 d-1), ρair is the air density (kg
layers are then exported to the MWSWAT, which is the
m-3), cp is the specific heat at constant pressure (MJ kg-1C-1),
MapWindow interface with the SWAT model. For generating
e oz is the saturation vapor pressure of air at height z (kPa), e
the database the following data products are used. They are
z is the water vapor pressure of air at height z (kPa), γ is the
1. Survey of India (SOI) topomaps on 1:50,000 scale.
psychrometric constant (kPa C-1), rc is the plant canopy
2. Indian Remote Sensing(IRS) satellite data and
resistance (s m-1), and ra is the diffusion resistance of the air
3. Other collateral data.
layer (aerodynamic resistance) (s m-1).
After exporting the data to MWSWAT, step by step procedure
PRIESTLEY-TAYLOR METHOD
is followed in order to get the output. The procedure is
Priestley and Taylor (1972) developed a simplified version of
described in section. The output is obtained in a notepad
the combination equation for use when surface areas are wet.
which is then verified with the original weather station data.
The aerodynamic component was removed and the energy
Methodology is described by steps shown below;
component was multiplied by a coefficient, α pet = 1.28,
 Study of PET
when the general surroundings are wet or under humid
 Collection of data for Sabarmati river
conditions
 Study of MWSWAT

 E O   pet   ( H net  G )  Exporting data to MWSWAT
 
 Model run
where λ is the latent heat of vaporization (MJ kg-1), E o is
 Result/Output
the potential Evapotranspiration (mm d-1), α pet is a
 Analysis
coefficient, ∆ is the slope of the saturation vapor pressure-
temperature curve, de/dT (kPa C-1), γ is the psychometric
RESULTS
constant (kPaC-1), H net is the net radiation (MJ m-2 d-1), and G
THE RESULTS ARE SHOWN IN THE FORM OF A TABLE
is the heat flux density to the ground (MJ m-2 d-1).
ACCORDING TO THE METHOD EMPLOYED
PET
HARGREAVES METHOD YEAR PENMANMON PRIESTLEYT HARGRE
The Hargreaves method was originally derived from eight T TH AYLOR AVES
years of cool- season Alta fescue grass lysimeter data from 1998 922.88 680.29 854.499
Z 972.596 714.662 895.558
Davis, California (Hargreaves, 1975). Several improvements 2000 964.661 746.799 921.247
were made to the original equation (Hargreaves and Samani, 2001 904.49 675.577 862.541
1982 and 1985) and the form used in SWAT was published in 2002 1020.063 730.969 910.006
1985(Hargreaves et al., 1985): 2003 847.495 658.591 855.508
2004 1028.106 740.152 889.266
 EO  0.0023  HO  (Tmx  Tmn)0.5  (T av  17.8) 2005 1001.597 714.863 911.628
Where parameters appearing have the same meaning as TOTL 7661.888 5661.903 7100.253
described in earlier methods. REFERENCES
K.S.Raju, A.K.Sarkar and M.L.Dash “Integrated water
Objectives resources planning and management”
To replace conventional hydrological methods with latest Remote sensing and GIS by Bhatta.
techniques like GIS tools. Manual for learning MWSWAT.

ISBN 978-81-921121-0-7
LOCATION OF WATER HARVESTING STRUCTURES IN SEETHA

RIVER BASIN OF COASTAL KARNATAKA BY RS AND GIS
PERSPECTIVE – A CASE STUDY
Mohandas Chadaga
Professor, Department of Civil Engg., M. I. T., Manipal –576104, Email: mdchadaga@yahoo.com
Ravindranath
Assoc. Professor, Department of Civil Engg., M. I. T., Manipal –576104.
Sreeniwas Poudel S.
Department of Civil Engg., M. I. T., Manipal –576104
Varun Kadian
Department of Civil Engg., M. I. T., Manipal –576104
ABSTRACT: Identifying site specific water harvesting structures in a river basin is a tedious job due to the
geologic and geomorphic variations of the basin and peoples need. Remote Sensing (RS) and Geographical
Information System (GIS) help us in this respect due to their established advantages. In this study, a river basin
‘Seetha’ situated in the Western Ghats foothills of coastal Karnataka is considered for the purpose of
identification and location of different water harvesting structures by using RS and GIS tools. Many thematic
maps like Slope map, Land use land cover map, soil map and runoff coefficient maps were generated and
overlaid by developing certain decision rules. SCS-CN method was used for the run-off analysis. Four different
types of water harvesting structures were considered in this study. The results demonstrate the capability of GIS
and its application for planning and locating water harvesting structures over large areas. In particular, farm
ponds and check dams were found to more relevant for this study area. Percolation ponds were not much
feasible as per the framed design rules. It can be concluded that the water harvesting potential of Seetha river
basin is very good and can be harnessed in economical way by this type of systematic studies for the benefit of
local community.
INTRODUCTION harvesting structures (Abbas, I.I. et. al., (2010),

Water harvesting is a general term to include all Girish Kumar, et.al., (2008), Durga Rao,K.H.V.
systems that concentrate, collect and store runoff from Venkateswara Rao,V. and Roy, P.S. (2007), Gupta,
small catchments for later use in smaller user areas. K.K. et.al., (1997), Sivaraman K.R., et. al., Mintesinot
Food and Agriculture Organization (FAO) of the B., Kifle W. & Leulseged). Studies have been done on
United Nations defines water harvesting as, “the many river basins because of the diversities in
process of collecting and concentrating surface runoff topography, geology and geomorphologies of these
from a runoff area into a run-on area, where the river basins. Modern techniques like Remote Sensing
collected water is either directly applied to the (RS) and Geographic Information Systems (GIS) can
cropping area and stored in the soil profile for be adopted for this purpose with fair accuracy. A case
immediate use of the crop or stored in a farm water study of analysis for the location of water harvesting
reservoir for future productive uses, i.e. domestic use, structures in a west flowing river basin of coastal
livestock watering, aquaculture and irrigation.” The Karnataka is presented in this paper with the results.
collected water is also used in groundwater recharge.
STUDY AREA: SEETHA RIVER BASIN
As a general rule the catchment area from where the
Seetha river basin is one of the major river basins of
water is drawn is larger than the command area,
the Udupi district of coastal Karnataka. River Seetha
where it is collected and used. Water harvesting is
is a west flowing river originating in the foothills of
essentially a traditional system used since many
Western Ghat. Major rivers of this basin are Seetha,
centuries, now being made over to meet the present
Goli Hole, Kollangar Hole, Megadde Hole, Markal
day needs.
Hole, Kudlutirtha Hole, Neemar Hole, Sunkadi Hole
In the present day scenario, the reasons for people
etc. (Fig.1). All these rivers originate at the Western
resorting back to water harvesting is mainly because
Ghat region and flow westwards joining Seetha River
of depletion of ground water, unpredictable rainfall
at various points of its length and later flow to join the
due to climate change and most importantly due to the
Arabian Sea. The basin is located between longitudes
increased scarcity of water because of the destruction
75˚ 41’24”E and 75˚9’0”E and latitudes 13˚ 21’36” N
of precious sources due to unplanned urbanization and
and 13˚30’0”. It is surrounded by the Arabian Sea on
increased water pollution. Following this, various
the west, Thunga basin on the east, Gangolli river
systems have been developed in individual,
system basin on the north and the Swarna basin on the
communal or watershed levels to effectively harvest
south. The basin is accessible by National highway
water (Sharda, V.N., Juyal,G.P.,).
Number-66 (NH-66) from Mangalore in the south and
Many studies have been conducted in river basins to
Goa in the north and also by Konkan railway running
find sites suitable for locating different types of water

in north-south direction. River Seetha originates at an generate land-use/land-cover maps. Google Earth
altitude of 980m near Kutrikal Gudda in the Western images are also used for this purpose.
Ghats and flows westwards. It has a total length of Digitization
flow of 82 km (Chadaga Mohandas (2009)). Digitization is the process by which the raster data are
converted to vector data. Digitization is adopted to
OBJECTIVES OF THE STUDY develop vector layers like drainage and contour
1. Digital database creation - includes development of (Lillesand and Kiefer, ITC educational textbook
contour maps, drainage maps, slope maps along with series). Digitization was done by overlaying an
the soil and Hydraulic Soil Group Maps. empty vector layer over the extracted study area
2. Land use/land cover mapping of the study area. toposheet and then tracing the rivers and contours.
3. Rainfall and runoff modelling of the study area. ERDAS-Imagine 8.5 was used for this purpose.
4. Site suitability analysis for locating water Creation of Base Maps
harvesting structures by analysing the above database. Slope map-The contour map was further processed to
create a slope map. The digitized contour map was
LITERATURE REVIEW exported ARCVIEW. Then using the inbuilt
Durga Rao, K.H.V. and Sathish Kumar, D., (2004), interpolation function of the software a Digital
developed a prototype Spatial Decision Support Elevation Model of the study area was created. From
System (SDSS) integrating Landuse/Landcover the Digital Elevation Model slope map of the study
derived from the remote sensing data, real-time area was derived in terms of percentage slopes (Fig.2.
hydrological data, GIS and a model based subsystem and Fig.3.). The classification of percentage slopes are
for computing soil loss, land capability classification shown in the Table 1.
and engineering measures. The developed SDSS can Hydraulic Soil Group (HSG) Map - Soil map of
help users to analyze soil erosion and suitable Dakshina Kannada region obtained from NBSS&LUP
watershed management practices. scanned and digitized. The study area was subset and
Modi, Gagendra P., Mishra Rajeev, and Rao Panakala each of the soils in the basin was identified from the
P. (1996), developed a method for determining runoff map index and was subsequently categorized as one
coefficients for various landuse/landcover with the of the four Hydraulic Soil Groups i.e. A, B, C or D
help of remote sensing data. For a given rainfall, with the help of literature (Fig. 4.).
overland flow was calculated using lumped parameter Land use/ Land Cover Map - The land use/land cover
model. The stream flow was computed by solving map is generated by using IRS-1C, MSS imageries
unsteady flow equation by finite difference equations. with Path-97 and Row-63. The imageries acquired in
Padmavathy et al. (1993) used ArcInfo for the March, 2003 and obtained from NRSC, Hyderabad
derivation of various thematic maps form IRS-1A were used for the purpose (Fig.5.). Supervised
imagery and Survey Of India (SOI) toposheets. Check classification is adopted as per the NRSA guidelines
dam sites were selected according to a suitability of 1987 (Level – 1). Following classes were
ranking considering certain criteria without estimating generated (Chadaga Mohandas, 2009).
runoff. 1. Agricultural Land
Tauer & Humborg (1992) used RS and GIS to 2. Forest
determine the potential sites for water harvesting. The 3. Water bodies
surface runoff was estimated for the drainage basin 4. Waste Land
during the wet season to determine the potential ARCView and ERDAS IMAGINE 8.5 softwares were
runoff irrigation areas using Landsat-TM and SPOT used for the analysis and development of various
data. thematic maps.
METHODOLOGY Runoff Analysis

Data Used Runoff is an important criterion for locating water
Survey of India (SOI) Toposheets- The basin is spread harvesting structures in any river basin. The intensity
over the SOI toposheets of 48 O/3, 48 O/4, 48 K/15 of runoff directly decides the type of water harvesting
and 48 K/11. structure a particular location. A runoff coefficient
Rainfall Data- Monthly average rainfalls of six map was prepared to be used in this analysis.
different influencing rainfall stations for a period of The major inputs of runoff analysis were the Rainfall
13 years (from January 1989 to January 2002) was Data, Land Use/Land Cover data, and Hydraulic Soil
obtained from the Minor Irrigation Department in Group map. Although a minimum of 25 years data is
Mangalore. These six stations were Ajekar, Karkal, required for accurate estimation monthly average
Kervase, Kota, Mulikar, Sitanadi and Udupi. rainfall data for the years 1989-2002 could only be
Soil map and Satellite Images-Soil maps obtained obtained from the Department of Minor Irrigation,
from the National Bureau of Soil Survey and Land Mangalore and this was used. Soil Conservation
Use Planning (NBSS&LUP) was adopted for Services (SCS) Curve Number method was adopted to
developing the soil layer. An IRS-IC Satellite image obtain the runoff coefficient map as indicated in the
from the same region of the year 2002 was used to flow chart (Fig.5, Fig.6. and Fig.7.). The Runoff

Coefficients were classified into three classes (Table have low salt tolerance in this region. Further
2). Integration of various thematic maps as per the uncultivable lands due to salt water intrusion could
various decision rules was done in Visual Basic 6.0 also after the construction of subsurface dykes be
and a site suitability map was developed. utilized for cultivation as the stored runoff would
dilute high salt content in these regions.
The creation of a Site Suitability Map was a multistep Percolation Ponds
process. Firstly those maps had to be created which These are illustrated in the map in Fig.10. The chief
would act as an input map for the Site Suitability function of a percolation pond is to store the water
Map. These maps are: from runoff after a storm and allow it to seep into the
1. Digital Elevation Model ground. Obtained result shows that the very less area
2. Slope Map especially in the plain regions is considered for
3. Land Use/ Land Cover Map locating this type of harvesting structures.
4. Curve Number Map (CN Map)
5. Runoff Coefficient Map SUMMARY AND CONCLUSION
6. Hydraulic Soil Group Map Four types water harvesting structures viz., check
Each of these maps is illustrated from Fig.3 to Fig.7. dams, farm ponds, sub surface dykes and percolation
Four types of water harvesting structures were ponds are considered for locating in this study. It can
selected to locate in the study area. Each structure be seen from the results that the extent of sites
could only be located at a point which fulfilled certain suitable for farm ponds greatly exceed the other
criteria known as Decision Rules (Table 4.). structures. Nine suitable sites were identified where in
check dams can be located to harvest water. Sub
Check Dams surface dykes can be located mostly in the western
There are about 9 locations in the Seetha river basin most areas of the basin. This is mainly due to the fact
where check dams can be built (Fig.8.). These are the that this is the only region where pervious soil are
located at the end regions of hill slopes indicating that available and sub-surface dykes serve to recharge
if check dams are built here it will also help in underground sources of water.
controlling soil erosion due to high velocity water Very less sites were identified where percolation
flowing from regions having higher slopes on the ponds could be located. However, all the regions
upstream side. suitable for these structures have very high water
Check dams will also lead to the formation of stream tables due to their close proximity to the Arabian Sea.
pools (a stretch of a river or stream in which the water Thus percolation ponds may not of more suitability
depth is above average and water velocity below for Seetha River basin. However, more detailed study
average) in the drain that it is located thus facilitate by including geological control points may be
surface water storage. required before arriving any final decision in this
Farm Ponds respect.
Farm pond regions are illustrated in the map in Fig. 9.
Seetha River basin is topographically more favorable SCOPE FOR FUTURE WORK
for locating farm ponds. Among the four selected A detailed analysis can be done by incorporating more
harvesting structures it can be seen that farm ponds data to achieve greater degrees of accuracy. For a
can be located over most of the regions. This is more accurate analysis of runoff a minimum of 25
basically due to the topography of the basin which is years of rainfall data is needed. Daily rainfall data
more or less plain for the major extent on the east may give a better result of runoff by SCS-CN method.
2 Existence of lineaments may be considered while
(about portion) after the initial highly sloping
3 framing the decision rules by preparing Lineament
regions of the Western. The site suitability map maps. The database prepared can be used to prepare
prepared for farm ponds was once again overlaid on soil erosion models and maps which can also be used
the Land Use and Land Cover map and it was by framing decision rules on all these.
observed that many of the regions identified for farm
pond were within or near the agricultural land. REFERENCES
Abbas, I.I. Muazu, K.M. and Ukoje, J.A., (2010),
Sub-Surface Dykes “Mapping LandUse-land Cover and Change
It can be seen from the map in Fig.8. that the sites Detection in Kafur Local Government,, Kastina,
suitable for locating subsurface dykes are very close Nigeria (1995-2008) Using Remote Sensing and
to the sea. Subsurface dykes could be constructed in GIS”, Research Journal of Environmental and
the obtained suitable region to effectively store the Earth Sciences 2(1): 6-12.
runoff water from the upstream and also prevent the Chadaga Mohandas (2009), River basin studies of
saline water infiltrating the land from the sea from Seeta-Swarna and Gangolli composite river
mixing with the stored water by acting as a suitable systems of coastal Karnataka, west coast of
barrier. This would help in cultivation of crops which India using remote sensing and geographic

information system (GIS) techniques. Ph D
thesis (unpublished) submitted to Manipal
University.
Girish Kumar, M. Agarwal, A.K. Rameshwor Bali,
(2008), “Delineation of Potential Sites for Water
Harvesting Structures using Remote Sensing and
GIS”, J. Indian Society of Remote Sensing 36,
323-334.
Durga Rao,K.H.V. Venkateswara Rao,V. and Roy,
P.S. (2007), “Development of spatial decision Fig.2. Digital Elevation Model
support system for water harvesting structures
using remote sensing inputs”, Journal of
Geomatics, 1(1),
Gupta, K.K. Deelstra, J. Sharma, K.D. (1997),
“Estimation of water harvesting potential for a
semi-arid area using GIS and remote sensing”,
Proceedings of Rambat Symposium S3, April
1997, IAHS publication no. 242,
Mintesinot B., Kifle W. & Leulseged T., “Fighting
famine and poverty through water harvesting in Fig.3. Slope map
Northen Ethiopia”, Comprehensive Assesment
Bright Spots Project Final Report
Sivaraman K.R.,”Rainwatre Harvesting Structures &
Technologies in different Geo-hydro-thermo
Regimes & Agro Climatic Zones-A case study”
Sharda, V.N., Juyal,G.P., “ Water harvesting
Techniques, Design of small dams and hydraulic
compliments” Fig.4.Hydraulic Soil Group (HSG) map
Ramakrishnan,D. Bandyopadhyay, A. and Kusuma,
K. N. “SCS-CN and GIS based approach for
identifying potential water harvesting sites in the
Kali Watershed, Mahi River Basin, India”,
Thesis work submitted to Department of Earth
Sciences, IIT-Bombay, Powai.
“Principles of remote sensing”, ITC educational
textbook series.
“Remote sensing and image interpretation”, Lillesand Fig.5. Land use/Land cover map
and Kiefer.
“Water Harvesting Structures for Ground Water
Recharge”, Watershed development training
manual
.
Fig.1. Drainage Basin Map
Fig.5. Flow Chart for the generation of runoff

coefficient

Table 1. Classification of Slopes
Slope Class Percentage Slope (%)
Nearly level 0-1 %
Very Gentle 1-3 %
Gentle 3-5 %
Moderate 5-10 %
Strong 10-15 %
Fig.6. Curve Number map Steep 15-35 %
Very Steep >35 %
Table 2. Rainfall Coefficient Classification

Class Runoff Coefficient
range (%)
Low <20 %
Medium 20-40 %
High >40%
Fig.7. Runoff Coefficient map
Table 3. Curve Numbers for Hydraulic Soil Cover
Complexes (Under AMC II conditions)
Hydraulic Soil Group
Land Use Land A B C D
Cover
Agriculture Paddy 95 95 95 95
Forest Dense 26 40 58 61
Waste Land - 71 80 85 88
Water Body - - - - -
Fig.8. Sites suitable for Check Dams
Table 4. Decision Rules for individual Water

Harvesting Structures
Harvesting Soil Runoff Permissible
Structure Permeability Coefficient Slope
Farm Low Medium/ 0-5 %
Ponds High
Fig.9. Sites suitable for Farm ponds Check Low Medium/ <15 %
Dams High
Sub- High Medium/ 0-3 %
surface Low
Dykes
Percolation High Low < 10%
Pond
Fig.10. Sites suitable for Sub-surface dykes
Fig.11. Sites suitable for Percolation Ponds

ISBN 978-81-921121-0-7
REVIEW ON SELF COMPACTING CONCRETE

Kumar Praveen
Professor, Department of Civil Engineering, IIT-Roorkee, 247667,pkaerfce@iitr.ernet.in
Asstt. Professor, Department of Civil Engineering, IIT-Roorkee, 247667, gdranfce@iitr.ernet.in
Jindal Abhishek
M.Tech. Student, Department of Civil Engineering, IIT-Roorkee, 247667,jindalabhishekpnrpce@iitr.ernet.in
ABSTRACT: Self compacting concrete (SCC) was introduced by Japanese researchers in 1980’s to overcome
the problem of the durability of the concrete structures without any aided compaction. Since the development of
SCC, a lot of research work has been performed, but still there’s an immense scope in the field of this
technology. Also the usage of SCC has been limited in the past mostly towards structural and building works,
the application of SCC towards pavement section is still limited, the objective of this paper is to share the light
on the technology of SCC in order to educate the professionals in the arena of construction industries and
infrastructure fields, so that research can be done in this direction and this technology can be successfully and
profitably be used in all construction fields. The absence of an established industrial standard for SCC allows
more creativity in fabricating a mix to specific job requirements.
INTRODUCTION Prof. Hajime Okamura proposed the concept of self

When large quantity of heavy reinforcement is to be compacting concrete in 1986, while the first prototype
placed in a reinforced concrete (RC) member, it is of self compacting concrete was developed in 1988 by
difficult to ensure that the formwork gets completely Prof. Ozawa at University of Tokyo. This type of
filled with concrete, that is, fully compacted without concrete fills each corner of formwork with congested
voids or honeycombs. Compaction by manual or by reinforcement without any aided compaction or
mechanical vibrators is very difficult in this situation. vibration. This increases productivity and reduces the
The typical method of compaction, vibration, labour cost and Noise levels and makes it easy to
generates delays and additional cost in the projects. work around. First paper on self compacting concrete
This problem can now be solved with self-compacting was authored by Ozawa et al in 1989 and first
concrete. A self-compacting concrete may be defined International workshop on self compacting concrete
as the type of concrete (sensitive mix, strongly held in Kochi, Japan in 1998.
dependent upon composition and characteristics of
constituents) which flows under its own weight and METHODS FOR MAKING SCC
has sufficient segregation resistance. Self compacting concrete was first developed by Prof.
Hajime Okamura in Japan during 1980’s.
This self-compacting concrete is advantageous for Manufacturing of a self compacting concrete requires
placing in structures with congested reinforcements, three main aspects to be fulfilled. They are as
without any compaction. Once applied, the self- follows:-
compacting property enables the concrete to fully 1. High amount of water reducing substance or
reinforce around the steel and completely fill the super plasticizer is added for obtaining high
spaces within the framework. The self-compacting of flowing characteristics.
concrete is achieved without losing any kind of
2. A type of aggregate mixture is added to get
strength, stability or change in properties. Self-
compacting concrete has been successfully used in the desired compactness. Note that the
France, Denmark, the Netherlands and UK, apart from aggregate content is of round shape and
Japan. proportional in size in order to increase the
locking tendency of the concrete.
DEVELOPMENT 3. Alteration of fluid properties is done to
Self compacting concrete was introduced by Japanese ensure a cohesive mix which will keep the
researchers in 1980’s. For several years beginning in
aggregate and paste together. These fluid
1983, the problem of the durability of the concrete
structures was a major issue. The creation of durable properties can be achieved by adding to a
concrete structures requires adequate compaction by high quantity of fine content such as cement
skilled workers. However, the gradual reduction in the fly ash or by adding Viscosity Modifying
number of skilled workers in Japan’s construction Admixtures (VMA).
industry has led to a similar reduction in the quality of
construction work. Thus, there was a need for a type The frequency of collision and contact between
of concrete which can be compacted without the help aggregate particles can increase as the relative
of skilled workers. distance between the particles decreases and then

internal stress can increase when concrete is Admixtures
deformed, particularly near obstacles. Research has Superplasticizers - Superplasticizers are a water-
found that the energy required for flowing is reducing admixture that causes a significant increase
consumed by the increased internal stress, resulting in in flowability with little effect on viscosity().
blockage of aggregate particles. To avoid this kind of The primary purpose behind using superplasticizers is
blockage, the level of coarse aggregate content is in achieving highly flowable concrete while
lowered as its energy consumption is intense. maintaining low water and high fines content. Low
Highly viscous paste is also required to avoid the water-to-cement ratio is a principal factor enhancing
blockage of coarse aggregate when concrete flows strength, durability, permeability and shrinkage.
through obstacles. When concrete is deformed, paste
with a high viscosity also prevents localised increases Combination of the most suitable W/C, and
in internal stress due to the approach of coarse Superplasticizer – Powder Ratio (Sp/P)
aggregate particles. High deformability can be
achieved only by the employment of a super
plasticizer, keeping the water powder ratio to a very
low value.
MATERIALS
The Materials used in self compacting concrete are
the same as in conventional concrete except that an
excess of fine material and chemical admixtures are
used. Also, a viscosity-modifying agent (VMA) will
be required because slight variations in the amount of
water or in the proportions of aggregate and sand will
make the self compacting concrete unstable, that is,
water or slurry may separate from the remaining
material.
To achieve a high workability and avoid obstruction
Optimum combination of
by closely spaced reinforcing, self compacting
superplasticizer and w/c ratio. (Okamura)
concrete is designed with limits on the nominal
maximum size (NMS) of the aggregate, the amount of
Different bases of New Generation super Plasticizers
aggregate and aggregate grading. However, when the
or High Water reducing agents (HRWRA) have
workability is high, the potential for segregation and
different water reduction capacities. The advantage of
loss of entrained air voids increases.
this water reduction can be taken either to increase the
The powdered materials are fly ash, silica fume, lime
strength as in high strength concrete or to obtain a
stone powder, glass filler and quartzite filler. The use
better flowability as in case of self compacting
of pozzolanic materials helps the self compacting
concrete.
concrete to flow better.
Following are bases which are commonly used as
Powder Materials
superplasticizers.
1. Hydraulic cements – Conforming to ASTM
 Modified Lignosulfonates(MLS).
C150, C595, C845, or C1157. Cement from one
 Sulfonated Melamine Formaldehyde (SMF).
cement source may provide excellent SCC while
 Sulfonated Naphthalene
changing to another source provides dramatically
Formaldehyde(SNF).
different results. Testing prior to production is
required. In applications where visual  Acrylic Polymer based(AP).
appearance is critical, cement of the same  Coplymer of Carboxilic Acrylic.
source, brand, and color, manufactured at the  Acid with Acrylic Ester(CAE)
same mill should be used throughout the project.  Cross Linked Acrylic Ploymer(CLAP).
2. Supplementary Cementitious Materials –  Polycarboxylatethers(PCE).
Pozzolanic or latent hydraulic powders,  Multicarboxylatethers(MCE).
conforming to ASTM C618, C989, or C1240,  Polyacrylates.
can provide additional workability, increased
 Combination of above.
strength, and reduced permeability and
efflorescence.
Other admixtures successfully used in SCC mixes
3. Inert powder – Limestone, dolomite, or
include air entrainers, water reducers, retarders,
granite dust finer than 0.150 mm (No. 100)
accelerators, and specialty admixtures such as
sieve.
corrosion inhibitors. The use of each of these

materials depends upon the specific application, and concrete, only that it will take more powder, more
pre-testing is required. When air-entraining VMA, or both to do so.
admixtures are used, quality of air system must be
confirmed, as mixes with lower stability may
experience loss of air due to low paste viscosity. MIX DESIGN
While there is no set definition for self compacting
The primary negative effect when adding concrete yet, for now the concrete construction
superplasticizer alone to concrete is that the mix may industry generally follows certain methods of
have a tendency to segregate and bleed. There are two measuring mix properties to define an SCC. The
paths to avoid this problem, involving modifying mix absence of an established industrial standard for SCC
proportions and use of viscosity modifying allows more creativity in tailoring a mix to specific
admixtures. job requirements.
Aggregates While Okamura and Ouchi described two methods for

design mix of self compacting concrete, these
Fine Aggregates - Fine aggregates should meet the methods are discussed as under:
requirements of ASTM C33 with the exception of
grading requirements. Variations in fine aggregate (1) Rational mix – design method
quality will have a major effect on the overall quality
of SCC. Again, water demand is of paramount Self-compactability can be largely affected by the
importance. All things being equal, the less water characteristics of materials and the mix proportions. A
needed to achieve a target slump flow, the better the rational mix-design method for self compacting
self compacting concrete is likely to be. Fine concrete using a variety of materials is necessary.
aggregate moisture control must be provided for. It is Okamura and Ozawa (1995) have proposed a simple
a critical factor in controlling self compacting mix proportioning system assuming general supply
concrete. from ready- mixed concrete plants. The coarse and
fine aggregate contents are fixed so that self
Coarse Aggregates - Aggregates should meet the compactibility can be achieved easily by adjusting the
requirements of ASTM C33 with the exception of water-powder ratio and superplasticizer dosage only.
grading requirements. The combined gradation of
aggregates is more important than individual  The coarse aggregate content in concrete is
gradations. fixed at 50% of the solid volume.
 The fine aggregate content is fixed at 40%
The maximum size of coarse aggregatesdepends on of the mortar volume.
the particular application and is usually limited to 20  The water-powder ratio in volume is
mm. It should not be larger than 1⁄2 (preferred) to 2⁄3 assumed as 0.9 to 1.0, depending on the
the spacing of the narrowest gaps between reinforcing properties of the powder.
bars. Mixes with higher maximum size relative to the  The superplasticizer dosage and the final
gap will be more sensitive to blocking, and will likely water-powder ratio are determined so as to
need either higher powder content or a VMA to ensure self-compactibility.
increase viscosity. In the mix proportioning of conventional concrete, the
water-cement ratio is fixed at first from the viewpoint
Coarse aggregates (and blends) having the lowest of obtaining the required strength. With self-
possible water demand will provide the best self compacting concrete, however, the water-powder
compacting concrete. When water demand is high, the ratio has to be decided taking into account self
water content in the paste is greater, requiring more compactibility because self-compactability is very
powder to keep the water-to-powder ratio low enough sensitive to this ratio. In most cases, the required
to avoid stability problems. Thus the more rounded strength does not govern the water-cement ratio
the aggregate the easier it will be to make quality self because the water-powder ratio is small enough for
compacting concrete. The lower the void content, the obtaining the required strength for ordinary structures
less mortar is needed to fill the voids. unless most of the powder materials in use is not
reactive.
Coarse aggregates with high-aspect ratios should be The mortar or paste in Self-compacting concrete
avoided if possible. Well rounded aggregates allow requires high viscosity as well as high deformability.
better packing (at equal gradation), and lower mortar This can be achieved by the employment of a
and lower powder content will be required. Care superplasticizer, which results in a low water-powder
should be taken that crushed aggregate is as blocky in ratio for high deformability.
shape as possible. This is not to say that more angular
aggregates cannot make good self compacting (2) Adjustment of water-powder ratio and
superplasticizer dosage.

 Self-compacting concrete was used for the wall
The characteristics of powder and superplasticizer of a large LNG tank belonging to the Osaka Gas
largely affect the mortar property and so the proper Company, whose concrete casting was
water-powder ratio and superplasticizer dosage cannot completed in June 1998. The volume of the self
be fixed without trial mixing at this stage. Therefore, compacting concrete used in the tank amounted
once the mix proportion is decided, self- to 12,000 m3.
compactability has to be tested by U-flow, slump flow  Application in under water construction : 40000
and funnel tests. Methods for judging whether the m3 of concrete placed under water by using the
water-powder ratio or superplasticizer dosage are Tremie method for the construction of a dry
larger or smaller than the proper value by using the dock.
test results, and methods for estimating the proper  The Toronto International Airport, where
values are necessary. The relationships between the concrete had to be pumped upwards from the
properties of the mortar in self-compacting concrete ground to form 101-foot tall columns.
and the mix proportion have been investigated and  In the US, a high-strength SCC was imperative
then formulated. These formulae can be used to for constructing tightly reinforced elements
establish a rational method for adjusting the water- poured in below-freezing weather forthe 68-
powder ratio and superplasticizer dosage to achieve story Trump Tower in New York City.
appropriate deformability and viscosity (Ouchi 1998).
ADVANTAGES
APPLICATIONS Self compacting concrete is both a time saver as well
as a money saver. On a large building job, even a five
After the development of the prototype of self- minute delay can cost the project thousands in
compacting concrete at the University of Tokyo, penalties because processes are interconnected and the
intensive research was begun in many places, more variables that can go wrong the higher the
especially in the research institutes of large likelihood that they will. The use of self compacting
construction companies. As a result, self-compacting concrete eliminates the need for additional machinery
concrete has been used in many practical structures. as well as manpower, both saving time and money.
The first application of self-compacting concrete was The advantage of self compacting concrete is that not
in a building in June 1990. Self-compacting concrete only is it self-leveling but it can also be mixed to dry
was then used in the towers of a prestressed concrete with a shiny finish that requires little else in terms of
cable-stayed Shin-Kiba Ohashi bridge in 1991. finishing. This saves time on a number of procedures
Lightweight self-compacting concrete was used in the including vibration, leveling, compaction and
main girder of a cable-stayed bridge in 1992. Since finishing with power floats. Finishing alone can be
then, the use of self-compacting concrete in actual quite time intensive as it requires a number of passes
structures has gradually increased. Self-compacting and if you wait too long or start too early you can
concrete has been successfully used in France, wave good-bye to having any type of decent finish.
Denmark, the Netherlands, Germany, USA and UK,
apart from Japan.  Higher installation performance since no
compaction work is necessary which leads to
 A typical application example of Self-
compacting concrete is the two anchorages of reduced construction times, especially at large
Akashi-Kaikyo (Straits) Bridge opened in April construction sites.
1998The volume of the cast concrete in the two  Reduced noise pollution since vibrators are not
ahchorages amounted to 290,000 m3. A new necessary
construction system, which makes full use of the  Higher and more homogenous concrete quality
performance of self compacting concrete, was
across the entire concrete cross-section,
introduced for this. The concrete was mixed at
the batcher plant beside the site, and was the especially around the reinforcement.
pumped out of the plant. It was transported 200  Improved concrete surfaces (visible concrete
meters through pipes to the casting site, where quality)
the pipes were arranged in rows 3 to 5 meters  Typically higher early strength of the concrete so
apart. The concrete was cast from gate valves that formwork removal can be performed more
located at 5 meter intervals along the pipes.
quickly.
These valves were automatically controlled so
that a surface level of the cast concrete could be
CONCLUSIONS
maintained. In the final analysis, the use of self-
 The development of SCC strongly supports an
compacting concrete shortened the anchorage
construction period by 20%, from 2.5 to 2 years. increased industrialization of the concrete
construction process. The fact that concrete can

be handled as a fluid gives in itself drastic Suksawang, N., Nassif, H., Najm, H., “Evaluation
process improvement possibilities. of Mechanical Properties for Self –
 Self-compacting concrete has two big Consolidating, Normal, and High –
advantages. One relates to the construction time, Performance Concrete.” Journal of the
which in most of the cases is shorter than the time Transportation Research Board, No. 1979,
when normal concrete is used, due to the fact that Transportation Research Board of the National
no time is wasted with the compaction through Academies, Washington, D.C., 2006, pp. 36-
vibration. The second advantage is related to the 45.
placing. Mukai, D., Altan, S., Ehrgott, J., “Early Strength of
 While there is no set definition for self Self – Compacting Concrete.” Transportation
compacting concrete yet, for now the concrete Research Record 1698, Paper No. 00-1043.
construction industry generally follows certain
methods of measuring mix properties to define an
SCC. The absence of an established industrial
standard for SCC allows more creativity in
tailoring a mix to specific job requirements.
 Due to its ability to guide itself into every nook
and cranny in the form, SCC can produce nearly
nil defects concrete. About 40 to 50% of cement
content can be replaced by supplementary
cementing material like fly ash; cost of the
concrete is greatly reduced.
 Although SCC finds its application in past in
many projects, still its limited to structural
elements such as bridge, columns, reinforced
walls etc. When talked of its usage in the field of
pavement construction, the application has been
limited to pavement repairs and rehabilitation.
REFERNCES
Okamura, H., Ouchi, M. (2003). “ Self –
Compacting Conrete.” Journal of Advanced
Concrete Technology Vol.1, No. 1, 5-15, April
2003.
Suresh, G., Bhanumathidas, N., Kalidas, N., “Self
Compacting Concrete, The Emerging Concrete
Practice for Critical Applications.” Master
builder.
NRMCA, “CIP – 37 Self Compacting Concrete
(SCC).” Concrete in practice, what, why &
how?.
Druta, C., “Tensile Strength And Bonding
Characterstics Of Self Compacting Concrete.”
Ouchi, M., Hibino, M., “Development, Applications
And Investigations Of Self-Compacting
Concrete.”
Abou-Zeid, M., Roushdy, M., “Performance and
Uniformity of Self – Compacting Concrete.”
Journal of the Transportation Research Board,
No. 1914, Transportation Research Board of
the National Academies, Washington, D.C.,
2005, pp. 71–80.

ISBN 978-81-921121-0-7
A STUDY ON SHOTCRETE TECHNOLOGY

Vipin Bansal
First-degree Student, Civil Engg. Dept., BITS Pilani, Rajasthan,(333031) India, Email: vipinbansal2008@gmail.com
Abhishek Choudhary
First-degree Student, Civil Engg. Dept., BITS Pilani, Rajasthan,(333031) India, Email: f2008449@bits-pilani.ac.in
Dipendu Bhunia
Assistant Professor, Civil Engg. Dept., BITS Pilani,, Rajasthan,(333031) India. Email: dipendubhunia@gmail.com
Ashutosh Parauha
First-degree Student, Civil Engg. Dept., BITS Pilani, Rajasthan,(333031) India, Email: ashutosh..parauha@gmail.com
Nikhil Shrivastava
First-degree Student, Civil Engg. Dept., BITS Pilani, Rajasthan,(333031) India, Email: nick310791@gmail.com
ABSTRACT
This paper will investigate and highlight critical aspects of sprayed concrete initially between dry and wet
shotcrete and then develop a suggested total system approach with the wet shotcrete process. It is important to
note that shotcrete is not a mystical combination of concrete constituents but is one of several ways to cast
concrete. The characteristics of concrete as well as basic concrete properties and technologies need to be
addressed whether it be shotcrete or cast concrete in the form of civil engineering construction. Aspects like
amount of cement, water-cement ratio, consistency, curing, and reinforcement need attention as with any type of
concrete construction. The one single main reason for shotcrete of poor quality being applied throughout the
world, and particularly in underground mining operations, is that it seems to be either forgotten or through lack
of training and expertise, that the basics of concrete technologies are paramount.
INTRODUCTION good adhesion to the surface. Within limits, the

The great advantages of sprayed concrete as not only material is capable of supporting itself in vertical or
a construction process but also as a rock support overhead applications.
process have been developed gradually since the early
1900’s in the United States. The improvement of Dry Shotcrete
materials, development of equipment as well as The cementitious material and aggregate are
application techniques have made sprayed concrete a thoroughly mixed and either bagged in dry condition,
very important part of underground civil and mining or mixed and delivered directly to the gun. The
works. Two spraying methods are common: dry mix mixture is normally fed to a pneumatically operated
and wet mix, and whereas this paper will particularly gun which delivers a continuous flow of material
cover wet mix shotcrete, a basic comparison as well through the delivery hose to the nozzle. The interior
as reasoning for this choice is important. It should of the nozzle is fitted with a water ring which
also be remembered that both methods have their uniformly injects water into the mixture as it is being
advantages and disadvantages and, depending upon discharged from the nozzle and propelled against the
project requirements, prevailing underground receiving surface.
conditions, accessibility and the experience and
training of personnel, then the best suited method Wet Shotcrete
should be chosen. The cementitious material, aggregate, water, and
Shotcrete is used in lieu of conventional concrete, in admixtures are thoroughly mixed as would be done
most instances, for reasons of cost or convenience. for conventional concrete. The mixed material is fed
Shotcrete is advantageous in situations when to the delivery equipment, such as a concrete pump,
formwork is cost prohibitive or impractical and where which propels the mixture through the delivery hose
forms can be reduced or eliminated, access to the by positive displacement or by compressed air.
work area is difficult, thin layers or variable Additional air is added at the nozzle to increase the
thicknesses are required, or normal casting techniques nozzle discharge velocity. [1]
cannot be employed. Additional savings are possible
because shotcrete requires only a small, portable plant SHOTCRETE VERSUS CONVENTIONAL
for manufacture and placement. Shotcreting CONCRETE
operations can often be accomplished in areas of Unlike conventional concrete, which is first placed
limited access to make repairs to structures. The and then compacted in the second operation, shotcrete
excellent bonding of shotcrete to other materials is undergoes placement and compaction at the same
often an important design consideration. The force of time due to the force with which it is projected from
the impact of this pneumatically propelled material on the nozzle. Shotcrete is more dense, homogeneous,
the surface causes compaction of the shotcrete paste strong, and waterproof than is possible to obtain by
matrix into the fine surface irregularities and results in any other process. For most bridges original concrete

will require additional repair long before shotcrete (4) The use of air-entraining admixtures (AEA) in
needs repair and it suffers little deteriorations. shotcrete is practical only in wet-mix shotcrete. When
Shotcrete is not placed or contained by forms. It can batched properly, AEA forms an air-void system
be impacted onto any type or shape of surface, suitable for providing frost resistance to wet-mix
including vertical or overhead areas. It forms an shotcrete. The formation of an air-void system in dry-
excellent bond with the substrate and can be given a mix shotcrete is not possible. However, dry-mix
variety of surface finishes. shotcrete, when properly proportioned and applied,
will have a compressive strength exceeding
Advantages of shotcrete approximately 7,000 pounds per square inch (psi). It
Shotcrete is used in lieu of conventional concrete, in has performed well in moderate exposures to freezing
most instances, for reasons of cost or convenience. and thawing.
Shotcrete is advantageous in situations when
formwork is cost prohibitive or impractical and where TYPES OF SHOTCRETE
forms can be reduced or eliminated, access to the
work area is difficult, thin layers or variable Fiber-Reinforced Shotcrete
thicknesses are required, or normal casting techniques Unreinforced shotcrete, like unreinforced
cannot be employed. Additional savings are possible conventional concrete, is a brittle material that
because shotcrete requires only a small, portable plant experiences cracking and displacement when
for manufacture and placement. Shotcreting subjected to tensile stresses or strains. The addition of
operations can often be accomplished in areas of fibers to the shotcrete mixture adds ductility to the
limited access to make repairs to structures.[7] material as well as energy absorption capacity and
impact resistance. The composite material is capable
Strength of bonding of sustaining post crack loadings and often displays
The excellent bonding of shotcrete to other materials increased ultimate strength, particularly tensile
is often an important design consideration. The force strength. Fibers used in shotcrete are available in three
of the impact of this pneumatically propelled material general forms: steel fibers, glass fibers, and other
on the surface causes compaction of the shotcrete synthetic fibers. Natural fiber, a fourth form, is not
paste matrix into the fine surface irregularities and commonly used in shotcrete and will not be discussed.
results in good adhesion to the surface. Within limits,
the material is capable of supporting itself in vertical Silica-Fume Shotcrete
or overhead applications. Silica fume is a very fine noncrystalline pozzolanic
material composed mostly of silica. Silica fume is
Comparison of dry-mix and wet-mix processes used in concrete and shotcrete to increase strength,
Shotcrete suitable for most requirements can be decrease permeability, and enhance cohesion and
produced by either the dry-mix or wet-mix process. adhesion. Specific advantages of silica fume in
However, differences in the equipment cost, shotcrete are the improved bond strength of shotcrete
maintenance requirements, operational features, to substrate surfaces, the improved cohesion of the
placement characteristics, and product quality may shotcrete, and the resulting ability to apply thicker
make one or the other more attractive for a particular layers of shotcrete in a single pass to vertical and
application. overhead surfaces. The material is more resistant to
"washout," where fresh shotcrete is subject to the
(1) Bond strengths of new shotcrete to existing action of flowing water, and rebound is significantly
materials are generally higher with dry-mix shotcrete reduced. Shotcrete containing silica fume may have
than with wet-mix shotcrete. Both shotcrete mixtures improved resistance to aggressive chemicals.
often provide significantly higher bond strengths to
existing materials than does conventional concrete. Polymer-Modified Shotcrete
(2) Typically, dry-mix shotcrete is applied at a much Polymers are incorporated into shotcrete in two ways.
slower rate than wet-mix shotcrete. Dry-mix shotcrete In one method, the entire binder is composed of a
is often applied at a rate of 1 or 2 cubic yards per hour polymer material. This is no longer a hydraulic-
compared to wet-mix shotcrete applied at a rate of up cement product but a polymer shotcrete. The more
to 7 or 8 cubic yards per hour. Depending on the common use of polymers is the addition of a polymer
application, the in-place production rate may be emulsion to the hydraulic-cement mixture, as with a
significantly lower because of obstacles, rebound, and partial replacement of the mixing water, or as total
other features which may cause delays. replacement, which disperses throughout the mixture
(3) Rebound is the shotcrete material that "bounces" forming a continuous polymer matrix. This is termed
off the shooting surface. Rebound for conventional polymer-portland-cement shotcrete.
dry-mix shotcrete, in the best of conditions, can be
expected to be at least 20 percent of the total material Accelerated Shotcrete
passed through the nozzle. Wet-mix shotcrete Accelerating admixtures are used extensively in
rebounds somewhat less than dry-mix shotcrete. shotcrete. Highly effective accelerators have been

developed for rapid setting of shotcrete. Often similar to a conventional 19.0-mm (3/4-inch) nominal
considered "super-accelerators," these are commonly maximum size aggregate, except for a reduction in the
used with dry-mix shotcrete. With the increasing use larger sizes to minimize rebound. Aggregate failing to
of silica fume, the use of accelerators may decline comply with these gradings may be used if
somewhat. In the past, these accelerators were preconstruction tests demonstrate that it gives good
exclusively powdered materials added to dry-mix results. However, a uniform grading is essential.
shotcrete materials. Now both powdered and liquid Coarse and fine aggregate should be batched
admixtures are used in both dry-mix and wet mix separately to avoid segregation.
shotcrete. The use of these accelerators with a wet- Fine aggregate for finish or flash coats and certain
mix process requires that the accelerator be added at other special applications may be finer than Grading
the nozzle rather than batched with the other No. 1. Finer fine aggregates, however, generally
materials.[5] produce shotcretes having greater drying shrinkage,
while coarser sands result in more rebound.
MATERIAL PROPRTIONING AND Lightweight-aggregate shotcrete is most practical for
PROPERTIES the dry-mix process. Since moisture and aggregate
a. Cementitious Materials contact is initiated at the nozzle, the severe
Portland cement. workability reductions common in conventional
Cement requirements for shotcrete are similar to those lightweight concrete production do not occur.
for conventional concrete. Portland cement must meet
the requirements of CRD-C 2011 (American Society Grading Limits for Aggregate
for Testing and Materials (ASTM) C 150), Type I or
II. Where the shotcrete will be exposed to soil or Percent by Mass Passing Individual Sieves
Sieve Size Grading No. 1 Grading No. 2 Grading
water high in soluble sulfates, Type II or V should be No. 3
used as described in EM 1110-2-2000. Blended
3/4-inch 100
cement must meet the requirements of CRD-C 203 1/2-inch 100 80-95
(ASTM C 595), Type IP or IS, and moderate sulfate 3/8-inch 100 90-100 70-90
0.19 inch (No. 4) 95-100 70-85 50-70
resistance may be specified by adding the suffix MS 0.093 inch (No. 8) 80-100 50-70 35-55
to the type designation. Where structural requirements 0.046 inch (No. 16) 50-85 35-55 20-40
require high early strength, Type III meeting the 0.024 inch (No. 30) 25-60 20-35 10-30
0.012 inch (No. 50) 10-30 8-20 5-17
requirements of CRD-C 201 (ASTM C 150) cement 0.006 inch (No. 100) 2-10 2-10 2-10
should be considered.[2,6]
c. Water
Pozzolan. Potable water should be used. If this is not available,
When added to a Portland-cement matrix, pozzolan the proposed water source should be tested according
reacts with the calcium hydroxide and water to to CRD-C 400. No special requirements are necessary
produce more calcium silicate gel. Consequently, for curing water applied to shotcrete (ASTM 1978).
shotcretes with pozzolan may exhibit improved long- Water for curing of architectural shotcrete should be
term strength performance and lower permeability. free from elements that will cause staining.
Pozzolan is sometimes added to wet-mix shotcrete to
enhance workability, improve pump ability, increase d. Chemical Admixtures
resistance to sulfate attack, and reduce expansion Use of admixtures.
caused by the alkali-silica reaction. Because of shotcrete equipment limitations, the use of
admixtures in shotcrete is not the same as in
Silica fume. conventional concrete. Admixtures should be tested in
Silica fume is an extremely fine, amorphous the field prior to use on large jobs to ensure that the
pozzolanic material which is a waste product of the desired properties are achieved. Chemical admixtures
silicon, ferrosilicon, or other silicon alloy production used in shotcrete should comply with the appropriate
in submerged-arc electric furnaces. The silica fume requirements given in CRD-C 625 (ASTM C 1141).
condenses from the exhaust gases forming extremely ACI 212.3R (paragraph A-1, ACI (1991a)),
minute spherical particles. The material is over 85 "Chemical Admixtures for Concrete," contains
percent silica dioxide, is approximately 100 times detailed information on general use in concrete.
finer than Portland cement, and has a specific gravity
ranging from 2.1 to 2.6. Air-entraining admixture (AEA).
The use of AEA’s in shotcrete is practical only in wet-
b. Aggregate mix shotcrete. Due to the loss of air during the
Aggregate should comply with the quality gunning process of the wet-mix shotcrete, the AEA
requirements of CRD-C 133 (ASTM C 33). Table 3-1 should be batched so that the measured air contents in
shows acceptable grading limits. Grading No. 1 the plastic mix prior to pumping are twice the desired
should be used if a mortar mixture is desired. Grading hardened shotcrete air content. The mixing process
No. 2 and 3 contain coarse aggregate; the latter is required to form the air bubbles does not occur in the

dry-mix process, hence a suitable air-void system is subsequently cut or chopped. Flat steel fibers are cut
not generated using the admixture. Air entrainment or slit from sheet of steel or by flattening wire. The
has slightly reduced rebound. melt-extraction process is used to "cast" fibers by
extracting fibers from a pool of molten steel.
Water-reducing and retarding admixtures. Consequently, fibers are round, flat, or irregular in
WRA’s meeting the requirements of CRD-C 87 shape. Additional anchorage is provided by
(ASTM C 494) are readily adapted to the wet-mix deformations along the fiber length or at the ends.
process but are not used in dry-mix shotcrete due to Deformations can be natural irregularities, crimps,
the ineffectiveness of the admixture when adding the corrugations, hooks, bulbs, and others. Collated fibers
admixture and water at the nozzle. Retarding and fibers with noncircular cross sections reduce the
admixtures are seldom used in shotcrete, except for handling and batching problems common with
near horizontal surfaces where subsequent finishing of straight, round fibers.
the shotcrete surface is required. [3]
PROPORTIONING OF SHOTCRETE
e. Reinforcing Steel a. Considerations.
Reinforcing bars for shotcrete should meet the same In general, conventional concrete technology may be
specifications as for conventional concrete. Because applied to shotcrete proportioning. Prior to mixture
of the placement method, the use of bars larger than proportioning, the following should be considered:
No. 5 or heavy concentrations of steel are not (1) Type of dry-mix or wet-mix shotcrete appropriate
practical. Large bars make it difficult to achieve for the work.
adequate build-up of good quality shotcrete behind (2) The specific job constraints on the shotcrete work.
the bar and heavy concentrations of steel interfere (3) The type of specification.
with the placement of shotcrete. In general, bar (a) Performance versus prescription.
spacings of 6 to 12 inches are recommended for (b) Contractor versus Government mixture
shotcrete reinforcement. It is often advantageous to proportioning.
specify as welded wire fabric, reinforcement either
uncoated, galvanized, or epoxy coated. Flat stock b. Mixture proportioning trial batching.
should usually be specified in lieu of rolled fabric. (1) Since shotcrete performance is highly dependent
Because of the rolled configuration, rolled welded on application procedures, trial batching and testing is
wire fabric is difficult to place at specified locations. a critical operation in verifying mixture performance.
Wire spacing should be as wide as possible to allow The batching and mixing of wet-mix shotcrete is
shotcrete to be built up behind. Spacing of 6 inches is practically identical to conventional concrete; only the
recommended; however wire spacing as low as 2 fabrication of specimens is different. However, dry-
inches has been used with 4 inches being more mix is a distinct process. It is normal procedure to
typical. In repair work, a thin shotcrete coating may obtain trial mixture proportions for shotcrete from the
not require reinforcement. When reinforcement is contractor.
exposed in the old concrete, but not severely (2) Test panels are particularly important for dry-mix
corroded, it may be the only reinforcement necessary. shotcrete because laboratory mixtures cannot
In other cases, additional reinforcement (bars or wire duplicate as-shot dry-mix shotcrete. Typically, a
mesh) may be required to replace corroded steel to performance specification of 12-hour, 7-day and/or
control temperature cracking, if not to satisfy 28-day compressive strengths will be specified, along
structural considerations with a grading for the aggregate. Both the wet- and
dry-mix methods will yield a higher as-shot cement
f. Fiber Reinforcement content and lower coarse aggregate content, due to
rebound of the aggregate.
Steel fiber reinforcement.
Steel fibers have been used in shotcrete to increase its c. Wet-mix proportioning.
ductility, toughness, impact resistance, and reduce Mixture proportioning procedures for the formulation
crack propagation. The fibers are commercially of conventional concrete for pumping applications are
available in lengths ranging from 1/2 to 3 inches. applicable for wet-mix shotcrete. The nominal
Typical fiber lengths for shotcrete range from 3/4 to maximum aggregate size is usually 3/4 inch or
1-1/2 inches and are used in the amount of 1 to 2 smaller. The batched cement content will typically
percent by volume of the shotcrete. The fibers have range from 500 to 700 pounds per cubic yard. Rich
little effect on compressive strength and produce only mixtures are common for shotcrete, especially, if
modest increases in flexural strength. However, they vertical or overhead shotcrete placement is required.
provide continued and, at times, improved load The limiting factor for cement content in a mixture is
carrying capacity after the member has cracked. often governed by the amount of cement necessary for
Steel fiber source the shotcrete to adhere to a wall or ceiling, not the
Steel fibers are manufactured in several ways. Wire specified compressive strength. It is not unusual for
fibers are produced from drawn wire that has been shotcrete used in vertical and overhead placement to

have 28-day strengths in excess of 4,500 psi, due only permeability is apparent at the lower dosages.
to the amount of cement necessary to make the Reductions in rebound and increases in cohesiveness
shotcrete adhere. for thick applications do not occur until silica-fume
(1) Workability. dosages exceed approximately 14 percent. Further
The slump for wet-mix shotcrete should be near the mixture adjustments to wet-mix shotcrete may be
minimum that the pump will handle. A 3-inch slump necessary to attain the required workability level.
should normally be considered the maximum slump to
be used. Excess slump will yield lower-strength g. Polymer-modified shotcrete
shotcrete which will tend to slough off of vertical and Polymer emulsions are typically 50 percent solids and
overhead surfaces. 50 percent water. The liquid portion of the emulsion
(2) Entrained air. replaces the equivalent volume of water, and the solid
If air entrainment is to be used, an air content ranging portion replaces the same volume of combined solids.
from 8 to 12 percent prior to pumping is typical. The Additional adjustments to attain desired workability
in-place shotcrete will have about one-half of the levels may be required.
entrained air that was recorded at the pump.
(3) Admixtures. PLACEMENT
Additional admixtures generally behave the same in
wet-mix shotcrete as they do in conventional concrete. Shotcreting can be a hazardous operation, especially if
Any admixture should be tested in the mixture the work must be conducted in a relatively confined
proportioning studies and on the test panels prior to area. The critical hazards include control of the
usage. material (shooting), rebound, plugged nozzle, hose or
coupling breaks, caustic materials, and dusting.
d. Dry-mix proportioning.
There is no established method of proportioning dry- Preparations
mix shotcrete. Since it is not practical to perform
laboratory trial mixtures for dry-mix shotcrete, field a. Unformed sections.
testing of dry-mix proportions is highly advisable, Where shotcrete is to be placed against earth surfaces,
especially if no field data exist for a given dry-mix. as in canal linings, the surfaces should first be
The in-place aggregate grading will be finer than the thoroughly compacted and trimmed to line and grade.
batched grading due to rebound, especially if larger Shotcrete should not be placed on any surface which
aggregate sizes are used. As with wet-mix shotcrete, is frozen, spongy, or where there is free water. The
the in-place cement factor will be higher also varies surface should be kept damp for several hours before
between 0.30 to 0.40. shotcrete is placed.
e. Fiber-shotcrete proportioning. b. Formed surfaces.

Steel fiber lengths for shotcrete are typically 1 inch Forms should be made of plywood sheathing,
but often range from 3/4 inch to 1-1/2 inches. The expanded metal lath, or other suitable material, true to
fiber should be at least 1/4 inch longer than the line and dimension. They should be adequately braced
diameter of the maximum aggregate size. Shorter to ensure against excessive vibration and constructed
fibers are more easily pumped through the system, to permit the escape of rebound and air during the
although more are required for equivalent gunning operation, particularly in the case of thick
performance. Fiber batch quantities are dependent on structural members. Columns should be formed only
required shotcrete properties. Typical fiber on two adjacent sides wherever possible. Satisfactory
proportions range from 0.5 to results may be obtained where three sides are formed,
2.0 percent by volume of shotcrete. Deformed fibers provided the width is at least 1.5 times the depth.
and fibers with end anchorage provisions produce Pilasters may be formed on two adjacent or opposite
shotcrete with properties equivalent to straight fibers sides. The soffit and one side of beams should be
at much lower fiber loadings. Since fibers tend to formed, and shores should be set below the soffit in
rebound at a greater rate than does aggregate, the fiber such a manner that no deflection will occur under the
batch quantity should be adjusted accordingly. load to be imposed. Short, removable bulkheads may
Proportioning mixtures using glass fibers is discussed be used at intersections. The forms should be oiled or
by the PCI (1981). Proportioning mixtures using dampened and should be clean just prior to gunning.
polypropylene fibers is discussed by Morganet al.
(1989). c. Work area access.
Safe, adequate scaffolding should be provided so that
f. Silica-fume shotcrete proportioning the nozzle man can hold the nozzle at the optimum
Silica fume is added to a shotcrete mixture as a angle and distance from the surface for all parts of the
cement replacement or in addition to cement. Batch work. The scaffolding should also provide easy access
quantities range from 7 to 15 percent by mass of to the shotcrete surface for screeding and finishing
cement. Strength enhancement and decreased operations.

in soft rock seams. A neat sharp edge, especially if it
d. Reinforcing steel repair. coincides with a change in the rock texture or a
Concrete around the reinforcing should be removed fracture line, is generally more appropriate than
until clean, uncorroded steel is reached. Concrete bleeding the shotcrete out or allowing overspray.
behind reinforcing steel should be removed to a depth However it should be accepted that feathering the
to allow proper placement from a nozzle angles not shotcrete may be required in order to achieve a good
more than 45 degrees to the surface. This depth is bond with the rock.
often more than 1.5 times the bar diameter but not less
than 1 inch. Reinforcing that is corroded so badly that Screening
its usable cross section is reduced must be replaced. Planting should be used to help screen the shotcrete
application.
f. Anchorage.
Anchorage of shotcrete follows practice for
conventional concrete. AVOIDING THE NEED FOR SHOTCRETE
g. Alignment control. The need for the use of shotcrete or any other better
A variety of alignment control devices are required to stabilization technique should be avoided. Besides
establish the limits of shotcrete placement, including being particularly unsightly, relative to an untreated
ground wires, guide strips, depth gauges, depth safe slope shotcrete is costly and high maintenance. It
probes, or forms. Ground wires are usually 18 or 20- is better to have designed the cutting so that
gauge, high-strength steel wires attached to a stabilization is not needed. The best time to avoid or
turnbuckle or spring to provide tension. Wires may be reduce the need for shotcrete is in the route selection
used to establish corners of the shotcrete work, and and refinement process. The best way to avoid the
may be spaced at 2- to 3-foot intervals for screed need for shotcrete is to allocate sufficient space for
guides for flat areas. Guide strips are wood lath the road and the cuttings. Liaison with geotechnical
attached to crosspieces at 2- to 3-foot intervals, used experts in the route selection stages will assist in
similarly to wires. Depth gauges, small metal or defining the space needed for the road corridor by
plastic devices attached to the surface to which advising on appropriate cutting slopes.
shotcrete is being applied, indicate the resulting
thickness of the shotcrete. They may be left in place, Principle
and slightly covered, if their presence is not a. Obtain sufficient land
damaging. Depth probes are stiff steel wire rods pre-
marked with the thickness of shotcrete, which are The need for shotcrete can be eliminated through
inserted into the shotcrete to check thickness. purchase of sufficient land so that batters can be laid
back to a stable grade. Where adjacent land is highly
MINIMIZING THE EXTENT OF SHOTCRETE valuable or threatened habitat, discussions should be
Precision design held with stakeholders to consider the relative merits
Aesthetically it is far better (and more cost effective) of the land needed versus the costs (financial and
to apply shotcrete precisely to unstable sections of aesthetic) of the shotcrete application.
cuttings than to apply a blanket covering. In order to
achieve this, the shotcrete application should be b. Avoid over steep cutting faces
planned and designed in advance so as to minimize
visual impact as well as stabilize the slope. This The need for batter treatment arises when cutting
planning need not be time consuming if the right faces are over steep for the combination of rock types,
expertise is employed and could be based upon fracturing and weathering patterns intersected. Ensure
photographs of the emerging cutting. that the stability and treatment consequences of
steepening cutting faces are given appropriate
Progressive stabilization consideration at the route selection and concept
Stabilization treatments (including shotcrete) should design stages. Where there is limited geotechnical
be applied progressively with excavation rather than information it is preferable to nominate flatter rather
left to the end. This tends to minimize usage which is than steeper batters in an EIS. Design for optimal not
a cost saving and satisfies both geotechnical and maximum gradients, then if there is a change in
urban design objectives. gradient or slope stability, shotcrete may still not be
Relationship with surrounding rock Finish or extend necessary. In most situations, slopes flatter than
the application of shotcrete up to distinct edges, 2(H):1(V) with 5m benches and setbacks from
natural joints or changes in the face of the cutting. carriageways will provide a stable cutting as well as
allow space for vegetation to establish.
Masking
Shotcrete should be controlled and applied only where c. Provide space for cuttings
required and masked off for other areas, for example

Obtaining sufficient space between the road and rock pre-construction decisions about batter slopes and how
cutting is by far the best way to avoid the need for they will be stabilized and maintained. Treatment
shotcrete. Every meter gained has a significant impact should be planned and preventative, rather than
on the risks posed by an unstable cutting. It also remedial, although some surprises may occur. For
allows planting to develop to help catch debris, existing slopes, you have what you have and the
improve the appearance of the road corridor and break treatment is almost always remedial. While the possible
up the expanse of concrete. options may be similar, the constraints on their use
(including costs) are different
d. Explore other sources of fill if required There should always be a clear purpose for the use of
any engineering measure and shotcrete is no exception.
In occasional situations cuttings are deepened to It must be understood in terms of its intended
balance additional fill requirements by steepening function(s) and comparisons made with alternatives
cutting faces rather than by widening the cutting which could replace it. It is not practical to dismiss the
footprint. In these circumstances consider borrowing use of shotcrete due to its appearance. The science of
material from other locations. rock durability is very complex and there are few
experts in Australia who could predict with a great
e. Explore alternative stabilization techniques degree of accuracy the durability of all types of rock
after exposure. Judicious use of shotcrete to ensure the
Where potential shotcrete stabilization needs are stability of the batter (or slope) and the safety of road
identified explore alternative solutions such as: users is inevitable when considering the extent of our
Reduce face heights and steepen slopes to reduce road network. Nevertheless shotcrete is sometimes used
erosion. Shotcrete application on the bench top only, in excess and applied when not always needed. Project
to minimize water penetration and ongoing erosion. managers and their teams need to apply control to the
Rock bolting (where possible rock bolt heads should applications so that it is applied with precision and
not be covered with shotcrete). Mesh netting or use of mindful of visual impacts.
bolted rock mesh. Fencing at key locations on the
benches and base of slopes to catch loose material. REFERENCES
Soft fall areas at the base of slopes to contain loose
material. Locally won rock gravity walls (e.g. Indian standard IS 9012
Woronora bridge project). Retaining walls or precast American Concrete Institute (ACI) Committee
panels over stabilized cuttings, to be considered in Report 506R-90,"Guide to Shotcrete"
urban areas with high land value and high quality
ACI 212.3R (paragraph A-1, ACI (1991a)),
finishes needed. Slopes stabilized by rock mattresses
or stone. "Chemical Admixtures for Concrete".
Engineering and design manual “EM 1110-2-
f. Be judicious in the use of shotcrete 2000”.
Pre stressed Journal (PCI 1981).
Consider the costs and benefits of the ‘do nothing’ CRD-C 201 (American Society for Testing
option. If there is sufficient space, allowing the slope and Materials (ASTM) C 150 ) Type I ,
to weather and erode safely may be the best option in
CRD-C 203 (ASTM C 595) , CRD-C 201
the long run.
(ASTM C 150), CRD-C 133 (ASTM C
g. Specify pre splitting 33)and similar codes.
Specify excavation methods that minimize the risk of www.commercialshotcrete.com
creating unstable slopes (e.g. Oak Flats interchange
where rock was ripped rather than presplit). Prepare
specifications and manage contracts such that the risk
of blasting damage to pre-split faces is minimized.
CONCLUSION
Today shotcrete has become a very useful material due
to its high strength, durability, low permeability, good
bond, limitless shape possibilities and ease of handling
in areas of difficult access. It also requires no
formwork, is highly cost effective and is particularly
useful where land space is limited. The circumstances
of its use may arise either as part of the original
construction or as remediation of existing batters. The
distinction is important. New work should allow
substantial control over geotechnical design, and hence

ISBN 978-81-921121-0-7
SOME STUDIES ON THE INFLUENCE OF CLAY ON CONSOLI-

DATION CHARACTERISTICS OF COHESIVE SOILS
S.V. Rao
Scientist ‘C’, Central Soil and Materials Research Station, Olof Palme Marg, Huaz Khas, New Delhi-110 016, Email:
camrad@sify.com
ABSTRACT: In the settlement behaviour of saturated soil mass, clay content plays an important role because
of the fact that the clay particles have more surface areas giving rise to more electrical unbalanced charges at the
surfaces. Every material undergoes certain amount of strain when a stress is applied. Soil undergoes strains
which are not proportional to the applied stresses. It is therefore necessary to study the stress-strain
characteristics of the soil in the foundation of structures in order to understand the settlement of the structure
and to predict and to make provision for the settlement which may occur during its lifetime. Since oedometer
experiments are time consuming, it will not be out of the way if the coefficients of the stress-strain relationship
are derived on basis of a few experiments within the zone of clay content and utilize the relationship for
estimating these parameters of the untested soil samples by interpolation. Consolidometers of fixed ring type
with capacity 16 kg/cm2 are used in this work. The soils taken for the studies were the cohesive soils collected
from three different dam sites having different percentages of clay. Based on the laboratory experimental studies
on consolidation characteristics of cohesive soils with different percentages of clay, the following conclusions
were arrived at: Coefficient of consolidation decreases linearly with increase in clay content. The relationship
between coefficient of compressibility (qv) and clay content at different stress level is linear. The strain recovery
increases with increase in plasticity index.
Keywords: Clay, Consolidation, Settlement, Oedometer
INTRODUCTION required weight of the sample to be prepared was

In the settlement behaviour of saturated soil mass, taken and the sample was thoroughly mixed with an
clay content plays an important role because of the initial moisture content of 24 percent. The inner
fact that the clay particles have more surface areas surface of the oedometer ring and the outer lateral
giving rise to more electrical unbalanced charges at surface of the loading pad were smoothened by
the surfaces. Every material undergoes certain amount applying high pressure-grease to eliminate any
of strain when a stress is applied. Soil undergoes friction between the sample and the ring and the
strains which are not proportional to the applied loading pad and the ring respectively. The porous
stresses. It is therefore necessary to study the stress- stones were boiled thoroughly to eliminate entrapped
strain characteristics of the soil in the foundation of air and subsequently placed in distilled water. The
structures in order to understand the settlement of the bottom porous stone thus prepared was kept inside the
structure and to predict and to make provision for the oedometer and a thoroughly wetted filter paper which
settlement which may occur during its lifetime. Since was made to the size of the porous stone was placed
oedometer experiments are time consuming, we will over the porous stone. The prepared soil was placed in
not be out of the way if the coefficients of the stress- the oedometer and the required prefixed thickness of
strain relationship is derived on basis of a few the sample was obtained by pressing the soil by static
experiments within the zone of clay content and compaction. The initial moisture content was kept at
utilize the relationship for estimating these parameters 24 percent for ensuring uniform compaction. Over
of the untested soil samples by interpolation. this surface of the soil, wetted filter paper and air free
porous stone were placed one over the other.
EXPERIMENTAL SET-UP AND TEST DETAILS
Equipment used is consolidometers of fixed ring type The oedometer mould was filled with distilled water
with capacity 16 kg/cm2. The soils taken for the upto its surface and the screws were tightened. Then
studies were the cohesive soils collected from the loading pad was kept gently and the oedometer
different dam sites having different percentages of was placed over the loading device. The extensometer
clay. The percentages of clay studies were 27.7, 45.3 was connected keeping sufficient travel for the
and 53.5 percent of 2 micron and less grain sizes. The measurement of compression during the experiment.
activities of these soil samples were 0.52, 0.66 and A small pressure of 0.25 kg/cm2 was applied as
0.81 respectively. setting load and the initial reading of the dial gauge
was noted. After 24 hours, final reading of the applied
Test Procedure setting load was noted. This final reading was the
The soil simple to be tested was dried in the oven for initial reading for the subsequent loading. The load
24 hours under the temperature of 105-1100C. After (stress levels) studied are 0.25-0.50 kg/cm2, 0.50-1.0
fixing the compaction dry density of 1.5 gm/cc, the kg/cm2, 1.0-2.0 kg/cm2, 2.0-4.0 kg/cm2, 4.0-8.0

kg/cm2 and in three cases the load (stress) increment N1 and N2 are dimensionless. The values of these
was upto 16 kg/cm2. The tests were carried out on constants during loading and unloading were
fully saturated samples with constant compaction evaluated for each of the soil sample. The values of
density of 1.5 gm/cc. these constants during loading are plotted against clay
contents in Figure 4 with a view to predict these
PRESENTATION AND DISCUSSION constants for other untested soil samples falling
The properties of the soil in the various samples tested within the tested clay content ranges.
are given in Table-I.
Stress - Strain Characteristics

The vertical strain during various stages of
consolidation was calculated as the ratio between the
settlement as recorded by the dial gauge and the initial
thickness of the sample. The relationship between the
vertical strain and the vertical stress both during
loading and unloading for all the tested samples are
shown is Figure 1. The figure shows a curvilinear
increase in vertical strain with increase in stress
during loading and a curvilinear decrease in stress
during unloading for all the samples tested. As the
percentage of clay increases, for the same vertical
stress of 4.0 kg/cm2, the strains observed are 0.136,
0.189 and 0.279 for the clay contents of 27.7, 45.3
and 53.5 percent respectively. Figure 2 shows the
variation between plasticity index and percentage
strain recovery. It can be seen that with increase in
plasticity index the strain recovery increases linearly.
A semi–log plot between vertical strain (1) on

arithmetic scale and vertical stress (v) on log-
arithmetic scale is shown in Figure 3. From this plot a
linear relationship for the stress and strain is arrived
as given below:
(v)1 = C1 (N1) 1 (during loading) (1)
(v)1 = C2 (N2) 1 (during unloading) (2)
In the above expressions the constants C1 and C2 have

the dimensions as that of stress whereas the constants

Variation in Compression Index (Cc) and Swelling which the curve becomes asymptotic is different for
Index (Cs) different clay contents. The values of stress levels are
From the present series of tests, the indices of 5.0, 3.5 and 2.5 kg/cm2 for the clay contents of 27.7,
compression and swelling were also evaluated for 45.3 and 53.5 percent respectively.
each of the tested sample at different stress levels and
it is seen that these values are varying slightly with Variation of (Cv ) with Stress Levels for different
stress levels. The average values of these indices Clay Contents
show increased values with increase in clay contents. The relationship between the coefficient of
consolidation and the clay content at different stress
levels is shown in Figure 7 which shows linear
decrement in the coefficient of consolidation with
increase in the clay content. The decrease in the
coefficient of consolidation (Cv) may be due to the
reduction in permeability which is caused due to
increase in clay content. It is interesting to note that
the straight lines for different stress levels are almost
converging at a clay content of around 52.8 percent.
This convergence may be envisaged as the combined
effect of increase in stress levels and clay contents.
The relationship between the coefficient of
consolidation and the clay content at different stress
levels can be expressed in the following from : y = mx
+ c, in which m and c are constants and the values of
these constants will depend upon stress levels.
Magnitudes of m and c are given in Table-II.
Table – II
Stress levels (m) cm2/Sec (c) cm2/Sec
0.25-0.50 (-) 0.049 0.026
0.50-1.00 (-) 0.041 0.022
1.00-2.00 (-) 0.027 0.014
2.00-4.00 (-) 0.015 0.008
4.00-8.00 (-) 0.013 0.007
Permeability Studies
Relationship between Coefficient of The coefficients of permeability were computed from
Compressibility (qv) and Clay Content the consolidation test results. As expected, the relation
Coefficients of compressibility obtained under various between the void ratio (e) and the coefficient of
stress levels are plotted against percentage clay permeability (k) was seen to be linear in a semi-log
contents. Figures 5 shows the (qv) values increase plot.
with increase in clay contents. It can also be seen that
at lower stress levels the increase in (qv) values are CONCLUSIONS
much more compare to higher stress levels. Based on the laboratory experimental studies on
It can be concluded from the above diagram that the consolidation characteristics of cohesive soils with
effect of clay content on coefficient of compressibility different percentages of clay, the following
is more during initial stages of loading and becomes conclusions were arrived at:
insignificant at higher loadings. (1) The strain recovery increases with increase in
plasticity index.
Coefficient of Consolidation (Cv) versus Average (2) The relationship between stress (v) and strain (1)
Vertical Stress for statically compacted cohesive soils in linear on a
The coefficient of consolidation decreases with semi-log plot. This relation can be expressed in the
increase in stress level and beyond a certain stress following forms:
level, the curve becomes asymptotic. This trend has (i) (v)1 = C1 (N1) 1 (during loading)
been observed for all the percentages of clay contents (ii) (v)1 = C2 (N2) 1 (during unloading)
which are shown in Figure 6. This indicates that the (3) The constant in the above expression (during
changes in coefficient of consolidation are loading) gives non-linear relationship with percentage
insignificant beyond a given stress level, which may clay content. The variation between these constants
be due to the fact that further decrease in void ratio is and clay contents shows convexity upwards (C1 ) and
comparatively negligible. The stress level beyond convexity downwards (N1).

(4) The relationship between coefficient of
compressibility (qv) and clay content at different stress
level is linear. It is found that during the initial stages
of loading the differences in (qv) values are significant
and at higher stress levels the variation of (qv) is
independent of clay contents.
(5) Coefficient of consolidation decreases linearly
with increase in clay content. The same trend has been
observed for all the stress levels studied. The plots are
converging towards increase in percentage of clay.
REFERENCES
Berry, P.L. and Wilkinson (1969), Radial
Consolidation of Clay Soil, Geotechnique, Vol.
XIX, No.2, pp. 260-262, June.
Booth, A.R. (1971), Compaction and Preparation of
Soil Specimens for Oedometer Testing, Soil
Specimen Preparation for Laboratory Testing,
ASTM, 78th ASTM- Canada pp. 216-225.
Fossberg, P.E. (1965) Some Fundamental Engineering
Properties of a Lime Stabilised Clay, Proceedings
of the VIth International Conference on Soil
Mechanics and Foundation Engineering, Vol. I,
pp. 221-223.
Kaul, B.K. (1973), Consolidation Characteristics of
Na Illite, Journal of the Indian National Society of
Soil Mechanics and Foundation Engineering, Vol.
89, No. SMI, pp. 115.
I.S. Code – 2720 Part XV – 1974.

ISBN 978-81-921121-0-7
NONDIMENSIONAL ANALYSIS OF A CIRCULAR PLATE RESTING ON

THE PASTERNAK FOUNDATION USING THE STRAIN ENERGY
APPROACH
Ashish Gupta
Assistant Professor, Department of Civil Engineering, B.I.E.T., Jhansi, E-Mail: shi_g2000@rediffmail.com
Dilip Kumar
Assistant Professor, Department of Civil Engineering, M.M.M.E.C., Gorakhpur
ABSTRACT: In this paper, nondimensional expression for the deflection of a thin circular elastic plate resting on
the Pasternak foundation by the strain energy approach has been adopted. A complete nondimensional analysis has
been done for the various parameters like modulus of subgrade reaction (ks*) and shear modulus (G*) for estimating
the deflection (w*) of the circular plate.
INTRODUCTION
The soil-foundation interaction problem is of REVIEW OF PAST WORKS
importance to many engineering situations, which A system of equations was developed by Reissner
include the study of raft foundations, pavement of (1945) for the theory of bending of thin elastic plates
highways, water tanks, airport runways, buried which takes into account the transverse shear
pipelines and pile foundations. A major proportion of deformability of the plate. Mindlin (1951) deduced
these studies are primarily concerned with elastic the three dimensional equations of elasticity in to a
analysis of the interaction problem. A complete two dimensional theory of flexural motions of
analysis of the interaction problem for elastic bodies isotropic elastic plates and computed the velocities of
generally requires the determination of stresses and straight-crested waves and found to agree with those
strains within the individual bodies in contact. obtained from the three-dimensional theory. Galletly
For generalized foundations, the model assumes that (1959) studied the circular plates on a generalized
at the point of contact between plate and foundation elastic foundation. Rodriguez (1961) presented the
there is not only pressure but also moments caused by three-dimensional bending of a ring of uniform cross-
interaction between the springs. Such a continuum sectional area and supported on a transverse elastic
model is often used to represent flexible structures foundation. The modified Bessel function for the
that are essentially one-dimensional in geometry. The three-dimensional case of a finite beam or plate was
ground may be represented by foundation model like given by Kerr (1964) by using the Kernal function.
the one-parameter Winkler model or the two- Weitsman (1970) analyzed the one parametric
parameter models proposed by Pasternak (1954) and Winkler and Reissner foundation under a single
Vlazov and Leont’ev (1966) or higher order models concentrated load under the assumption that tensile
like those developed by Kerr (1964) and Reissner stresses could not be transmitted across the interface
(1958). between beams or plates and their supporting sub
In recent years, efforts have been made to analyze the grade. This assumption forms no contact regions
thin circular elastic plate resting on a two parameter under the beam or plate. Due to no contact region the
Pasternak foundation under concentrated and beam or plate lifts up away from its foundation and
distributed loads adopting approximate technique for the sub grade does not support the structure. The lift
determining approximate plate deflection. The off phenomenon under various boundary conditions
analysis also considers the effect of tensionless based on the beam or plate was also reported.
characteristics of the Pasternak foundation, that is, lift Celep (1988) presented a phenomenon that a thin
off of the plate from the foundation surface. It was circular plate subjected to eccentric concentrated load
reported that lift off appears when the slopes of the and moment as well as uniformly distributed load has
foundation surface and that of the plate are not equal. been supported on an elastic foundation of Winkler
A circumferential shear force develops along the type foundation reacts in compression only and he
contour of the plate due to the slope discontinuity of reported that the plate will lift off when the
the plate and soil surfaces. It also reflects the lift off foundation stiffness is low. It has been also reported
the plate and, consequently, the soil reactions develop that, if the edge of plate is not anchored against
within the range of contact radius. lifting off, the plate may have a partial contact with
the foundation.

Celep and Turhan (1990) analyzed the axisymmetric analysis has been presented the circular plates on
vibrations of a circular plate subjected to a tensionless Winkler foundation; that reacts in the
concentrated dynamic load at its center. The solution compression only.
given by them was the extended solution given by Wang et al. (2001) presented the exact axisymmetric
Weitsman (1970). The solution was carried out by buckling solutions of Reddy circular plates on the
using the Galerkin’s method and the modal functions Pasternak foundation. The solution was the extended
of the completely free plate. This problem was a solution of Kirchhof’s buckling solution of circular
nonlinear one but linear theory of plate was used in plates without elastic foundation.
the problem. Guler (2004) analyzed a thin circular plate resting on
Guler and Celep (1993) reported the extended study a two parameter Pasternak type foundation under
of axisymmetric vibrations of a circular plate on concentrated central and distributed loads. It was
tensionless elastic foundation by Celep and Turhan analyzed the effect of tensionless characteristics of
(1990). The solution was given by using Galerkin’s the Pasternak foundation system. For the analytical
method using the modal functions of the completely expressions, lift off (plate takes lift from the
free plate. Various contact curve and time variation foundation surface) had been considered. The system
with the central and edge deflection were shown. The of nonlinear algebraic equations accomplished by
most valuable result was given that increasing the using an iterative method. Galerkin technique has
load values, keeping loading eccentricities value been adopted for the approximate analytical
constant, an increase in the displacements without solutions.
changing the extent of the contact. Shukla at al. (2011) derived an expression for the
Utku et al. (2000) presented a new formulation for deflection of a thin circular elastic plate resting on the
the analysis of axisymmetrically loaded circular Pasternak foundation under concentrated load by
plates supported on elastic foundation. In this adopting the strain energy approach. Nondimensional
formulation the circular plate was modeled as a series expression for plate deflection is also reported for
of concentric annular plates in such a way that each convenience in this study; which is shown by
annular plate is considered as simply supported along Equation 1. With the help of the complete derivations
the inner and outer edges, hence no contact develops by shukla et al. (2011), an effort has been done for
between the plate and the foundation. The deflection analyzing the circular plates resting on two-parameter
and radial bending moment were chosen as the foundation soil in the present study with some design
solution variables at each interior support. This charts.
DERIVATION FOR THE NONDIMENSIONAL ANALYSIS OF THE CIRCULAR PLATE RESTING ON

PASTERNAK FOUNDATION UNDER THE CONCENTRATED LOADING CONDITION
The response function for the Pasternak foundation is given by:
  2w 1 w 
q  k s w  GH    (1.1)
 r
2
r  r 
where H = thickness of the shear layer. On reproducing the deflection of the circular plate; It can be shown as:
w = C0 +C2.r2 (1.2a)
or,
w
 2C 2r (1.2b)
r
1 w
 2C 2 (1.2c)
r r
and,
2w
 2C 2 (1.2d)
r 2
By substituting Eq. 1.2(a), (c) and (d) in to Eq. (1.1),

q  k s ( C 0  C 2 r 2 )  4 C 2 GH (1.3)
The strain energy U1 of thin circular elastic plate resting on Pasternak foundation can be calculated as:
2
U 1  4  D (1   ) C 2 a 2 (1.4)
The work done W by the load acting on the circular plate on Pasternak foundation can be shown as:
W  Q C 0 (1.5)
The strain energy U2 owing to the deformation of the Pasternak foundation is calculated as:
2a
1
U2    2 q w  r dr d 
0 0
(1.6a)
By substituting Eqs. (1.2a) and (1.3) into Eq. (1.6a),

2 a
1
U2    2 k C
0 0
s 0  
 C 2 r 2  4 C 2 GH C 0  C 2 r 2 rdrd  
2 a
  2 k C 
1
 s 0  C2r 2  2

 4 C 2 GH C 0  C 2 r 2 rdrd 
0 0
2 a
1
  2 k C   
2 2 2
 s 0  C 2 r 4  2C 0 C 2 r 2  4G C 0 C 2 H  C 2 r 2 H rd rd 
0 0
a
2
 k C 
2 2 2
 s 0 r  k s C 2 r 5  2 C 0 C 2 k s r 3  4 C 0 C 2 GrH  4 C 2 r 3 GH dr
2 0
2 2 a
 2 r
2
r 4
k s C 2 r 6 4 GHC 0 C 2 r 2 4 GHC 2 r 4 
  k s C 0  2k sC0C2    

 2 4 6 2 4 0
2 2
 2 a a 4 k sC2 a 6 2 
   k s C 0  2k sC 0C 2   2 GHC 0 C 2 a 2  GHC 2 a 4 
 2 4 6 
 ks 2

6
2 2
a 3C 0  3C 0 C 2 a 2  C 2 a 4  G Ha 2 2C 0 C 2  C 2 a 2  2
  (1.6b)
The total potential energy of the thin circular elastic plate-foundation system is:
  U 1  U 2  W  (1.7)
By substituting Eqs. (1.5), (1.6a) and (1.6b) into Eq. (1.7),
k s 2
2
  4C2 D(1 )a 2 
6
2
 2
 2

a 3C0  3C0C2a 2  C2 a 4  GHa2 2C0C2  C2 a 2  QC0 
(1.8)
For the stable equilibrium, potential energy should be a minimum. Thus,

 
 0 (1.9a)
C 0
and,
 
 0 (1.9b)
C 2
From Eq. 1.9 (a) and (b),

0
C0
k s 2
6
 
a 6 C 0  3 C 2 a 2  2  GHa 2 C 2  Q  0
4
Q  k sa 2C0    k2sa  2GHa2  C2
 
or,
C 
Q   k sa 2C 0  (1.10a)
2 4
  k sa 2 
  2G H a 
 2 

0
C 2
 k s 6   k s 4 
 8  D 1   a 
2
a  2  GHa 4  C 2   2  GHa 2
 a C 0
 3   2 
(1.10b)
By substituting Eq. (1.10a) into Eq. (1.10b),

 
 
 ks 6 4 
Q  k s a 2C0    k 
8D1  a  3 a  2GHa    k a 4
2
  2GHa2  s a 4  C0
    2 
  s  2GHa2  
  2  
 
 
 ks 4 2
  Q   k sa C 0      a 2  2 GH  k s a 2 
 8 D 1     3 a  2 Ga H    k a 2
2
s    2 
   2 GH  C 0 
  2  
2
 k a2 
  2 GH  s   a 2
Q  2 
  k sa 2 
C0  k a4 
 8 D 1     s  2 Ga 2 H 
 3 
or,

 
 
 
 
 
Q  1  (1.11a)
C 0  2  2 
a   k sa 2
 
 2 GH  
 2 
 k    
s 4
  k sa  
  8 D 1     2 Ga 2
H  
  3  
By substituting Eq. (1.11a) into Eq. (1.10a),

 
 
 
2
 k s a Q   
   
Q  a 2   1 
C2    2 
 k s a 4 2   k sa 4 2   k a 2
 
  2 Ga H    2 Ga H   2GH  s

 2   2   2  
ks   
  ksa 4  
  8D 1      2Ga 2 H  
  3 
  
  
  
  
  
Q   1  (1.11b)
C2  2
1 ks 2 
 k a    k a  2
 a 2  s  2 GH     2 GH  s  
 2    2  
  ks  4 
 k sa
  8 D 1      2 Ga H  
2
  3 
By substituting Eq. 1.11(a) and (b) into Eq. (1.2a),
   
   
   
    
     
Q 1   Q r2  ks 
w 2  2   2   2
1
a   k a2   a2  ksa 2GH   ksa2  
2GH s   2    2GH  
  2       2  
ks     ks   
 ks a4   ksa4 
 8D1   8D1 
2 2
2Ga H 2Ga H 
 3   3 
(1.12)
A special case of Eq. (1.12) for G = 0 reduces to the following expression:

 
 
Q 1  1  Q .r 2 (1.13)
w 
 .k s .a 2   1 32 .D .1       1 16 .D .1    
   4
   .k s .a 4   
 3 k s .a   6 k s .a 4 
Eq. (1.13) is the expression for the case of a thin circular plate resting on the Winkler foundation.
On dividing Eq. (1.13) by radius of the circular plate a, it follows:
   
   
   
    
     
w Q  1   Qr2   ks 
 3 2   2   2
1
a π.a  
 
2GH s 
k a2   πa3 ksa  2GH  
 2GH 
ksa2 
 
      2     
 2      2 
ks   ks  
 ksa4 2   ks .a4 
 8D1 ν   2Ga H  8D1 ν  2
 2Ga H 
3   3 
   
   
   
    4 
     ks a 
Qa  1   Qar2  D 
 4  2   2   2
1
4
a 1 k a2  a  ks a  a4  k sa2 
D  2GH s   D  2GH  2GH   
D D 2    D  2  
  ks .a 4 2
D  2  
ks     
 k s a4 2Ga2H   D ksa4 2Ga2H 
 81     81   
 3D .D   3D .D 
   
   
   
   4 
   kSa 
2
1Qa  1  1Qa  r  1  D 
   2       4 2  2
1
π D  2GaH ks a   π D a ks a 2Ga H
2 4 2
2GaH ks a  4
          
ks a4
D 2D   2D D 
 kS.a4
D 2D 
   4

2     4

2 
 D 81ν ks.a 2G.a H  D 81ν ks a 2G.a H 
 3D D   3D D 
(1.14)
Considering nondimensional parameters as

w
w    (1.15a)
a
r
r    (1.15b)
a
 Qa 
Q    (1.15c)
 D 
 Ga 2 H 
G     (1.15d)
 D 
  k a4 
k s   s  (1.15e)
 D 
Eq. (1.14) reduces to,
   
   
   
   
  2  

 Q  1  Q r   ks 
w   2     2
 1

  k s  2G   
 
   ks    ks  
 2G      2G  
  2    2   2   
 k s      
ks   
 
 ks   k 
 81      2G    81     s  2G  
 3   3 
or,
 
 
 
  
2   2 

 Q r Q  1  1  r k s 
w     
 ks 
   2  k s 
 
   k    
 
  2G   2G   s   2  2G  
 2     2     
ks   

 ks 
 8 1      2G  
3
(1.16)
Eq. (1.16) provides an expression for deflection of the thin circular elastic plate resting on Pasternak foundation
under concentrated loading in nondimensional form.
CONCLUSIONS been found that a large number of studies and various

Estimation of deflection of circular plates is of great analytical methods and the numerical techniques
importance as it is associated with many geotechnical have been adopted for the analysis of circular plates
engineering structures. As described earlier, it has resting on the different foundation models under

concentrated loading condition. Various efforts have Journal of Sound and Vibration, Vol.183,
been made to analyze a thin circular elastic plate No.2, pp.185-195.
resting on the Winkler foundation and Pasternak
foundation by adopting different approaches in order
Kerr, A.D. (1964). “Elastic and Viscoelastic
to present the analytical expressions for the plate Foundation Models.” Journal of Applied
deflection as well as for the study of the plate lift off. Mechanics, Transactions ASME, Vol.31,
Deflection can be directly calculated by putting all No.3, pp.491-498.
the parameters in this equation. Eq. (1.16) shows the indlin, R.D. (1951). “Influence of Rotatory
deflection of the circular plate resting on the
Pasternak foundation under concentrated loading in Inertia and Shear on Flexural Motions of
their nondimensional form. Isotropic, Elastic Plates.” Journal of
Applied Mechanics, Transactions ASME,
NOTATIONS Vol.18, pp.31-38.
a Radius of the plate (m) Pasternak, P.L. (1954). “Fundamentals of a
D Flexural rigidity of the plate (Nm)
G Shear modulus of the Pasternak foundation (N/m2)
New Method of Analyzing Structures on
H Thickness of the Pasternak shear layer (m) an Elastic Foundation by Means of Two
ks Modulus of subgrade reaction of the Paste nak Foundation Moduli.” (In Russian),
foundation (N/m3) Gosudarstvenuoe izdatel’slvo literary po
q Reaction for the foundation to applied concentrated stroital’stvu I arkhitekture, Moskva-
load (N/m2)
Q Concentrated load applied at the centre of the plate Leningrad (as cited in Vlasov and
(N) Leont’ev 1966).
r radius vector of any point on the plate (m) Rodriguez, D.A. (1961). “Three-
rb Contact radius (= b/a) (dimensionless) dimensional bending of a ring on an
w Deflection of the plate (m) elastic foundation.” Journal of Applied
υ Poisson’s ratio of the plate material (dimensionless)
Mechanics, 28, 461-463.
REFERENCES
Utku et al. (2000). “Circular Plates on
Celep, Z. (1988). “Circular Plates on Elastic Foundations Modeled with
Tensionless Winkler Foundation.” Annular Plates.” Computers and
Journal of Engineering Mechanics, Structures, Vol.78, pp.365-374.
ASCE, Vol.114, No.10, pp.1723-1739. Vlazov, V.Z., and Leontiev, U.N. (1966).
Celep, Z. and Turhan, D. (1990). Beams, plates and shells on elastic
“Axisymmetric Vibrations of Circular foundations, Israel Program for Scientific
Plates on Tensionless Elastic Translations, Jerusalem (translated from
Foundations.” Journal of Applied Russian).
Mechanics, ASME, Vol.57, pp.677-681. Wang, C. M., Xiang, Y., and Wang, Q.
Galletly, G D. (1959). “Circular Plates on a (2001). “Axisymmetric buckling of reddy
Generalized Elastic Foundation.” Journal circular plates on Pasternak foundation.”
of Applied Mechanics, Transactions Journal of Engineering Mechanics,
ASME, Vol.26, pp.297. 127(3), 254-259.
Guler, K. (2004). “Circular Elastic Plate Weitsman, Y. (1970). “On Foundations That
Resting on Tensionless Pasternak React in Compression Only.” Journal of
Foundation.” Journal of Engineering Applied Mechanics, ASME, Vol.37, No.
Mechanics, ASCE, Vol.130, No.10, 1, pp.1019-1031.
pp.1251-1254.
Guler, K. and Celep, Z. (1993). “Static and
Dynamic Responses of a Circular Plate
on a Tensionless Elastic Foundation.”

ISBN 978-81-921121-0-7
OPTICAL SENSOR BASED CRACK OPENING DISPLACEMNETS

STUDIES IN CONCRETE
K.Samrajyam
Research Scholar, Department of Physics, NIT, Warangal
B.Sobha
Associate Professor, Department of Physics , NIT, Warangal
T D Gunneswara Rao
Associate Professor, Department of Civil Engineering , NIT, Warangal
ABSTRACT: Cracking is one of the frequent cause of complaints in concrete industry .Cracking is known to
occur due to fatigue, over loading and damage caused either by loading or environment. The failure of bridges,
structures, buildings and dams is due to the cracking. This underlines the importance of estimation of crack
width as predictor for the health of the concrete structures .Here we report on preliminary results obtained using
a plastic optical fibre intensity based technique. The results so obtained in a three point bend test on plain
concrete specimens are compared with the data obtained from the INSTRON.
INTRODUCTION Sample preparation

Many public structures that are well designed need Materials
periodic maintainence and regular inspection to Cement conforming to IS 12269 – 1983 with specific
ensure their health. In this connection crack
gravity 3.15 is used for the concrete mix. Natural river
monitoring and detection plays an important role
.Crack widths ranging from 0.2 mm to 2mm caused sand with specific gravity 2.60 is used as fine
either by exposure to environment , natural hazards aggregate. Crushed coarse aggregate of size 20mm
are known to signal the failure of the structure. was used (60%) and also 10 mm size was used (40%)
They exist both optical ( based on optical fibre with specific gravity 2.78 .
interferometric techniques , FBGs[1]). And also non
optical methods for estimation of the crack mouth Mix proportions
opening displacement in concrete structures. M20 grade concrete mix design was used for the
However the excellent performance of these methods casting and the mix ratio: 1:1.5:3 was used for the
is offset by their expensive instrumentation preparation of concrete samples[3].
requirements for signal processing. In most cases the
high measurement resolution provided be these Casting
methods proves to be of no direct advantage given Cubes of 100mm size were used to determine the
the fact that more economical methods are provided compressive strength of concrete. Cylinders with 150
by using plastic optical fibres. These intensity based mm diameter and 300 mm length were used to
devices are of low cost.And it is more easier to determine the Modulus of Elasticity of concrete.
measure the optical power than the phase /state of Prisms of 100mm x 100mm x 400mm (B x D x
polarization of light. It is more advantageous with L),Beams of dimensions 700*150*100 (B*D*L), size
multimode fibres as the phase and state of polarization were casted. A needle vibrator was used for
is lost when it propagates through this kind of fibres. compaction. All the specimens were cured for 28 to
It was our intention to choose an economical, 40 days in the curing tank. After curing time the
inexpensive technique that would fit into our needs in samples were separated from the curing tank and
the long run. Hence we opted to use plastic optical notch depth (a/d=0.2,0.3)was used for the prisms and
fibres for the investigation of our crack width beams .Notch was made in the specimens using notch
experiments cutter.
Description of a measurement system
Advantages of Plastic Optical Fibre
Characterisitics We used a standard multi-mode plastic fibre having a
In addition to the usual advantages offered by silica core diameter of 980 µm made with PMMA resin
optical fibres the other advantages include : a) capable surrounded by a fluorinated polymer cladding of
diameter 1000 µm . A LED source injects a
either of being surface mounted or be embedded in
large civil structures and composite materials to asses modulated light beam into the fibre whose light output
their reliability b)large inherent strain sensitivity is received by a highly-sensitive NPN
(6%-12%) c) have high fracture toughness d) provide phototransistor. The photo transistor ‘s current is
large core size , large numerical aperture , flexibility, converted to voltage using a transimpedance
easy handling, less costly.(2,4) amplifier . We developed a simple, inexpensive
transimpedance amplifier that converts the current
EXPERIMENTAL DETAILS generated by the photo transistor in to a voltage

which is then logged into a computer using a data
logger
Preparation of the sensor

The surface of the POF sensor is abraded using a fresh
razor blade causing the removal of a predetermined
length of the cross section.( we refer to this as guage/
sensitization length , approximate range = 2 to 5 cm).
This process is adopted to increase the amount of light
lost in the fibre in the bending process –result is
increase in the responsivity.
The sensor so prepared is glued to a prenotched
concrete specimen.using a suitable cynoacrylate
adhesive and after intial preparation of the concrete
specimen surface as outlined in [5]
Test Procedure
The figure [1] below shows the photograph of the
experimental set up used for the 3-point bend
specimen test arrangement in the Materials
Laboratory of the Civil Engineering Department , at
NIT, Warangal.
Fig. 2 Typical Graph between Stroke (mm)v/s

norm.pof voltage/Load

The experimental runs were performed on five
concrete beams with dimensions (40*15*10) cm,
(40*10*10) cm using INSTRON. Initially when the
beam was at rest there no change in the pof voltage
was observed .When the crack generation started, the
pof voltage varies and shows a reduction beyond the
maximum stroke value. It can be deciphered that the
measured displacement or crack width range using
pof sensor lies in the range 0.02 to 0.1 mm and this
displacement is equal to the stroke value obtained
from the instron. It is known that stroke is related to
Fig. 1 Experimental setup for the three point bend test displacement and displacement is related to Crack
of concrete beam on 1000KN instron mouth opening displacement (CMOD).
The pof sensor was pasted across the notch with a CONCLUSION
cyanoacrylate adhesive . The load was applied on the A concept of using plastic optical fibers for the
beam with a cross head displacement rate is 0.002 measurement of crack width (range 0.02 to 0.1) mm
mm per sec. During the loading The data obtained in concrete beams is explored in this paper. Further
from the POF sensor was logged using a data logger( calibration tests in this direction are necessary to
sampling speed of 1 sample/ sec). The normalised reinforce the adoption of plastic optic fiber sensors in
POF voltage is observed to vary according to the the health monitoring of the structres
crack opening displacement (in the range 0.01 to 0.1
mm)(COD) and is shown in Fig 2 below ACKNOWLEDGEMENTS: The authors would like
to gratefully acknowledge the partial financial
assistance received from the AP State Council of

Science and Technology in the form of R&D and Concrete Technology, by M.L.Gambhir, Tata-
special projects sanctioned in 2009 McGraw Hill , New Delhi
Plastic optical fibers: an introduction to their
REFERENCES technological processes and applications. Zubia
Fiber-optic Bragg grating sensors for bridge J, Arrue J., Optical Fiber Technology,2001,
monitoring Cement Concrete Components , 7(2): 101–140
Maaskant R, Alavie A T, Measure R M, Tadros Crack detection and vertical deflection monitoring in
G, Rizkalla S Hand Guha-Thakurta A, 1997, concrete beams using plastic optical fibre
vol.19 , pp21–33. sensors, Kuang K S C, Akmaluddin, Cantwell
Plastic Optical Fibre Sensors for Structural Health W J, Thomas C Measurement Science and
Monitoring: A Review of Recent Progress, K. S. Technology,2003, vol.14, no.2, pp205–216
C. Kuang, S. T. Quek, C. G. Koh, W. J.
Cantwell, and P. J. Scully, Journal of Sensors,
Volume 2009 (2009), Article ID 312053,

ISBN 978-81-921121-0-7
DYNAMIC ANALYSIS OF NAILED OPEN CUTS- A CASE STUDY

Bhishm Singh Khati
Assistant professor, CED, G B Pant Engg. College, Pauri, Uttarakhand, bhishm_khati@yahoo.co.in
Dilip Kumar
Assistant professor, CED, NIMS University,Jaipur, Rajsthan, India, jhadilip@sify.com
Swami Saran
Emeritus Professor, Dept. of Earthquake Engineering, IIT Roorkee, India, Email: saranfce@iitr.ernt.in
ABSTRACT: Soil nailing method is used for stabilizing the tunnel portal and shafts, slopes and to support the
excavations and also for remedial uses like repair of unstable old retaining walls, stabilization of failed slopes
etc. by increasing the apparent cohesion of the soil. This method is very easy and effective method in earthquake
prone areas. At present, only static and Pseudo-static analysis of a nailed open cut has been reported in the
literature. Dynamic analysis is most significant for soil nailing projects in earthquake prone regions of India.
In this paper dynamic analysis of nailed open cuts is presented in order to obtain the displacements of the cut. It
is assumed that nailed soil mass corresponding to a static factor of safety equal to unity moves during the
earthquake. The soil behind this mass is replaced by series of springs and the nails being replaced by stiffeners.
Equation of motion works out to be a two degree differential equation, which on integration and putting proper
boundary conditions yield the displacement of the cut.
Keywords: Open Cuts, Soil Nailing, Dynamic Displacements, Static Analysis
INTRODUCTION with a permanent cast-in-place concrete facing. Soil

Soil nailing reinforces the soil within a failure zone to nailing is not appropriate at sites consisting of clays
make it behave as a monolith, much the same way as which creep, clean sands which will not stand up or
lumber acts after having been nailed together. The areas below the water table. The nails are either
process is done from the top down, that is, soil is placed in drilled bore holes or grouted along their
excavated and secured, or “nailed”, in small lifts, length to form ‘Grouted nails’ or directly driven into
starting at the surface. Residual stresses in the soil of the ground as ‘driven nails’. It is seen that the nailing
the slip plane act on the grouted mass of bar in grout techniques is useful for making underground cut and
and “nails” itself. The great advantage of the also for stabilization of the loose soils.
technique is that it allows soil retention in very tight
Pseudo-static Analysis
areas at any configuration and at competitive costs to
A vertical nailed open cut of height H has been
other proven techniques such as tiebacks Once a lift is
analyzed to evaluate the critical failure surface using
secured it becomes self-supporting, able to
pseudo – static analysis. The length of the nail used is
permanently hold itself in position as the excavation
L and with a vertical and horizontal spacing of Sv.
continues. The basic concept of soil nailing (Patra and
The failure surface with a Factor of Safety (F.S) of
Basudhar, 2001) is to reinforce and strengthen the
unity is considered as critical. A computer program
existing ground by installing closely spaced steel bars,
has been developed for the pseudo – static analysis to
called "nails," into a slope or excavation as
find the critical failure surface using the MATLAB
construction proceeds from top to down. This process
software.
creates a reinforced section that is itself stable and
The pseudo-static analysis is based on the following
able to retain the ground behind it. The
assumptions: (After Saran, 2005)
reinforcements are passive and develop their
1. The deformations of the soil in the active
reinforcing action through nail ground interactions as
zone are sufficient to fully mobilize the
the ground deforms both during and following
shear strength of the soil over the entire
construction. Nails work predominantly in tension,
failure surface.
but are considered by some to also work in
2. The mass of the nail is neglected because it
bending/shear under certain circumstances. Generally,
is very less compared to the mass of the
the soil nails significantly increase the apparent
soil.
cohesion of the soil through their ability to carry
3. The failure is along a surface defined by
tensile loads. A construction facing is also usually
the arc of a logarithmic spiral passing
required, and is typically shotcrete reinforced by
through the toe and the ground at right
welded wire mesh or chicken mesh. For permanent
angle. The centre of the spiral is located on
walls, the shotcrete construction facing is typically
a straight line which rises at an angle Ф
covered in cast-in-place concrete facing. The walls
and passes through the point where the
consist of soil nails spaced approximately 5 ft. (1.5 m)
failure surface meets the ground.
on center with a shotcrete facing which is covered

4. The shear resistance of the nail due to nail
bending stiffness is taken care of by using
the plastic analysis method suggested by
Jewell and Pedley (1990). The shear
resistance mobilized in the nail is
calculated by limiting the soil bearing
pressure to the safe value given by the
equation
 b  v {(1  Ka ) / 2) tan( 2   4 )}
{  ) tan }
2
*e
(1)
where
σb = Limit bearing stress between the soil and
the nail
σv = Vertical stress
Ф = Angle of internal friction of the soil
Ka = Coefficient of active earth pressure
5. The normal stress on the nail surface in the
horizontal direction is assumed to be Ka Fig. 2(a) Nailed Open Cut of height H
times the normal stress in the vertical
direction.
6. The internal failure mode of the wall is either
by pull-out or rupture or excessive bending
leading to the formation of a plastic hinge in
the nail whichever is critical. The horizontal
and vertical seismic forces are taken in terms
of horizontal and vertical seismic coefficients
αh and αv.
The inputs for the program for pseudo-static analysis

are as follows
i) H – Height of the nailed open cut, m.
ii) L – Length of the nail used in the cut,
m.
iii) D – Diameter of the nail used in the cut,
m.
iv) Sv – Vertical spacing of the nail, m
v) γ – Unit weight of the soil, kN/m3
vi) Ф – Angle of internal friction of soil,
deg
vii) αh – Horizontal seismic coefficient
viii) αv – Vertical seismic coefficient
ix) C – Cohesion of the soil, kN/m2 Fig. 2(b) Mathematical Model of Nailed Open Cut
x) f1 – Limit bond stress of the nail, kN/m2
xi) fy – yield stress of the nail, kN/m2 The assumptions on which dynamic analysis is based
are the same as that of the static analysis discussed in
DYNAMIC ANALYSIS TO FIND THE section 2. In addition to these assumptions the
DISPLACEMENTS following assumptions are also made
Dynamic analysis of the nailed open cut has been (i) The earthquake motion may be considered as
done to evaluate its displacement at various an equivalent sinusoidal motion with
heightsThe soil in the failure surface is considered to uniform peak acceleration and the total
be the moving mass (Fig. 2(a)). The mass of the soil displacement is equal to residual
in the failure surface is lumped at its CG. The soil displacement per cycle multiplied by number
behind the failure surface is replaced by closely of cycles.
spaced independent elastic springs as shown in Fig. (ii) Soil stiffnesses for displacement of the open
2(b). The nails behind the failure surface are replaced cut towards the backfill and away from the
by elastic stiffeners as shown in Fig. 2(b). backfill are different.

(iii) Damping of soil and base friction are evaluated by linear regression. The equations of the
neglected. curve for each nail length are given in Table 1.
(iv) The critical failure surface is the failure
surface at which the factor of safety (F.S) is Table 1 Equation of the Curves to find Stiffener
unity which is found by pseudo-static Constants
analysis discussed in section –2.
Diam Lengt Equation of the curve
eter, h, L
FORMULATION USED FOR THE DYNAMIC D m
ANALYSIS mm
The inputs for the program are as follows 16 3.0 K = -0.038 q3 + 1.5 q2 – 2.2 q
i) H – Height of the nailed open cut, m. + 320
ii) L – Length of the nail used in the cut, m. 2.5 K = -0. 32 q3 + 1.5 q2 – 5.8 q +
iii) D – Diameter of the nail used in the cut, 250
m. 2.0 K = -0.014 q3 + q2 – 8.3 q +
iv) Sv – Vertical spacing of the nail, m 250
v) nh – Modulus of sub grade reaction of 1.5 K = -0.048 q3 + 3.6 q2 – 70 q +
the soil, kN/m3 600
vi) γ – Unit weight of the soil, kN/m3 25 3.0 K = -0.049q3 – 2.4 q2 + 61 q +
vii) F – Excitation force, kN 140
viii) f – Excitation Frequency, Hz 2.5 K = -0.026 q3 + 0.58 q2 + 30 q
+ 120
EQUATIONS OF MOTION 2.0 K = -0.038 q3 + 2.2 q2 – 19 q +
The soil in the failure surface is considered to be the 440
moving mass (Fig. 2(a)). The mass of the soil in the 1.5 K = 0.0075 q3 – 0.1 q2 + 5.2 q
failure surface is lumped at its CG. The soil behind + 310
the failure surface is replaced by closely spaced
independent elastic springs. The nails behind the These equations are further generalized for different
failure surface are replaced by elastic stiffeners. by correlating the coefficients the above equations
The spring constant values at various division
in terms of length of the nail as follows,
points were For D = 16 mm,
1 2 K si  (0.25L 2
- 1.2L - 1.2)q 3
 (2.2L 2 - 11L  15)q 2
K1  h Sv (2 a)
6  (-58L 2  300L - 390)q  (420L 2
- 210L  270)
2
K 2  h Sv (2 b) (3)
2
K 3  2 h S v (2 c) For D = 25 mm,
2 3 2 2
2 K si  (0.12L - 0.51L - 0.51)q  (5.5L  23L - 22)q 
K i  ( i  1)  h S v (2 d)
2 2
(55L  210L  180)q  (110L  330L  100)
1 2
(4)
Kn  ( 3 n  1)  h S v (2 e)
6 The equations of motion of the nailed cut using
where K1 and Kn are the spring constants of top and D’Alemberts principle can be written as follows
bottom most points, Ki, the spring constant at any .. n n n
intermediate divisions. (Saran, 2006) M x  (  K i   K si ) x  (  K i [(H - h) - (i - 1) Sv] 

i 0 i 0 i 0
The stiffener constants of the nails are determined
n
by carrying out pull out tests in the laboratory. 1
Gosavi (2006) carried out 135 pull out tests with  K si [(H - h) - (i - 2 ) Sv])   Fo sin  t
i 0
different surcharge values for the length of the nails
(5)
of 1.0 m, 1.5 m, 2.0 m, 2.5 m and 3.0 m and load –
displacement curves for different surcharges were n n
.. 1
plotted. In the present study the loads corresponding J   ( K i [(H - h) - (i -1) Sv]   K si [(H - h) - (i - ) Sv]) x 
to displacements of 1 mm, 2 mm, 3 mm and 4 mm i 0 i0 2
were noted for all the tests and the stiffness were n n
calculated by dividing the load by the 1
(  K i [(H - h) - (i - 1) Sv]2   K si [(H - h) - (i - ) Sv]2 )   0
corresponding displacements. Mean stiffness for i0 i 0 2
each of these tests were taken and plotted against (6)
the surcharges used for that nail length. The Where
equation of the curve for different nail lengths was M – Mass of the soil in failure wedge

J – Polar mass moment of inertia of the soil about VALIDATION OF THE PROGRAMS
the axis of rotation The results from the dynamic analysis were
w – Frequency of the excitation force validated using a three dimensional model of nailed
H – Height of the cut cut using ANSYS. (Fig. 3)
H – Height of CG from the base of the cut
x – Translatory displacement
θ – Rotational displacement
The solution of the equations can be written as
x  X sin  t (7}
   sin  t (8)
where X and β are arbitrary constants.
Solving Eqs. (5) and (6), we get

ao
X (9)
b2
(a  2 ) 
r 2 (c  2 )
ao (10)
 
2 r2 2
( a   )( c   ). b
b Fig. 3 Three Dimensional Model of Nailed Cut
Therefore displacement is given by
ao (11) PROBLEM DESCRIPTION
x sin  t The soil-nail model is a two degree of freedom system
b2
(a   2 )  2 with combined translational and rotational motion.
r (c   2 ) The height of the cut is H (Y-direction), width of the
ao cut B (X – direction) is taken as 0.8 times the height.
 sin  t (12) The length of the cut is kept equal to the length of the
2 r2 2 nail L (Z – direction).
( a   )( c   ). b
b
The displacement at the top of the cut is given by BOUNDARY CONDITIONS
x top  x  ( H  h )  (13) The soil – nail model is restrained in all directions at
the base. There are no restrains at any other directions
Where,
n n
( K Static Analysis
i  0
i  
i  0
K si )
Static analysis of the open cut is performed using the
 a (14)
M following loads to evaluate the displacements of the
cut.
F (a) Self weight of the cut
o
 a o
(15) (b) Surface loads
M (c) Acceleration due to gravity
n n
1 Dynamic Analysis
{Ki{(Hh)(i1)Sv}Ksi{(Hh)(i )Sv)} Response Spectra analysis of the cut is performed
i0 i 0 2
b using the standard response spectra from IS- 1893:
M 2002 (Part-1). At the start modal analysis is
(16) performed to find the frequency of the structure. The
n mode extraction was done using Block Lanczos
{ Ki {(H  h)  (i 1) Sv }2 method to find twelve modes. The modes were
i 0 combined using Square root of sum of squares (SRSS)
 method. Using these frequencies response spectra
J analysis was carried out to find the dynamic
n (17) displacement of the structure. For response spectra
 Ksi{(H  h)  (i 1) Sv / 2)}2} analysis, single point excitation response spectrum
i 0 was used.
c
J

The results of displacement of the cut from ANSYS
and MATLAB for the following parameters were (c) Height of the cut = 12 m
carried out (Table 2).
Height of the Cut, H = 12 m
Table 2 LIST OF PARAMETERS USED 40
Parameters Values considered

30
Displacement (mm)
Height of the Cut, (H) 6 m, 12 m, 14m
Length of the Nail, (L) 4 m, 5m, 6 m 20
Diameter of the Nails, (D) 16 mm, 25 mm, 32

10
mm
ANSYS
Spacing of the Nails, (S) 0.4 m, 0.5 m, 0.6 m MATLAB
0
0 2 4 6 8 10 12
Z is the height at which the displacement is measured. Z (m)
It is the height at which the nails are placed. Fig.4 (c) Variation in Displacement for Height
= 12 m
Comparison for Different Heights of the Cut, H
L=5m D = 25 mm S = 0.5 m Fig. 4 (a), Fig. 4 (b) and Fig. 4 (c) show the variation
of displacement with height of the cut for the height
(a) Height of the cut = 8 m of the cut H = 8 m, 10 m and 12 m respectively, using
Height of the Cut, H = 8 m MATLAB and ANSYS. It can be observed that as the
4
height of the cut increases displacement also increases
which is as expected. Further the results obtained
3
from the developed MATLAB code is in very good
Displacement (mm)
agreement with that using ANSYS, therefore verifies

2
the code. It can be observed that the agreement is very
good for H = 12 m.
1
ANSYS Comparison for Different Length of the Nails, L

MATLAB
0 H=8m D = 25 mm S = 0.5 m
0 2 4 6 8 10
Z (m)
(a) Length of the nails L = 4 m
Fig. 4(a) Variation in Displacement for Height = 8
Length of the Nail, L = 3 m
m
3
(b) Height of the cut = 10 m

Displacement (mm)
Height of the Cut, H = 10 m

12
1
10
D isp lacem en t (m m )
ANSYS
8
MATLAB
0
0 2 4 6 8 10
6
Z (m)
4 Fig. 5(a) Variation in Displacement for Length = 3

m
2
ANSYS
MATLAB
(b) Length of the nails L = 4 m
0
0 2 4 6 8 10
Z (m)
Fig.4 (b) Variation in Displacement for Height =
10 m

Length of the Nail, L = 4 m
8 Diameter of the Nail, D = 16 mm
3.5
3
6
Displacement (mm)
Displacement (mm)
2.5
4 2
1.5
2
1
ANSYS
MATLAB 0.5 ANSYS
0
0 2 4 6 8 10 MATLAB
0
Z (m) 0 2 4 6 8 10
Fig. 5(b) Variation in Displacement for Length =
Z (m)
4m
Fig. 6(a) Variation in Displacement for Diameter D
(c) Length of the nails L = 5 m = 16 mm
(b) Diameter of the Nail = 25 mm
Length of the Nail, L= 5 m
4 Diameter of the Nail, D= 25 mm
4
3
Displacement (mm)
3
Displacement (mm)
ANSYS
1
MATLAB
0
ANSYS
0 2 4 6 8 10
MATLAB
Z (m) 0
0 2 4 6 8 10
Fig. 5(c) Variation in Displacement for Length = 5
m Z (m)
Fig. 6(b) Variation in Displacement for Diameter D

Fig. 5 (a), Fig. 5 (b) and Fig. 5 (c) show the variation = 25 mm
of displacement with length of the nails for length of
the nails L = 3 m, 4 m and 4 m respectively, using (c) Diameter of the Nail = 32 mm
MATLAB and ANSYS. It can be observed that as the
length of the nail increases displacement decreases Diameter of the Nail, D = 32 mm
3
which are as expected. Further the results obtained
from the developed MATLAB code is in very good
2.5
agreement with that using ANSYS, therefore verifies
the code. The agreement is excellent for L = 5 m.
Displacement (mm)
Comparison for Different Diameters of the Nails, 1.5

D
H=8m L=5m S = 0.5 m 1
(a) Diameter of the Nail = 16 mm

0.5
ANSYS
MATLAB
0
0 2 4 6 8 10
Z (m)
Fig. 6(c) Variation in Displacement for Diameter D

= 32 mm

Fig. 6(a), Fig. 6(b) and Fig. 6(c) show the variation of
Spacing of the Nails, S = 0.6 m
displacement with diameter of the nails for diameter 4
of nails D = 16 mm, 25 mm and 32 mm respectively,
using MATLAB and ANSYS. It can be observed that
as the diameter of the nail increases displacement 3
decreases which are as expected. Further the results
Displacement (mm)
obtained from the developed MATLAB code is in
very good agreement with that using ANSYS,
2
therefore verifies the code. The agreement is
reasonably good for all diameters.
1
Comparison for Different Spacing of Nails, S. ANSYS

MATLAB
0
H=8m L=5m D = 25 mm
0 2 4 6 8 10
(a) Spacing of the Nails = 0.4 m Z (m)
Spacing of the Nails, S = 0.4 m Fig. 7(c) Variation in Displace

3
-ment for Spacing = 0.6 m
Fig. 7 (a), Fig. 7 (b) and Fig. 7 (c) show the variation
Displacement (mm)
2 of displacement with spacing of the nails for spacing

of the nails S = 0.4 m, 0.5 m and 0.6 m respectively,
using MATLAB and ANSYS. It can be observed that
1
as the spacing of the nail increases displacement also
increases which are as expected. Further the results
ANSYS
obtained from the developed MATLAB code is in
MATLAB very good agreement with that using ANSYS,
0
therefore verifies the code. This agreement is good for
0 2 4 6 8 10
Z (m)
S = 0.6 m.
Fig. 7(a) Variation in Displace CONCLUSIONS

-ment for Spacing = 0.4 m Displacement of the open cut at various heights
obtained from dynamic analysis (MATLAB) match
(b) Spacing of the Nails = 0.5 m very well with those obtained by three dimensional
modeling using ANSYS.
Spacing of the Nails, S = 0.5 m Displacement of the Nailed Cut depends on the soil
4
and Nail parameters. From the present study it can
be concluded that Displacement of the Nailed Cut
3
1. Increases with the increase in height of the
Displacement (mm)
cut.
2. Decreases with the increase in length of the
2 nail.
3. Decreases with the increase in diameter of
the nail.
1 4. Increases with the increase in spacing of the
ANSYS
nail.
MATLAB Among the nail characteristics, length of the nail is
0 the most significant parameter to reduce the
0 2 4 6 8 10
displacement. An increase in length of nail from 3 m
Z (m) to 5 m (for a height H = 8 m) may decrease the
Fig. 7(b) Variation in Displace displacement of the cut by more than 50 percent.
-ment for Spacing = 0.5 m
Future Directions
(c) Spacing of the Nails = 0.6 m In the present analysis only displacement was the
output but some other parameters such as stresses can
also considered.
In the present analysis, only one acceleration of 0.3g
is considered. Hence the effect of different
accelerations can be studied.

Analysis for real earthquake time history can be
carried out.
Damping of soil has been neglected and the same
shall be considered for the analysis.
The soil behavior has been assumed linear but a
nonlinear soil model may be considered.
REFERENCES
Babu G.L.S and Singh V.P., (2008), “Numerical
Analysis of Performance of Soil Nail Walls in
Seismic Conditions”, ISET Journal of
Earthquake Technology Vol. 45, No. 1-2, pp 31
– 40.
“Criteria for Earthquake Resistant Design of
Structure”, (Part – 1), Fifth Revision Bureau of
Indian Standards.
Gosavi M.S., (2006), “ Behavior of Nailed Open
Cuts” – Ph.d Thesis, Department of Civil
Engineering, IIT Roorkee
Juran I., and Elias, V., (1991),”Ground Anchors and
Soil Nails in Retaining Structure”, Chapter-26,
pp 868-905, Foundation Engineering
Handbook, 2.nd Edition.
Lee, K.L and Chan, K, (1972),”Number of equivalent
significant cycles in strong motion
earthquakes”, Proceedings, International
Conference on Microzonation, Seattle,
Washington, vol. II, pp. 609 - 627
Mittal S. and Biswas A.K. (2006),”River Bank
Erosion Control by Soil Nailing” –
Geotechnical and Geological Engineering Vol.
24: pp 1821 - 1833.

ISBN 978-81-921121-0-7
COMPARATIVE STUDY OF OPTIMAL CROPPING PATTERN USING

LINEAR PROGRAMMING AND GENETIC ALGORITHM
Ingle Visha
Post Graduate Student, Dept. of Civil Engg., Govt. College of Engg, Aurangabad-431005, Maharashtra, E-mail:
inglevv@gmail.com
D.G.Regulwar
Faculty, Dept. of Civil Engg., Govt. College of Engg, Aurangabad-431005, Maharashtra,E-mail: dgregulwar@rediffmail.com
ABSTRACT: Multiplicity of cropping systems has been one of the main features of Indian agriculture which is
attributed to rain-fed agriculture and prevailing socio-economic situations of farming community. Due to rapidly
growing population and industrialization, water allocation for irrigation is being diverted for other purposes. In
future more and more water will be required to the cities and industries. In these circumstances, optimum water
allocation will be extremely important to meet ever increasing demands and to irrigate maximum area. Irrigation
planning and management has gained significance due to growing social needs and increasing demands for food
grains while the resources have remained limited. The irrigation management involves optimal cropping pattern
for a given land area and water availabilities with objective function to maximize agricultural economic returns.
This paper presents on the development and comparison of result of LP model using LINGO (Language for
Interactive General Optimization) and Genetic Algorithm (GA).On the concept of distributing the released water
to different crops with their command area for their cultivation are considered according to irrigation policy. The
objective function of this study is to determine the optimal cropping pattern with adequate water supply that will
maximize income. The model is applied to a case study of Jayakwadi reservoir- stage 1, Aurangabad district,
Maharashtra State, India. The optimization model attempts to find out feasible combinations of resources and their
allocation with suitable combinations which maximize agricultural economic return subject to given constraints
expressed as LP model expression/equation. By comparing the results of cropping pattern model using LP and
GA, the optimal cropping pattern is determined.
KEY WORDS: Genetic Algorithm, Linear Programming and Cropping Pattern.
INTRODUCTION be linear. For this purpose, it is very essential to

The utilizable water resources on the earth’s surface linearise the conditions or constraints that may be
cannot be expanded to meet the growing needs for non-linear in nature by making use of an
agriculture, industries and domestic sectors. Especially approximation procedure. Several modified
for developing countries like India, where the formulations based on LP were suggested and their
distribution of the available water resource with respect applications were reported in the literature. Some of
to space and time is quite uneven and also the growth the important applications of LP algorithm and
rate of population is high, the importance of optimal various special forms of LP algorithms such as mixed
utilization and the need for conservation of the integer programming, goal programming, stochastic
available resources cannot be under-estimated. In the linear programming, linear physical programming,
past decades, the methodology of system Analysis has successive LP (SLP) are reviewed.
been introduced and has been widely accepted through Raju and Nagesh Kumar have also discussed on
a number of research works carried out in the planning irrigation planning using genetic algorithms. They
and management of water resources. Gurav and have used GA technique to evolve efficient cropping
Regulwar (2010) have worked on fuzzifying a model to pattern for maximizing benefits for an irrigation
deal with irrigation planning involving uncertainty and project in India. The constraints include continuity
vagueness. Kamodkar and Regulwar (2011) have equation, land and water requirements, crop
worked on cropping pattern for a given land area and diversification and restrictions on storage.
water availabilities with objective to maximize
economic returns using linear programming and Genetic Algorithm
genetic algorithm. Wardlaw and Sharif have presented several alternative
formulations of a genetic algorithm for reservoir
LINEAR PROGRAMMING systems, which were evaluated using the four-
General reservoir, deterministic, finite-horizon problem. The
Linear programming (LP) is applied to water results demonstrate that a genetic algorithm could be
resources problem for various purposes mainly due to satisfactorily used in real time operations with
the simplicity in solution and due to the availability of stochastically generated inflows. A more complex
standard algorithm. The main requirement of the ten-reservoir problem was also considered, and results
formulation of the LP models is that both the produced by a genetic algorithm were compared with
objective functions as well as the constraints should previously published results. The genetic algorithm

approach is robust and is easily applied to complex The fitness function F(x) = f(x) for maximization
systems. It has potential as an alternative to stochastic problem.
dynamic programming approaches.
F(x) = 1 / [1+f(x)] for minimization
Regulwar et al (2002) have discussed the use of problem.
genetic algorithm for optimal reservoir operation for
irrigation. The objective of the study was to maximize Reproduction
the irrigation release. The constraints for this Reproduction, also known as selection operator, is an
optimization were reservoir capacity, hydropower, operator where by an old chromosome is copied into a
and storage continuity. The approach has been applied mating pool according to its fitness value. Highly fit
to Jayakwadi project, Maharashtra is taken as a case chromosomes receive higher number of copies in the
study. next generation. Copying chromosomes according to
their fitness means that the chromosomes with a
Goals of Optimization higher fitness value have higher probability of
The most important goal of optimization is contributing one or more offspring in the next
improvement. Optimization seeks to improve generation. This is this is according to Darwin’s
performance towards some optimal point or points. It evolution theory of survival of the fittest fit; the best
basically aims of seeking improvement to approach ones should survive and create new offspring. The
towards the destination of optimum point itself. While commonly used reproduction operator is the
trying to accomplish optimization, we long for proportionate reproduction operator, where a string is
perfection to describe and attain what is ‘best’ by selected for the mating pool with a probability
knowing how to measure and alter what is good or proportional to its fitness. In this process, each
bad. Optimization theory encompasses the individual string is assigned a probability of being
quantitative study of optima and methods for finding selected for copying as:
them. k
f i   fi
Difference between GA and Traditional Methods i 1
By abstracting nature’s adaptation algorithm of choice Where,
in artificial form, GA differs from conventional fi = The fitness or objective function value of the ith
optimization and search techniques in four ways – individual
1. GA work with a coding of the parameter set, k = Population size.
not the parameter themselves. The following selection methods are described by
2. GA’s search from a population of points, not Goldberg:
a single point. 1. Deterministic sampling.
3. GA’s use payoff (objective function) 2. Remainder stochastic sampling without
information, not derivatives or other auxiliary replacement.
knowledge. 3. Stochastic sampling without replacement.
4. GA’s use probabilistic transition rules and 4. Remainder stochastic sampling with
stochastic operators, not deterministic rules. replacement.
Table 1a : Comparison of Natural and GA 5. Stochastic sampling with replacement.
Terminology 6. Stochastic tournament (Wetzel ranking).
Natural GA Perhaps the easiest is to create a biased Roulette
Chromosome String Wheel where each current string in the population has
Gene Feature, character or a roulette wheel slot sized in proportion to its fitness.
Allele detector e.g.: Suppose the sample population of 4 strings has
Locus Feature value objective or fitness function values ‘f’ as shown in
Genotype String position table.
Phenotype \structure
Epistasis Parameter set Table 1b Strings and Fitness Values
Non linearity. No. String (x) Fitness Percentage of
(x2) total
Phases of Genetic Algorithm 1 01101 = 13 169 14.4
Genetic algorithm consists of three phases: 2 11000 = 24 576 49.2
1. Initialization. 3 01000 = 8 64 5.5
2. Evaluation. 4 10011 = 19 361 30.9
3. Genetic Operation. Total 1170.00 100.00
If F(x) is the fitness function and f(x) is the objective The percentage of population and total fitness is
value function, then- calculated in above table. The approximated weighted
roulette wheel for this generation’s reproduction is
shown in figure below.

Flip bit mutation
A mutation operator that simply inverts the value of
the chosen gene (0 goes to 1 and 1 goes to 0) can only
1 01101 = 13 be used for binary genes.
2 11000 = 24
Example: -
3 01000 = 8
Original String: 1 0 1 1 0 0 1
4 10011 = 19
Assume mutation site: 4
String after mutation: 1 0 1 0 0 0 1
Boundary mutation
Fig. 1 Roulette wheel.
A mutation operator that replaces the value of the
Crossover
chosen gene with either the upper or lower bound for
1. Simple or Single Point Crossover.
that gene (chosen randomly) can only be used for
2. Multipoint Crossover.
integer and float genes.
3. Uniform Crossover.
Uniform mutation
Single point crossover
A mutation operator that replaces the value of the
In this method crossover is carried out at a single
chosen gene with a uniform random value selected
point. This is illustrated in the following example.
between the user-specified upper and lower bounds
Let Par1 and Par2 be the two parents selected for
for that gene. This mutation operator can only be used
crossover. Assume the strings Par1 and Par2 be the
for integer and float genes. It permits the value of a
two parents selected for crossover. Assume the strings
gene to be mutated randomly within its feasible range
Par1 and Par2 as below:-
of values, possibly resulting in significant
Par1 : 1 1 0 0 | 0 1 0 1
modification of otherwise good solutions. There are
Par2 : 1 0 1 1 | 0 1 1 1
two types of uniform mutations. Modified uniform
Now, a crossover site is selected randomly as an
mutation and Random uniform mutation.
integer between 1 and string length. Here string length
is taken as 8, let the crossover site be 4.
Modified uniform mutation
Then children Child1 and Child2 are generated as
Modified uniform mutation permits modification of a
below:
gene by a specified amount, which may be either
Child1 : 1 1 0 0 0 1 1 1
positive or negative.
Child2 : 1 0 1 1 0 1 0 1
Random uniform mutation
Multi point crossover
In random uniform mutation a small proportion of the
This method is similar to Single Point Crossover
genes from the parent population are randomly
except that more than one crossover site is randomly
selected and replaced with other genes also generated
selected and contents of Child1 and Child2 are
at random. This step ensures that all possible genes
selected alternatively from Par1 and Par2 with change
remain available for the construction of better
from one parent to another at the crossover sites.
solutions, even when the evolving population drives
some potentially useful blocks to extinction through
Uniform crossover
the selection mechanism. Random uniform mutation
In this method, crossover is performed over the entire
is neutral in the sense that it never takes advantage of
length of string of bits. For this a “mask” is generated
the current position of the parent population in the
randomly. This “mask” is a string of bits of value ‘0’
search space nor is it influenced by the effect that
or ‘1’ and sizes same as the chromosome string
previous mutations have had in the value of the
length. With the information in the mask we can
objective function.
generate the new population as below:
Par1 : 1 1 0 0 1 0 1 1
Par2 : 0 1 0 0 0 1 0 0 Non-uniform mutation
A non-uniform mutation is a mutation operator that
Mask : 0 0 1 0 1 1 0 1
increases the probability that the amount of the
Child 1: 1 1 0 0 0 1 1 0
mutation will be close to 0 as the generation number
(If mask = 0, Child 1 = Par1 and Child2 = Par2)
increases. This mutation operator keeps the
Child2 : 0 1 0 0 1 0 0 1
population from stagnating in the early stages of the
(If mask = 1, Child 1 = Par2 and Child2 = Par1)
evolution then allows the genetic algorithm to fine
Mutation
tune the solution in the later stages of evolution. This
1. Flip Bit Mutation.
mutation operator can only be used for integer and
2. Boundary Mutation.
float genes. In non-uniform mutation, the amount by
3. Uniform Mutation.
which genes are mutated can also be reduced as a run
4. Non-Uniform Mutation.
progresses, and can therefore help in the later
5. Gaussian Mutation.
generations to fine tune the solutions. This operator is

particularly suited to problems where high precision is  Mutation is made to prevent falling GA into
required. local extreme, but is should not occur very often,
because the GA will change to random search.
Gaussian mutation
A Gaussian mutation is a mutation operator that adds Population size
a unit Gaussian distributed random value to the Population size says how many chromosomes are in
chosen gene. The new gene value is clipped if it falls population (in one generation). If there are too few
outside of the user-specified lower or upper bounds chromosomes, GA has a few possibilities to perform
for that gene. This mutation operator can only be used crossover and only a small part of search space is
for integer and float genes. It also ensures that the explored. If there are too many chromosomes, GA
search space does not remain limited to that defined slows down.
by the parent population, but it does so in a less
disruptive manner than the random version. In this
Start
procedure, a number of parent individuals are selected
and small changes are performed in all of their
components so that the resulting chromosome is
located somewhere within a small neighborhood of its Initial Population
parents. The alterations performed to the genes are
normally distributed with mean zero and a standard
deviation, appropriate for the problem at hand. In this Selection
way, every gene could have an independent
distribution allowing the user to define the appropriate
magnitude of the change for each variable involved. Mating
This mutation procedure places the offspring in the
hyper-ellipsoids of constant probability defined by the
multivariate normal distribution specified, allowing Crossover
the individuals to align and take steps in directions
that, hopefully, maximize improvement.
Mutation
Parameters of GA
There are three basic parameters of GA.
1. Crossover probability. No
2. Mutation probability. Terminat
3. Population size.
Yes
Crossover probability
Stop
Crossover probability says how often will be
crossover performed. If there is no crossover,
offspring is exact copy of parents. If there is a Fig. 2 Algorithm for GA Technique
crossover, offspring is made from parts of parent’s
chromosome. Methodology
 If crossover probability is 100%, then all The methodology discussed here is Genetic Algorithm. It
offspring are made by crossover. is a search procedure based on the mechanics of natural
 If it is 0%, whole generation is made from selection and natural genetics, which combines an
exact copies of chromosomes from old population. artificial survival of the fittest among string structures
Crossover is made in hope that new generation will with structured yet randomized information exchanged
have good parts of old chromosomes & may be the with genetic operators abstracted from nature with some
chromosomes will be better. However, it is good to innovative flair of human search. The operation of genetic
leave some part of population survive to next algorithms is straightforward. Genetic algorithms use
generation. random choice to guide highly exploitative search. They
operate on populations of strings, with the string coded to
Mutation probability represent some underlying parameter set. Reproduction,
Mutation probability says how often will be parts of crossover and mutation are applied to successive string
chromosome mutated. If there is no mutation, populations. These operators involve random number
offspring is taken after crossover without any change. generation, string copying, and partial string exchange.
If mutation performed, part of chromosome is
changed. The success and performance of genetic algorithms are
 If mutation probability is 100% whole dependent on population size, number of generation, and
chromosome is chromosome is changed, if it is 0%, the probabilities of crossover and mutation. Goldberg
nothing is changed. suggested that good GA performance require the choice

of high crossover probability, a low mutation probability
(inversely proportional to the population size) and a
moderate population size. Populations of 50-100 Need for
individuals taken through 10-20 generations have a high optimization
probability of creating optimal or near-optimal solutions.
The mutation operator is applied to the entire population
on a bit by bit basis. Choose design
variables
The GA is composed of the following steps:
1. Randomly generate an initial population of
binary strings representing possible designs. Formulate
2. Decode the binary strings to obtain decimal- constraints
valued decision variables for each design in the
population. Using a simulation model, calculate the
fitness of each member of the current generation Formulate objective
based on the objective function value and possible function
constraint violations. In the process of evaluating the
fitness of an individual design the constraints of the
management model are examined. The amount of any Set up variable
constraint violation is multiplied by a weight factor bounds
and added to the fitness value as a penalty:
p
fi  f ( x)    δj(j )2 Choose an
j 1
optimization
Where, algorithm
fi is the fitness value and f(x) is the objective function
value, p is the total number of constraints,  is – 1 for
maximization and +1 for minimization problem, δj is Obtain solution
the penalty coefficient or weight factor and Øj is
amount of violation of the jth constraint.
3. Starting with the current population of
designs, produce subsequent generations in three steps Fig. 3 Optimal design formulation process
using the genetic operations of reproduction,
crossover, and mutation applied to current generation. The first step is to realize the need for using
4. Repeat steps 2 and 3 until convergence or the optimization in a specific design problem. Thereafter,
maximum number of generations is exceeded. the designer needs to choose the important design
variables associated with the design problem. The
OPTIMAL PROBLEM FORMULATION formulation of optimal design problems involves
Optimization procedure must only be used in those other considerations, such as constraints, objective
problems where there is definite need of achieving a function. As shown in the figure, there is usually a
quality product or a competitive product. It is hierarchy in the optimal design process; although one
expected that the design solution obtained through an consideration may get influenced by the other.
optimization procedure is better than other solutions
in terms of the chosen objective i.e. cost, efficiency, SYSTEM DESCRIPTION
safety etc. it is impossible to apply a single The Jayakwadi project built across river Godavari in
formulation procedure for all engineering design Aurangabad district of Maharashtra state. The dam is
problems. The purpose of the formulation procedure situated at Latitude 180 46’ N’ Longitude 75020 E,
is to create a mathematical model of the optional Jayakwadi village, Taluka Paithhan district
design problem, which then can be solved using an Aurangabad. The catchment area of the reservoir upto
optimization algorithm. The following figure shows dam sit is 21774 km2. The length of the dam is 10280
an outline of the steps usually involved in an m and its maximum height is 37.73 m. The gross
optimal design formulation process. storage of reservoir is 2909 Mm3and live storage is
2170 Mm3. Total installed capacity for power
generation is 12.0 MW (Pumped storage plant).
Irrigable command area is 1,41,640 Ha Area of
irrigation in Kharif is 22%, in Rabi 45% Two seasonal
28%, Hot weather 3% and Perennial crops 4.5% Map
of Maharashtra State showing Jayakwadi project is
shown in map.

Season wise :
Maximum sowing area constraints:
Kharif
KC PC
k
 A   A  CA
i i i  CAip
Where,
CAk i = Comand area for kharif season for ith crop (ha)
CAp i = Command area for perennial season for ith crop
(ha)
Rabi
RC PC
Fig. 4 Map of Maharashtra
 Ai   Ai  CAiR  CAiP
Formulation of the Problem
Maximum agricultural return CARi = Command area for Rabi season for ith crop
(ha)
KC RC
HWS Perennial
AER =  A .NBC   A .NBC
i i i i
PC HWC
PC TSC P
 A   A  CA  CAiHW
+  A .NBC   A .NBC
i i i i
i i i
HWC
+  A .NBC i i
CAiHW = Command area for hot weather season for ith
crop (ha)
AER = Agricultural Economic Return (Million Rs)
NBCi = BCi – ICI Affinity constraint :
i = Crop Index (Upper limits using the existing cropping pattern)
x1 = Sugarcane (P) Perennial,
x2 = Banana (P)
x3 = Chillies (TS) A1PC  CAiP
x4 = L.S. Cotton (TS) A2PC  CAiP
x5 = Sorghum (K)
x6 = Paddy (K) Two seasonal,
x7 = Sorghum (R)
x8 = Wheat (R) A3TSC  CAiTC
x9 = Gram (R) TSC
A4  CAiTC
x10= Ground Nut (HW)
Where,
KC = Kharif Crop
A1 = Area under crop No. 1 (Perennial Crop)
RC = Rabi Crop
A2 = Area under crop No. 2 (Perennial Crop)
PC = Perennial Crop
A3 = Area under crop No. 3 (TSC)
TSC = Two Seasonal Crop
A4 = Area under crop No. 4 (TSC)
HWC = Hot Weather Crop
Ai = Area or ith Crop, (Ha)
Kharif,
NBCi = Net Benefit Co-efficient for ith Crop,
IC i = Input Cost for ith Crop.
A5KC  CAiK
Subjected to Constraints :
A6KC  CAiK
Total Sowing Area Constraints :
Rabi,
KC RC PC TSC HW
A7RC  CAiR
 A   A   A   A   A  CA
i i i i i A8RC  CAiR
A9RC  CAiR
Where, CA = Total command area for all sector for all
crops (Ha) Hot weather,

A10 HWC  CAiHW 3. Water Availability Constraint
OPTIMIZATION MODEL 3.16 X1 +3.49 X2 + 0.399 X3 +0.456 X4 +0.099 X5

To work out the optimal cropping pattern; the + 0.65 X6 + 0.526 X7 +0.99 X8 + 0.498 X9+ 1.592
following cropping pattern models are developed and X10 < 67000;
solved using LINGO. In this cropping pattern model
the increase in the yield due to the sufficient water to The above cropping pattern model is developed for
crop is taken into consideration using Genetic 100% water availability.
Algorithm model. In the model i.e. (Cropping pattern Water availability constraint:
model) the variables X1…X10 indicates the various
crops according to varying water availability. KC RC PC
 Ai .IWRi   Ai .IWRi   Ai .IWRi

Table 2 Total Water Amount and Profit Values for
All Plants TSC HWC
Varia Plant and Season Total   Ai .IWRi   A .IWR
i i  TWA
ble Water amount Water
(%) Amount
(mm) Non-Negativity Constraints:
X1 Sugarcane (100) Perennial 3160
X2 Banana (100) Perennial 3492 X1,X2,X3,X4,X5,X6,X7,X8,X9,X10  0
X3 Chillies (100) Two Season 399
X4 LSCotton(100) Two Season 456 Performance Analysis
X5 Sorgum (100) Kharif 99 The objective of this study is to maximize the revenue
X6 Paddy (100) Kharif 650 by considering adequate supply of water to a specific
area that has to be cultivated considering Linear
X7 Sorgum (100) Rabi 526
Programming model, from the available constraints
X8 Wheat (100) Rabi 990
i.e.; available land and water for irrigation. Form
X9 Gram (100) Rabi 498 optimization of cropping pattern the optimal cropping
X10 Groundnut (100) Hot Weather 1592 pattern is obtained.
Cropping Pattern Model Analytical Analysis for Optimal Cropping Pattern
Objective Function using LINGO for LP model
The cropping pattern is solved considering
Maximize=AER; agricultural economic returns according to linear
programming considering 100% water availability.
AER=66820X1+115200X2+128000X3+1420X4+574 Results are shown in Table. The maximized revenue
0X5+7200X6+4620X7+11770X8+7020X9+10320X1 obtained is Rs.1683478.00 from the project.
0;
Table 3 Considering the Water Availability Model,
Subjected to Jayakwadi Stage I
1. Total sowing area constraint Sr. Varia Crop Area in Ha (for
No. ble water
X1 + X2+ X3 + X4 + X5 + X6 + X7 + X8 + X9 X10 Availability
<141640 100%)
1 X1 Sugarcane 4249.2
2. Maximum sowing area constraint
2 X2 Banana 2124.6
3 X3 Chillies 4249.2
X1 + X2+ X3 + X4 + X5 + X6 < 77193.8
X1 + X2+ X3 + X4 + X7 + X8 + X9 < 109771 4 X4 L.S.Cotton 0
X1 + X2+ X10 < 10623 5 X5 Sorgum(K) 16996.8
X1 < 4249.2 6 X6 Paddy (K) 6456.65
X2 < 2124.6 7 X7 Sorgum(R) 0
X3 < 4249.2 8 X8 Wheat (R) 35410.00
X4 < 35410 9 X9 Gram (R) 7082.00
X5 < 16996.8 10 X10 Groundnut 0
X6 <14164 (HW)
X7 <21246 Net cropped Area (Ha) 76568.450
X8 <35410 Net Revenue (Rs) 1683478.00
X9 <7082 Water Used (Ha.m) 67000
X10 < 4249.2

4 X4
L.S.Cotto 0 4251.2
n
5 X5 Sorgum(K 16996.8 2767.2
)
6 X6 Paddy (K) 6456.65 2128.8
Area (Ha)
7 X7 Sorgum(R 0 5.2500
)
8 X8 Wheat (R) 35410.00 35410.00
9 X9 Gram (R) 7082.00 4251.5
10 X10 Groundnut 0 4449.2
(HW)
Crop Net cropped Area (Ha) 76568.450 63886.15
Net Revenue (Rs) 1683478.00 154070.00
Graph 1 Optimal Cropping Pattern using LP Model Water Used (Ha.m) 67000 67000
Table 4 Computational Analysis for Optimal
Cropping Pattern using GA
Sr. Variable Crop Area in Ha
No. (for water
Availability
100%)
1 X1 Sugarcane 4249.2
2 X2 Banana 2124.6
3 X3 Chillies 4249.2 Area (Ha)
4 X4 L.S.Cotton 4251.2
5 X5 Sorgum(K) 2767.2
6 X6 Paddy (K) 2128.8
7 X7 Sorgum(R) 5.2500
8 X8 Wheat (R) 35410.00 Crop
9 X9 Gram (R) 4251.5
10 X10 Groundnut (HW) 4449.2 Graph 3 Comparison of Optimal Cropping Pattern
Net cropped Area (Ha) 63886.15 Using LP and GA
Net Revenue (Rs) 1540700.00 JUSTIFICATION FOR ERRORS
Water Used (Ha.m) 67000 It is observed that solution obtained buy LP and GA
are reasonable close. This amount of percent
difference is justified since; the linear programming
technique is suitable for linear programming problem
formulation as it is deterministic in nature. For the
model developed for optimal cropping pattern, linear
programming is used. In case of linear programming
problem, the search for optimum values start forum
Area (Ha)
single point and in a particular direction. As it is

deterministic in nature, the global optimum is found.
The genetic algorithm is search technique based on

the Darwin’s principle i.e. survival of fittest fit. In
Crop GA, the search for optimization value start from more
number of points (single point in case of LP).
Graph 2 Optimal Cropping Pattern using GA Model Therefore we can get more number of feasible
solutions for the same mathematical model developed
Table 5 Comparison of Optimal Cropping Pattern (Number of solution depends on population size). The
Using LP and GA (X1…X10) GA is suitable for complex, nonlinear problems. Still,
Sr. Variable Crop Area in Ha Area in Ha for comparison of solution, linear programming is
No. (for water (for water compared with GA. As GA is search technique,
Availabilit Availability suitable for nonlinear problem, gives more number of
y 100%) 100%) Using feasible solutions for the same problem, (near global
Using LP GA optimum).
1 X1 Sugarcane 4249.2 4249.2
2 X2 Banana 2124.6 2124.6
3 X3 Chillies 4249.2 4249.2

CONCLUSIONS Accepted: 18 November 2010, Springer Science
 The optimized revenue and optimized + Business Media B.V. 2010.
cropping pattern is decided from the solution for eight Goldberg, D.E., “Genetic Algorithms is Search,
different cases by varying the water availability. Optimization and Machine Learning”. 1989,
Under the optimal cropping pattern by applying the Addison Wesley, Reading Mass, pp1-50, 245-260.
water stress to crop the values of revenue are 1663907 Jyothirmai, U.L., Shareef, S.M., and Raju, V.T.
rupees and cropped area is 141640 Ha for water “Optimal Farm Plans for Increasing income
availability. These results of the study shows that Water use and Employment in Three Regions of
even with very low water supply, it is possible to keep Nagarjunasagar Command Area in Andhra
the farm revenue at high levels and the land to be Pradesh”, 4th International R & D Conference on
irrigated also more. Water and Energy ofr 21st Century, CBIP, 28-31
 The liner programming approaches can be January 2003, Aurangabad, Maharashtra, India.,
successfully applied to determine the optimal Vol-II, pp.592-604.
cropping pattern and to derive the agricultural returns Kamodkar and Regulwar, “Comparison between
for irrigation using LINGO. This methodology is Linear Programming and Genetic Algorithm for
demonstrated through application to Jayakwadi Cropping Pattern”, National conference paper
project, in Aurangabad district. presentation EPM 2011.
 With the emergent water problems in the Regulwar, D.G. Ghangrekar, M.M., and Anand Raj
world, countries prefer the methods for optimum P,. “Optimal Reservoir Operation for Irrigation
utilization from their water potentials. Obviously, using Genetic Algorithms”. Proc. International
provision of agricultural irrigation water to plant Conference on advances in civil Engineering,
according to their requirements will ensure maximum ACE-2002, IIT-Khargpur, January 2002 , pp.259-
production per unit area. However, unnecessary 265.
irrigation by farmers in adequate water conditions
leads to soil salivation and an increase in water table.
 GA is found to be an effective optimization
tool for irrigation planning and the results can be
utilized for efficient planning of any irrigation system.
SCOPE FOR FUTURE WORK

LINGO programming model and GA model is
developed for optimization of agricultural economic
returns for single objective i.e. releases either for
irrigation. However other objectives like optimization
for water supply, navigation purpose flood control etc.
can be considered and these objectives can be
fuzzified to develop Multiobjective fuzzy Linear
Programming (MOFL) Model.
APPLICATIONS
In the present study, a LINGO programming model is
developed to find optimal cropping pattern by
developing logic in case of water availability for a
season. However for better results a water yield
relation of plants, field capacity an intra-seasonal
irrigation i.e. (irrigation during different growth stages
of crop) approach can be applied for an exact values
of agricultural economic returns factor in actual farm
of a suitable size.
ACKNOWLEDGMENT
The authors are thankful to Command Area
Development Authority, Aurangabad, Maharashtra
State, India for providing necessary data for the
analysis.
REFERENCES
Gurav and Regulwar, “Irrigation Planning Under
Uncertainty—A Multi Objective Fuzzy Linear
Programming Approach”, 28 March 2010 /

ISBN 978-81-921121-0-7
A CRITICAL EVALUATION OF FOLLOWING PHENOMENA IN INDIAN

MIXED TRAFFIC CONDITIONS
K.V.R. Ravi Shankar
Associate Professor, Department of Civil Engineering, GMR Institute of Technology
Rajam -532127, Andhra Pradesh, INDIA, e-mail: kvrrshankar@gmail.com
Tom V. Mathew
Associate Professor, Department of Civil Engineering, Indian Institute of Technology Bombay
Powai, Mumbai-400 076- INDIA, e-mail: vmtom@civil.iitb.ac.in
ABSTRACT: Car-following behaviour is one of the building blocks of traffic simulation models. Several studies
analysed following behaviour leading to the development of distinct models. However, most of them are confined to
following behaviour of similar vehicle-types. A study on vehicle-type dependent following behaviour is required for
accurate modelling of mixed traffic conditions where the static and dynamic characteristics of vehicles differ
significantly. In this paper, vehicle-type dependent following behaviour is analysed using data collected for different
vehicle-type combinations. The results point to the need for considering vehicle-type behaviour particularly in mixed
traffic conditions. Future simulation models should adopt the vehicle-type dependent behaviour for a realistic
representation of mixed traffic conditions.
INTRODUCTION
Modeling the behaviour of one vehicle following another
vehicle is the fundamental concept of many of the
existing traffic simulation tools. Several research studies
have been conducted and theories were developed to
represent this behaviour (Gazis et al., 1961; Yang and
Koutsopoulos, 1996; Gipps, 1981; Treiber et al., 2000;
Wiedemann, 1974; Wiedemann and Reiter, 1992;
Fritzsche, 1994; Bando et al., 1995; Chakroborty and
Kikuchi, 1999). However, the following behaviour in
mixed traffic conditions poses a new challenge and the
recent studies proved the vehicle-type effect on following Fig. 1 Vehicle-type dependent following behaviour in
behaviour. The manoeuvrability of vehicles and the sight mixed traffic conditions
distance of the drivers’ effect the driving behaviour
largely in mixed traffic conditions. For example, driver
DATA COLLECTION
behaviour may be different in the case of bus following a
For analysing the vehicle-type dependent behaviour, nine
passenger car compared to passenger car following a bus
lead-following vehicle-type combinations were
(as shown in Figure 1). The static and dynamic
considered for data collection. Two major urban arterials
characteristics of these vehicles differ significantly and
in Mumbai, India, are selected for data collection. A 4.6
influence the driver behaviour. Some of the recent studies
km long straight section on Eastern express highway and
also point to the effect of vehicle-type on driver following
5.9 km stretch on Western express highway in Mumbai
behaviour (Punzo and Tripodi, 2007; Ye and Zhang,
are selected. Even though, these stretches are referred as
2009; Brackstone et al., 2009). There is a need to
express highway, they are a part of the main arterials of
realistically represent the mixed traffic flow behaviour. In
Mumbai city and comprise of mixed traffic. A snapshot of
spite of the several simulation models being used, there is
the eastern express highway stretch is shown in Figure 2.
no common consensus as to which model or simulation
These sections are eight lane divided urban highways with
performs better under given traffic conditions. The
typical traffic volumes ranging from 833 to 1066
driving behaviour also changes from region to region and
vehicles/hr/lane. These stretches are continuous sections
in countries like India, the diversity in vehicle and driving
with no major intersections and represent a typical urban
characteristics will have a considerable influence. Hence
traffic. Data was collected using vehicles equipped with
there is a need to study the vehicle-type dependent
GPS receivers and the procedure is described below.
behaviour and suggest models which can represent the
Experiments were conducted with nine combinations of
vehicle-following behaviour rather than the conventional
vehicles comprising of passenger car (length 4.7 m), three
car-following models.
wheeled auto-rickshaw (length 2.6 m) and bus (length 9.4
m). Data was collected using GPS (Global Positioning
System) equipped vehicle.

is more prominent in the case of a bus following a
passenger car (C-B). Thus vehicle-type also has a
considerable influence on following behaviour. Small
sized vehicles were observed to follow the lead vehicle at
a lesser headway compared to large vehicles. The higher
manoeuvrability of small sized vehicles allows the drivers
to maintain lesser safety distance.
The relative speed and following distance plot for the nine
vehicle-type combinations considering the entire data set
is shown in Figure 4. The following distance values on y-
axis represent the distances between lead and following
vehicles for a particular vehicle-type combination. It can
be seen from this figure that a considerable difference
exists in the following distance between vehicles for
various vehicle-type combinations. The points are
Fig. 2 A snapshot of the eastern express highway section concentrated in a narrow range for the A-A combination
and widely spread for passenger car following passenger
With the help of beacon receiver, the real time car (C-C) and bus following passenger (C-B)
differentially corrected GPS data was recorded. Data was combinations.
obtained from a series of experiments carried out along
the selected stretches under real traffic conditions from CONCLUSION
October, 2007 to March, 2008. As the car-following Traffic flow modeling is one of the much sought after
behaviour may change in rainy conditions and at night studies in transportation engineering and has wider
time due to less visibility, experiments were conducted on applications in capacity analysis, driver behaviour
sunny days and during peak and off-peak time. assistance systems etc. In the past decades, considerable
Preliminary surveys were conducted for assessing the research has been carried out to understand and model the
satellite availability and traffic characteristics. Leading driving behaviour in terms of certain mathematical
vehicle driver was asked to follow the traffic stream and relationships. However, a realistic modeling of driving
the following vehicle driver to follow the lead vehicle behaviour poses a challenge to the research community,
without overtaking. Drivers selected for the data particularly in mixed traffic conditions. Car-following
collection are regular drivers and were not aware of the behaviour was analysed for different lead-following
objectives of the study. They were in the age group of 25 vehicle-type combinations. Extensive field data was
to 30 years. Data sets containing lane changing or collected using GPS receivers for nine vehicle-type
extrusion of vehicles between lead and following vehicles combinations comprising passenger cars, three wheeled
were excluded. auto-rickshaws, and buses from two major arterials of
Mumbai city in India. This study addresses the influence
VEHICLE-TYPE DEPENDENT BEHAVIOUR of the type of vehicle on the behaviour of mixed traffic.
The data was analysed to see how the vehicle-type This study established the existence of distinct car-
influences the following behaviour. For this, the space following behaviour for various lead and following
headway i.e. distance between the lead and following vehicle-type combinations.
vehicles less vehicle length is determined from the
latitude and longitude values obtained from the GPS
instruments. As the following behaviour changes from REFERENCES
driver to driver, data was collected using different drivers Bando, M., Hasebe, K., Nakayama, A., Shibata, A. and
for each of the lead-following vehicle-type combination. Sugiyama, Y. (1995), “Dynamical model of traffic
Since the following-distance is a function of speed, congestion and numerical simulation”, Phys. Rev. E,
analysis was conducted separately for various speed Vol. 51, American Physical Society, pp. 1035–1042.
categories (multiples of 5 m/s). The cumulative frequency Brackstone, M., Waterson, B. and McDonald, M. (2009),
distribution of following distance for various drivers and “Determinants of following headway in congested
the vehicle-type combinations is shown in Figure 3. It can traffic”, Transportation Research Part F: Traffic
be seen from this figure that the space headway Psychology and Behaviour, Vol. 12, pp. 131–142.
distribution varies significantly with the vehicle-type. Chakroborty, P. and Kikuchi, S. (1999), “Evaluation of
Moreover, the driver variability is less significant than the the general motors based car-following models and a
vehicle-type variability for almost all speed ranges. Thus proposed fuzzy inference model”, Transportation
vehicle heterogeneity has a significant effect on car Research Part C: Emerging Technologies, Vol. 7, pp.
following behaviour in addition to driver heterogeneity. 209–235.
This result is also consistent with a recent study showing Fritzsche, H. T. (1994), “A model for traffic simulation”,
that following behaviour is affected by the type of lead Transportation Engininering Contribution, Vol. 5, pp.
vehicle (Brackstone et al., 2009). However, the variation 317–321.

Gazis, D. C., Herman, R. and Rothery, R. W. (1961),
“Nonlinear follow-the-leader models of traffic flow”,
Operations Research, Vol. 9, pp. 545–567.
Gipps, P. G. (1981), “Behavioural car-following model
for computer simulation.”, Transportation Research,
Part B: Methodological, Vol. 15 B, pp. 105–111.
Punzo, V. and Tripodi, A. (2007), “Steady-state solutions
and multiclass calibration of gipps microscopic traffic
flow model”, Transportation Research Record 1999,
Journal of the Transportation Research Board,
Washington D.C., pp. 104–114.
Treiber, M., Hennecke, A. and Helbing, D. (2000),
“Congested traffic states in empirical observations and
microscopic simulations”, Phys. Rev. E, Vol. 62,
American Physical Society, pp. 1805–1824.
Wiedemann, R. and Reiter, U. (1992), Microscopic traffic
simulation: The simulation system mission,
background and actual state, final report, Technical
report, Brussels.
Yang, Q. and Koutsopoulos, H. N. (1996), “A
microscopic traffic simulator for evaluation of dynamic
traffic management systems”, Transportation Research
Part C: Emerging Technologies , Vol. 4, pp. 113–129.
Ye, F. and Zhang, Y. (2009), Vehicle-type-specific
headway analysis using freeway traffic data, presented
at the 88th annual meeting of the transportation Fig. 3. Space headway profiles for the nine vehicle-type
research board, number Report no.:09-2997, combinations and different drivers
Washington D.C.
Fig. 4. Typical following distance vs relative speed

profiles for various vehicle-type combinations

ISBN 978-81-921121-0-7
FEM FORMULATION OF DIRECT METHOD OF CONCRETE GRAVITY

DAM - NONLINEAR FOUNDATION INTERACTION
A Burman
Asst. Professor, Dept. of Civil Engineering, Birla Institute of Technology, Mesra, India
D Maity
Assoc. Professor, Dept. of Civil Engineering, Indian Institute of Technology, Kharagpur, India
A Kumar
UG Student, Dept. of Civil Engineering, Birla Institute of Technology, Mesra, India
S Bhushan
S Sankrit
INTRODUCTION: Considerable research has been conducted in the area of dynamic interaction analysis of
structure-foundation coupled system. The structure is considered to be linear, isotropic and the foundation is
assumed to follow elasto-plastic yield behavior. In the present work, the free field response is obtained first for the
foundation domain subjected to ground motion. The free field responses are used to determine the soil-structure
interaction forces which are exerted on the coupled soil-structure system along with the externally applied
earthquake excitations. The responses thus found are added to the free-field responses to calculate the total
displacements, velocities and accelerations of the system considering the effects of SSI. The algorithm presented
here is robust in terms of runtime as well as memory allocation considerations.
THEORITICAL FORMULATION
Modeling of Dam The Drucker-Prager Yield Criterion
In the present work, Koyna concrete gravity dam has
been chosen as the structure. The dam body is
considered to be homogenous and isotropic. For plane
 
strain problems, the constitutive matrix D expressing
the relationship between stress   and strains  for
the dam monolith is:
   D   (1)
where,
   Fig. 1 Geometric representation of the Mohr-
 1 1 
0 
  (2) Coulomb and Drucker-Prager yield surfaces in
D  E 1     1 0  principal stress space
1  1  2  1  
 1  2 
 0 0
21  
An approximation to the Drucker-Prager law was
 presented by Drucker and Prager (1952) as a
Here E and  are the Young’s modulus and the modification of the Von Mises yield criterion. The
influence of a hydrostatic stress component on
Poisson’s ratio of the material of the dam body
yielding was introduced by inclusion of an additional
respectively.
term in the Von Mises expression to give
Modeling of Foundation 1
In the present work, the foundation material is assumed J 1   J 2 '  2  k ' (3)
to be of rock in nature. A nonlinear constitutive model
This yield surface has the form of a circular cone. In
has been chosen to simulate the stress vs. strain
order to make the Drucker-Prager circle coincide
behavior of soil/rock which is nonlinear in nature. The
with the outer apices of the Mohr-Coulomb
framework of elasto-plasticity has been chosen to
hexagonat any section, it can be shown that (as in
represent the foundation material properties. A popular
Fig. 1):
and widely accepted yield criterion for geo-materials
namely Drucker-Prager criterion has been adopted for
the purpose of analysis.

K cc  K cc  K cc
2 sin  6c cos  s  f 
 ,k'  (4) (7)
33  sin   33  sin   By solving Eq. (6) using Newmark’s integration
Coincidence with the inner apices of the Mohr- method, one would obtain the absolute
Coulomb hexagon is provided by: displacements, velocities and accelerations of the
2 sin  , 6c cos  (5) coupled SSI problem. The free-field solution is
a k' 
33  sin   33  sin   designated by the free-field displacements v,
velocities v . Here, u g
However, the approximation given by either the inner and accelerations v is the
or outer cone to the true failure surface can be poor for ground acceleration vector. By a simple change of
certain stress combinations. variables, it becomes possible to express the absolute
Solution Scheme for Coupled Dam-Foundation
System displacements U , velocities U and accelerations U 
In dam-foundation interaction problems, the foundation in terms of displacements u, relative to the free field
and the structure do not vibrate as separate systems displacements v. Or,
under external excitations, rather they act together in a U S   vS   uS 
coupled way. The most common SSI approach used is       
 U c    vc    uc 
based on the "added motion" formulation. To develop U f   v f   u f 
the fundamental SSI dynamic equilibrium equations,      
the soil-structure system, as shown in Fig. 2 is U S   v S   u S 
considered.       
U c    vc    u c 
U f   v f  u f 
     
U S   v S   u S 
     
U c    v c    u c  (8)
U f   v f  u f 
     
The above equation can be written as
 M ss M sc 0   us  Css C sc 0   us 
     
 M cs M cc M cf   uc   Ccs Ccc Ccf   uc 
 0
 M fc M ff  uf   0 C fc C ff  u f  (9)
Fig. 2 Soil- structure interaction model
 K ss K sc 0   us 
The SSI model here is divided into three sets of node   
  K cs K cc K cf   uc   R  F
points, viz. the common nodes at the interface of the
structure and soil are identified with the subscript “c”;
 0
 K fc K ff  u f 
the nodes within the structure are with “s” and the where:
nodes within the foundation are with “f” respectively. Mss Msc 0  vs  Css Csc 0  vs 
From the direct stiffness approach in structural analysis,      
R   Mss Mcc Mcf   vc   Css Ccc Ccf   vc 
the dynamic force equilibrium of the system is given in  0 M fc M ff  vf   0 C fc C ff  v f 
terms of the absolute displacements, U, by the      
following sub-matrix equation: Kss Ksc 0  vs 
  
Mss Msc 0 us  Css Csc 0 us   Kss Kcc Kcf  vc 
       0 K fc K ff  v f 
Mcs Mcc Mcf  uc Ccs Ccc Ccf  uc    
 0 Mfc Mff  uf   0 Cfc Cff  u f  (6) M ss Msc 0 usg 
        (10)
& F  M ss Mcc Mcf ucg 
Kss Ksc 0 us   Mss Msc 0  usg 
       0 M fc M ff ugf 
Kcs Kcc Kcf  uc  M\cs Mcc Mcf  ucg    
 0 Kfc Kff  uf   0 Mfc Mff  ugf  To calculate the free field responses (i.e. free field
      displacements, velocities and accelerations), only the
Where the mass and stiffness at the contact nodes are foundation part without the presence of
the sum of the contributions from the structure (s) and superstructure is solved. Hence, the corresponding
foundation (f), are given by values of the displacement, velocity and acceleration
M cc  M cc s   M cc  f  for the structural part are taken as zero. This involves
the introduction of the following change of variables:
s  f 
C cc  C cc  C cc

U S   0   uS 
      
 U c    vc    uc 
U f   v f  u f 
     
U S   0   u S 
      
U c    v c    u c 
U f  v f  u f 
     
U s   0   u s 
      (11)
 U c    vc    u c 
U f  v f  u f 
      Fig. 3 Viscous dashpots connected to each degrees of
The foundation domain is subjected to freedom of a boundary node
earthquake motion and the free-field displacement for
the common and other foundation nodes are obtained. A way to eliminate waves propagating outward from
the structure is to use Lysmer and Kuhlemeyer
+ (1969) boundaries. This method consists of simply
connecting dashpots to all degrees of freedom of the
boundary nodes and fixing them on the other end
+ (12) (Fig. 3).
Effect of Hydrodynamic pressure
and Assuming the reservoir water to be inviscid and
M ss M sc 0 0 incompressible and its motion to be of small
 
R   M ss M cc ( s)
0  vc   amplitude, the governing equation for hydrodynamic
pressure is as follows (Westergaard, 1933):
 0 0 0   0  (13)
2 p  0 (15)
Css Csc 0  0   Kss Ksc 0  0
C C ( s)      
 ss cc 0 vc    Kss Kcc (s )
Kcf  vc  NUMERICAL RESULTS
 0 0 0  0   0 0 0   0  Selection of an optimum Mesh Size
A dam (Fig. 4) of height 15.0 m, crest-width 2.0 m
After obtaining the free field response, the interaction and base width 10.0 m discretized with isoparametric
force R is calculated using Eq. (13). The added linear quadrilateral elements. This particular dam-
responses (displacement, velocity and accelerations) of foundation system was originally solved by Yazdchi
coupled dam-foundation domain are calculated using et al. (1999) using coupled FEM-BEM technique
Eq. (14). The added responses are added to the free considering soil-structure interaction effects. The
field responses to get the absolute responses of the dam and the foundation are assumed linear
coupled soil and structure domain. elastic with the following material properties.
Poisson’s ratio 0.2; modulus of elasticity Ed = 3 × 107
kN/m2 and mass density as 2600 kg/m3. The
Poisson’s ratio and the mass density of the
foundation were assumed the same as those of the
dam. In addition, a horizontal static load of 1000 KN
is applied on the rightmost crest point of the dam
shown in Fig. 5. It is observed that the times and the
horizontal displacements at the rightmost crest point
converges sufficiently for a mesh division of 4 × 6
(horizontal × vertical) in the dam portion as well as
for a mesh division of 12 × 4 for the foundation
R+F (14) region. However, a mesh division of 4 × 10
(horizontal × vertical) in the dam portion has been
chosen for the validation purpose.
Absorbing Boundary

Table 1 Comparison of maximum horizontal crest
displacements & accelerations (a/g)
Fig. 4 The geometry of dam-foundation prototype In spite of differences in the methodologies used for
(Yazdchi et al, 1999) solving this particular problem, appreciable
Validation of Proposed Algorithm agreement in the obtained values of horizontal crest
The results of the present model are compared with the displacements is achieved by the present method. In
results of dam model (Fig. 6) analyzed by Yazdchi et. the present analysis, the side nodes at the boundary of
al. (1999) for Koyna ground motion. The foundation the foundation domain are attached to viscous
size of 100 m × 50 m has been considered for validating dashpots and the nodes at the base are fixed apart
the algorithm with the dam prototype solved by from the middle node.
Yazdchi et al. (1999) considering the effect of soil- Response of Koyna Dam
structure interaction. After arriving at a suitable mesh The seismic response of Koyna dam has been
grading for both the dam and the foundation domain, an investigated considering the interaction behavior of a
attempt is made to compare the present results with the linear concrete dam and an elasto-plastic foundation
result obtained by Yazdchi et al. (1999). While solving subjected to Koyna earthquake (1967)
the dam-foundation model as used by Yazdchi et al. by acceleration. The foundation material is assumed to
the proposed scheme, the effect of viscous damping and be of hard rock. The width and the depth of the
the hydrodynamic pressure is also considered. In foundation are assumed 350.0 m and 100.0 m
addition, initially the dam has been analyzed respectively. The geometry of the dam-foundation
considering the effects of its self-weight and the system chosen for the analysis is shown in Fig. 5.
hydrostatic pressure, which produced initial The material properties of the dam are as follows:
acceleration in the dam body. The Young’s modulus, Poisson’s ratio and mass
In the present work, the hydrodynamic pressures at the density of dam body are considered as 3.15e+10
upstream face of the dam are calculated These N/m2, 0.235 and 2415.816 kg/m3 respectively.
hydrodynamic forces are applied on the upstream face Similarly, the Young’s modulus, Poisson’s ratio and
of the dam. Table 1 shows the comparison between the mass density of foundation are considered as
results of Yazdchi et al (1999) and that of the proposed 1.75e+10 N/m2, 0.2 and 1800.0 kg/m3.The values of
method for different Ef/Ed (impedance ratio) ratios. The cohesion and angle of internal friction for foundation
maximum crest displacement of the dam under seismic
excitation by both the method has been tabulated in the
are C=150 KN/m2,   25 respectively (Deb,
Table 4 for a comparison purpose. The present results 2006).
are also compared with the results obtained by Reddy et
al. (2008). The obtained displacements by the proposed
interaction scheme are in very close agreement with the
results obtained both by Yazdchi et. al. (1999) and
Reddy et al. (2008). The slight discrepancy between the
two results obtained in both cases might be due to:
i) Use of different methods and numerical
tools for the solution of the coupled system.
ii) Different mesh sizes considered for the
problem.
Fig. 5 The geometry and boundary condition of the
Koyna dam-foundation system

It is observed from the convergence study that the
4
results converged sufficiently for a mesh grading of
M i n o r p r i n c i p a l s tr e s s (M P a )
2
0
8  5 for the dam domain and for a mesh grading of -2
12  5 for the foundation domain with sufficient degree -4
-6
of accuracy. In the following section results of the -8
-10
comparison of the dam foundation interaction analysis -12
assuming linear and non-linear behavior of foundation -14

-16
Linear dam-linear foundation interaction
Linear dam-nonlinear foundation interaction
domain is presented. In these analyses, to propound the -18
0 1 2 3 4 5 6 7 8
effect of soil structure interaction upon the rigid Time (sec)
analysis, the self-weight of the dam and foundation
have not been considered. In the interaction analysis, Fig. 8 Minor Principal Stress variation at node A
the analysis is done by assuming a semi-infinite
boundary condition by fixing with dashpots at the ends, CONCLUSION
and the remaining nodes at the base of the foundation The paper presents a methodology for the analysis of
have been fitted with rollers allowing only horizontal concrete gravity dam subjected to seismic excitations
translation. It is to be noted all these analyses are using direct method of soil structure interaction
carried out for the horizontal components of Koyna considering material non-linearity of foundation.
(1967) earthquake acceleration and the effect of vertical Dashpots are added at the boundaries of the
components is not taken into account foundation for representing semi-infinite domain.
Fig. 6 shows the comparison between the variation of The Dam domain is considered to be consisting of
crest displacements, obtained by dam-foundation linear elastic material. The foundation domain is
interaction analysis assuming linear and non-linear considered to be consisting of hard rock mass having
behavior of foundation domain. Fig. 7 and Fig. 8 non-linear behavior and is analyzed using Drucker-
display the variation of major and minor principal Prager failure criterion. Drucker-Prager yield criteria
stresses plotted at neck node ‘A’ (as shown in Fig. 7) represent the behavior of foundation more efficiently
for linear and nonlinear foundation material used in the because it is a dedicated yield criterion for rock
interaction analyses. It is evident that the values masses. The proposed method is validated from the
obtained for the horizontal crest displacements as found literature, which shows the accuracy of the developed
out from non-linear analysis is more than those algorithm. The numerical results presented here
obtained assuming linear analysis. Also the stresses at prove the efficiency of the present algorithm to solve
the neck node have increased when the non-linear a soil-structure coupled problem of massive
behavior of foundation domain is considered. structures such as concrete gravity dams. The values
1.00E-01
obtained for the horizontal crest displacements as
8.00E-02 found out for non-linear analysis is more than those
H o r i z o n ta l c r e s t d i s p (m )
6.00E-02
4.00E-02
obtained assuming linear analysis. The stresses at the
2.00E-02 neck node and heel node have increased when the
0.00E+00
-2.00E-02
non-linear behavior of foundation domain is
-4.00E-02 considered.
-6.00E-02 Linear dam-linear foundation interaction
-8.00E-02 Linear dam-nonlinear foundation interaction
-1.00E-01
0 1 2 3 4 5 6 7 8
REFERENCES
Time (sec)
Deb, D (2006), “Finite Element Method – Applications in
Geomechanics”, Prentice Hall of India.
Fig.6 Variation of Crest Displacement for Koyna Drucker, D. C. and Prager, W. (1952). Soil mechanics and plastic
Excitation analysis for limit design. Quarterly of Applied Mathematics,
vol. 10, no. 2, pp. 157–165.
20 Lysmer J. and Kuhlemeyer R. L. (1969), “Finite dynamic model
Linear dam-linear foundation interaction for infinite media”, Journal of Engineering Mechanics
M a j o r p r i n c ip a l s tre s s (M P a )
15 Linear dam-nonlinear foundation interaction Division, ASCE, Vol. 95 (EM4), pp. 859-877.
Reddy B. V., Burman A. and Maity D. (2008), “Seismic Response
10 of Concrete Gravity Dams Considering Foundation
Flexibility”, Indian Geotechnical Journal, Vol. 38, pp. 187-
5
203.
0
SAP 2007 The user’s manual.
Westergaard H. M. (1933), “Water pressures on dams during
-5 earthquakes”, Trans. ASCE, Vol. 98, pp. 418-472.
0 1 2 3 4 5 6 7 8 Yazdchi M., Khalili N., Valliappan S. (1999) “Dynamic soil-
Time (sec) structure interaction analysis via coupled finite element-
boundary element method”. Soil Dyn. And Earthquake
Fig. 7 Major Principal Stress variation at node A Engng., Vol. 18, pp. 499-517.

ISBN 978-81-921121-0-7
DEVELOPMENT OF A STIFFNESS ANALYSIS FOR A

STRUCTURAL ELEMENT IN A BUILDING INFRA-
STRUCTURE
Pradeep Kumar
Associate Prof., Deptt. of Civil Engg., HBTI, Kanpur
Swatantra K. Porwal
Assistant Prof., Deptt. of Mechanical Engg., VSGOI, Kanpur, swatantraporwal@yahoo.com
ABSTRACT: Buildings play a key role in the infrastructure development. Safety against the mechanical
parameters in the building needs extra attention with the conventional and seismic analysis & design of a building. It
is equally important to understand the behavior of failure in the members of the building. Strength and Stiffness are
two main criteria for design of failure of a structural member such as beam in any building. For calculation of the
stress & stiffness value, the beam is discretized into the smaller elements like rectangular, triangular, hexahedral and
tetrahedral; and for equal discretization, isoperametric element is used. In this paper, the beam as structural element,
subjected to load is considered. Its deformation from its original position is assessed. As the value of the stress
becomes maximum, the structure starts failing. For protection of the structural element from failure, this work
proposes to discretize the beam into the smaller elements called finite elements. The results are calculated and
compared with the values obtained by other conventional methods of analysis.
Keywords - Strength, Stiffness, Deflection, Finite Element, isoparametric element.
A Beam can be defined as a structural element which

INTRODUCTION has one dimension (length) considerable larger
Beam Slab and column are the most important compared to the other two directions i.e. breath &
components of a building. As per the strength criterion of depth and is supported at few points . For the
the design, they should be strong enough to resist bending calculation of the stress & stiffness value in the
moment and shear force or in other words ,the bending beam, it is discritized into the smaller elements and
stresses and shear stresses but as per the stiffness criterion for the discritization, the elements like rectangular,
, they should be stiff enough to resist the deflection and triangular, hexahedral and tetrahedral are available
not to deflect more than the permissible limit under the and for equal discritization, isoparametric elements
action of the loading. In actual practice some are used.
specifications are always laid to limit the maximum
deflection of a cantilever or a beam for its smaller span. Finite Element Formulation
Structural design engineering process includes material Finite Element Analysis (FEA) was first developed in
selection for construction; determine appropriate 1943 by R. Courant, who utilized the Ritz method of
structural system for a particular case. If the stress value numerical analysis and minimization of variational
is less than the allowable stress value then the structure is calculus to obtain approximate solutions to vibration
safe but for greater than the allowable stress value, systems. Shortly thereafter, a paper published in 1956
structure fails. The complexity of modern structures often by M. J. Turner, R. W. Clough, H. C. Martin, and L.
requires a great deal of creativity in order to ensure the J. Topp established a broader definition of numerical
structural support and resist the loads they are subjected analysis. The paper centered on the "stiffness and
to. At the time of manufacturing, lot of stresses act on the deflection of complex structures".
structural element due to the axial loading, bending
moment, whose value is greater than the expected value . Isometric Elements
In this work , the calculation of the stress , using the x  N1' x1  N 2' x2  ........N n' xn
stiffness value with the concept of Finite Element
z  N1' z1  N 2' z2  ........N n' zn
Analysis , is done a program is developed which first
calculates the stiffness value and later, the stress value by
using the concept of FEM. y  N1' y1  N 2' y2  ........N n' yn  x   N '   xn 
STRUCTURAL MEMBER : A CASE STUDY where [N’] are the shape function of the parent
rectilinear element &{xn} are the nodal coordinates

of the element.The shape function will be expressed Thus this transformation relates a unit square in r & s
through the natural coordinate system r,s& t.. The shape coordinates to an arbitrary quadrilateral in Cartesian
function [N’] used in the above transformation thus help (x,y) coordinates system whose shape & size are
us to define the geometry of the element in the Cartesian determined by the eight nodal coordinates
coordinate system. If these shape function [N’] are the x1,y1,x2,y2,x3,y3,x4,y4.The above relation also helps to
same as the shape functions [N] used to represent the determine the x,y coordinates of any point in the
variation of displacement in the element, these elements element when the corresponding natural coordinates r
are called ‘isoparametric’ element. & s are given.
 x   N '   xn   /  r   x /  r  y / r  / x   /  x 
     J  
 /  s   x /  s  y / s  / y   /  x 
Four-Noded Two- Dimensional Element
Consideingr a quadrilateral two dimensional element the where [J] is the Jacobian matrix. Hence the
parent element is a rectangle mapped into a square in derivatives with respect to Cartesian coordinate
natural coordinates & This in turn is transformed into an system can be given as
arbitary quadrilateral element with straight boundaries.the
shape function used for represeting the variation of  /  x  1  /  r 
   J   
displacement for a four noded Rectangular ele ment given  /  y   /  s 
by equation can be used to describe the geometry of the
arbitrary quadrilateral in Cartesian system. However from 4 4
computer programming point of view the terms are x   N i' xi & y   N i' yi
i 1 i 1
rearranged to suit the description of the vector {xn} as-
 x1 y1 
 N /  r  N 2 /  r  N 3 /  x  N 4 /  x   x2 y2 
 J    N1 /  s 
 N 2 /  s  N 3 /  s  N 4 /  s   x3 y3 
 1
 
 x4 y4 
& noting that Ni is a function in (r,s) the Jacobian [J]

can be evaluated as
 x1 y1 
  1  s  / 4 1  s  / 4  1  s  / 4  1  s  / 4   x2 y2 

J    
  1  r  / 4 1  r  / 4  1  r  / 4  1  r  / 4   x3 y3 
 
 x4 y 4 
Let the inverse of  J  is required-
 x1  1 J J12 
J    11 
y 
 1  J 21 J 22 
N
 1 0 N 2 0 N 3 0 N 4 0   x2 
x     It should be observed here that in order to transform
    N1 0 N 2 0 N 3 0 N 4 0   y2  the x & y coordinates into r & s coordinates the
y   . 
   inverse of  J  must exist. Hence the determinant of
. 
 
the Jacobian J  must be non zero at every point of
.  (r,s).
y3 where Ni (i= 1,2,3,4) are given by equation –  J   det  J    x /  r  y /  s   x /  s  y /  r  0

N1  1  r 1  s  / 4 N 2  1  r 1  s  / 4
N 3  1  r 1  s  / 4 N 4  1  r 1  s  / 4

 u /  x   J11 J12 0 0   u /  r   K   hwi w j  B T C  B  J 
 u /  y   J J 0 0  


  21 22
    u /  s  Where K= stiffness value
  v /  x  0 0 J11 J12   v /  r  wi w j = Gauss weights
 v /  y  0 0 J 21 J 22 
  v /  s  h = Thickness of the Element
B = Strain-Displacement Matrix
The strain-displacement relations can be expressed in J = Jacobian
Matrix form as
 u /  x
 x  1 0 0 0   
CONSTITUTIVE MATRIX
     v /  y 
    y   0 0 0 1    To determine the stresses in the members of a
  0 1 1 0   v /  x  structure or in a deformable solid it is necessary to
 xy   
 v /  y  know the components of stress as a function of the
components of strain & vice versa.We assume that
 u /  r  material is elastic & obey’s the hook’s law According
 J11 J12 0 0  
  u /  s  to hook’s law the six components of stress may be
   0 0 J 21 J 22   v /  r  expressed as a linear function of six component of
 J 21 J 22 J11 J12    strain.
 v /  s 
ox  C11 C12 .......... C16   x 
4 4 o    
'
x  N u & y  N v '
 y  C21 C22 .......... C26   x 
i i i i
i 1 i 1 o  C C .......... C36   x 
 z    31 32  
Txy  C41 C42 .......... C46   xy 
u1    
 
Tyz  C51 C52 .......... C56   yz 
v1   
u /r N1 /r 0  N2 /r 0  N3 /r 0 N4 /r 0  u2  T  C61 C62 .......... C66   zx 
   zx 
u /s N /s 0  N /s 0  N3 /s 0  N4 /s 0  v2 
   1 2
   
v/r 0  N1 /r 0 N2 /r 0 N3 /r 0 N4 /r  u3  1-   0 
  
v/s 0  N /s 0  N2 /s 0 N3 /s 0 N4 /s  v3 
 1
  C  E / 1   1  2     1-  0 
u4   
v   0 0 1-2  / 2 
 4
So we will use
The stiffness matrix for the element is given by-
T
Experimental Program
 K   h     B  C  B  dxdydz This work will develop a software in c language and
this software will calculate stiffness value and then
In the case of the present example of the two dimensional compaired the given result.
element it is given by-
T Literature survey represents that most of the
 K   h   B  C  B  dxdydz sciencetist have been worked in the field of structural
engineering & Finite Element Analysis & the main
[k] = h∫ [B]T [C] [B] dx dy dz aim of all sciencetist is to design a safer structural
element or to protect the failure of the structural
where h is the thickness of the element. It can be shown element which occurs during the manufacturing stage
that the elemental area in the Cartesian coordinates (x-y) & loading condition it has been discussed that during
can be expressed in terms of the area in the area in the the manufacturing stage more residual stresses occurs
local coordinates (r-s) as- & the calculation of these stresses is must. So to
protect the failure of the structural element during
dxdy   J  drds manufacturing stage & loading condition analytical
work is done. This analytical work includes the finite
 K   h   B T C  B  J  drds element analysis for every time dependent effects &
manufacturing stage.At the time of loading, two

criteria is more important these are Stiffness & Strength CONCLUSION
or Stress which are applying on it. so the calculation of The software developed, provides an insight into the
these values strength & stiffness is must & for the values of stiffness and deflection that are important to
calculation of these values this work provides a software protect the failure of the structural elements . As
which will calculate the strength & stress value by using discussed ,during the manufacturing stage, more
the concept of FEM. residual stresses occur and during the loading
condition, structural element failure occurs due to the
RESULT FOR COMPOSITE ELEMENT failure of the strength & stiffness value and by the
Element no. Deflection(mm) deflection value, which is calculated at the point
Stiffness(N/mm.) where the stress is maximum. To calculate all these
Element no.1 3.719254*10-30 8.888889*101 values, an analytical work is done and discussed by
using the FEM by which the analysis of any part or
Elemen t 7.528492*10-27 8.913123*101
element can be done for the building infrastructure .
no.2
By this very reason, the Finite Element Analysis has
Element no.3 1.417448*10-22 8.937357*101
been included for every Time dependent effects &
Element no.4 -99.53938 8.961591*101 Manufacturing stage. This work paves a path for
Element no.5 -70.1099 8.985825*101 structural engineering fraternity for providing the
Element no.6 2.247546*10 -30
8.901006*102 value of stiffness, deflection and also protects the
failure of the structural elements which will be
Element no.7 1.153879*10-26 0.901006*102
important for both, a user of the building and the
Element no.8 -34.66768 0.905853*102 developer who builds the structure.
Element no.9 -45.76091 0.908276*102 References
Element 2.515676*10-18 0.910700*102 [1]Zhigang Yu , XiaoliGuo , Fule Chu ,(2010), ”A
no.10 multivariable hierarchical finite element for static and
Element 9.086976*10-31 0.913123*102 vibration analysis of Beams”,Vol 46,No. ,pp 625-
no.11 631
Element -73.70749 0.915546*102 [2]M. Lezgy-Nazargah , S.B.Beheshti Aval ,
no.12 M.Shariyat,(2010) “A refined mixed global–local
Element -41.55019 0.917970*102 finite element model for bending analysis of
no.13 multilayered rectangular composite beams with small
Element -34.56874 0.915546*102 widths”,Vol 49,No. 2011,pp. 351-362
no.14 [3] Newmark NM, Siess CP, Viest IM. Tests and
Element 3.839850*10-19 0.913123*102 analysis of composite beams with incomplete
no.15 interaction. Proc Soc Exp Stress Anal 1951;9(1):75–
Element 5.390944*10-30 0.910700*102 92.
No16
[4] Spacone E, El-Tawil S. Nonlinear analysis of
Element 7.471256*10-27 0.908276*102
steel–concrete composite structures: state of the art. J
No17
Struct Eng ASCE 2004;30(2):159–68.
Element -39.84877 0.905853*102
No18 [5] Leon RT, Viest IM. Theories of incomplete
Element -93.23479 0.903429*102 interaction in composite beams. In: Proceedings of
No19 the composite construction in steel and concrete III,
Element -37.8328 0.901006*102 Irsee, Germany, 9–14 June. p. 858–70.
No20 [6] Faella C, Martinelli E, Nigro E. Steel and
Element 4.779204*10-30 8.985825*101 concrete composite beams with flexible shear
No21 connection: ‘‘exact’’ analytical expression of the
Element -70.97516 8.961591*101 stiffness matrix and applications. Comput Struct
No22 2002;80:1001–9.
Element 1.668300*10-23 8.937357*101 [7] Ranzi G, Bradford MA, Uy B. A direct stiffness
No23 analysis of a composite beam with partial interaction.
Element -101.95116 8.913123*101 Int J Numer Meth Eng 2004;61:657–72.
No24 [8] Girhammar UA, Pan D. Dynamic analysis of
Element 5.787271*10-18 8.888889*101 composite members with interlayer slip. Int J Solids
No25
Struct 1993;30(6):797–823.
[9] Dall’Asta A, Zona A. A Finite elements for the
analysis of composite members with interlayer slip.

In: Proceedings of CTA XVIII (Italian Workshop on
Steel); 2001.
[10] Griffiths DW, Nethercot DA, Rockey KC, Evans
HR. Finite element method: a basic introduction. 2nd ed.
Halsted Press; 1975.
[11] Weaver W, Gere JM. Matrix analysis of framed
structures. 3rd ed. Chapman & Hall; 1990.
[12] Ranzi G, Bradford MA. Time-dependent analysis of
composite beams with partial interaction using the direct
stiffness approach. UNICIV Report R-423, School of
Civil and Environmental Engineering, The University of
New South Wales, Australia; 2004.
[13] Girhammar UA, Gopu VKA. Composite beam-
columns with interlayer slip-exact analysis. J Struct Eng
ASCE 1993;119(4):1265–81.
[14] Ranzi G, Bradford MA. Analytical solutions for the
time-dependent behaviour of composite beams with
partial interaction. Int J Solids Struct 2006;43:3770–93.
[15] Ansourian P. Experiments on continuous composite
beams. Proc Inst Eng 1981;71(Part 2):25–51.
[16] Oehlers DJ, Bradford MA. Composite steel and
concrete structural members: fundamental behaviour.
Oxford: Pergamon Press; 1995.
[17] Ranzi G. Partial interaction analysis of composite
beams using the direct stiffness method. PhD thesis,
UNSW; 2003. 1214 G. Ranzi, M.A. Bradford /
Computers and Structures 85 (2007) 1206–1214

ISBN 978-81-921121-0-7
EMERGING TECHNOLOGIES FOR COLOUR REMOVAL
Shiva Shankar Y
Assistant Professor, Department of Civil Engineering, Hindustan College of Science & Technology, Mathura-281122.
ysivashankar@yahoo.com.
Abhishek Kumar
Manager, Muncipal Solid Waste Department, Jamshedpur Utilities & Services Company Ltd, Jamshedpur-831001.
abmishra30@gmail.com.
Harshit Jain
UG Student, Department of Civil Engineering, Institute of Technology, Banaras Hindu University, Varanasi 221005.
hjair4617@gmail.com
Devendra Mohan
Professor, Department of Civil Engineering, Institute of Technology, Banaras Hindu University, Varanasi-221005.
devmohan9@yahoo.com.
ABSTRACT: Coloured industrial effluents discharged into surface-water bodies has become a serious
problem in India because a sizeable population depends on surface water-resources for supply. With rapid
growth of industrialization and use of synthetic dyes, conditions of surface water bodies are deteriorating fast to
the level of serious concern and sometimes, these may even be rendered unfit for certain uses. Therefore, proper
treatment of coloured effluents is one of the most urgent tasks before an environmental engineer and a number
of directions are being explored for an effective control. This paper presents a brief review on emerging
technologies that can be used for enhanced colour removal from industrial effluents.
KEY WORDS: Colour removal, emerging technologies, chitosan, coagulation and flocculation, adsorption,
sago industry, ionizing radiation.
INTRODUCTION disposal of coloured water is into the streams or

Coloured water is indicative of dissolved organic nearby water bodies, chemically polluting the water
material, inadequate treatment, high disinfectant bodies and affecting the ecosystem, which constitutes
demand and the potential for the production of excess the second level impacts. Further this leads to third
amounts of disinfectant by-products. Inorganic level impacts, i.e., further degradation in quality due
contaminants such as certain metals are also common to chemical and biological subsequent transformation
causes of colour. In general, the point of consumer in the ecosystem conditions.
complaint is variable over a range from 5 to 30 colour Synthetic colouring agents are harmful to human
units, though most people find colour objectionable beings and animals because of their toxicity
over 15 colour units (USEPA 2011). carcinogenicity (Birhali & Ozmen 2005; Onur &
Industries like textile dyeing, pulp and paper mills and Erinc 1998), mutagenicity (Abdou et al. 2001),
chemicals manufacturing units generate a large teratogenicity (Marks et al. 1981) and certain other
amount of coloured water. Synthetic dyes have characteristics, which affect the human health
increasingly been used in textile and dyeing industries adversely.
because of their ease of application and cost- The methodologies used for colour removal can be
effectiveness in synthesis and high stability to light, broadly classified as physico-chemical and biological.
temperature, detergent and microbial attacks. More Commonly used methods are coagulation and
than 10,000 chemically different dyes are being flocculation, adsorption and advanced oxidation
manufactured. The world dyestuff and dye processes. With increasing concentration of
intermediates production are estimated to be around pollutants, application of these methods is not
7×108 kg per annum (Fatih & Sengul 2011). At satisfying the purpose completely because even traces
present, synthetic colouring materials have become a of remaining pollutants can affect the aquatic
big problem for the environment. Their synthetic ecosystems adversely. Therefore, a combination of
origin may cause interference with the natural above-mentioned methodologies with biological
degradation of such materials. techniques can be a feasible alternative. This paper
The Environmental impacts can be broadly classified presents a review of various methodologies proposed
into three types i.e., first level impacts, second level recently for enhanced colour removal.
impacts and third level impacts. These are classified
based on sequence of their manifestation or EMERGING TECHNOLOGIES
occurrence after disposal. The first level Application of Bio-Polymer
environmental impact is colouration of surface with Renault et al. (2009) have studied the use of chitosan
which it comes into contact. The common practice of a partially deacetylated polymer obtained from the

alkaline deacetylation of chitin, a biopolymer that achieved a 90%. The combined process also
extracted from shellfish sources. In particular, the decreases COD values as it implies the use of lower
development of chitosan-based materials as useful resin concentration than coagulation–flocculation.
coagulants and flocculants is an expanding field in the
area of water and wastewater treatment. Their Use of Alum Sludge
coagulation and flocculation properties can be used to Chu (2001) has effectively achieved dye removal
remove particulate inorganic or organic suspensions, from textile industry wastewater using recycled alum
and also dissolved organic substances. Chitosan sludge. The removal of dyes from such wastewater by
possesses several intrinsic properties such as its non- recycled alum sludge (R.A.S.), generated by the
toxicity, its biodegradability and its outstanding coagulation process itself, has been studied in detail
chelation behaviour that make it an effective and optimized. One hydrophobic and one hydrophilic
coagulant and/or flocculant for the removal of dye were used as typical compounds to examine the
contaminants in the dissolved state. It has the physico- performance of this process.
chemical characteristics of both coagulants and It was found that R.A.S. is a good material for
flocculants, i.e., high cationic charge density and long removing hydrophobic dye from wastewater, while
polymer chains, leading to bridging of aggregates and simultaneously reducing the fresh alum dosage. It was
precipitation. Numerous works have demonstrated found that about one third of the fresh alum dosage
that chitosan and its derivatives (in particular grafted could be saved. The back-diffusion of residual dye
biopolymers) can be a potential substitute for metallic from the recycled sludge was detected, but it could be
salts and synthetic polyelectrolytes in the treatment of easily controlled by addition of small amount of fresh
wastewater for the removal of both particulate and alum to the system. Use of R.A.S. is not
dissolved substances. recommended for the removal of hydrophilic dyes,
since high solubility characteristics of such dyes can
Combination of Physico-Chemical Methodologies cause deterioration in the water quality during
Guendy (2010) has proposed a combination of recycling.
coagulation and adsorption processes for Acid Red Bosibuyuk et al. (2004) have studied the use of sludge
dye removal using aluminum sulfate (Alum) and obtained from water treatment plant for treatment of
ferric chloride as coagulants and granular activated dye wastes. Potential for using this sludge for the
carbon (GAC) as adsorbent. The results revealed that treatment of reactive, direct, disperse, acidic, and
application of coagulation before adsorption is most basic dyestuffs by coagulation and sorption has been
effective in removing nearly 71- 97.7% using alum at studied in detail. The sludge acted as a coagulant and
pH 6- 8 and 53.7- 98.9% using FeCl3 at pH 4- 6 at a removed colour with excellent efficiencies in case of
lower dosage of both coagulants in the range of 70- basic, disperse and direct dyes. The optimum
110 ppm and 20- 90 ppm dye concentration. Removal conditions were: pH value of 5 and a sludge dosage of
of the dye by adsorption increased from 75- 90% by 2000 mg/L. Intermediate results were obtained for
decreasing its initial concentration from 50 to 10 ppm acidic and reactive dyes. Efficiency of the sludge was
for 2.5 gm GAC and from 80- 95% for 5 gm GAC at also compared with alum and ferric chloride for the
pH 8 and equilibrium time of shaking 120 min. at same group of dyes.
room temperature (25± 20C). The sludge was also used as a coagulant to treat the
Riera-Torres (2010) studied use of nanofiltration, wastewater from a textile industry. At dosage varying
coagulation–flocculation to remove colour from five from 2000-4000 mg/L, the sludge was as effective as
reactive dyes namely Remazol Black 133 B (C. I. ferric chloride and alum removing COD. Sorption
Reactive Black 5), Procion Crimson H-EXL, Procion tests have shown that disperse and reactive dyes could
Navy H-EXL, Procion Yellow H-EXL and Procion not get adsorbed onto the sludge
Blue H-EXL all of them with the monochloro tri azine Moghaddam et al. (2010) have described application
reactive group and chromophore azo. Nanofiltration of coagulation and flocculation processes for dye
was carried out using an energy saving nanofiltration removal using sludge from water treatment plant.
membrane in a 0.5 L pilot plant at 10 bar. They have also worked on optimization through
Coagulation–flocculation experiments were response surface methodology for ferric chloride
performed in a jar test apparatus using an anionic (FCS) sludge for the removal of Acid Red 119
polyelectrolyte flocculant, FeCl3 as a coagulant agent (AR119) dye from aqueous solutions. The dye
and with two different resins melamine–urea– removal of 96.53% was observed and it was found
formaldehyde and polyamine resin. Results proved that reusing the FCS as a low cost material into the
that nanofiltration removed in the order of 40 and coagulation/flocculation processes in wastewater
80% of colour for the five dyes while coagulation– treatment plants can offer some advantages such as
flocculation reached between 85 and 95% of colour high efficiency for the dye removal and economic
removal for four dyes except in the case of polyamine savings on overall treatment plant operation costs.
resin for RB5. The combination of these techniques
dramatically improved the removals reaching over
98% of colour removal for all dyes except for RB5

Ionizing Radiation carbon deposit and change of soil matrix are possible.
Abdou et al. (2011) found that use of ionizing Advantage of combining the effluents was that the pH
radiation such as electron beam and gamma gets neutralized and therefore, there remains no need
irradiation could be utilized in the decolourization and for adding caustic/lime. Similarly, the presence of
degradation of the dye waste water after dyeing cotton starch in the sago industry wastewater reduces
fabrics with four different dyestuffs based on azo and demand for the addition of nutrients during the
anthraquinone structures. The dyes used in study are process. The sago effluent was utilized as co-substrate
Azo Direct Blue, Azo Reactive Yellow, to enhance the degradation of textile dye house
Anthroquinone Reactive Blue and Anthraquinone effluent which is recalcitrant in nature. The colour and
Direct Green. Influence of parameters like irradiation COD removal efficiency was studied by feeding the
dose, H2O 2 concentration, pH of the solution and dye combined wastewater of sago and textile dyeing
concentration on the percentage decolourization was industry at different mixing proportions of 90:10,
studied. Radiation technology, as an advanced 80:20, 75:25, 70:30, 65:35 with COD ranging from
oxidation process, has been recognized as a promising 5200 to 6320 mg/l. The maximum COD removal was
process for hazardous organic wastewater treatment. about 53.1% in acidogenic reactor (90:10 mixing
It is well known that most coloured materials undergo ratio) and 88.5% in methanogenic reactor (70:30
bleaching or colour changes when exposed to ionizing mixing ratio). In methanogenic reactor, at organic
radiation. Therefore, gamma and electron beam loading rate of 5.6 kg COD/m3 / d maximum COD
irradiation can be considered as potential methods for (88.5%) and colour (91.3%) removal were observed.
the treatment of wastewater from textile and other
industries using dyes for different purposes. Usage of Clayey Soils
Electron beam irradiation can decompose effectively Hassan & Hameed (2011) have proposed use of Fe-
dye molecules, but large dosages are necessary to clay as effective heterogeneous Fenton catalyst for
completely degrade dye molecules to small fragments. decolourization of Reactive Blue 4. In the study
Presence of H2O2 can greatly promote degradation of decolourization of anthraquinone dye Reactive Blue 4
dye solutions and increase COD removal efficiency. (RB4) in an aqueous solution was investigated by
Results proved that combination of radiation using Fe-ball clay (Fe–BC) as heterogeneous catalyst.
technology and other treatment process may be a In heterogeneous solid catalysts, iron ions are
potential approach for dye wastewater treatment. The immobilized on the solid supports or within the
results have indicated that dye solutions are easily structure and in the pore/interlayer of the catalysts. As
decolourized by using electron beam in presence of a result, the catalysts can maintain its ability to
H2O2 than in its absence. The results showed that generate hydroxyl radicals from hydrogen peroxide,
decolourization of dye solution mainly depends on the and iron hydroxide precipitation is also prevented.
pH media in presence of H2O2, structure of dye and to Besides showing limited leaching of iron ions, the
some extent the concentration of the dye. Complete catalysts can be easily recovered after the reaction,
decolourization and degradation of dyes has been and remain active during successive operations.
achieved at pH 9 and dye concentration of 1 g/L in Use of clays as catalysts for Fenton-like reactions is a
case of an azo dye. Data of COD and TOC removal promising alternative for decontamination of soils,
confirmed relatively higher efficiency of electron groundwater, sediments and industrial effluents
beam with respect to gamma irradiation. because they are natural, abundant, inexpensive and
environment-friendly. Wide usefulness of clay
Reuse of Industrial Waste Water minerals is essentially a result of their high specific
Senthil Kumar et al. (2011) have studied on surface area, high chemical and mechanical stabilities
wastewaters from textile dyeing industries for the and a variety of preferable surface and structural
decolourization and removal of degradable organics properties.
with sago wastewater as a co-substrate in a pilot scale Experiments were carried out with varying
two-phase upflow anaerobic sludge blanket (UASB) combinations of Fentons reagent, ball clay and Fe-BC
reactor. The UASB reactor was inoculated with seed catalyst on Reactive Blue dye. The results indicated
sludge from an anaerobic digester treating sago that Fe–BC has been proved to be a superior
wastewater. heterogeneous catalyst for decolourization of RB4 in
Release of coloured compounds into natural water an aqueous solution. The best operation parameters
bodies is undesirable because of their impact on for Fenton oxidation of RB4 were 0.80 wt.% of iron
photosynthesis of aquatic plants, carcinogenic nature ions loading, 5.0 g/L of catalyst dosage, and 8 mM of
of many of these dyes and their break-down products. H2O2 for 50 mg/L initial RB4 concentration at an
Sago effluent possesses high COD and it is highly initial pH 3.0 at 30°C temperature. Under these
acidic and with high organic contents in nature. As a conditions, 99% decolourization efficiency of RB4 in
result of discharge of sago wastewater, depletion of aqueous solution was achieved within 140 min of
dissolved oxygen may take place in water sources of reaction time. Fe–BC catalyst demonstrated not only
the affected area. It also causes an increase in air good catalytic activity, but also a reasonably small
pollution in terms of stringent foul odour, inorganic iron leaching. This characteristic makes possible for

the catalysts to have long term stability, without decolorization in first step and TOC reduction of 95–
generating significant iron hydroxide sludge. 97% in the two steps.
Electro Coagulation
Bahadir et al. (2011) have studied the electrochemical CONCLUDING REMARKS
decolourization of textile dye and removal of metal In view of the facts mentioned above, it becomes clear
ions from [textile dye|metal ion] binary mixtures of that use of emerging techniques involving
[Acid Blue 29|Cu(II)], [Reactive Red 2|Cu(II)], [Acid combination of methodologies, use of environment-
Red 97|Ni(II)], and [Reactive Blue 4|Ni(II)] using iron friendly materials like chitosan, clays and others,
anodes and Na2SO4 electrolyte in a batch reuse of sago industry wastewater and application of
electrochemical reactor. In this work, the ionizing radiations can enhance the colour removal
electrochemical process was found independent of significantly. The colour removal by application of
type of textile dye and metal ion in binary mixtures various commonly used physico-chemical methods
due to the similar removal trends were observed in all like coagulation and flocculation, adsorption, etc.
the cases. Depending on electrochemical reaction varies between 60- 70%. Application of properly
conditions, textile dye decolourization and metal ion selected certain techniques mentioned above can raise
removal were obtained between 70.6–99.7% and the percentage upto 90-99%. In addition to these,
64.9–100% for textile dye decolourization and metal application of biological methodologies can add
ion removal, respectively. At response surface advantages of enhanced colour removal with less
optimized operating conditions, 89.4–98.2% textile severe problem of sludge disposal.
dye decolourization and complete metal ion removal
were achieved. REFERENCES
In the electro chemical treatment, rapid Abdou, L. A. W., Hakeim, O. A., Mahmoud, M. S.
decolourization achieved in the study indicated that and El-Naggar, A. M. (2011). Comparative study
the acid and reactive dyes were completely degraded between the efficiency of electron beam and
into smaller and colorless organic and inorganic gamma irradiation for treatment of dye solutions,
molecules without any coagulant addition or any Chemical Engineering Journal, Vol. 168, 752–
further physicochemical processes in order to enhance 758.
the treatment performance. It was concluded that Abdou, R. F., Abdel-Aal, S. A., Waly, E. A. and
metal ions were removed from binary mixtures by Khalil, Z. H. (2001). Effect of cyanine dyes
precipitation with OH- ions produced at the cathode, analogues in breaking dormancy and their
by coagulation/flocculation with Fe(OH)2 and/or mutagenic effects in garlic (Allium Sativum L.),
Fe(OH)3 clusters, or by direct reduction to solid metal ISHS Acta Horticulture 208: VI International
on the cathode surface. Results of this study have Symposium on Medicinal and Aromatic Plants.
indicated that this electro chemical reactor could be a Detailed Report Available at http://www
feasible and an efficient alternative treatment system .actahort.org/books/208/ 208_17.htm.
for the electrochemical textile dye decolourization and Birhali, A. and Ozmen, M. (2005). Evaluation of the
removal of metal ions from [textile dye|metal ion] toxicity and teratogenenicity of six commercial
binary mixtures. textile dyes using the frog embryo teratogenesis
assay xenopus, Journal of Drug and Chemical
Combination of Fungal and Bacterial Treatment Toxicology, Vol. 1, 51-65..
Cenek et al. (2011) have studied combined fungal and Bosibuyuk, M., Yilmaz, T., Kayranli, B., Yuceer, A.
bacterial treatment for colour removal in textile and Forster, C. F. (2004). The use of wasteworks
wastewater. Low efficiency of dye removal by mixed sludge for the treatment of dye wastes, Journal of
bacterial communities and high rates of dye Environmental Technology, Vol. 1, 51-65.
decolourization by white-rot fungi suggested a Cenek, N., Katerina, S., Oldrich, B., Olga, K.,
combination of both processes to be an option for Andreas, H. and Werner, F. (2011). Potential of
treatment of textile wastewaters containing dyes and combined fungal and bacterial treatment for
high concentrations of organics. In this two-step colour removal in textile wastewater, Bioresource
approach, fungal application provides the Technology, Vol. 102, 879–888.
decolourization and the bacterial step the removal of Chu, W. (2001). Dye removal from textile dye
organic carbon pollution from the effluents. I. lacteus wastewater using recycled alum sludge, Wat. Res.
formed sustainable biofilms usable for long periods Vol. 35, Number 13, 3147–3152.
and was capable of rapid decolourization of various Fatih, D. and Sengul, K. (2011). Removal of basic red
dye structures. It maintained effective levels of 46 dye from aqueous solution by pine tree leaves,
laccase during prolonged applications and both could Chemical Engineering Journal, doi: 10.1016
decolorize and detoxify the textile wastewater. /j.cej. 2011.03.029.
Sequential two-step application of Fungal Trickling Guendy, H. R. (2010). Treatment and reuse of
Filter and bacterial reactors resulted in efficient wastewater in the textile industry by means of

coagulation and adsorption techniques, Journal of Renault, F., Sancey, B., Badot, P. M. and Crini, G.
Applied Sciences Research, Vol. 6, 964-972. (2009). Chitosan for coagulation/flocculation
Hassan, H. and Hameed, B. H. (2011). Fe-clay as processes – an eco-friendly approach, European
effective heterogeneous fenton catalyst for the Polymer Journal, Vol. 45, 1337–1348.
decolorization of reactive blue 4, Chemical Riera-Torres, M., Gutierrez-Bouzan, C. and Crespi,
Engineering Journal, doi:10.1016/ j. cej. 2011. M. (2010). Combination of coagulation–
04. 040. flocculation and nanofiltration techniques for dye
Marks, T. A., Gupta, B. N., Ledoux, T. A. and removal and water reuse in textile effluents,
Staples, R. E. (1981). Teratogenic evaluation of Desalination, Vol. 252, 53–59.
2-nitro-p-phenylenediamine,4-nitro-o-phenylene Senthilkumar, M., Gnanapragasam, G., Arutchelvan,
diamine and 2, 5-toluenediamine sulfate in the V. and Nagarajan, S. (2011). Treatment of textile
mouse, Journal of Teratology, Vol. 24, Number dyeing wastewater using two-phase pilot plant
3, 253-265. uasb reactor with sago wastewater as co-
Moghaddam, S. S., Moghaddam, M. R. A. and Arami, substrate, Chemical Engineering Journal, Vol.
M. (2010). Coagulation/Flocculation process for 166, 10–14.
dye removal using sludge from water treatment USEPA (2011). Details are available online at
plant: optimization through response surface http://water.epa.gov.
methodology, Journal of Hazardous Materials,
Vol. 175, 651–657.

ISBN 978-81-921121-0-7
COMPARATIVE ANALYSIS OF BED LOAD FORMULAS FOR

INTENSE BEDLOAD CHANNELS
Sumit Talukdar
Research Scholar, Dept. of Civil Engg., Indian Institute of Technology Guwahati, Guwahati-781039, Email:
s.talukdar@iitg.ernet.in
Bimlesh Kumar
Assistant Professor, Dept. of Civil Engg., Indian Institute of Technology Guwahati, Guwahati-781039, Email:
bimk@iitg.ernet.in
Subashisa Dutta
Associate Professor, Dept. of Civil Engg., Indian Institute of Technology Guwahati, Guwahati-781039, Email:
subashisa@iitg.ernet.in
ABSTRACT: The study on Bedload transport behaviour is widely explored from the last few decades and
many semi-empirical or empirical equilibrium transport equations are developed. The phenomenon is a very
complex due to its varied physical properties like velocity, depth, slope, particle size in the alluvial system.
Alluvial channels having Φhave been considered as intense bed load channels. Φ is called as Einstein
transport function. In practical applications, these formulae have appreciable deviation from each other in
derivation and also their ranges of applications are different. It is imperative to know the best predictive
equations in for the field applications. Present work, based on large database of different bed load
measurements, does a comparative analysis in terms of prediction ability of different bed load equations.
Keywords: flow depth, flow velocity, sediment, incipient motion, shear stress, critical shear stress, bedload,
transport rate, alluvial river.
INTRODUCTION Engelund & Hansen (1967), Ackers-White (1973),

Bedload transport in alluvial rivers is the principle Engelund & Fredsøe (1976) and van Rijn (1984).
link between river hydraulics and river form and is Selection of an appropriate formula is very difficult
responsible for building and maintaining the channel because of availability of numerous equations on
geometry (Parker 1979; Leopold 1994; Goodwin bedload transport. Field practitioners must be able to
2004). Bedload prediction is of primary importance accurately predict the sediment capacity of river
for river engineering, fluvial geomorphology, eco- channels in order to assess physical and biological
hydrology, environmental surveys and management, function and stability. Hence, an attempt has been
and hazard prediction (Recking, 2009). Bedload made to analyze some of the most used equations for
transport can be described as a random phenomenon their prediction capability for the field applications.
that is generated by the interaction of turbulent flow
structure with the materials of the bed surface DATA SOURCE
(Einstein 1950). This interaction is very complex and Recking (2006) has compiled an extensive database
as a result, attempts to model this process have for bedload measurements and same has been used
largely resulted in limited or qualified success. There here. The data are accomplished herein termed as
is no bedload transport until the critical variable such intense bedload transport, (Einstein parameter)is
as shear stress, stream power, discharge, or velocity higher than approximately 0.4 (Graf, 1998). Table 1
has been exceeded by the flow conditions and that shows the ranges of the observations used in the
the bedload transport rate increases in proportion to present work.
the increase in the flow condition beyond the critical Table 1 Summary of Hydraulic Properties
value. Also numerous studies, both in natural Variable Average Minimum Maximum Std. Dev.
channels and in flumes are based on this concept and
a large number of transport rate equations are b(m) 0.55 0.10 2.00 0.38
developed in the literature as use depend on the flow d(m) 0.01 0.00 0.04 0.01
conditions. The probabilistic approach to bedload σ(SD) 1.58 1.00 8.46 1.25
movement (Einstein 1950) offered new insight into
bedload transport processes. However, the level of ρs(kg/m3) 2650.37 1250.00 2810.00 127.35
complexity made application of this method to S 0.02 0.00 0.20 0.04
natural channels very difficult (Yang, 1996). A large
u(m/s) 0.94 0.22 2.88 0.47
number of sediment transport models designed to
describe bedload have been formulated e.g. Meyer- y(m) 0.07 0.01 0.86 0.08
Peter & Müller (1948), Bagnold (1956), Bagnold (m2/s) 0.00 0.00 0.00 0.00
(1966), Einstein (1950), Yalin (1963), Chang et al.,
C(g/m3) 82.83x103 314.20 13.56x105 18.35x104
(1967). Some of the many other transport models
which can be mentioned in this context are: Φ 4.58 0.04 264.05 17.96

BEDLOAD TRANSPORT EQUATIONS Equal-Mobility Approach
Hans Albert Einstein defined a bedload formula as Based on field data from Oak Creek (Milhous, 1973),
"an equation linking the rate of bedload Parker et al. (1982) introduced the equal-mobility
transportation with the properties of the grain and of theory, which states that the threshold conditions for
the flow causing the movement" (Einstein, 1942). each size fraction is independent of grain size. As a
The various equations analyzed in the present work consequence of equal mobility, the bedload transport
out of which some of the representative approaches rate of a certain flow can be calculated by a single
have been discussed as follows: representative grain diameter as the d50 (Parker et al.,
1982).
Shear Stress Approach
The basic assumption is that, when the flow RESULTS AND DISCUSSIONS
conditions exceed the criteria for incipient motion, Although based on different approaches, these
sediment particles on the streambed start to move. equations rely primarily on the same general
The transport of bed particles in a stream is a assumptions. The fluid and sediment properties are
function of the fluid forces per unit area (the tractive steady and uniform having infinite and continuous
force or shear stress) acting on the streambed (Du supply of sediment for which the bedload is
Boys 1879, Shields 1936 & Kalinske 1947). transported and the sediment stored in a reach can be
neglected (Graf 1971; Gomez and Church 1989; Reid
Energy Slope Approach and Dunne 1996). The ability of each approaches in
Sediment transport phenomenon is the result of many predicting bedload rate are calculated against
interacting parameters and one or the other parameter experimental data and the correlation coefficients
can be replaced by a function of some other ones. (R2), root mean squared errors (RMSE, σ) and slope
One example is the wall shear stress, which depends of lines of best fit (slope) are shown in Table 2. The
directly on the slope and the bed roughness. Such a better performance for R2, σ and slope are exist in
methodology has been used in energy slope Graf and Suszki (1987), Einstein-Brown (1950), Gill
approach, which is primarily dependent on the (1972), Van Rijn (1993), Julien (2002) and Wilcock
energy slope (Meyer-Peter et al. 1934 & Meyer-Peter & Crowe (2003). The value R2 of Du Boys (1879),
and Müller 1948). Wilson (1966), Graf (1971), Fernandez-Luque &
Beck (1976), Wiberg & Smith (1989), Soulsby
Discharge Approach (1997), Ribberink (1998), Wong & Parker (2006a,
Bedload transport is defined as the bed material b), Mayer-Peter & Müller (1948), Parker (1979),
weight per unit width per unit time, gb (kg/sec/m) Ashida & Michue (1972), Engelund-Fresoe (1976),
or the volume of the transported bed material per Madsen (1991) and Wilcock (2001) curves are
unit width per unit time, qb (m3/sec/m), given by similar in nature. While Shields (1936), Ashmore
Yalin (1977). Simultaneously, Schoklitsch (1934, (1988), Low (1989), Meyer-Peter (1934), Schoklitch
1943) developed an expression for the critical (1934), Yalin (1963), Engelund & Hassan (1967),
discharge applicability of the bed-load transport Nilson (1992), Fredsoe & Deigaard (1992), Brown &
models. Julien (1995), Camenen & Larson (2006) and
Abraham & Gao (2006) performs separate from
Probabilistic Approach others. Finally, Graf and Suszki (1987) method has
A paradigm change became evident as the turbulence the optimum values of R2, σ and slope which predicts
had to be incorporated in a more explicit form. It was better Φ values consistently close to the experimental
just not enough to take care of the turbulent values. Figure 1 shows the best four comparison of
fluctuations by a mean wall shear stress. It is the the curves (Φcal versus Φobs) proposed by different
merit of Einstein (1942, 1950) who realized this gap investigators with data reported in the various studies
in the transport theories and started incorporating for sediment bedload transport at intense condition.
turbulence. Einstein’s law is defined by observations The figure shows considerable compatibility with the
that the probability that hydrodynamic forces move a observed Φ.
single grain on the bed depends on grain form and
the instantaneous flow condition of the near-wall The experimental data shows a marginally scatter
flow field and the particle moves when the resulting and could be explained as a set rather than well
forces acting on the grain are strong enough defined curves. The discrepancy is primarily due to
(Einstein-Brown 1950) . particle shape, random nature of the entrainment
process, and the difficulty with defining criteria that
Regression Approach adequately capture this feature, but other factors may
Limitation of defining complex process into precise also play a role. These comparisons show that the
mathematical expression, which governs the bedload estimation for bedload transport may be a standardize
transport, has compelled researchers to use the data phenomenon for studying the transport of sediments
driven model (regression, neural network) for over the surface.
explaining bedload transport process (Rottner 1959).

Table 2 Comparison of different approaches for
180
predicting intense bedload rate. 160 Julien (2002)
2
R =0.957, =5.517, slope=0.837
Statistical Parameters 140
Sl. No. Formula R2 σ Slope 120
Calculated
100
1. Du Boys (1879) 0.948 1.379 1.950 80
60
2. Shield’s (1936) 0.936 0.005 5.3x10-4
40
3. Wilson (1966) 0.940 3.910 0.412 20
0
4. Graf (1971) 0.940 2.590 0.275 0 20 40 60 80 100 120 140
Observed 
180
5. Fernandez-Luque and 0.940 1.857 0.196 Van Rijn (1993)
160
van Beck (1976) 2
R =0.954, =5.660, slope=0.832
140
6. Graf and Suszka 0.960 7.202 1.084
(1987) 120
Calculated
100
7. Ashmore (1988) 0.930 1.060 0.095
80
-4 -5
8. Low (1989) 0.890 1.8x10 1.6x10 60
9. Wiberg and Smith 0.940 1.857 0.196 40
(1989) 20
10. Soulsby (1997) 0.940 1.681 0.176 0

0 20 40 60 80 100 120 140
Observed 
11. Ribberink (1998) 0.948 3.510 0.438
180
12. Wong and Parker 0.945 1.568 0.187
(2006a) 160 Julien (2002)
2
R =0.957, =5.517, slope=0.837
13. Wong and Parker 0.940 1.278 0.136 140
(2006b)
120
14. Meyer-Peter (1934) 0.915 0.267 0.027
Calculated 
100
15. MPM (1948) 0.940 2.606 0.275
80
16. Schoklitsch (1934) 0.916 6.5x10-5 6.5x10-6 60
17. Einstein-Brown 0.965 24.448 3.775 40
(1950) 20
18. Gill (1972) 0.956 2.253 4.063
0
0 20 40 60 80 100 120 140
19. Parker (1979) 0.941 3.338 0.375 Observed 
180
20. Yalin (1963) 0.924 3.426 0.343 Van Rijn (1993)
160 2
R =0.954, =5.660, slope=0.832
21. Engelund and 0.930 12.512 1.358 140
Hansen (1967)
120
22. Ashida and Michue 0.944 4.275 0.507
Calculated 
(1972) 100
23. Engelund-Fredsoe 0.941 5.257 0.585 80
(1976) 60
24. Madsen (1991) 0.941 0.281 0.031

40
25. Nilson (1992) 0.939 4.032 0.417 20
26. Fredsoe-Deigaard 0.902 5.628 0.528 0

0 20 40 60 80 100 120 140
(1992)
Observed 
27. Van Rijn (1993) 0.954 5.660 0.832
28. Brown and Julien 0.932 5.123 0.531 Fig. 1 Comparison between Measured and Observed
(1995) Bedload Transport Rate for best four formulae at
29. Nino and Garcia 0.941 14.627 1.629 Intense Condition.
(1998)
30. Julien (2002) 0.957 5.517 0.837 CONCLUSIONS
31. Camenen and Larson 0.768 5.297 0.165 Bedload transport formulae are based on the idea that
(2006) a relationship exists between hydraulic conditions,
32. Abrahams and Gao 0.928 4.582 0.512 the sediments present and the sediment transport rate
(2006) (Gomez and Church, 1989). However, there are
33. Wilcock (2001) 0.941 0.326 0.037 many factors, primarily related to the temporal and
34. Wilcock and Crowe 0.950 0.206 0.028
spatial resolution and accuracy of observations in
(2003) real rivers that confound this relationship for
example local shear stress and description of
sediment and surface structure. Hassan and Church

(2000) comment that in low transport regimes, Brownlie W. R. (1981). Compilation of alluvial
transport rate is "exceedingly sensitive" to bed channel data: laboratory and field. Report No.
surface grain size and structure. Consequently, KH-R-43B, W. M. Keck Laboratory of
transport equations are often considered unreliable Hydraulics and Water Resources, California
and the results variable. This paper compares most Institute of Technology, Pasadena, California.
widely used empirical bedload methods based on the Camenen B., Bayram A., Larson M. (2006).
concept of sediment transport for predicting the Equivalent roughness height for plane bed under
intensity of bedload rate. The primary aim of the steady flow. Journal of Hydraulic Engineering,
present investigation is to study the degree of 132(11), 1146-1158.
deviancy of calculated bedload rate with the Chang F.M., Simons D.B. and Richardson E.V.
measured value. Towards this objective, an extensive (1965). Total Bed-Material Discharge in Alluvial
set of existing experimental data set on bedload Channels. U.S. Geological Survey Water-Supply
sediment are used to find the most successful Paper 1498-I.
prediction for measured bedload transport rate at Du Boys M.P. (1879). Le Rhone et les Rivieres a Lit
intense conditions. Of these proposed curves it may affouillable. Mem. Doc., Ann. Pont et Chaussees,
be indicated that Graf and Suszki (1987) curve ser. 5, 18.
having high R2=0.960, a comparatively σ =7.202 and Einstein H.A. (1942). Formulas for the transportation
slope=1.084. From the above thirty three methods it of bedload. Trans. Am. Soc. Civil Engrs., 107.
may be concluded that the proposed methods are Einstein H.A. (1950). The Bedload Function for
statistically precise with the observed (measured) Sediment Transportation in Open Channel Flows.
bedload sediment transport. U.S. Dept. Agric., Soil Conserv., T.B. no. 1026.
Engelund F. and Hansen E. (1967). A monograph on
ACKNOWLEDGEMENTS sediment transport in alluvial streams. Technical
The authors gratefully acknowledge the financial University of Denmark, T. Forlag, editor, 62.
support that was received from the Department of Engelund F. and Fredsoe J. (1976). A sediment
Science and Technology, Govt. of India (SERC- transport model for straight alluvial channels.
DST: SR/S3/MERC/005/2010) to carry out the Nordic Hydrol, 7, 294-298.
research work presented in this paper. Fernandez L.R. and Beek R. (1976). Erosion and
transport of bedload sediment. J. Hydraulic Res.,
REFERENCES 14(2), 127-144.
Abrahams A. D. and Gao P. (2006). A bed-lod Garde R.J. and Ranga raju K.G. (2008). Mechanics
transport model for rough turbulent open-channel of Sediment Transportation and Alluvial Stream
flows on plane beds. Earth Surface Processes and Problems. New Age International (P) Limited
Landforms, 31, 910-928. Publications, Third Edition, 181-225.
Ackers P. and White W.R. (1973). Sediment Gill M. A. (1972). Erosion of sand beds around spur
transport: A new approach and analysis. J. dikes. J. Hyd. Eng. Div. ASCE, 98(9), 1587–
Hydraul. Div. ASCE, 99(11), 2041–2060. 1602.
Ashida K. and Michiue M. (1972). Study on Gomez B. and Church M. (1989). An assessment of
hydraulic resistance and bedload transport rate in bed load transport formulae for gravel bed rivers.
alluvial streams. Transactions, Japan Society of Water Resour. Res., 25(6), 1161–1186.
Civil Engineering, 206, 59-64. Gomez B. (2006). The potential rate of bed-load
Ashmore P. E. (1988). Bedload transport in braided transport. PNAS, 103(46), 17170–17173.
gravel bed stream models. Earth Surface Goodwin P. (2004). Analytical solutions for
Processes and Landforms, 13, 677-695. estimating effective discharge. J. Hydraul. Eng.,
Bagnold R. A. (1956). The flow of cohesionless 130(8), 729–738.
grains in fluids. Philos. Trans. R. Soc. London, Graf W. H. (1971). Hydraulics of sediment transport.
Ser. A, 249, 235– 297. New York.
Bagnold R.A. (1966). An approach to the sediment Graf W.H. (1998). Hydraulics of Sediment
transport problem from general physics. US Transport. Water Resources Publications, Third
Geological Survey Professional Paper 422-I. Printing, 123-159.
Beheshti A.A. and Ashtiani B. A. (2008) Analysis of Graf W. H. and Suszka L. (1987). Sediment transport
threshold and incipient conditions for sediment in steep channels. Journal of Hydrosciences and
movement. Coastal Engineering, 55, 423–430. Hydraulic Engineering, 5(1), 11-26.
Bhattacharya B., Price R.K. and Solomatine D.P. Hassan M.A. and Church M. (2000). Experiments on
(2007). A machine learning approach to modeling surface structure and partial sediment transport on
sediment transport. ASCE Journal of Hydraulic a gravel bed. Water Resources Research, 36(7),
Engineering, 133(4), 440-450. 1885-1895.
Brown C. B. (1950). Sediment transportation. Julien P. Y. (1995). Erosion and sedimentation.
Engineering Hydraulics, H.Rouse, New York, Cambridge University Press, 280.
Wiley, 769-857.

Julien P. Y. (2002). River Mechanics. Cambridge Recking A. (2010). A comparison between flume and
University Press, 434. field bedload transport data and consequences for
Kalinske A.A. (1947). Movement of Sediment as surface-based bedload transport prediction. Water
Bed Load in Rivers. AGU, 28, 4. Resources Research, W03518, 46, 1-16.
Leopold L. B. (1994). A view of the river. Harvard Reid L.M. and Dunne T. (1996). Rapid evaluation of
University Press, Cambridge, Mass. sediment budgets. Cremlingen, Catena.
Low H.S. (1989). Effect of sediment density on bed- Reid I., Bathurst J. C., Carling P. A., Walling D. E.
load transport. Journal of Hydraulic Engineering, and Webb B. W. (1997). Sediment erosion,
115, 124–138. transport and deposition, in Applied Fluvial
Madsen O.S. (1991). Mechanics of Cohesionless Geomorphology for River Engineering and
Sediment Trans-port in Coastal Waters. Proc. Management. edited by C. R. Thorne, R. D. Hey,
Coastal Sediments. ASCE, New York, USA, 15- and M. D. Newson, John Wiley, New York, 95-
27. 135.
Meyer-Peter E., Favre H. and Einstein H. A. (1934). Ribberink J.S. (1998). Bed load transport for steady
Neuere Versuchsresultate über den and unsteady oscillatory flow. Coastal
Geschiebetrieb. Schweizerische Bauzeitung, Engineering, 34, 59–82.
Zürich, Switzerland, 103(13), 147-150 (in Rottner J. (1959). A formula for bed load
German). transportation. La Houille Blanche, 14(3), 285-
Meyer-Peter E. and Muller R. (1948). Formulas for 307.
Bed-Load Transport. 2nd Conference IAHR Soulsby R.L. (1997). Dynamics of Marine Sands. A
Congress, Int'l. Assoc. for Hydraul. Res., Manual for Practical Applications. Thomas
Stockholm, 39-64. Telford, London, UK.
Milhous R. T. (1973). Sediment transport in a gravel- Soulsby R.L., Damgaard J. S. (2005). Bedload
bottomed stream. Ph.D. dissertation, Dep. of Civ. sediment transport in coastal waters. Coastal
Eng., Oreg. State Univ., Corvallis, 232. Engineering, 52, 673–689.
Nino Y. and Garcia M.H. (1994). Gravel saltation-2, Schoklitsch A. (1934). Der Geschiebetrieb und die
Modeling. Water Resources Research, 30(6), Geschiebefracht. Wasserkraft und
1915-1924. Wasserwirtschaft., 29(4), 37-43.
Paintal A.S. (1971). Concept of critical shear stress Sheilds A. (1936). Anwendung der
in loose boundary open channel. Journal of Ahnlichkeitsmechanik und Turbulenz-forschung
Hydraulic Division, ASCE, New York, USA, 112, auf Geschiebebewegung. Mitteilungen der
8. Preuss. Versuchsanst. f. Wasserbau u. Schiffbau,
Paphitis D. (2001). Sediment movement under Heft, Berlin, 26.
unidirectional flows: an assessment of empirical Van Rijn L.C. (1984a). Sediment Transport, Part I-
threshold curves. Coas. Eng., 43, 227–245. Bed Load Transport. Journal of Hydraulic
Parker G. (1979). Hydraulic geometry of active Engineering, 110(10), 1431-1456.
gravel rivers. Journal of Hydraulic Engineering, Van Rijn L. C. (1984b). Sediment transport, Part III-
105(9), 1185-1201. Bed Forms and alluvial roughness. Journal of
Parker G., Klingeman P.C. and Mclean D.G. (1982). Hydraulic Engineering, 110(12), 1733-1755.
Bedload and size distribution in paved gravel-bed Van Rijn L. C. (1987). Mathematical modelling of
streams. J. Hydraul. Div., 108, 544–571. morphological processes in the case of suspended
Parker G. (1990a). Surface-based bedload transport sediment transport. Doc. Thesis, Delft University
relation for gravel rivers. Journal of Hydraulic of Technology, Delft, the Netherlands.
Research, 28, 417-436. Van Rijn L.C. (1993). Principles of sediment
Parker G. (1990b). The ACRONYM series of transport in rivers, estuaries and coastal seas.
PSACAL programs for computing bedload Aqua Publications, Amsterdam, the Netherlands.
transport in gravel rivers. External Memorandum Wong M. and Parker G. (2006). Re-analysis and
M-220, St. Anthony Falls Laboratory, University correction of bed load relation of Meyer-Peter
of Minnesota, 124. and Muller using their own database. Journal of
Pica M. (1972). Su alcuni aspetti del trasporto solido Hydraulic Engineering, Submitted, 132, 1159-
in alvei torrentizi. L’Energia Elettrica, 8, 497– 1168.
508. Wiberg P. and Smith J. D. (1989). Calculation of the
Recking A. (2006). An Experimental Study of Grain critical Shear stress for motion of uniform and
sorting effects on Bedload. Ph.D. Thesis, French heterogeneous sediments. Water Resources
engineering university - INSA Lyon. Research, 23(8), 1471-1480.
Recking A. (2009). Theoretical development on the Wilcock P.R. (2001). Toward a practical method for
effects of changing flow hydraulicson incipient estimating sediment-transport rates in gravel-bed
bed load motion. Water Resources Research, rivers. Earth Surf Process Land, 26, 1395–1408.
W04401, 45, 1-16.

Wilcock P. R. and Crowe J. C. (2003). Surface-based
transport model for mixed-size sediment. Journal
of Hydraulic Engineering, 129(2), 120-128.
Williams G. P. (1970). Flume width and water depth
effects in sediment transport experiments. U. S.
Geological Survey, Professional Paper 562-H.
Wilson K. C. (1966). Bed load transport at high shear
stress. Journal of the Hydraulics Division, ASCE,
6, 92.
Yalin M. S. (1977). Mechanics of Sediment
Transport. 2nd edn., Pergamon Press, Oxford,
UK.
Yang C.T. (1996). Sediment Transport: Theory and
Practice. McGraw-Hill Series in Water Resources
and Environmental Engineering, ISBN 0-07-
114882-5, Singapore.

ISBN 978-81-921121-0-7
ANALYSIS OF ROUGHNESS AND CRACKING ON SELECTED

FLEXIBLE PAVEMENT IN BANGALORE CITY
Prathima G
IV Sem M.Tech, DSCE, Bangalore, prathivijay76@gmail.com
ABSTRACT:The road surfaces are assumed to be good and even when it is free from any undulations, depressions
and any kind of cracks. These pavements undergo continuous progressive changes from the day one of its
construction since it will be opened to traffic immediately. Along with the traffic volume the climatic factors are
considered as the important parameters influencing the pavement performance condition. With the influence of these
parameters the pavement deteriorates progressively over a period of time. Thereby both the structural and functional
adequacy of the pavement to serve the required traffic is an essential criterion. The structural deterioration of the
pavement is evaluated with respect to various kinds of distresses like crack, patch, ravel etc while the functional
evaluation of the pavement is conducted with the unevenness measurements, riding comfort of the people. The
present study is planned for the future prediction of pavement distresses both structurally and functionally. The
variables such as crack area and unevenness are considered in the study. The various field studies are conducted
periodically on the selected stretch of road during different seasons of the year. The data collected are rebound
deflection studies, traffic volume increments in terms of commercial vehicles per day, crack area and other surface
distress measurements, the unevenness and riding quality measurements are made up in the analysis.
The collected data was analyzed by the development of empirical relationship (empirical model). The crack criteria
on the selected test pavement when observed for ‘Do – Nothing’ strategy showed the response within the
permissible limits while the roughness parameter displayed a wide range beyond the permissible limit. The
empirical relationship cannot be relied on all condition data. The real data observed for long term future predictions
is an essential criterion.
INTRODUCTION high stresses in the plane, due its roughness. It may or

Pavement Distress may not be accompanied by structural failure.
The road pavements in India are quickly leading to the
end of their design life due to the increase in wheel load Pavement Performance and Deterioration
applications and as well the increase in wheel load The ‘Pavement Performance’ by itself is considered
repetitions. Thereby this has lead to the increase in rate as one of the complex phenomenon since it involves a
of deterioration of the pavements. Along with the large variations in many of the features such as material
structural deterioration the functional condition of the characteristics, traffic volume, the climatic and
pavement surface also needs to be evaluated from time environmental factors, the quality of construction and so
to time which hereby indicates user’s point of view. on. These varying features are termed as ‘performance
From the user’s point of view it should provide safe, variables’. The performance variables enhances the
easy, comfort and fast movement of the vehicles at a deterioration process such as roughness, pavement
very low vehicle operation cost. The periodical deflection and different types of surface distress viz.,
evaluation of the structural and functional deterioration cracks, potholes, bleeding, patchwork etc.
of the flexible pavements leads to the strong pavement ‘Pavement deterioration’ is a sensitive observation
surface exhibiting low signs of failure. that represents the progressive change in the road
It is important to ascertain whether certain types of pavements without any abrupt failure of the structure.
pavement distress are progressive, leading to eventual This means pavement deteriorates over a period of time
failure of the road, or whether they are non-progressive. under the influence of loading and hence it is called as
Distinction will be made between two different types of ‘Structure - Sensitive - System’. The complete reliance
failure. The first of its kind is structural failure, includes on pavement performance represents a static condition
a collapse of the pavement structure or a breakdown of wherein one must await for a relatively long period of
one or more of the pavement components of such time before new applications are involved. The other
magnitude to make the pavement incapable of alternative involves the theoretical equations that are
sustaining the loads imposed upon its surface. The based upon simplified assumptions and at times do not
second, type functional failure wherein the pavement apply to conditions as they exist in the field. It is
can no longer carry out its intended function without thereby good to rely upon both approaches.
causing discomfort to passengers or without causing

Objectives and Scope of the Present Study initiated, immediate remedial action should be taken
Objectives -The main objectives of the study are: else they increase in area ultimately leading to the
 To study the performance of flexible pavements, disintegration of the surface. The cracks develop due to
that are inserviceable condition in terms of the repeated application of load on to the surface which
parameters such as rebound deflection, crack area causes excess of tensile strain in the bituminous layers.
and surface unevenness condition
 To develop empirical equations for structural and Functional Condition of the Pavement
functional condition deterioration of the pavements The performance of the flexible pavement is defined as
 Application of linear and non-linear curves for a functional characteristic of the road surface that is
evaluation of pavement performance for roughness primarily related to level of serviceability that provides
and crack parameters observed over a period of to the users. The user’s opinion on the quality of the
time and traffic volume pavement is described by the surface parameters.
Scope - Thereby pavement unevenness is the principle measure
The present study is planned for pavement performance of pavement condition. This is identified by pavement
evaluation by the analysis of pavement distress roughness parameter. The pavement roughness is
parameters- roughness and cracking on the selected generally defined as an expression of irregularities in
stretch. As the two distress parameters influence both the pavement surface that adversely affect the ride
structurally and functionally on the pavement surface quality of a vehicle and thus the user. The roughness is
the variations are observed under the influence of traffic considered to be an important pavement characteristic
volume, over period of time by the application of because it affects not only ride quality but also vehicle
empirical relationship. delay costs, fuel consumption and maintenance costs
(vehicle operation costs).
Literature Review
Review of Earlier Work on Pavement Performance Road roughness characteristic: It is a feature applied
Reddy B.B. (1996)(1) developed the pavement all along the length of the road and this is defined by a
deterioration models of various parameters such as summary index that measures directly the longitudinal
deflection, roughness and crack criteria for the analysis profile and then applied for a mathematical analysis and
of pavement performance. The crack models for two thereby reducing the profile to the roughness statistic.
types of pavement crust such as (BM+PC) AND One such roughness measure is the International
(BM+BC) were adopted. The functional evaluation of Roughness Index (IRI).
the pavement was also taken into consideration with the
development of deterioration models as Riding Comfort METHODOLOGY
Index model (RCI). Reddy B.B. along with The measure of distress can be either subjective or
Veeraragavan (2001)(2) presented a rational approach objective. The subjective measurement may be a rating
for the maintenance of flexible pavements that are in of high, medium, or low based on visual inspection.
serviceable condition by development of a priority Objective measurements, are generally expensive to
ranking model and ranked the pavements with respect to obtain, it is the use of different types of automated
budget criteria. Reddy B.B. and Sharma along with distress detection equipment i.e. pavement condition
Veeraragavan (2004)(3) put forth the development of rating van for condition surveys.
flexible pavement deterioration models which reflects as
a methodology for pavement preservation. This Crack Measurements: The cracks are measured in
pavement preservation aims at preserving investments terms of Crack Area Measurements. This involves the
which enhances the pavement performance by measurement of crack length and width. They are
extending the pavement life. This is one of the cost- classified into three classes as low, medium and high
effective strategies planned for the pavement treatments based on the cracking area per sqm. If the area is
conducted on the existing roadways. between (0 to 5mm/sqm) it is assumed to be low, if
between (5 to 10mm/sqm) it is medium and (>10mm
STRUCTURAL AND FUNCTIONAL CONCEPTS /sqm) it is taken as high.
Structural Condition of the Pavement
There are many kinds of distresses appearing on the Unevenness Measurements: The Bump Integrator is
pavement surface, but the most prevailing mode of a response type road roughness measuring equipment.
distress in flexible pavement is considered to be fatigue The road roughness measurement made by Bump
cracking. A crack is a common defect or a fault Integrator is a manual method. While the roughness and
observed in bituminous surfaces. This weakens the pavement condition estimation by a pavement condition
surface structure gradually due to the percolation of rating van i.e. road survey equipment called ‘IRSM
water into the surface area. Once these cracks are HAWKEYE 2000 Network Survey Vehicle’

DETERIORATION MODELS
Deflection Criteria Crack /Deflection
0.25
Deflection mm
0.2
1.08
1.06 0.15
y = 0.000x - 0.097
1.04 R² = 0.75
0.1 y = 0.827x - 0.665
1.02
R² = 0.965
1
0.05
0.98
0.96 0
D
C 0.94
0.92
0.1
0.05
0
1700
1700
1800
1800
1900
Traffic Volume cvpd
The performance of deflection is evaluated by using

linear equation. The linear equation is of the form Y =
0.0006x – 0.0979, where Y = Deflection, mm and x =
Traffic Volume predicted as per equation- A = P ( 1 + r
1900
Traffic Volume cvpd

2000
)x . Future predicted deflection data is obtained by this

linear equation thereby varying the traffic volume. The
data is tabulated in table 6.1.
Crack Criteria
(Y)( crack )ti = ( 0.0005xi )traffic + ( 0.8273xi )deflection + (
0.202xi )time – 1.46 --------- (1)
Crack/Traffic
0.25
0.2
0.15
y = 0.000x - 0.845
R² = 0.891
2000
Crack /Time
0.25
0.2
0.15
0.1
0.05
Time
0.75
1
2
3
4
5
6
7
8
9
10
0
(years)
0.25
0.50
0.9
1893
1989
2025
2177
2340
2516
2705
2908
3126
3361
3613
3884
0.95
0.2
Traffic
Volume
(cvpd)
1797
(mm)
0.95
0.95
1.06
1.12
1.21
1.31
1.41
1.53
1.65
1.78
1.92
2.07
2.23
1
Deflection mm
0.4
Time Years
Table.1 Crack Model Data Summary

Deflection
1.05
y = 0.202x + 0.049
R² = 0.891
0.6
The distress parameter crack is dependent on three

1.1
0.8
Crack
(%)
0.11
0.13
0.21
0.68
1.03
1.40
1.77
2.17
2.57
2.98
3.43
3.88
4.35
independent variables i.e. traffic volume, deflection and
515
period of time. The above graphs shows the linear
variation of crack with these three independent
variables. The linear equations obtained from these Roughness/Deflection
2450
graphs helps to build an empirical relationship of the
2400
form 1. By this equation 1 future prediction of crack is
obtainable. The future predicted crack values in percent 2350
are tabulated in the table as shown below. 2300
2250 y = 1436.x + 906.4
The crack percentage seen from the modeled data shows R² = 0.629
2200
a gradual rise in distress to the influence of traffic over a
period of time. 2150
0.9 0.95 1 1.05 1.1
Comments
From the model, the structural parameter crack is Deflection mm
designed for ‘Do-Nothing Strategy’. Do-nothing option
puts no maintenance to the selected road surface from
the period of last overlay. As per the Intervention Levels Roughness/Time
for Urban Roads(12) the maximum permissible cracking 2450
for arterial roads is defined to be 5 percent. This gives a 2400
comparison to the Table 1 of Crack Model data. The 2350
model data shows it is within the permissible limit. Out
of the total test stretch area (say about 58500 m2) the 2300
maximum crack observed at the prediction of tenth year 2250 y = 456x + 2095.
is about 4.35 % that indicates only about 2544 m2. This R² = 0.983
2200
cracked percent displays the Serviceability Indicator of 2150
the test surface. The level of serviceability is good
(Level 1) with respect to cracking criteria. 0 0.5 1
Time Years
Roughness Criteria
(Y)( roughness )ti = (1.1875 xi )traffic + (1436.4xi )deflection + The distress parameter roughness is dependent on three
(456xi )time + 3078 (2) independent variables i.e. traffic volume, deflection and
period of time. The above graph shows the linear
Roughness/Traffic variation of roughness with these three independent
variables. The linear equations obtained from these
graphs helps to build an empirical relationship of the
2450
form eqn 2. By this equation 2 future prediction of
2400 y = 1.187x + 75.72 roughness is obtainable. The future predicted roughness
2350 R² = 0.983 values in terms of unevenness index are tabulated in the
2300 table 2 as shown below.
2250 Table.6.3 Roughness Model Data Summary
2200 Time Traffic Deflection Roughness
2150 (years) Volume (cvpd) (mm) (mm /km)
0.25 1797 0.95 2201
R 1750 1800 1850 1900 1950 2000
Traffic Volume cvpd

0.50
0.75
1
2
3
4
5
6
7
8
9
10
1893
1989
2025
2177
2340
2516
2705
2908
3126
3361
3613
3884
0.95
1.06
1.12
1.21
1.31
1.41
1.53
1.65
1.78
1.92
2.07
2.23
2341
2429
7548
8313
9107
9915
10767
11637
-
-
-
-
516
COMMENTS the higher range i.e. beyond permissible limit at the 1st
From the model, the functional parameter roughness is year as per the analysis.
designed for ‘Do-Nothing Strategy’. Do-nothing option This is a clear indication that either on the basis of
puts no maintenance to the selected road surface from deflection or on roughness value the maintenance
the period of last overlay. As per the Intervention Levels strategy has to be built upon. On priority, the choice of
for Urban Roads (12) the maximum permissible selection of distress parameter for maintenance aspect
roughness for arterial roads is defined to be 2000 depends on the agency who has involved in the
mm/km. When this is compared to the table 2 of contractual work.
roughness model data it shows a wide range in the
output value. The empirical model developed for LIMITATIONS
roughness prediction indicates this large variation in the Any road study cannot be the work of couple of months
data is due to many random features that are influential or a year. The study needs periodical evaluation due to
parameters responsible for rise in roughness. various influencing factors such as environmental
The environmental factors (non-structural effects) impact, seasonal change, and the gradual rise in
influencing roughness includes along with the movement of vehicles etc observed not less than 6
temperature the other primary aspects like moisture months interval of time. Since the present study is
variations and the sub grade movements which cause based on future predictions of pavement distresses by
volume changes or distortion in the pavements. These the development of empirical relationship, it seems to
factors are not directly related to traffic or the pavement be acceptable criteria for certain parameters while for
strength but they do influence roughness progression other parameters it cannot be relied upon. There by the
substantially and are quite difficult to quantify.The real data obtained with due course of observations can
calculated model values indicate Level 3 of the be more reliable.
serviceability indicator. With these high values the
surface cannot be utilized without any maintenance FUTURE SCOPE
aspect. This shows requirement of maintenance This available data can be made use of, in future for
strategies to be undertaken during the first year of its developing the generalized model relationship which
usage. helps for long term prediction of highway projects. If
the work is further continued periodically, the regular
DISCUSSIONS observations help in developing the reliable model.
With the help of mathematical relationship an effort is
made to predict the probable future distresses REFERENCES
(roughness and crack parameter) of the selected Reddy B.B. (1996), “Development of Failure Criteria
pavement surface with change in time and traffic for Flexible Pavements” , Ph. D Thesis, Bangalore
volume. University, Bangalore, December 1996.
The observations showed that a reliable crack value is Reddy B.B. and Veeraragavan, A, (2001), “Priority
obtained from the mathematical relationship that could Ranking Model for Managing Flexible
assist the highway engineers or the pavement Pavements at Network Level”, Paper No. 478,
management studies in designing the probable Journal of Indian Roads Congress, Volume 62-
maintenance strategies. The crack criteria is within the 3, December 2001.
permissible limits as per the Intervention Level for Reddy B.B, Sharma R.S, Veeraragavan A, (2004),
Urban Roads(12). This indicates the pavement surface “Development of Flexible Pavement Preservation
will be in good condition for the future predicted Framework for an Integrated Asset Management”,
duration. Paper No.505, Journal of Indian Roads Congress,
While the observations made with respect to the Volume 65-2.
roughness parameter shows a higher range as compared S.S. Naidu, Dr.P.K. Nanda, Pawan Kalla, Dr. P.K.
to the Intervention Level for Urban Roads(12). It is Kanchan and K. Sitaramanjaneyulu (2005),
necessary the surface needs maintenance repairs to “Pavement Maintenance Management System for
provide the serviceability level. Urban Roads Using Software HDM – 4- A Case
When compared among parameters i.e. deflection, Study”, Journal of Indian Roads Congress, Volume
crack, and roughness values of model data it is observed 66-3, October 2005.
that the deflection reaches the permissible limit at 0.75 Maj. Shailendra Singh and Veeraragavan. A, (2007), “
years with traffic volume equal to 1989 cvpd, the crack Preventive Maintenance: Cost – Effective
is within the permissible limit even at the 10th year Technique to Preserve Indian Highways”, Highway
which means there is no need for immediate Research Bulletin, Journal No. 77, Indian Roads
maintenance while the roughness of model data shows Congress, 2007

Indian Roads Congress (1983), “Geometric Design IRC: 81-1997, Indian Roads Congress, New
Standards For Urban Roads In Plains”, IRC: 86- Delhi.
1983, Indian Roads Congress, New Delhi. Indian Roads Congress (2001),“Guidelines for the
Ralph Haas, W. Ronald Hudson, and John Zaniewski Design of Flexible Pavements”, IRC:37- 2001,
(1994), “Modern Pavement Management”, Krieger Indian Roads congress, New Delhi.
Publishing Company, Malabar, Florida, 1994. Ministry of Road Transport & Highways (2001),
“Manual for Construction and Supervision of
“Pavement Performance Study”, Study on Existing Bituminous Works”, Indian Roads Congress, New
Pavement Sections, Ministry of Surface Transport, Delhi.
Volume-2, September 1994, Central Road Research Ministry of Road Transport & Highways (2004),
Institute, New Delhi. ‘Guidelines for Maintenance Management of
Indian Roads Congress (1997), “Guidelines for Primary, Secondary and Urban Roads’, Indian
Strengthening of Flexible Road Pavements Roads Congress, New Delhi.
Using Benkelman Beam Deflection Technique”,

ISBN 978-81-921121-0-7
INVESTIGATIONS OF MIXING IN MECHANICALLY STIRRED TANK:

COMPARISON OF CD-6 IMPELLER AND RUSHTON TURBINE
Thiyam Tamphasana Devi
Research Scholar, Department of Civil Engineering, Indian Institute of Technology Guwahati
Guwahati-781039, Assam, India. d.thiyam@iitg.ernet.in
Bimlesh Kumar
Assistant Professor, Department of Civil Engineering, Indian Institute of Technology Guwahati
Guwahati-781039, Assam, India.bimk@iitg.ernet.in
ABSTRACT: For proper designing of stirred tank, it is necessary to know the detail mechanism of the
hydrodynamics of fluid inside the stirred tank. The study on the non standard (i.e., CD-6) impeller is found very less
in literature as compare with standard impeller like Rushton turbine. This study contributes on the comparative study
of CD-6 impeller and Rushton turbine. The hydrodynamics of mechanically baffled stirred tank (T=27 cm diameter)
of CD-6 impeller and Rushton turbine (d=9.3cm diameter) mounted at T/3 from tank bottom was being investigated
using commercial software (Fluent®) of computationally fluid dynamics (CFD). The results of the CFD simulation
was compared with published literature studied on Rushton turbine. Multiple reference frames (MRF) impeller
model and realizable k-ε turbulence model coupled with first order upwind differencing discretization schemes were
applied in this simulation. The nature of the flow field and turbulence field at various radial were investigated. It is
learnt that the presented numerical model in this study is underpredicted from the experimental observation but good
accepatable pattern is achieved. From the comparative study of this numerical model, it is concluded that CD-6
impeller gives good mixing property than the Rushton turbine.
Key words: Hydrodynamics, Stirred tank, CFD, Rushton turbine, CD-6, MRF
1. INTRODUCTION much easier and economical means than the use of

In the industrial process like chemical, mineral and experiments (Rao and Sivashanmugam 2010). The
wastewater treatment industries, mechanically stirred accuracy of CFD model depends mainly on various
tanks are widely used for either simple liquid mixing or factors like impeller model, turbulence model,
for more complex multiphase processes. Even a simple discretization scheme, convergence criteria, etc. and
tank is geometrically complex and under most hence for better results one has to select these factors
operating conditions the flow is turbulent. In order to suitably and wisely.
understand the complex phenomena that occur in such
tanks, it is necessary to investigate the flow fields as The property of mixing can be defined indirectly in
well as turbulence field in the vessels. The total mixing terms of flow field and turbulence field. The magnitude
in a stirred tank depends on the mean flow and of these fields depend on the various geometric
turbulence, which in turn also depends on the shape of parameters of stirred tank like shape & size of tank and
impeller (Dewan et al. 2006). In recent years, a more impeller, number of impeller blade, rotational speed of
fundamental approach has been taken to the impeller, impeller submergence depth, type of fluid,
understanding of the mixing process through detailed shape, size & number of baffles, etc. It is well known
studies of the flow pattern created by an impeller that baffled stirred tank possess better mixing than the
(Shekhar & Jayanty 2002). The geometry of an unbaffled stirred tank. This is because of the
impeller of stirred tank plays major role in optimum improvement in mixing of the swirling motion of the
design of stirred tanks (Driss et al. 2010). The optimum fluid producing complex recirculating turbulent flow in
design and the efficiency of mixing operations are the tank due the interaction of the fluid flow with the
important parameters on product quality and production stationary baffles. The velocity and turbulent field were
costs. investigated by Wu and Patteron (1989) experimentally
based on baffled Rushton turbine using Laser Doppler
Computational Fluid Dynamics (CFD) has become a Velocimeter (LDV). They found from their study that
very powerful tool in the process industry not only for the mean velocities, pumping capacities, and turbulence
the research and development of new processes but also intensities in the impeller stream were approximately
for the understanding and optimization of existing ones proportional to the impeller speed. And if these values
(Aubin et al. 2004). Using CFD, one can understand normalized with impeller tip velocity then it become a
the mechanism of mixing of fluids in mixing tanks with function of position in the tank only, not a function of

impeller speed. In this study, the experimental values of The continuity equation is a statement of conservation
axial and radial velocities of Wu and Patterson (1989) of mass.
were compared with the simulated values studied on  
Rushton turbine and CD-6 impeller. The main objective  ( U i )  0 (1)
of this study is to carry out a comparative study on
t  x i
Rushton turbine and CD-6 impeller as the published The momentum equation is a statement of conservation
literature based on CD-6 impeller has been found very of momentum in each of the three component
less as compared to Rushton turbine. directions. Apart from momentum transport by
convection and diffusion, several momentum sources
2. GEOMETRY OF STIRRED TANK are also involved.
The detail of the stirred tank used in this study is taken  ( U i )  p 
 ( U i U j )   
same with the dimension used by Wu and Patterson t xi x i x j (2)
(1989) and is shown in figure 1(a). The diameter of the
stirred tank (T) is 27cm and impeller diameter is (d) U i U j 2U k
[ (    ij )]  g i  Fi
9.3cm mounted at a height of (C) T/3 from tank bottom x j x i 3x k
and open at top. The fluid used is water and filled at a In equation (2), the convection terms are on the left.
height (H) equal to the tank diameter. (i.e., 27cm). The The terms on the right hand side are the pressure
speed of the impeller is 200rpm. The generated gradient, a source term; the divergence of the stress
unstructured grid (835515 cells) is shown in figure 1 tensor, which is responsible for the diffusion of
(b). momentum; the gravitational force, another source
term; and other generalized forces (source terms),
respectively.
k-ε turbulence model is a widely used model in CFD

because of its reasonable accuracy for a wide range of
turbulent condition in an agitated vessel (Rao and
Sivashanmugam 2010). In the turbulent regime,
fluctuations in the mean velocity and other variables
occur, and their effect needs to be incorporated into the
CFD model in order for the model to be able to provide
meaningful results. This is done through the use of a
turbulence model. A process of time averaging the
conservation equations for turbulence is included in
Navier-Stokes equations. When the turbulence is
included the transported quantity (velocity), is assumed
to be the sum of an equilibrium and a fluctuating
'
component, U i  u i . After time-averaging over many
cycles of the fluctuation, terms containing factors of the
(a) (b) fluctuating component average to zero. Hence equation
(2) becomes
Fig. 1. (a) Geometry of stirred tank used in this study (Ui )  p 
(b) Unstructured grid of stirred tank.  (UiU j )    
t xi xi x j (3)
3. METHODOLOGY Ui U j 2Uk 
[(   ij )]  (ui' u jj )  gi  Fi
CFD simualtion was performed to annalyze the velocity x j xi 3xk x j
and turbulence field of stirred tank and velocity fields
Equation (3) is called Reynolds-Averaged Navier-
were compared with the Wu and Patterson
Stokes, RANS equation for momentum.
experimental results. The theory involved in CFD
' '
simulation is discussed in section 3.1. The new term, u i u j (bar indicates average) is called
the Reynolds stresses. The two transport equations that
3.1. Theory need to be solved for the kinetic energy of turbulence,
The two equations involved in computational fluid k, and the rate of dissipation of turbulence, ε:
dynamics (CFD) modelling are conservation of mass
and conservation of momentum (Reynolds-averaged
Navier-Stokes).

 ( k )    k frame and the tank walls are at rest in the stationary
 ( U i k )  (  t )  Gk   frame. The grid used for an MRF solution must have a
t xi xi  k xi
perfect surface of revolution surrounding each rotating
(4) frame. The snapshot model (Ranade 1996) is a steady
state approach that captures the flow field at a single
 (  )    t    2 instant in time, when the impeller position relative to
 ( U i  )  (  )  C1 G k  C 2 
t xi xi   xi k k the baffles is fixed (Marshall & Bakker 2002). The SM
model is a time-dependent solution approach in which
the grid surrounding the rotating components
(5) physically moves during the solution.
C1, C2, σk and σε are empirical constants. Gk is the
generation term; the product of velocity gradients and The order of discretization refers to the convective
depends on turbulent viscosity: terms in the governing equations. The CFD available
discretization schemes are central differencing scheme,
U i U j U j first order and second order differencing scheme, Quick
Gk   t (  )
U j xi x i differencing scheme. Central differencing scheme is
(6) suitable for when the LES (large eddy simulation)
turbulence model is used. First order upwind
k2 differencing scheme is suitable when the convection
 t  C 
 dominates and the flow is aligned with the grid. Second
(7) order is modified version of first order and QUICK is
The solutions for k and ε are used for solving the also similar with second order upwind differencing
turbulent viscosity, μt. Using the values of μt and k, scheme (Marshall & Bakker 2002).
the Reynolds stresses can be calculated in equation (3)
3.2. CFD Simulation
and the new velocity components are used to update the
The CFD software Fluent ® was used to analyze the
turbulence term, Gk (eqn. 6).
fluid flow and turbulence flow of stirred tank. Full tank
dimension was used for the generation of unstructured
The three families of k-ε turbulence model are standard,
grid (835515 cells). The theoretical background
renormalization group (RNG) and realizable. The
involved in CFD simulation has been mentioned in the
standard is a semi-empirical and the two transport
previous section (3.1). The impeller model adopted in
equations that need to be solved for this model are for
this study is MRF as it is suitable for baffled stirred
the kinetic energy of turbulence, k, (eqn. 4) and the rate
tank. Realizable k-ε model is used for turbulence model
of dissipation of turbulence, ε (eqn. 5). RNG is a
as it is suitable for swirling turbulent motion of fluid
modified version of the k-ε model improving the results
flow with first order upwind differencing discretization
for swirling flows and flow separation. The Realizable
scheme. First order upwind differencing discretization
k- ε model is the recently developed model addition to
scheme is a simple model, it does not take time to
the family of k-ε model. The realizable k-ε is
simulate and memory usage is also very less as
considered to be a better model than the standard k-ε
compared with other discretization models. The
for stirred tank flows, as better accounts for flow
convergence criterion for all the equation (continuity,
features such as strong streamline curvature, vortices
velocity, k and ε) considered in this study is 10-4.
and rotation (Gimbun et al. 2009). It contains a new
formulation for the turbulent viscosity, μt (eqn. 7) and a
4. RESULT AND DISCUSSION
new transport equation for the dissipation rate, ε. The
The mean radial and axial velocity normalized by
turbulent viscosity, μt is computed using a variable for
Vtip=πNd measured at radial distance, r=5 & 9cm is
the quantity of Cμ.
shown in figure 2(a-b) and figure 3 (a-b) respectively.
The typical pattern of radial velocity achieved by
Multiple reference model (MRF), snapshot model,
experimental results isobserved similar with the
sliding model (SM) are the most commonly used
simulated values. In case of the experimental and
impeller models in CFD. They mentioned about the
predicted results of Rushton turbine, it is observed that
three main generalized impeller model as MRF
the simulated results is underpredicted from the
(Gentric et al. 2005, Deglon & Meyer 2006, Singh et al.
experimental result mostly at the central and tip region
2007), SM (Shekher & Jayanty, 2002, Zadghaffari et al.
of impeller. This may be due to the creation of
2010) and snapshot (Ranade et al. 1996, 2001). In MRF
complexity of turbulence flow particularly in this
approach, a rotating frame is used for the region of the
region. If we consider only the simulated values of
rotating components while a stationary frame is used
radial velocity for CD-6 impeller and Rushton turbine,
for stationary regions. Impeller is at rest in the rotating

more magnitude is observed in case of CD-6 impeller. In case of axial velocity, the typical pattern achieved
The variation in the values of radial velocity due to the experimentally is little different from the simulated
variation in radial distance can be seen from figure 2(a- result. The CD-6 impeller shows more convincing
b). The maximum magnitude of radial velocity results than the Rushton turbine (figure 3a & 3b). There
measured at r=5cm (exactly near the impeller tip) is is always some discrepancy in CFD simulation when it
found greater than the magnitude measured at r=9cm is compared with experimental results. Axial velocity
(farther distance from the axis of shaft) for both the also found vary with the variation in radial distance and
case of CD-6 impeller and Rushton turbine. The the curve nature is seen near the impeller tip (r=5cm).
maximum mgnitude of radial velocity at radial When the radial distance nearer to the tank wall, the
distance, r=5cm for experimental is 0.75 while the nature of axial velocity tends to straightness. This is
simulated values are 0.5 Vtip amd 0.6 Vtip for Rushtone due to the resistance imposed by the stationary wall of
turbine and CD-6 impeller respectively. tank. It resists the freely movement of fluid particles
inside the tank. Due to the assumption of no slip
condition the flow field near the tank wall is expected
to be neglible and is observed in figure 3b as its radial
distance (r=9cm) is likely to approach near the tank
wall.
(a) r=5cm
(a) r=5cm
(b) r=9cm
Fig. 2. Mean radial velocity at radial position (a) r

=5cm and (b) r=9cm
This discrepancy of radial velocity in the CFD (b) r=9cm

simulation may be corrected by selecting appropiate
convergence criteria or other parameters like impeller Fig. 3. Mean axial velocity at radial position (a) r=5cm
model, turbulence model, discretization scheme, etc. and r=9cm

(a) top view at z=10.5cm (CD-6) (b) top view at z=10.5cm (Rushton)
(c) front view at y=0 (CD-6) (d) front view at y=0 (Rushton)
Fig. 4. Vectors of turbulent kinetic energy, k (m2/s2) showing (a-b) top veiw at just above the impeller blade (i.e.,z=10.5cm
from bottom of tank) and (c-d) front view at the centre of tank (i.e.,y=0) for CD-6 impeller and Rushton turbine
respectively.
The vectors of turbulent kinetic energy, k (m2/s2) just above 5. CONCLUSION

the impeller blade (z=10.5cm) and exactly half of the tank Numerical prediction of Rusthon turbine is comparable with
(y=0) for CD-6 impeller and Rushtone turbine were earlier published experimental results. The radial velocity
presented in figure 4(a-d). turbulent kinetic energy is one of and turbulent kinetic energy around impeller shows the
the most important parameter for defining the mixing mixing capability of that particluar impeller. Non-standard
capacity and performance of a stirred tank (see Wu & impeller such as curved blade shows appreciable increase in
Patterson 1989). The formation of a pair of trailing vortices radial velcoity when compared with Rushton turbine.
behind the blades of Rushton turbine is well known in Simulation also predits more turbulent kinetic energy for
baffled stirred tanks mentioned by Alcamo et al. (2005) and CD-6 than Rushton turbine, which implies that CD-6
also witnessed in this simulation (figure 4d). The magnitude impeller has more mixing capability than Rusthon turbine.
of turbulent kinetic energy in these figures 4a-d shows that
the CD-6 impeller produces more kinetic energy but the ACKNOWLEDGMENT
strong flow path (clear flow lines) is seen in the case of This work was supported by a research grant from Indian
Rushton turbine. Institute of Technology Guwahati, Guwahati
(SG/CE/P/BK/01).

Notation blade turbines design effect on the stirred tank flow
C=height of impeller from tank bottom, m characteristics. European Journal of Mechanics
C1, C2, Cμ, σk, σε =turbulence model constant B/Fluids, 29, 236-245.
CD-6=chemineer disk Gentric, C., Mignon, D., Bousquet J. and Tanguy ,P.A.
CFD=computational fluid dynamics (2005) Comparision of mixing in two industrial gas-
d=impeller diameter, m liquid reactors using CDF simulations. Chemical
Fi=net force in the i direction, N Engineering Science, 60, 2253-2272.
g=gravitational acceleration (m/s2) Gimbun, J., Rielly, C.D. and Nagy, Z.K. (2009) Modelling
Gk=generation term for turbulence, kg/ms3 of mass transfer in gas-liquid stirred tanks agitated by
H=height of fluid (water), m Rushton turbine and CD-6 impeller: A scale-up study.
k=turbulent kinetic energy, m2/s2 Chemical Engineering Research and Design, 87, 437-
MRF=multiple reference frame 451.
N=impeller rotation speed, s-1 Marshall, Elizabeth Marden and Bakker, Andre. (2002)
rpm=revolution per minute Computional Fluid Mixing. Fluent, Incorporated, USA.
T= tank diameter, m Ranade, V.V. and Dometti, S.M.S. (1996) Computational
t=time, s snapshot of flow generated by axial flow impellers in
Vz=axial velocity, m/s baffled stirred vessels. Trans. IChemE, 74(A), 476-484.
ui ’=fluctuating velocity component in the direction i, m/s Ranade, V.V., Perrard, M., LeSauze, N., Zubeb, C., and
Ui=velocity components in the direction i, m/s Bertrand, J. (2001) Trailing vortices of Rushton turbine
Uj= velocity components in the direction j, m/s PIV Measurements and CFD Simulations with Snapshot
Uk=velocity components in the direction k, m/s Approach. Chemical Engineering Research and Design,
Vr=radial velocity, m/s 79(A), 3-12.
Vtip= πdN, m/s Shekhar, S.M. and Jayanti, S. (2002) CFD study of power
z=vertical height, m and mixing time for paddle mixing in unbaffled vessels.
r=radial distance, m Trans IChemE, 80(A), 482-498.
δij=kronecker delta Singh, K.K., Mahajani, S.M., Shenoy, K.T., Patwardhan,
ε=turbulent dissipation rate, m2/s3 A.W. and Ghosh, S.K. (2007) Chemical Engineering
μ=dynamic viscosity of the fluid, Pa.s Science, 62, 1308.
μt=turbulent viscosity, kg/ms Wu, H., Patteron, G.K. (1989) Laser Doppler measurements
π=3.14 of turbulent-flow parameters in a stirred mixer.
ρ =density of fluid, kg/m3 Chemical Engineering Science, 44, 2207-2221.
Zadghaffari, R., Moghaddas, J.S. and Revstedt, J. (2010)
REFERENCES Large eddy simulation of turbulent flow in a stirred tank
Alcamo, R., Micale, G., Grisafi, F., Brucato, A. and driven by a Rushton turbine, Computers & Fluids, 39,
Ciofalo, M. (2005) Large-eddy simulation of turbulent 1183-1190.
flow in an unbaffled stirred tank driven by a Rushton
turbine. Chemical Engineering Science, 60, 2303-2316.
Ankmma, R.D. and Shivashanmugam, P. (2010)
Experimental and CFD simulation studies on power
consumption in mixing using energy saving turbine
agitator. Journal of Industrial and Engineering
Chemistry, 16, 157-161.
Aubin, J., Fletcher, D. F. and Xuereb, C. (2004) Modeling
turbulent flow in stirred tanks with CFD: the influence
of the modeling approach, turbulence model and
numerical scheme. Experimental Thermal and Fluid
Science, 28, 431–445.
Deglon, D.A. and Meyer, C.J. (2006) CFD modelling of
stirred tanks: Numerical considerations. Minerals
Engineering, 19, 1059-1068.
Dewan, A., Buwa ,V. and Durst, F. (2006) Performance
optimizations of grid disc impellers for mixing of single-
phase flows in a stirred vessel. Chemical Engineering
Research and Design, 84(A8), 691-70.
Driss, Z., Bouzgarrou, G., Chtourou, W., Kchaou, H., and
Abid, M.S. (2010) Computational studies of the pitched

ISBN 978-81-921121-0-7
ESTIMATION OF HEAVY METAL CONTAMINATION FROM

MUNICIPAL SOLID WASTE LANDFILL AT KOLKATA USING
EPACMTP MODEL
Sanjukta Basak
Master of Engineering Student, Department of Civil Engineering, Jadavpur University, Kolkata -700032, India.
Email: sanjukta_86@yahoo.co.in.
Subhasish Chattopadhyay
Ph.D. Scholar, Department of Civil Engineering, Jadavpur University, Kolkata -700032, India.Email:
subharpita@rediffmail.com.
Swapan Kumar Mukhopadhyay
Reader, Department of Civil Engineering, Jadavpur University, Kolkata -700032, India. Email: swapanju2003@yahoo.co.in.
Amit Dutta
Reader, Department of Civil Engineering, Jadavpur University, Kolkata -700032, India.
Email: amitt55@yahoo.com; amittdutta@civil.jdvu.ac.in.
ABSTRACT: In developing countries like India, rapid population growth and economic development is resulting in
significant increase in municipal solid waste (MSW). Presently a very crude system of waste management is
followed throughout the country. Disposal of solid waste is generally done in open dumps with no liner or leachate
collection systems. Kolkata; one of the four metropolitan cities in India is generating around 3000 MT/day of MSW.
The city with sufficient precipitation during monsoon coupled with unlined landfill sites give birth to considerable
amount of polluted leachate having high organic and inorganic content, which has high probability of contaminating
soil, surface water and ground water and render the water resource unfit for municipal use. EPA’s Composite Model
for Leachate Migration and Transformation Products (EPACMTP) model has been used to evaluate migration of
heavy metal through the ground-water pathway from MSW landfill site at Dhapa, Kolkata to down-gradient
arbitrary drinking water wells situated in the shallow aquifer. Landfill without any cover or liner system is treated as
a finite depleting source. The model uses Monte-Carlo method to generate a probability distribution of well
concentrations that reflects the variability in the various modeling parameters. Leachate generation from open
dumping ground at Dhapa having an area of 21.4 hectare and almost 20 m height is (900-1000) L/m2/year as
estimated by EPA’s Hydrologic Evaluation of Landfill Performance (HELP) Model. For Kolkata the aquifer
saturated thickness is set as 10 m and the unsaturated zone (silty clay loam soil) thickness or depth to ground water
table as 20m. The aquifer hydraulic conductivity is set at 1890 m/year, hydraulic gradient as 0.0057. MSW of
Kolkata is polluted with heavy metals like Hg (1mg/kg), Cd (3mg/kg), total Cr (43.9mg/kg), Cu (54.4mg/kg) and Pb
(788mg/kg). Average heavy metal concentration in leachate from Dhapa dumping ground is 0.05mg/L Cr, 0.045
mg/L Cd, 0.0055mg/L Hg, 0.075mg/L Pb and 0.2mg/L Cu (KEIP, 2005). Considering leak as 986 L/m2/year and
recharge into the subsurface strata from surrounding as 456 L/m2/year, at a receptor well situated 500m distance
away from the landfill and at 25m depth below ground level, peak concentration of Cr, Hg, Cd, Cu and Pb are
found to be 0.2585 mg/L, 0.002843 mg/L, 0.02326mg/L 0.1034mg/L and 0.03877mg/L respectively and these
concentration remains in the receptor well for ~ 1419, 876.7, 1140, 5144 and 9600 years respectively and then starts
decreasing. When leachate quantity is reduced to 43 L/m2/year by providing good top cover comprising vegetative
soil, drainage layer and barrier clay layer then peak concentration of all these heavy metal are found to be well
within their respective DWS.
Key words: Heavy Metal Pollution, MSW, Landfill Leachate, Groundwater Pollution, EPACMTP Model.
INTRODUCTION country which does not provide liner or leachate

In developing countries the management of municipal collection (Singhal & Pandey, 2001), (Kumar, 2005)
solid wastes is becoming a major problem (Visvanathan and treatment facility and thus pose a reasonable
& Glawe, 2006), (Glawe et al., 2005). Rapid population probability of adverse effects on health and
growth, economic development and modernization are environment (Sahu, 2007).
leading to generation of huge amount waste. India with
1.21 billion populations is generating MSW at the rate India receives huge amount of rainfall and almost 70
of 100 gpcd in small towns and 500 gpcd in large % of the annual rainfall occurs during the monsoon.
towns. Open dumping of waste is done all over the Thus undoubtedly landfill sites produce huge amount

of polluted leachate having high organic and inorganic collection and 50–55% open vats are used in the
content, which can contaminate soil, surface water and present collection system (Chattopadhyay S et al.,
ground water and render the water resource unfit for 2009). Almost all of the waste generated is collected
municipal use (Rao & Shantaram, 2003), (Saha et al., and transported to the disposal site but a very crude
2003). Dumping of electronic goods, electro plating disposal system is followed.
waste, painting waste, used batteries, etc. in MSW In Kolkata also open dumping (without liners and
landfills and is resulting in increased concentration of without a leachate management facility) is carried out
heavy metal in leachates which in turn is polluting the in the Dhapa dumping ground situated in the eastern
groundwater. Electronic goods contribute around 70% fringe of the city. It has been in use for around 20
of the heavy metals in landfills. India is importing years and now it has exhausted. The Dhapa dumping
electronic waste specially computers having high lead, ground has an average area of 21.4 ha. (KEIP, 2005).
mercury and cadmium concentration from Singapore, The waste pile with no intermittent or top cover is
South Korea and US for disposal (Esakku et al., 2003). almost a vertical structure with 20 m average height
Further some small scale industries dumps their wastes (Chattopadhyay S et al., 2007b). Currently some area
in MSW landfills which is also an eminent source of adjacent to this site has been acquired by KMC and
heavy metals. Heavy metals like Cr, Hg, Pb, Cd, Cu dumping is going on over there.
found in leachate are of great concern as they are toxic,
carcinogenic, non biodegradable and can undergo The physical composition of MSW of Kolkata is given
bioaccumulation and biomagnifications in Table 1 (Chattopadhyay S et al., 2009). Density of
(Tchobanoglous. et. al., 1993), (Oman & Junestedt, MSW is 850 kg/m3 (CPHEEO, 2000). The specific
2008). In India almost 61% of the water needs are gravity of the waste is calculated as 1.58, total porosity
fulfilled with ground water but this limited available of the MSW has been calculated as 0.694 (v/v)
water source is getting polluted by migrating leachate (Murthy, 1989) and moisture content as 46 %.
(Iqbal & Gupta, 2009), (Murray et al., 1981),
(Varshney, 2008), (Jhamnani & Singh, 2009), (Mor et Table 1 Average physical composition of municipal
al., 2006), (ESG, 2011), (Sabahi et al., 2009a,b), solid waste of Kolkata
(Yusoff & Al-Hawas , 2008), (Niininen et al., 1994), Components Wet Components Wet
(Bhalla et al., 2011). Weight % Weight %
Compo-sable 50.56 Rubber 0.68
To find out the extent of contamination of ground leather
water, conducting field test is costly and time Paper 6.07 Rags 1.87
consuming. Therefore use of validated models (Mohan Plastic 4.88 Wood 1.15
Glass 0.34 Coconut 4.5
& Muthukumaran, 2004), (Mirbagheri et al., 2009),
Metal 0.19 Bones 0.16
(Rouholahnejad & Sadrnejad, 2009) for this purpose is Inert 29.6
becoming a common practice. EPA’s Composite Model
for Leachate Migration and Transformation Products
MSW of Kolkata is polluted with heavy metals like Hg
(EPACMTP) is one such validated model which has
(1mg/kg), Cd (3mg/kg), total Cr (43.9mg/kg), Cu
been used around the world (Isukapalli, 1999), (Li &
(54.4mg/kg) and Pb(788mg/kg). Kolkata situated at
Benson, 2009), (Reinhart & McCreanor, 1999). 22.340 N latitude and longitude 88.240 E receives
plenty of rainfall and produces huge amount of
Kolkata, one of the four metropolitan cities of India is leachate containing high concentration of organic and
not an exception to the above stated situation
inorganic pollutants (Saha et al., 2003). Natural
(Chattopadhyay et. al., 2009) yet not much work has
leachate collected around the old dumping grounds at
been done on quantitative analysis of leachate and its
Dhapa is found to have 0.43-0.85 mg/L Cr, 0.04-0.05
impact on ground water. EPACMTP has been used to
mg/L Cd, 0.002-0.009 mg/L Hg, 0.07-0.08 mg/L Pb
evaluate migration of metal through the ground-water
and 0.06-0.28 mg/L Cu. Other than heavy metals
pathway from MSW landfill site at Dhapa Kolkata to natural leachate contains high BOD (375-425mg/L),
down-gradient arbitrary drinking water wells situated in COD (2455-2545 mg/L), Cl- (3520-3943mg/L), TKN
the shallow aquifer. (228.6-585.5mg/L), TDS (9912-10186mg/L) and
phenolic compounds < 1mg/L(KEIP, 2005).
STUDY AREA
Kolkata with 9 million populations generates Therefore to find the effect of leachate percolating into
3000T/day of MSW as on 2006 analysis and is
the ground water, the sub surface soil strata of the city
estimated to reach 3750 T/day by the year 2025
has been studied. The vadose zone below the ground
(Chattopadhyay S et al., 2007a). There is no source
surface generally contains 10 to 25m silty clay loam
segregation arrangement, only 60% house-to-house

and its hydraulic conductivity varies between 10-8 to 10- recharge rate was determined for silty loam soil type
5
cm/sec. Below this a fine sand bed is found which is considering average annual rainfall of Kolkata to be
the first shallow aquifer 10 to 20 m thick having 1850 L/m2 using the HELP model. The constant
hydraulic conductivity of around 10-3 cm/sec. Though recharge rate has been set as 456 L/m2/year. For
the second aquifer at a depth of 80 - 90 m is the most Kolkata the aquifer saturated thickness is set as 10 m
potential and exploited aquifer of Kolkata, yet many and the unsaturated zone (silty clay loam soil)
parts of surrounding areas still use the shallow aquifer thickness or depth to ground water table as 20m
as drinking water source. The ground water flow in the considering the base of the landfill on the ground
Dhapa area is predominantly from east to west i.e. from surface (KEIP, 2005). The aquifer hydraulic
landfill to city (KEIP, 2005), (Sengupta, 2009). conductivity is set at 1890 m/year and the hydraulic
gradient is assigned a value of 0.0057. Since Kolkata
The analysis of water samples from the shallow aquifer region has sand and gravel aquifer (KEIP, 2005) non
(19-25 m) around the Dhapa dumping ground clearly carbonate ground water type which represents
brings out the inherent poor chemical quality of the unconsolidated sand and gravel aquifer with a natural
groundwater. The tube wells tapping deeper ground pH of 7.4 has been used.
water at 80-90m below in the Dhapa area have high
concentrations of phenolic compounds at places, high Since the depleting source scenario is most appropriate
Fe, Mn, high TDS (Ca, Mg and Cl content ) and for a landfill waste management scenario, where the
occasionally high Cr (KEIP, 2005). waste accumulates during the active life of the unit,
but leaching may continue for a long period of time
after the unit is closed, therefore metals have been
METHODOLOGY modeled using linearized MINTEQA2 isotherm that
EPACMTP simulates the fate and transport of dissolved calculates a single value of Kd from a nonlinear
contaminants from a point of release at the base of a isotherm (EPA, 2003b).
different waste disposal unit (e.g. landfill), through the The hypothetical drinking water well is assumed to be
unsaturated zone and underlying groundwater, to a located at a distance 500 m away from the edge of the
receptor well at an arbitrary downstream location in the waste unit because as per Solid Waste Management
aquifer. It takes into account various site specific, and Handling Rules in case of landfills no
constituent specific, unsaturated zone specific and development buffer zone should be 500m around the
aquifer specific parameters. Deterministic or Monte landfill (CPHEEO, 2000). Since the contaminant
Carlo methodology can be used to model continuous or plume in the ground water follows a Gaussian plume
finite source (EPA, 200a). pattern, highest concentration is always found along
the plume centre line so distance from the plume
Landfill without any cover or liner system is treated as centerline is kept constant at zero. The well position
a finite depleting source. Monte-Carlo method has been has been fixed at a constant depth i.e. midpoint of the
used to generate a probability distribution of well saturated zone.
concentrations that reflects the variability in the various
modeling parameters. Though Monte-Carlo analysis has *Note : The distance of the well has been varied to
been done yet site specific infiltration, recharge and soil find concentration versus distance curve. Also the
characteristics has been given as a constant value. Thus depth has been varied to find its effect on
a combination of Monte Carlo and Deterministic concentration. Concentrations are found at midpoint of
analysis has been done by setting Logical variable for aquifer having depth 10m.Therefore depth below
Regional Site-based Analysis as False as needed for ground level = (unsaturated zone depth + 5 m) (EPA,
Nationwide Analysis deterministic analysis. Therefore 1997).
Monte-Carlo method has been used with a combination
of constant, derived and distribution of input variables
to find the contaminant concentration at the receptor OBSERVATIONS
well. Chromium concentration in MSW is taken as 43.9
mg/kg and in leachate as 0.5mg/L as reported by KEIP
Since a finite source is being simulated, both the for Dhapa, Kolkata (KEIP, 2005). Therefore ratio of
unsaturated and saturated zone transport modules are chromium concentration in waste to leachate i.e.
implemented in the transient mode and flow in steady Nratio is 280 L/kg. WHO and BIS recommended
state. Infiltration or leakage from 20m high landfill with maximum allowable concentration in drinking water
21.4 ha. area (at Kolkata’s landfill site Dhapa) without for Cr6+ as 0.05 mg/L.
any top cover or liner (KEIP, 2005) is estimated to be
986 L/m2/year using HELP model (EPA,1994a,b), The

The 90 percentile value is taken for further study as it observed at the well. EPACMTP gives the highest
assures that the ground water concentration will be less peak concentration within a time period and we got the
than this value for at least 90 percent time considering highest peak concentration of 0.2858 m/L and the time
the range of variability associated with waste sites. of its occurrence upto 1419 year even when the time
Therefore it provides a large degree of confidence that period or Tcutoff was set to 10,000 years. Therefore it
the results are adequately protective of human health can be inferred that 0.2858 mg/L concentration
and the environment. The percentiles versus peak remains upto 1419 years and after that it stars
chromium concentration are plotted and are shown in decreasing.
the Figure 1.
If concentration verses time is plotted then the curve
It is observed that peak chromium concentration first increases and then becomes constant for a long
appears in the receptor well after a certain period ~ 8 time and then again would decreases thus taking the
years as it takes some time to travel through the entire shape of a plateau as shown in Figure 2.
unsaturated zone and then through the aquifer before
reaching the well.
10 0
80
60
Percentile
40
20
0
0 .1 8 0 .2 0 0 .2 2 0 .2 4 0 .2 6 0 .2 8
P e a k C o n c e n tr a t io n ( m g /L ) Fig. 2 The concentration breakthrough curve for
chromium
Fig. 1 Graph showing percentile versus peak chromium
concentration (mg/L) curve Peak, 30 year and 70 year average receptor well
concentration of chromium with the passage of time is
Therefore as the contaminated ground water plume shown in Figure 3.
reaches the well chromium concentration is observed
and from then concentration starts increasing rapidly
(Bhalla et al., 2011). At around 9.5 years it crosses the 70 Y e ar A ver age
30 Y e ar ave rag e
maximum limit for DWS and then reaches the highest 0 .3 0 P ea k
concentration of 0.2585mg/L at around 30years. After 0 .2 5
that the highest peak concentration i.e 0.2585mg/L

Concentration (mg/L)
0 .2 0
remains constant upto 1419 year. The reason behind is
that the landfill contains huge amount of chromium 0 .1 5
(high Nratio) which takes a long time to deplete. As the 0 .1 0

landfill is considered as a depleting source, it takes
almost 3485.6 years to exponentially reduce leachate 0 .0 5
concentration to 0.05 mg/L i.e DWS. Thus 0 .0 0

0 10 20 30 40 50 60 70 80
contaminants are contributed to the leachate for many Ye ar
years even after the site is closed (Farquhar, 1988). The
decrease of leachate concentration emanating from the Fig. 3 Peak, 30 year average and 70 year average
base of the landfill is very slow and so high leachate chromium concentration variation with time
concentration remains for several years. This time can
be reduced only when total waste concentration in the At an early stage when the curve shows a sharp
landfill is reduced by reducing Nratio, since all other increase a substantial difference between peak and two
variables are constant. average concentrations is obtained as shown in Figure
3. As usual 30 year average is found to be more than a
The landfill almost acts as a continuous source of 70 year average as the exposure period is always
chromium and so due to a constant inflow of chromium selected to be centered about the time when peak
to the ground water a steady state concentration is receptor well concentration occurs. Since a flat curve

remains for a long time during this period the steady The initial chromium concentration is changed by
state concentration, peak concentration and 30 year and changing the Nratio keeping all other parameter
70 year average is found to be same. unchanged; it is observed that ground water
concentration remains same as the initial leachate
As the position of the receptor well is shifted away concentration. Only when Nratio is higher,
from the landfill the chromium concentration also concentration remains in the receptor well for longer
decreases but even at 800 m distance away from the period and vice-verse. If chromium concentration in
landfill it is above the DWS with a concentration of the waste is taken as 140mg/kg i.e. the average
0.1873 as shown in Figure 4. concentration found in Chennai MSW (Esakku et al.,
2003) but in leachate it is kept at 0.5mg/L then the
0.5
50 m
up to 1 41 1 yr
only difference in result will be that 0.2858 mg/L
concentration will remain upto 4450 years and after
0.4 that it will decrease
Peak Concentration (mg/L)
0.3
Mercury concentration in MSW is <1mg/kg and in
500 m
u p to 14 1 9 yr leachate is 0.0055mg/L as reported by KEIP for
0.2
800 m
Dhapa, Kolkata (KEIP, 2005). Mercury concentration
0.1
u p to 14 2 2 yr in MSW is taken as 0.29 mg/kg which is the
average concentration found in Chennai (Esakku et al.,
0.0
0 10 20 30 40 50
2003). EPA (EPA, 2011) and BIS recommended max.
Y ea rs allowable concentration in drinking water for Hg2+ as
0.001mg/L. At around 12 years it crosses the
Fig. 4 Variation of peak concentration of chromium at maximum limit for DWS and then reaches the highest
different distance with time concentration of 0.002843mg/L at around 30years.
After that the highest peak concentration i.e 0.002843
Depth to ground water affects the amount of dispersion mg/L remains constant upto 876.8year and then starts
and dilution that occurs between the base of the landfill decreasing. It takes almost 1,557.79 years to reduce
and the well, and therefore impact concentrations at the leachate concentration to 0.001 mg/L i.e DWS. At a
monitoring well. Decrease in receptor well distance of 800m receptor well concentration is
concentration with increasing depth to ground water is 0.002058 mg/L which is above the DWS.
shown in Table2.
Lead concentration in MSW is 788 mg/kg and in
Table 2 Variation of chromium concentration with leachate as 0.075mg/L as reported by KEIP for Dhapa,
depth from ground level Kolkata (KEIP, 2005). EPA recommended Maximum
Contaminant Level (MCL) i.e the highest level of a
Depth below Peak contaminant that is allowed in drinking water is 0.015
ground level conc. mg/L (EPA, 2011) for lead but BIS recommended it as
(m)* (mg/L) 0.05 mg/L. Lead appears in the receptor well after a
16 0.2758 very long about 200 years from the closer of the
20 0.2743 landfill. After 300 years it peak concentration at the
25 0.2585 well starts increasing rapidly and reaches 0.0285 mg/L
45 0.2459 within 400 years but then it concentration increases
85 0.2456 very slowly and reaches the steady state concentration
105 0.2454 of 0.03877mg/L at around 2000 year and remains
constant upto 9600 year as for a depleting source,
At 500 m distance away from the landfill and at (20+5) leachate concentration decreases exponentially with
m depth below ground level no concentration is found time. It takes almost 73450 years to reduce leachate
if saturated hydraulic concentration of the aquifer is concentration to 0.05 mg/L. S. But at a distance of 500
reduced to as low as 1×10-3 cm/sec, then slowly m lead concentration is 0.03877 mg/L which is within
receptor well concentration increases as flow rate is the acceptable limit.
increased and reaches a peak of 0.2847 mg/L at 1×10-2
cm/sec. After that it again starts decreasing probably Cadmium concentration in MSW is 3 mg/kg and in
due to a dilution effect caused by fast moving waster at leachate as 0.045mg/L as reported by KEIP for Dhapa,
higher hydraulic conductivities (Rouholahnejad & Kolkata (KEIP, 2005). BIS recommended maximum
Sadrnejad, 2009). allowable concentration in drinking water for Cd2+ is
0.01 mg/L but EPA recommended MCL is 0.005 mg/L

(EPA, 2011 for cadmium. Cadmium appears in the From the output obtained from running EPACMTP
receptor well after 10 years from the closer of the model it is observed that metal concentration appears
landfill. At around 20 years it crosses the maximum in the receptor well after a lag period and then
limit for DWS and then increases rapidly and reaches increases rapidly and reaches the peak but remains for
0.02202 mg/L within 40 years but then it concentration several hundred or thousand years. Organic
increases very slowly and reaches the steady state concentration on the other hand remains lesser period
concentration of 0.02326 mg/L at around 100 year and for around 25 years after the closer of the landfill and
remains constant upto 1140 years. It takes 1729.7 years then starts decreasing, as by then the waste almost gets
to exponentially reduce leachate concentration to 0.01 stabilized. Therefore during the active period
mg/L i.e DWS. At a depth of 25 m below the ground providing leachate collection and treatment and bottom
level and a distance of 800m receptor well liner system is a necessity as. However, several
concentration is 0.01685 mg/L which is slightly above researchers have documented that liners eventually
the DWS of 0.01 mg/L. leak. Maximum life of liners (50-100 years) (Lee &
Lee, 1993) thus lined MSW landfills will do no more
Copper concentration in MSW is 54.4 mg/kg and in than postpone groundwater pollution and rather remain
leachate is 0.2 mg/L as reported by KEIP for Dhapa, a threat to groundwater quality forever from heavy
Kolkata (KEIP, 2005). Thus as such copper metals point of view. But if leakage rate is reduced to
concentration in leachate from Kolkata’s dumping 43 L/m2/year by providing complete cover system then
ground is within BIS maximum allowable standard in metal concentrations are found to be safe in ground
drinking water i.e 1.5mg/L and also well within EPA’s water.
MCL of 1.3 mg/L (EPA, 2011).EPACMTP has been
run to check weather the ground water concentrations But if such a landfill is closed and properly covered
remains within the desirable limit of 0.05mg/L as set by infiltration and leachate generation reduces thereby
BIS. Copper also appears in the receptor well after 80 reducing the threat to the quality of groundwater
years from the closer of the landfill. After 100 years it connected to the landfill area.
peak concentration at the well starts increasing rapidly
and it crosses the desirable limit of 0.05 mg/L after 170 CONCLUSIONS
year. It reaches 0.0758 mg/L within 700 years but then Kolkata with sufficient precipitation especially during
it concentration increases very slowly and reaches the monsoon with unlined landfill sites is generating ~
steady state concentration of 0.1034 mg/L at around (900-1000) L/m2/year i.e. ~ 2.14 × 108 L/year leachate
1300 year and remains constant upto 5114 year. from Dhapa. This leachate is highly contaminated with
Therefore it takes almost 6501.2 years to exponentially heavy metals which remain in the receptor well for
reduce leachate concentration to 0.05 mg/L i.e DWS. hundreds or thousands of years. Thus for the future
At a depth of 25 m below ground level and a distance of engineered landfills and for existing open dump sites
800m receptor well concentration is 0.0749 mg/L which throughout the country including the dumpsite at
is above the desirable DWS of 0.05 mg/L. Dhapa, it is recommended that a complete cover
system comprising 60cm barrier clay on the top of
All the above results are obtained at 986 L/m2/year waste followed by 15 cm lateral drainage sand layer
leakage rate. If by providing the complete top cover and 45 cm vegetative soil cover on top of it with fair
system leachate is reduced to 43 L/m2/year then at vegetation should be provided as it reduces leachate
500m distance away from the landfill and at 25 m depth generation by ~ 96% after the closer of the landfill in
below ground level then metal concentrations is found order to prevent contamination of precious
to be safe in the receptor well as shown in Table 3. groundwater due to solid waste disposal.
Table 3 Safe peak rreceptor well concentration of REFERENCES

different metal at 43 L/m2/year leakage rate Bhalla, G., Kumar and A, Bansal, A. (2011).
Leakage Rate 43 L/m2/year Assessment of Groundwater Pollution near
Municipal Solid Waste Landfill. Asian Journal of
Metal DWS Peak Receptor Well
Water, Environment and Pollution,
(mg/L) Concentration (mg/L)
Environmental and Resources Conservation and
Chromium 0.05 0.048 Recycling, Volume 8, Number 1, Pgs 41-51 2011,
Mercury 0.001 0.00053 IOS Press, ISSN 0972-9860 (Print) 1875-8568
Lead 0.05 0.005314 (Online).
Cadmium 0.01 0.0043 Chattopadhyay, S., Dutta, A. and Ray, S. (2007a).
Copper 0.05 0.2227 × 10-5 Existing municipal solid waste management of
Kolkata – Deficiencies and its solutions. Journal of

Indian Association for Environmental Management Mavallipura Illegal Solid Waste Dump Near
34 (3), 161–167. Bangalore –Special Focus on Heavy Metals in
Chattopadhyay, S., Dutta, A. and Ray, S. (2007b). Wastes.
Sustainable Municipal Solid Waste Management for www.esgindia.org/campaigns/Mavallipura/docs/Ma
the City of Kolkata. International Conference on vallipuraSWM.pdf, 1.04.2011.
Civil Engineering in the New Millennium: Farquhar, G. J. (1988). Leachate: production and
Opportunities and Challenges (CENeM-2007), 11–14 characterizaton, Can. J. Civ. Eng. 16.317 - 325
January, 2007, Bengal Engineering and Science (1989).
University, Shibpur, India. Glawe, U., Visvanathan, C. and Alamgir, M. (2005).
Chattopadhyay, S., Dutta, A. and Ray, S. (2009). Solid Waste Management in Least Developed Asian
Municipal solid waste management in Kolkata, India Countries – A Comparative Analysis. International
– A review. International Journal of Integrated Waste Conference on Integrated Solid Waste Management
Management Science & Technology, Waste in Southeast Asian Cities, 5-7 July 2005, Siem Reap,
Management 29 (2009), 1449–1458. Cambodia,
CPHEEO (Central Public Health and Environmental http://static.esgindia.org/campaigns/Mavallipura/doc
Engineering Organization) (2000). Manual on s/MavallipuraSWM.pdf, 26.05.2011.
municipal solid waste management, Ministry of Iqbal, M.A. and Gupta, S.G. (2009). Studies on Heavy
Urban Development, Govt. of India, New Delhi. Metal Ion Pollution of Ground Water Sources as an
Urban Development, Govt. of India, New Delhi. Effect of Municipal Solid Waste Dumping. African
EPA (Environmental Protection Agency) (1994a). The Journal of Basic & Applied Sciences 1 (5-6): 117-
Hydrologic Evaluation of Landfill Performance 122, 2009, ISSN 2079-2034.
(HELP) Model: Engineering Documentation for Isukapalli, S.S. (1999). Uncertainty Analysis of
Version 3. EPA/600/R-94/168b, September 1994. Transport-Transformation Models. Doctor of
EPA (Environmental Protection Agency) (1994b). The Philosophy Thesis, Graduate School—New
Hydrologic Evaluation of Landfill Performance Brunswick Rutgers, The State University of New
(HELP) Model: User’s Guide for Version 3. Jersey.
EPA/600/R-94/168a, September 1994. JJhamnani, B. and Singh, SK. (2009). Groundwater
EPA (Environmental Protection Agency) (1997). EPA’s Contamination due to Bhalaswa Landfill Site in
Composite Model for Leachate Migration with New Delhi. International Journal of Civil and
Transformation Products (EPACMTP): User’s Guide, Environmental Engineering 1:3, 2009.
Office of Solid Waste (5305W),Washington, DC KEIP (Kolkata Environmental Improvement Project)
20460.EPA\R08-99.217.wpd, www.epa.gov/osw. (2005). Rapid Environmental Impact Assessment
EPA (Environmental Protection Agency) (2003a). Report and Environmental Management Plan of
EPA’s Composite Model for Leachate Migration Engineered Landfill at Dhapa. Kolkata Municipal
with Transformation Products (EPACMTP): Corporation, India.
Parameters/Date Background Document, Office of Kumar, S. (2005). Municipal Solid Waste
Solid Waste (5305W), Washington, DC 20460. Management in India: Present Practices and Future
EPA530-R-03-003, www.epa.gov/osw. Challenge. NEERI, August 2005.
EPA (Environmental Protection Agency) (2003b). Lee, G.F. and Lee, A.J. (1993). Groundwater Quality
EPA’s Composite Model for Leachate Migration Monitoring at Lined Landfills: Adequacy of Subtitle
with Transformation Products (EPACMTP): D Approaches. G. Fred Lee & Associates, El
Technical Background Document, Office of Solid Macero, CA (530)753-9630 May 1993.
Waste (5305W), Washington, DC 20460. EPA530-R- http://www.gfredlee.com/Groundwater/GW-
03-006, www.epa.gov/osw. MONITpaper93.pdf, 16.02.2011.
EPA (Environmental Protection Agency) (2011). Li, L., Benson, C.H. (2009). Assessment of the
Water: Basic Information about Regulated Drinking Roadway Module in IWEM Version 2 (Beta). Final
Water Report 09-03, Recycled Materials Resource Center,
Contaminants.http://water.epa.gov/drink/contaminant University of Wisconsin-
s/basicinformation/benzene.cfm, 16.04.2011. Madison,www.recycledmaterials.org, 4.12.2010.
Esakku, S., Palanivelu, K. and Joseph, K. (2003). Mirbagheri, S. A., Monfared, H. and Kazemi, H.R.
Assessment of Heavy Metals in a Municipal Solid (2009). Simulation modelling of pollutant transport
Waste Dumpsite. Workshop on Sustainable Landfill from leachate in Shiraz landfill. Journal of
Management, 3–5 December, 2003, Chennai, India, Environmental Earth Sciences, Volume 59, Number
pp.139-145. 2 / November, 287-296, 2009, ISSN 1866-6280
ESG (Environment Support Group Trust) (2011). (Print) 1866-6299 (Online), Publisher Springer
Report on water quality in the vicinity of the Berlin / Heidelberg.

Mohan, Dr S. and Muthukumaran, M. (2004). 2009, ISSN 1553-345X, © 2009 Science
Modelling of Pollutant Transport in Ground Water. Publications.
IE (I) Journal.EN, Vol 85, September 2004. Saha, R., Chattopadhyay, A. and Naidu, B.R. (2003).
Mor, S., Ravindra, K., Dahiyaa, R. P. and Chandra, A. Pollution Potential of Leachate from East Kolkata
(2006). Leachate Characterization and assessment of Solid Waste Dumpsite. Workshop on Sustainable
groundwater pollution near municipal solid waste Landfill Management, 3–5 December, 2003,
landfill site. Environmental Monitoring And Chennai, India, 303-307.
Assessment, 2006 Jul; 118 (1-3):435-56, Centre for Sahu, A.K. (2007). Present Scenario of Municipal
Energy Studies, Indian Institute of Technology, New Solid Waste (MSW) Dumping Grounds in India.
Delhi,India, Proceedings of the International Conference on
http://www.ncbi.nlm.nih.gov/pubmed/16897556, Sustainable Solid Waste Management, September
27.02.2011. 2007, Chennai, India, 327-333.
Murray, J.P, Rouse, J.V, Carpenter and A.B. (1981). Sengupta, P.K. (2009). Groundwater In Kolkata - An
Groundwater contamination by sanitary landfill Overview of Spatial Distribution Pattern of
leachate and domestic wastewater in carbonate Chemical Parameters. Jadavpur Centre for study of
terrain: Principal source diagnosis, chemical transport Earth Science, Jan,2009,
characteristics and design implications. Water ww.jcses.blogspot.com/2009/01/hydrogeology-of-
Research kolkata.html, 4.02.2011.
Volume 15, Issue 6, 1981, Pages 745-757. Singhal, S. and Panday, S. (2001). Solid Waste
Murthy, V.N.S. (1989). Soil Mechanics & Foundation Management in India: status and future directions.
Engineering. Sai Kripa Technical Consultants, Third TIMES (TERI Information Monitor on
Edition, Vol.I. Environmental Science), Volume 6, Number 1, June
Niininen, M., Kalliokoski, P. and Pàrjàlà, E. (1994). 2001.
Effect of organic contaminants in landfill leachates Tchobanoglous, G., Theisen, H. and Vigil, S. (1993).
on groundwater quality in Finland. Groundwater Solid Waste Management Engineering Principles
Quality Management. Proceedings of the GQM 93 and Management Issues. McGraw-Hill International
Conference held at Tallinn, September 1993, IAHS Ed., P.5.
Publ. no. 220, 1994. Varshney, V. (2008). Ankleshwar's groundwater high
Oman, C. B. and Junestedt, C. (2008). Chemical on heavy metals. Down To Earth, Issue: May 31,
characterization of landfill leachates – 400 2008 www.downtoearth.org.in/Science &
parameters and compounds. Waste Management 28 Technology, 10.10.2010.
(2008) 1876–1891.Published by Elsevier Ltd. Visvanathan, C. and Glawe, U. (2006). Domestic Solid
Rao, J.K. and Shantaram, M.V. (2003). Soil and Waste Management in South Asian Countries – A
Water Pollution due to Open Landfills. Workshop on Comparative Analysis. Presented at 3 R South Asia
Sustainable Landfill Management 3–5 December, Expert Workshop, 30 August - 1 September, 2006
2003; Chennai, India, pp. 27- 38. Kathmandu, Nepal.
Reinhart, D.R. and McCreanor, P.T. (1999). Yusoff, M.K. and Al-Hawas, I.A. (2008). Effect of
Implications of Time/Space Variable Leachate Head Landfill Leachate on Groundwater Quality. The
on Liner Leakage. ICFAI Journal of Environmental Sciences, Vol. 2,
www.floridacenter.org/publications/time_space_var_ No. 1, pp. 32-43, February 2008.
99-9.PDF, 20.01,2011.
Rouholahnejad, E. and Sadrnejad, S.A. (2009).
Numerical simulation of leachate transport into the
groundwater at landfill sites. 18th World IMACS /
MODSIM Congress, Cairns, Australia 13-17 July
2009, http://mssanz.org.au/modsim09.
Sabahi, E.A., Rahim, S.A., Zuhairi, W.Y.W., Nozaily,
F.A. and Alshaebi, F. (2009a). Leachate Composition
and Groundwater Pollution at Municipal Solid Waste
Landfill of Ibb City, Yemen. Sains Malaysiana
38(3)(2009): 295–304.
Sabahi, E.A., Rahim, S.A., Zuhairi, W.Y.W., Nozaily,
F.A. and Alshaebi, F. (2009b). The Characteristics of
Leachate and Groundwater Pollution at Municipal
Solid Waste Landfill of Ibb City, Yemen. American
Journal of Environmental Sciences 5 (3): 256-266,

ISBN 978-81-921121-0-7
BEHAVIOUR OF SUCTION CAISSON AS FOUNDATION SYSTEM

FOR OFFSHORE WIND TURBINE IN SANDY SOILS
Suchit Kumar Patel

Post-Graduate student, Department of Civil Engineering, IIT Guwahati, Guwahati-781039, Email: p.suchit@iitg.ernet.in
Baleshwar Singh
Associate Professor, Department of Civil engineering, IIT Guwahati, Guwahati-781039, Email: baleshwar@iitg.ernet.in
ABSTRACT: Suction caisson can be used as an alternative foundation for offshore wind turbine. Both
monopod and tripod or tetrapod foundations are considered. The main challenging problems encountered are
due to large horizontal forces and overturning moments. This review is about the development of design
methods for the suction caisson as a foundation in sandy soil for offshore wind turbine. Here the attention is
given on installation and design for in-service performance. Some field trials and laboratory studies are reported.
Here also some relevant theoretical works are mentioned. The calculation procedure depends on first predicting
the suction beneath the caisson lid and then further calculating tensile load. Also variation of stiffness with
loading level of foundation is observed.
INTRODUCTION with less steel than that required for an equivalent

With increase in energy demand all over the world, pile foundation resulting in lower material cost.
offshore wind energy is becoming a suitable However the main cost benefit is that the installation
alternative to fulfil the energy demand. Offshore method is much quicker and simpler than for other
wind energy industry is very rapidly expanding solutions. Installation is significantly not weather
sector in the world today. Due to limitation of non dependent. Suction caisson can be flexibly adapted to
renewable energy resources, it has very wide scope a variety of structural forms which maximise its
to fulfil energy demand in future. This energy is accessibility and can accommodate a variety of water
environmental friendly and pollution free. The wind depth. Also at the end of wind turbine life, suction
speed is typically higher and steadier in offshore than caisson may be removed completely from seabed
onshore and become steadier with increase in (Byrne et al. 2002).
distance from shore. There is large continuous area
available to expand wind farm industries. Also there Design of suction caisson foundation for offshore
is less wind turbulence which allows the turbine to wind turbines differ in many ways than that of
harvest energy more effectively. It also reduces the typical oil and gas industries foundations (Houlsby et
fatigue load effect on turbine. In offshore case there al. 2005). Firstly the structures are installed in
is not any chance for noise and visual problems shallow water depth up to 40m. Also the structures
which affect the local people. But there is huge are relatively light. In wind turbine case vertical
investment and high risk involve in this project. So it loading on foundation is relatively small, while
is of greater need for development of economically horizontal and moment loadings are much larger and
and technically optimum solution for the installation also dynamic in nature. Due to shallow water depth
of it. There are different types of foundation system sometimes local scour and sediment movement occur
that have been developed like monopile, gravity base which is caused by shallow water wave effect. The
and suction caisson. dynamic interaction between the wind-induced
loading of turbine rotor and the wave induced
Suction caissons have been extensively used as loading on the structure result in complicated
anchor and foundation for small number of offshore dynamic loading on foundation.
platforms for oil and gas industries. They are
presently being considered as a possible Suction caisson is of two type monopod and tripod or
economically alternative foundation option for tetrapod. The monopod resists the overturning
offshore wind turbines as discussed by Houlsby & moment directly by its rotational fixity in the seabed.
Byrne (2000). Suction caissons are large cylindrical As turbine becomes larger, monopod design may
structures open at the base and closed at the top. The become sufficiently large and become uneconomical.
installation of suction caisson is done in two steps. The alternative solution found for this is tripod or
Initially it is penetrated under its self-weight in tetrapod which has multiple smaller foundations. In
seabed. Afterwards suction is applied to penetrate it either of these foundation configurations the
to desired depth. The suction is created by pumping overturning moment of the structure is resisted by
out the water from the cylindrical compartment. It push-pull action of opposing vertical loads on the
creates pressure differential on the lid of caisson upwind and downwind foundations. The key design
which effectively increases the downward force on issue for monopod is the overturning moment, whilst
foundation. These are simple steel structures deigned for tripod or tetrapod it is tension in upwind leg. It is

important to recognise that the design of a turbine
foundation is not usually governed by consideration The frictional terms are calculated in similar way as
of ultimate capacity, but is typically dominated by in pile design. The end bearing is taken as the sum of
consideration of stiffness of foundation and Nq and Nγ term in conventional way and it is
performance under fatigue loading. In sand applied assumed that solutions for a strip footing of width t
suction also generates flow within the soil along with are appropriate for the caisson rim (Houlsby et al.
downward movement of skirt of foundation. Pore 2005). If no account is taken of the enhancement of
pressure gradients are beneficial to the installation vertical stress close to the pile due to frictional forces
process and must be accounted for in design further up the caisson, the vertical load on caisson for
calculation. Under dynamic loading accumulated penetration to depth h in the absence of suction is
pore pressure even leads to liquefaction. With given as
degradation of the sand layer strength and drainage
of pore water the foundation and sand surrounding it
will sink gradually, even pore pressure may further
rise which lead to liquefaction of sand surrounding Outside friction Inside friction
foundation. (1)
End bearing
This paper is a review of main problem studied
during design of suction caisson in sandy soils. The If the effect of stress enhancement is ignored, it leads
two main problems studied for design of suction to an underestimation of the force and suction
caisson as foundation are (a) installation and (b) in- required for full penetration. If this effect is
service performance. considered with uniform effective stress assumed
within the caisson and outside, then Eq. (1) can be
INSTALLATION expressed as
The principal difference between installation of
suction caisson for an offshore wind turbine and for
previous application is that the turbines are likely to
be installed in much shallower depth. The installation
process can be broken in two components. (a) Self-
weight penetration and (b) suction installation. The + (2)
self weight penetration is important to insure seal at
edge of foundation in order to complete suction where
installation adequately. For design it is important to
understand the interaction between soil density, the
skirt wall thickness and the effective vertical load
acting on foundation. The designer will have to Suction Installation
predict the required suction as a function of Let the pressure in caisson be s with respect to
penetration depth. This can be used to assess pump seabed water pressure. If pa is atmospheric pressure,
capacity and rate of suction to be applied. Also absolute pressure inside caisson is pa + γwh w – s.
designer should be aware of any design limitation Excess pore pressure developed at caisson tip is as
like maximum aspect ratio to avoid the possibility of where a is pore pressure factor. Absolute pressure at
liquefaction. the tip is pa + γw (hw + h) – as. This leads to
downward hydraulic gradient of as/γwh on the outside
Self-weight Penetration of caisson and upward hydraulic gradient of (1 –
A typical outline of suction caisson is shown in Fig. a)s/γwh on the inside. Due to this γ’ in Eq. (2) is
1 . The resistance on the caisson is calculated as the replaced by γ’ + as/h outside and by γ’ – (1- a)s/h
sum of outside and inside friction and end bearing inside the caisson during the calculation of vertical
capacity. stress as in Eq. (3).
(3)
Limit of Suction Installation

With increase in suction the upward hydraulic
gradient inside the caisson reaches a critical value at
Fig. 1 Outline of suction caisson (after Houlsby et al. which vertical effective stress inside the caisson at
2005) tip falls to zero and condition for liquefaction is

generated. Further increase in the suction may
induce local piping with a major water inflow to (5)
caisson without further significant penetration of
caisson. At this condition γ’h – (1 – a)s = 0 which where h = height of load acting above foundation, D
gives s = γ’h/(1- a). Substituting this value in Eq. (3) = caisson diameter and m and p are index for model
leads to and prototype. Each test had three phases.
1. Installation phase: The caisson was installed

with suction and CPT tests were performed
(4) before and after the caisson installation.
2. Loading phase: Wind turbine tower was
By solving Eq. (4), the maximum depth of mounted on caisson top and the loading on
penetration for a given suction can be determined. caisson was done by pulling the tower
horizontally. The combined loading was
Comparison with Field Observation controlled by changing loading height.
A first major structure was installed using suction 3. Dismantling phase: The caisson was removed by
caisson in dense sand at Draupner in the North Sea applying overpressure inside caisson.
by Statoil in 1974. Fig. 2 shows the theoretical
prediction of suction compared with the range of The moment loading test was done in two ways.
field measurement as obtained on the site. It can be Firstly small amplitude loading with relatively high
seen that there is a good correlation found between frequency was applied. Secondly larger amplitude
the case record and the calculated suction pressure. but lower frequency cycles were applied. Fig. 3
shows the hysteresis loop obtained from series of
cyclic tests at different amplitudes. It is clear that as
cycle becomes larger the stiffness reduces but
hysteresis increases.
Further jacking test was performed with much larger

amplitude of cycle under quasi-static conditions. At
larger amplitude of rotation the secant stiffness
decreases and hysteresis increases as the amplitude
increases. The unusual shape of hysteresis loop at
very large amplitude is due to gapping occurring at
the side of the caisson. Fig. 4 shows the moment
Fig. 2 Calculated and measured suction at Draupner rotation responses of jacking test.
E (after Houlsby and Byrne, 2005)
IN-SERVICE PERFORMANCE
A large number of tests have been carried out to
study the performance of a caisson under moment
loading at relatively small vertical loads as relevant
to wind turbine design.
Field Test on Monopod

The largest test involved the instrumented Vestas
V90 3.0 MW prototype turbine at Frederikshav,
Denmark (Houlsby et al. 2005). The test program
involving the prototype was focusing on long term
deformations, soil structure interaction, stiffness and
fatigue. Only model analysis was used to analyze the Fig. 3 Moment rotation response of caisson at
structure behavior of wind turbine during various progressively higher amplitudes of loading SEMV
operational conditions. The large scale tests test (after Houlsby et al. 2006)
employed loading by applying a horizontal load at
fixed height, under constant vertical load. The large
scale tests were complemented by laboratory studies.
The laboratory and large scale test were intended to
model the prototype. Scaling of the tests was
achieved by Eq. (5).

Fig. 4 Hysteresis loop from hydraulic jacking test
(after Houlsby et al. 2005) Fig. 6 Load against displacement during pullout test
(after Houlsby et al. 2006)
The secant stiffness deduced from both test were
combined in Fig. 5 and are plotted against the
amplitude of cyclic rotation. The two tests give a
consistent reduction pattern of shear modulus with
strain amplitude, similar to that obtained from
laboratory tests.
Fig. 7 Measured and computed friction during

pullout (after Houlsby et al. 2006)
Vertical cyclic load tests were performed on multiple

caisson and the performance was investigated under
cyclic vertical loading (Kelly et al. 2004). Fig. 8
shows the caisson performance under different
Fig. 5 Shear modulus against rotation amplitude vertical cyclic loading. Small amplitude cycles show
(after Houlsby et al. 2005) stiff response while large amplitude cycles show
more hysteresis and more accumulated displacement
At the end of jacking tests pullout tests were per cycle. As soon as cycle goes to tension, a much
performed. The test result is shown in Fig. 6. The softer response is absorbed and hysteresis loops
tensile load decreases rapidly. The proportion of acquire banana shape. During design this soft
tensile load generated by suction pressure inside the response should be avoided. These observation
caisson is also shown. The difference between total means that tension must be avoided in design of
load carried by caisson and suction load represents tripod or tetrapod foundation for wind turbine. In
friction acting on caisson skirt. The frictional load shallower depth avoiding the tension means that
generated on the skirt during pullout of sand is either foundation must have large spacing between
shown in Fig. 7 along with an estimate of friction the footings or the ballasting must be used.
computed using Eq. (6) (Houlsby et al. 2005).
(6)
Field Test on Tetrapod or Tripod

In multiple footing the overturning moment is
resisted by opposing push-pull action on the upwind
and downwind side of foundation. Tripod is the most
obvious multiple footing designs to use and has
advantage of simplicity. But approximate calculation
indicates that the tetrapod structure is usually a more
favourable configuration to avoid tension as it Fig. 8 Cyclic vertical loading of model caisson (after
requires somewhat less materials. Kelly et al. 2004)

CONCLUSIONS
In this paper design calculation is presented which
can be used to assess the suction caisson foundation
installation in sand. The overall design includes self-
weight installation, installation with suction and
limitation of suction installation process. In-service
performance of suction caisson foundation is
analyzed separately for monopod and tripod or
tetrapod. The small vertical load, high horizontal
load and moment are features of offshore wind
turbine. It is found that monopod foundation design
is dominated by moment loading, while in case of
tripod or tetrapod tensile loading is governing factor.
With different loading tests it is found that as vertical
loading amplitude increases, stiffness is reduced and
hysteresis increases. When tension is reached there is
sudden reduction in stiffness. Also with increase in
moment loading, stiffness decreases whereas
hysteresis increases. Pullout test indicates that a
sizeable ultimate tensile resistance can be generated
but is accompanied by extremely large displacement.
REFERENCES
Byrne, B., Houlsby, G., Martin, C. and Fish, P.
(2002). Suction caisson foundation for offshore
wind turbines. Wind Engineering, 26(3), 145-155.
Houlsby, G.T. and Byrne, B.W. (2000). Suction
caisson foundation for offshore wind turbines and
anemometer masts. Wind Engineering, 24(4), 249-
255.
Houlsby, G.T. and Byrne, B.W. (2005). Design
procedure for installation of suction caisson in
sand. Proceeding ICE, Geotechnical Engineering,
158(GE3), 135-144.
Houlsby, G.T., Ibsen, L.B. and Byrne, B.W. (2005).
Suction caisson for wind turbines. Proceeding of
the International Symposium on Frontiers in
Offshore Geotechnics, Perth, Australia, 2005, 75-
94.
Houlsby, G.T., Kelly, B.R. and Byrne, B.W. (2005).
The tensile capacity of suction caisson in sand
under rapid loading. Proceeding of the
International Symposium on Frontiers in Offshore
Geotechnics, Perth, Australia, 2005, 405-410.
Houlsby, G. T., Kelly, B.R., Huxtable, J. And Byrne,
B.W. (2006). Field trial of suction caisson in sand
for offshore wind turbine foundation.
Geotechnique, 56(1), 3-10.
Kelly, R.B., Byrne, B.W., Houlsby, G.J. and Martin,
C.M. (2004). Tensile loading on model caisson
foundation for structure on sand. Proc. ISOPE,
Toulon, Vol. 2, 638-641.

ISBN 978-81-921121-0-7
DYNAMIC ANALYSIS OF MULTISTORIED BUILDING

USING FLAT SLAB
M.S.Janbandhu
Student, M.Tech. Structural Engg.Y.C.C.E Nagpur , Email: maheshsj_1@yahoo.co.in
V.G.Meshram
Asst. Professor, Civil Department, Y.C.C.E Nagpur , Email: meshramvg@gmail.com
ABSTRACT: The seismic effect on the structure is the most important factor while designing the
civil Engineering structures which will cause adverse effect later on. In present study efforts being
taken while analyzing the structure for Earthquake resistance. In this we have taken a live project a
five storied building with flat slab and dynamic analysis is done by Response Spectrum Method
taking all load combinations as per IS-1893.We have used software SAP 2000 for this purpose. Main
emphasis is given on the structural response when flat slab is provided. Behavior of the structure is
studied for four types of models like a flat slab only, Flat slab with head, with drop and with both
head and drop with equal thickness as well unequal thickness. The results of all four types are then
compared for time period, axial forces, shear forces, moments in each direction and the slab panel
moments. The structure is designed as per IS-456-2000 and IS-13920 guidelines.
1.0 INTRODUCTION
In zone of high seismicity, the flat-slab systems are
designed such that slab-column space frame supports
1.1General
gravity loads and the shear walls provide resistance
Flat slab is a reinforced concrete slab supported by
to lateral load (Wey and Durani 1992; Robertson and
columns with, or without drops. The columns may be
Durani1992; Moehle and Diebold 1985). However, it
with, or without, column heads. Flat-slab is one of
is required by the building codes [IS: 456; ACI:
the most widely used systems in reinforced concrete
building 1989] that the gravity load subsystem must
construction because of its high degree of structural
be able to deform with the lateral load resisting
efficiency. It use simple formwork and reinforcing
system without any loss of its load carrying capacity.
arrangements, and requires the least storey height.
Thus, in realty the two subsystems act together.
Although efficient in resisting gravity load, the flat
Furthermore, since the design seismic force
slab system is inherently flexible and can have
recommended by the codes are generally much less
excessive lateral drift when subjected to seismic
than what the structure would experience during a
loading. Its susceptibility to severe damage during
major earthquake , a certain degree of nonlinear
strong earthquakes is well documented (Rosenblueth
response is to be expected.
1986; Hawkins 1980).
Fig 1.1 Without Drop & No Column Fig. 1.2 With Drop & No Fig. 1.3 Without Drop & with
Head Column Head Column Head

1.2 Advantages 1.3.4 Thickness of Flat Slab
Some of the advantages of Flat Slab are:- a) The thickness of the flat slab shall be generally
1. Rapid construction controlled by considerations of span to Effective
2. Maximum design flexibility depth ratios.
3. Economy b) For slabs with drops span to effective depth ratios
4. Headroom height requirements large given in 23.2 shall be applied directly; otherwise the
5. Minimum storey heights span to effective depth ratios obtained in accordance
6. Controlled deflection with Provisions in 23.2 shall be multiplied by 0.9.
7. Optimum clear span For this purpose, the longer span shall be considered.
8. Sufficient lateral displacement capacity The minimum thickness of slab shall be 125 mm.
1.3 Terminology Related With Flat Slab 1.3.5 Drop

a) The drops when provided shall be rectangular in
1.3.1 Column Strip plan, and have a length in each direction not less than
Column strip means a design strip having a width of one third of the panel length in that direction.
0.25 l1, but not greater than 0.25 l2, on each side of b) For exterior panels, the width of drops at right
the column centerline, where I1, is the span in the angles to the non-continuous edge and measured
direction moments are being determined, measured from the centre-line of the columns shall be equal to
centre to centre of supports and l2, is the-span one-half the width of drop for interior panels.
transverse to l1, measured centre to centre of
supports. 1.3.6 Column Head
Where column heads are provided, that portion of a
1.3.2 Middle Strip column head which lies within the largest right
Middle strip means a design strip bounded on each of circular cone or pyramid that has a vertex angle of 90
its opposite sides by the column strip. degree and can be included entirely within the
outlines of the column and the column head, shall be
1.3.3 Panel considered for design purposes.
Panel means that part of a slab bounded on-each of
its four sides by the centre-line of a column or
centre-lines of adjacent-spans.
Fig. 1.4 Flat slab with drop & column

Head Structure
Fig. 1.5 Slab beam Structure

2.0 ANALYSIS OF FLAT SLAB BY RESPONSE
SPECTRUM METHOD
2.1 Introduction to Problem 0.50 0.50 0.50 0.50 0.50 0.50
0.26
C1 C2 C3 C3 C2 C1
FIFTH
1 2 3 4 5 6 FLOOR
0.26
Panel no. 16 Panel no. Panel no. Panel no. Panel no. 6.6m
C5 17 C6 18 19 20
C4 C6 C5 C4 FOURTH
FLOOR
7 8 9 10 11 12 0.26
Panel no. 11 Panel no. 12 Panel no. 13 Panel no. 14 Panel no. 15
6.6m THIRD
C7 C8 C9 C9 C8 C7 FLOOR
0.26
13 14 15 16 17 18
SECOND
Panel no.6 Panel no. 7 Panel no. 8 Panel no. 9 Panel no.10 6.6m FLOOR
0.26
C4 C5 C6 C6 C5 C4
19 20 21 22 23 24
FIRST
FLOOR
GROUNDLEVEL
Panel no. 1 Panel no. 2 Panel no. 3 Panel no. 4 Panel no. 5 6.6m
C1 C2 C3 C3 C2 C1
25 26 27 28 29 30
5.6m 5.6m 5.6m 5.6m 5.6m
Fig.2.1 Line Plan of Structure Fig. 2.3 Sectional Elevation of Flat-Slab

with Drop Structure
0.50 0.50 0.50 0.50 0.50 0.50
0.26
Internal Drop ExternalDrop

0.50 0.50 0.50 0.50 0.50 2.2x2.0m 0.50 1.1x1.0m
FIFTH 0.26
FLOOR
0.26
FIFTH
FLOOR
0.26
FOURTH
FLOOR
0.26
FOURTH
FLOOR
THIRD 0.26
FLOOR
0.26
THIRD
FLOOR
SECOND 0.26
FLOOR
0.26
SECOND
FLOOR
FIRST 0.26
FLOOR
GROUNDLEVEL
FIRST
FLOOR
GROUNDLEVEL
Fig. 2.2 Sectional Elevation of Simple

Flat-Slab Structure
Fig. 2.4 Sectional Elevation of Flat-Slab with
Head Structure

0.50 0.50 0.50 0.50 0.50
Internal Drop
2.2 x2.0m 0.50
External Drop
1.1 x1.0m
2.2.3 Seismic Force Calculation
0.26
FIFTH
FLOOR
Zone II for Nagpur city Z=0.1
0.26
Hence spectra for medium soil \
FOURTH
Importance factor =1.5
FLOOR
0.26 Sa/g=1+15T0.00  T  0.101
Damping 5%
THIRD
FLOOR
0.26
= 2.5 0.1  T  0.55
Soil type medium
SECOND
FLOOR
= 1.36/T 0.55  T  4.0
0.26
R for SMRF=5 & OMRF=3
FIRST
1. Time period in long direction Tx =0.09H/d
FLOOR
GROUNDLEVEL =0.09x21/28 =0.357sec
2. Time period in short direction Ty=0.09H/d
=0.09x21/26.4 =0.368sec
Take Sa/g for both direction = 2.5
Fig. 2.5 Sectional Elevation of Flat-Slab
with Drop & Column Head Structure 2.2.4 Scale Factor
2.2 Modeling of flat slab for Response Spectrum 1 I

f  x =0.15
 Column size (500x500)mm 2 R
 Slab thickness 200 mm without drop
 Live load on all floors 6 kN/m2 (50% for Where, I= Importance factor =1.5
earthquake) R=Response reduction factor=5
 Grade of concrete M25 & steel Fe 415N/mm²
 Brick wall external & internal 230 mm 2.2.5 Response Spectrum for Zone II:-Soil Type II
 Density of brick wall including plaster 20kN/m3 (Medium Soil)
 Density of brick wall including plaster 20kN/m3 Table 2.2 Time Period Vs Acceleration
 Drop size external (1100x1000)mm Period(Sec) Acceleration
 Drop size internal (2200x 2000)mm 0 0.1
 Depth of Head 400mm 0.1 0.25
 Head at bottom (500x500)mm 0.55 0.25
 Head at top (1200x1200)mm 0.8 0.17
1 0.136
1.2 0.1133
2.2.1 Mathematical Model 1.4 0.0971
1.6 0.085
Table 2.1 Mathematical Model 1.8 0.0756
Sr.No Perticulars 2 0.068
1 The mathematical model includes five 2.5 0.0544
storey building with 30 column in each 3 0.0453
floor and 20 slab section 3.5 0.0389
4 0.034
2 All columns are modeled as beam element 4.5 0.034
and slab is modeled as plate element 5 0.034
having area type shell 5.5 0.034
3 Each plate is sub divided into 9*8 =72 6 0.034
plate element 6.5 0.034
4 For dynamic analysis of structure 7 0.034
response spectrum method was used 7.5 0.034
8 0.034
2.2.2 Soil Strata 8.5 0.034
9 0.034
It is assumed that the footing is resting at 1.5m from 8.5 0.034
GL 10 0.034
Medium soil

Head & Drop seems to be low due to which Sa/g will
be less and lateral forces will also be less
RESPONCE SPECTRA
0.3
0.25
0.2
0.15 Accler…
0.1
0.05
0
A 1
2
3
4
5
6
7
8
9
10
11
12
0
1.60
1.56
1.47
0.47
0.46
0.43
0.23
0.23
0.21
0.14
0.14
0.12
10
PERIOD (Sec)
2.3 Results and Discussion
1.61
1.57
1.50
0.47
0.46
0.43
0.23
0.23
0.21
0.13
0.13
0.12
20
Fig. 2.6 Response Spectra for Zone II
2.3.1 Comparison of Time Period (sec)
Table 2.3 Comparison of Time Period (sec)
Time Period for all Flat Slabs (sec)

Modes Flat Head Drop
1.65
1.61
1.53
0.49
0.48
0.44
0.24
0.23
0.21
0.14
0.14
0.12
Drop
head
1.66
1.62
1.53
0.49
0.48
0.44
0.42
0.24
0.23
0.23
0.21
0.21
Fig. 2.7 Comparison of Time Period (sec)
Description & Interaction (Time Period)

As it is seen from SAP-Results there are many modes
out of which if we consider the first two modes it has
been seen that the time period of flat slab without

TYPE
C1
C2
C3
C4
C5
C6
C7
C8
C9
C2
C3
C4
C5
C6
C7
C8
C9
Fig.2.8 Comparison of Axial Force (kN) First
FLAT
SLAB
593
728
708
722
836
833
715
838
869
42
42
40
63
70
44
73
77
Storey
Table 2.4 Comparison of axial Force of first storey
DROP
573
708
689
703
819
810
695
815
879
20
40
38
39
62
72
38
71
74
WITH
HEAD
581
716
695
712
825
821
708
827
888
Description & Interaction (Axial Force)
21
38
41
40
65
63
38
70
78
DROP
WITH
HEAD
569
699
681
693
809
799
689
806
876
As we go from fifth Story to the first story we can

see the first story have long column and we have
maximum axial force to be developed in the column
of first story with drop and head.
Table 2.5 Comparison of Shear Force of first storey
TYPE
C1
FLAT
SLAB
21
DROP
WITH
HEAD
DROP
WITH
HEAD
19
40
35
36
60
70
39
65
72
542
Fig 2.9 Comparison of Shear Force (kN) First Storey
Fig 2.11 Comparison of Moment in X- Direction
Description & Interaction (Shear Force) (kN-m) First Storey
From SAP Result it is seen that due to lateral forces Description & Interaction (Moments)
developed are in decreasing order from flat to drop
with head & there is not much differences in the From the graph it is found that the moments are
models with only drop & only head. directly proportional to the stiffness. Since the
stiffness is going to increase from flat to drop with
Table No.2.6 Comparison of Moment (kN-m) First head & if we compare Slab with the head & Slab
Storey with the drop. The stiffness will not very much there
Flat slab Drop With head Drop with fore the moment variation are not much.
head
TYPE M2 M3 M2 M3 M2 M3 M2 M3 2.3.2 Comparison of Slab Moment (kN-m/m) for
C1 76 107 80 120 84 110 112 101 Panel No. 7
C2 335 38 300 30 285 30 261 35 Table 2.7 Comparison of Slab Moment in X-
C3 295 31 357 32 306 32 330 32 Direction (kN-m/m) for Panel No. 7
C4 22 196 25 231 32 221 21 231 Panel X- Direction
C5 23 30 45 41 40 31 27 35 Column Strip Middle Strip Column strip
C6 24 32 40 32 40 41 28 51 dn up
C7 21 298 46 328 25 307 34 293 M11 M22 M11 M22 M11 M22
C8 31 32 25 31 62 37 36 41 Flat Slab
Max.Positive 26 15 19 29 25 16
C9 31 32 26 32 45 42 40 48
Max.Negative -147 -145 -23 -31 -142 -149
Drop
Max.Positive 26 15 19 29 25 16
Max.Negative -147 -144 -23 -31 -142 -149
Head
Max.Positive 26 15 19 29 25 16
Max.Negative -147 -144 -23 -31 -142 -149
Drop with Head
Max.Positive 41 -31 100 124 42 31
Max.Negative -362 -164 -77 -123 -362 -164
Fig 2.10 Comparison of Moment in X- Direction

(kN-m) First Storey

Table 2.8 Comparison of Slab Moment in Y-
Direction (kN-m/m) for Panel No. 7 COMPARISON OF TIME PERIOD
Panel Y- Direction
Column Strip Middle Strip Column strip 2.5
L R
T IM E P E R I O D (S E C )
M11 M22 M11 M22 M11 M22 2
FLAT
Flat Slab
1.5 HEAD
Max.Positive 28 29 26 24 28 27
Max.Negative -141 -149 -31 -33 -147 -145 1 DROP
Drop DROP HEAD
0.5
Max.Positive 19 29 26 24 18 27
Max.Negative -147 -149 -31 -33 -147 -146 0
Head 1 2 3 4 5 6 7 8 9 10 11 12
Max.Positive 18 29 26 24 18 27 MODES
Max.Negative -146 -149 -31 -33 -147 -145
Drop with Fig. 3.1 Comparison of Time Period (sec)
Head
Max.Positive 225 464 94 153 223 464
Max.Negative -355 -164 -95 -95 -356 -169 4. DESIGN OF FLAT SLAB BY IS 456:2000
3. ANALYSIS OF FLAT SLAB USING The term flat slab means a reinforced concrete slab
DIFFERENT THICKNESS BY RESPONSE with or without drops, supported generally without
SPECTRUM METHOD beams, by columns with or without flared column
heads. A flat slab may be solid slab or may have
Table No. 3.1 Different Slab Thickness recesses formed on the soffit so that the soffit
comprises a series of ribs in two directions. The
Type Flat Column Drop Drop recesses may be formed by removable or
Slab Head (mm) with permanent filler blocks.
(mm) (mm) head
(mm) 4.1 Components of flat slab design
Thickness 240 220 200 180
of slab
3.1 Comparison of Time Period (sec)
Table 3.2 Comparison of Time Period (sec)

Modes Flat Head Drop Drop
Head
1.00 1.60 1.70 1.85 2.00
2.00 1.56 1.66 1.80 1.95
3.00 1.47 1.58 1.69 1.81
4.00 0.47 0.49 0.52 0.55
5.00 0.46 0.48 0.51 0.55 Fig. 4.1 Panel, columns trips and middle strips in Y-
6.00 0.43 0.45 0.47 0.54 direction
7.00 0.23 0.23 0.24 0.48
8.00 0.23 0.23 0.24 0.31 4.1.1 Column Strip
9.00 0.21 0.21 0.21 0.28 Column strip means a design strip having a width of
10.00 0.14 0.13 0.13 0.24 0.25 I1, but not greater than 0.25l 2, on each side of
11.00 0.14 0.13 0.13 0.24 the column centre-line, where I, is the span in the
direction moments are being determined, measured
12.00 0.12 0.12 0.13 0.21
centre to centre of supports and 11, is the -span
transverse to 12, measured centre to centre of
supports.
4.1.2 Middle Strip

Middle strip means a design strip bounded on each of
its opposite sides by the column strip.

4.1.3 Panel 4.1.6 Depth of Flat Slab
Panel means that part of a slab bounded on-each of The thickness of the flat slab up to spans of 10 m
its four sides by the centre -line of Column or centre- shall be generally controlled by considerations of
lines of adjacent-spans. span (L) to effective depth (d) ratios given as below:
Division into column and middle strip along: Cantilever 7; simply supported 20; Continuous 26.
For slabs with drops, span to effective depth ratios
Table 4.1 Span Calculations given above shall be applied directly; otherwise the
Longer span Shorter span span to effective depth ratios in accordance with
above shall be multiplied by 0.9. For this purpose,
L1=6.6m, L2=5.6m L1=5.6m,L2=6.6m the longer span of the panel shall be considered. The
(i) column strip (i) column strip minimum thickness of slab shall be 125 mm.
= 0.25 L2= 1.65m = 0.25 L2 = 1.4m Taking effective depth of 25mm
But not greater than 0.25 But not greater than 0.25 Overall depth D = 220 +25 = 245mm >125 mm
L1 =1.4 m L1= 1.4m (minimum slab thickness as per IS: 456)
(ii) Middle strip (ii) Middle strip
It is safe to provide depth of 240 mm.
=6.6-(1.4+1.4)=3.8m =5.6-(1.4+1.4)=2.8m
4.1.7 Estimation of Load Acting On The Slab
4.1.4 Drop Dead load acting on the slab = 0.240 x 25 = 6 kN/m2
When provided shall be rectangular in plan, and = Wd1
have a length in each direction not less than one- Floor finishes etc. load on slab = 1kN/m2= Wd2
third of the panel length in that direction. For exterior Live load on slab = 6kN/m2= Wl
panels, the width of drops at right angles to the non- Total dead load = Wd1+ Wd2=7 kN/m2
continuous edge and measured from the centre -line The design live load shall not exceed three times the
of the columns shall be equal to one -half the width design dead load.
of drop for interior panels.
Since the span is large it is desirable to provide drop. Wl
Check: =0.85 < 3 ok
Drop dimensions along: Wd
Total design load= Wd + Wl =6.0+7=13kN/m2
Table 4.2 Drop Dimension
4.2 Shear in Flat Slab
Longer span Shorter span The critical section for shear shall be at a distance
L1=6.6m, L2=5.6m L1=5.6m,L2=6.6m d/2 from the periphery of the column /capital/ drop
Not less than L1/3=2.2m Not less than L1/3=1.9m panel, perpendicular to the plane of the slab where d
is the effective depth of the section. The shape in
plan is geometrically similar to the support
Hence provide a drop of size 2.2 x 2 m i.e. in internal immediately below the slab.
column strip width.
Hence provide a drop of size 1.1 x 1m i.e. in external Check for shear stress developed in slab
column strip width. The critical section for shear for the slab will be at a
distance d/2 from the face of drop.
4.1.5 Column Head Perimeter of critical section = 4 x 2240 = 8960mm
Where column heads are provided, that portion of a (2000+240/2+240/2)
column head which lies with in the largest right Total factored shear force: (1.5x13(L1xL2-(2.24)
circular cone or pyramid that has a vertex angle of (2.24))
90”and can be included entirely within the outlines = (1.5x13(6.6x5.6-(2.24)
of the column and the column head, shall be (2.24))
considered for design purposes. = 623kN
Nominal shear stress = τv
Column head dimension along:
vu 623000 =0.29N/mm2
= =
Table 4.3 Column Head Dimension bd 8960X 240
Longer span Shorter span
Shear strength of
2
L1=6.6m, L2=5.6m L1=5.6m,L2=6.6m concrete=  c=0.25 fck =0.25 25 =1.25N/mm
Not less than L1/3=2.2m Not less than L1/3=1.9m τv <  c (safe design ok)
If τv <1.5  c then the slab should be re-designed
Adopting the diameter of column head = 1.20 m Check for shear in drop
=1200 mm as per IS (456:2000) b0=  (D + d0) = 3.14(1.2+.46) =5.21m
V= (1.5x13 (5.6x6.6-(1.2) (1.2)) = 693kN

Nominal shear stress = τv Detailing of C3 Column for First Storey
vu 693000
= = =.75N/mm2
bd 2000 x 460
Shear strength of
concrete=  c=0.25 fck =0.25 25 =1.25N/mm2
τ v <  c (safe design ok)
4.3 Maximum Design Moments for Flat Slab as

Per SAP 2000 Version 9.03
Table 4.4 Slab Moments in X- Direction

Slab Maximum Moments For Panel No.7
Panel Load X- Direction

Combination Column Middle
Strip Strip Column strip
M11 M22 M11 M22 M11 M22
Drop
with
Head Figure 4.2 Detailing of C3 columnas per IS
Max.
Positive 1.5(DL+LL) 308 866 92 104 31 0 13920:1993.
Max.
Negative 1.5(DL+LL) -568 -161 -67 -105 -388 -956
Table 4.5 Slab Moments in Y- Direction

Slab Maximum Moments For Panel No.7
Panel Y- Direction
Load Column
Combination Strip Middle Strip Column strip
M11 M22 M11 M22 M11 M22

Drop
with
Head
Max.
Positive 1.5(DL+LL) 303 866 80 134 303 866
Max.
Negative 1.5(DL+LL) -388 -956 -9 -184 -388 -956
Table 4.6 Steel Calculation

Panel Longer span Shorter span
Column strip -956kN-m -568kN-m
Middle strip +134 kN-m +104 kN-m Figure 4.3 Detailing of C1 column as per IS
Column strip -956 kN-m -956 kN-m 13920:1993
Reinforcement 4664mm2 3512mm2
–ve steel 1938mm2 1508 mm2
Reinforcement 20mm  @130c/c 20mm  @130c/c
(Note: - Design of column for 1% steel is safe for all
+ve steel
16mm  @100c/c 16mm  @130/c structure in all checks as per IS 456:2000 and
Spacing –ve
steel detailing is done by IS 13920:1993)
Spacing +ve
steel 5. CONCLUSIONS
1. Drops are important criteria in increasing the shear
Table 4.7 Design of column by IS-456-2000 and IS- strength of the slab.
13920 2. By incorporating heads in slab, we are increasing
Type Main Main Stirrups Stirrups rigidity of slab.
R/F R/F IS 456-2000 IS 13920- 3. The dynamic analysis results indicate that the
IS 456- IS 1993 lowest mode of vibration i.e. third mode was the
2000 13920-
1993 torsional mode. This seems to be a typical of flat-slab
C1 6- 6- 8mmDia. 1)8mmDia. building with a central core of shear walls.
25Ф+2- 25Ф+2- 2L@180mm 2L@250mm 4. Modeling of flat plate slab with diaphragm and
16Ф 16Ф without diaphragm in case of response spectrum
2)8mmDia.
3L@80mm
method there is no variation in axial force, shear

force and moment as moment of inertia of slab is “Enhancing Performance of Slab-Column
very high it acts as rigid. Connections” , Journal of Structural
5. The negative moment’s section shall be designed Engineering, Vol. 134, No. 3, March 1, 2008.
to resist the larger of the two interior negative design L. Cao and C. J. Naito, “Design of Precast
moments for the span framing into common Diaphragm Chord Connections for In-Plane
supports. Tension Demands” , Journal of Structural
6. Enhance resistance to punching failure at the Engineering, Vol. 133, No. 11, November 1,
junction of concrete slab & column by providing 2007.
drop with head. M. Altug Erberik and Amr S. Elnashai “Loss
7. Drop with head & Flat slab head is very good Estimation Analysis of Flat-Slab Structures” ,
combination to reduce the moment with less Natural Hazards Review, Vol. 7, No. 1, February
thickness of slab. 1, 2006.
8. In earthquake zone we shall provide only flat slab M.Anitha, B.Q.Rahman, JJ.Vijay, “Analysis and
drop with head & ductile detailing for all structure. Design of Flat Slabs Using Various Codes”,
International Institute of Information
Technology, Hyderabad (Deemed University)
REFERENCES
April 2007.
Abdel Wahab, El-Ghandour, Kypros Pilakoutas, and
M. H. Haraj, K. A. Soudki, and T. Kuds,
Peter Waldron, “Punching Shear Behavior of
“Strengthening of Interior Slab-Column
Fiber Reinforced Polymers Reinforced Concrete
Connections Using a Combination of FRP
Flat Slabs: Experimental Study.” , Journal of
Sheets and Steel Bolts”, Journal of Composites
Composites for Construction, Vol. 7, No.3,
for Construction, Vol. 10, No.5, October 1,
August 1, 2003.
2006.
ACI Manual of Construction Practice 1995 Part – 3
Paul Arthur Whitehead and Timothy James Ibell
Page No. 318/318R-(201-228).
“Rational Approach to Shear Design in Fiber
Ahmed Ghobarah, “Performance-Based Design in
Reinforced Polymer-Prestressed Concrete
Earthquake Engineering: State Of Development”
Structures”, Journal of Composites for
, Engineering Structures Elsevier Science
Construction, Vol. 9, No.1, February 1, 2005.
23,874-884, March 17, 2001.
Softwares:-SAP-2000V0.9 & AUTOCAD-2004.
Austin P.D, Moehle J.P, “Interior Slab Column
Y.H.Luo, A.J.Durrani, J.P.Conte, “Equivalent Frame
Connection of Building,” 2000
Analysis of Flat Plate Buildings for Seismic
B. A. Izzuddin, X. Y. Tao; and A. Y. Elghazouli,
Loading”, Journal of Structural Engineering,
“Realistic Modeling of Composite and
Vol. 120, No. 7, July 1994.
Reinforced Concrete Floor Slabs under Extreme
Loading.” , I: Analytical Method Journal of
Structural Engineering, Vol. 130, No. 12,
December 1, 2004.
Criteria for Earthquake Resistant Design of
Structures Part 1 General Provisions and
Buildings (Fifth Revision) IS 1893(part 1):2002.
Dr. Ram Chandra, Virendra Gahlot, “Elements of
Limit State Design of Concrete Structures”,
Scientific Publishers (India), 2004.
Dr. R .K. Ingle “Excel Spreadsheet for Design of
Colum by IS 456-2000 & IS 13920”, Applied
Mechanics Department, VNIT, Nagpur.
Ductile Detailing of Reinforced Concrete Structures
to Seismic Forces IS 13920:1993
Durrani A. J.; Mau S. T., AbouHashish A. A., and Yi
Li, “Earthquake Response of Flat-Slab
Buildings” , Journal of Structural Engineering,
Vol. 120, No. 3, March, 1994.
Ehab El-Salakawy, Khaled Soudki, and Maria Anna
Polak, “Punching Shear Behavior of Flat Slabs
Strengthened with Fiber Reinforced Polymer
Laminates”, Journal of Composites for
Construction, Vol. 8, No. 5, October 1, 2004.
Indian standard “Plain & Reinforced Concrete Code
of Practice” Fourth Revision IS: 456:2000.
Joo-Ha Lee, Young-Soo Yoon, Seung-Hoon Lee,
William D. Cook, and Denis Mitchell,

ISBN 978-81-921121-0-7
TECHNO-ECONOMIC STUDY OF REMOTE AREA SMALL HYDRO

POWER PLANTS
Mon Prakash Upadhyay
M.Tech II Year, Alternate Hydro Energy Centre, Indian Institute Of Technology, Roorkee-247667
Rahul Bhatt
M.Tech II Year, Alternate Hydro Energy Centre, Indian Institute Of Technology, Roorkee-247667
Nitin Kumar Sahu
Assistant Professor, Iimt Institute Of Engineering And Technology MeerutE-mail: mon.u67@gmail.com
ABSTRACT: Electricity is most convenient form of energy, which is utilized for lighting, heating and
industrial production. Also generation and consumption of electricity through small hydro does not pollute the
environment and therefore can play an important role for protection of the environment and in maintenance of
ecological balance especially we can say for hilly areas also.Hydro power Development Corporation of
Arunachal Pradesh Limited (A Govt. of Arunachal Pradesh Enterprises) is entrusted with the design and
construction of the power projects in Arunachal Pradesh State. HPDCAPL has assigned the task to Alternate
Hydro Energy Center (AHEC), IIT, Roorkee to prepare detailed project report and detail engineering design and
drawing work for the Sumbachu small hydro project. In this paper we have done the techno-economic study of
remote area small hydro power plants (2×1500 kW)” in Zemithang town of Tawang district of Arunachal
Pradesh. The discharge measured on the Sumbachu Stream on 18th July, 2010 was found to be 1.85 cumec with
gross head of 279 m. Two unit of 1500 kW each has been proposed. The design calculation for all civil
components and electromechanical components is carried out.
INTRODUCTION 1. Improve basic living conditions and

The renewable sources of energy which could make education standard.
substantial contribution to meet the rural energy needs 2. Establishment and sustenance of small-scale
are solar, biogas, wind and water power. In Arunachal and rural based industries.
Pradesh, the potential of hydropower schemes 3. Improve agricultural productivity by stable
identified is about 560 MW in its different river power supply for the irrigation need.
basins. Though the rate of grid power remains same 4. Improve and further promote tourism in
irrespective of the point of withdrawal but the fact is valley.
that the supply of power in the remote and hilly areas
and at the tail end has been extremely poor and erratic SOCIO ECONOMIC AND ENVIRONMENTAL
with low voltage with frequent and very long IMPACT OF THE SCHEME
shutdowns. Electrification of the area will bring numerous
With the incentives, and right environment through development activities. The working hours of rural
policies available for hydro power generation, folks will increase with the availability of electric
HPDCAPL decided to install a hydropower plant on supply in the morning and evenings hours in their
perennial stream to meet the existing demand of the domestic/household area. Modern commercial and
Zemithang area. Sumbachu Small Hydro Scheme is entrainment facilities like Radio and Television will
proposed to be located on the right bank of bring the entire population of the area in mainstream
Sumbachu river with the purpose of tapping the of development. Use of electricity in cooking and
renewable source of energy at competitive prices and heating will reduce dependence on forest wood. Water
with a view to improve the socio- economic supply shall be augmented with the installation of
conditions of the region. There is great need of rural pumps. The availability of dependable power will
electrification as an infra-structure for development create avenues for cottage/mini industries giving a
and government is laying considerable priority for the boost to the economic development of the area.
same. By the execution of such Small Hydro Further with the creation of modern facilities for
Schemes, the area will usher into new era of living, migration of locals will also be reduced.
development. The proposed scheme will supply Sumbachu Small Hydro Power Scheme envisages the
energy to nearby area. This scheme is very suitable to power generation from Sumbachu Nallah and this
meet the power demand in villages like Zemithang development is without any pondage. Mini/Small
etc.Near the proposed site. At present there is 33kV
Hydro Power development projects, while sharing all
grid at a distance of about 30 km from power house
site. Several areas are being provided electricity the benefits of hydro electric generation, harnesses a
through D.G. set which is very costly. The unit cost renewable source of energy in extremely
from D.G. supply in this area comes out to be more environmentally benign manner. Being small it does
than Rs. 10.00 per unit. Thus development of this
scheme shall contribute to-

not involve any submergence or violation of the appropriate to choose for this project. Two nos. of
sanctity of forests. synchronous generators are proposed to be installed.
Availability of reliable electrical energy is expected to Synchronous generator of 1875 KVA (1500 kW) at
3.3 KV, 0.80 PF, 3 Phase A.C., 50 Hz, 1000 rpm.
reduce over dependence on nature depleting fuel and
Each Turbine shall be equipped with suitable PLC
will also result in saving of petroleum products like based electronic type of governor, which
diesel and Kerosene. switches/adjust load so as to match available
The scheme would also help in activating some generation. The main inlet valve of the turbine shall
industrial development activities by the local people be of butterfly type for connecting to penstock.
due to regular and bulk availability of electricity and Electrical control, protection and metering system will be
will also improve their living conditions. based on state of art technologies. Microprocessor based
automation systems for the governing and control of
The magnitude of construction activity will not induce
power plant will be adopted. The power generated will
migration of labor to this area, as sufficient local labor be stepped up to 33 KV through 2 numbers,
is available in the area, and thus local ecology will not 2000KVA 3.3/33 KV step up transformers and shall
be pressurized. be transmitted to the state grid. Provision for
In addition, it is proposed to do plantation in the emanating one no. 33 KV line from this project for
scheme area wherever possible. It will further add to connection with grid with full protection and metering
systems has been made in this project. However the
environment and overall outlook of scheme.
cost of 33KV line is not included in the project report.
The project has been identified at such a place which
CIVIL WORK STUDY may be developed as tourist centre in future. The
Keeping in view the availability of material the power supply to the circle head quarter at Zemithang
Trapezoidal trench Type weir is proposed to be and adjoining nearby areas along with the military
provided to divert the water from nallah to intake establishments may be improved with proper
channel. Such weirs are also suited for mountainous distribution system. Construction of 33 KV line from
streams as they do not much interfere with the regime the switchyard of the project to distribution
of the stream. The water fed from diversion weir is substations has not been covered in the estimate of the
lead to desilting tank through a rectangular project. All the following equipment shall be provided
R.C.C.channel (bed width 1.00 m x water depth 1.00 in the switchyard with terminal and auxiliary
m & bed slope=1/200). At Every interval of 50.0 m structures and proper earthing. Suitable control panels
suitable man holes shall be provided by 2 no. having protection and metering equipment etc shall be
removable RCC slabs of size 1.0 x 0.50 m for housed inside the power house at appropriate location.
maintenance & cleaning purpose. To allow the water i. 2 x 2000 kVA 3.3/ 33 kV Step up Transformers
to escape from forebay tank in case of emergency ii. 250 kVA 33/0.415 kV Station Transformer
shutdown of the machines, 5 m wide spillway is iii. Vacuum circuit breakers
provided in the upstream of forebay tank to divert the iv. Current transformers
water into the nallah through a monsoon drain already v. Potential transformers
existing on site which meets the Nallah. vi. Line isolator
vii. Lightning arrestors
Powerhouse building is a simple structure housing the viii. Shielding
generating units, auxiliary equipment, control panels ix. Pale fence etc.
and suitable outlet for tail water discharge. Turbine
discharge shall be disposed off to Nyaritang chhu TECHNO-ECONOMIC STUDIES
river through the tail race channel from each unit i.e. i. Flow Duration
for a discharge of 0.75cumec with the width of Based on the discharge data collected for the year
channel is 1.00 meter with a depth of water as 0.60 m 2001 and 2002, flow duration analysis has been
and free board 0.30 m. The length of tail race channel carried out. The water available has been carried out.
is approximately 30 meter and channel is provided The water available for percentage of time equaled or
with the slope suitable to the terrain. The channel is exceeded. The various discharges equaled or
proposed to be constructed in stone masonry. exceeded are as follows:
100% = 0.47 /s
ELECTROMECHANICAL WORK STUDY
75% = 0.91 /s
As per the standard selection charts, the turbine for 50% = 1.70 /s
generating 1500 kW at 268 meters net head is
available with 1000 rpm speed. The specific speed of 25% = 5.10 /s
the turbine at 1000 rpm comes out to 30.2, which falls
in the range of Standard Pelton Turbine. Hence ii. Energy generation
standard Pelton turbine with double jet system may be

Based on the discharge available, power potential has and Viswanadham, B.V.S. (1998). Model studies
been worked out at various dependability. For on mechanics of engineering. Journal of
computation of power potential, the efficiency of Geotechnical and Geoenvironmental Engineering,
turbine and generator has been taken as 0.88 and 0.95 ASCE, 90(3), 229-248.
respectively. Power generation may vary depending Detail project report of Sumbachu small Hydel
upon the actual equipment purchased and their Project (2×1500 kW) in Zemithang town of
efficiencies. Standard turbine runner diameter and Tawang district of Arunachal Pradesh prepared by
generator capacity have been taken into consideration “AHEC, IIT Roorkee”.
for finding out the power potential and possible
energy generation.
iii. Installation capacity and no. of units
The power potential at different dependability is

broadly as follows.
Dependability Power Potential
100% 1046 kW
75% 2017 kW
50% 3782 kW
25% 11329 kW
iv. Energy generation

The provisional installed capacity of 3000 kW at
about 65% dependability may be assumed for further
exercises. Therefore for 3000 kW capacity at
75% Plant Load Factor = 19.17 Million Units
60% Plant Load Factor = 15.77 Million Units
CONCLUSIONS
The site of Sumbachu Small Hydro Electric scheme is

suitable for development of hydropower on Sumbachu
Nallah by utilizing the existing fall in the nallah.
Installation of 2 unit of 1500 kW each by utilizing the
net head of 271 meter and discharge of 1.485 cumecs
is considered most appropriate taking into account the
load demand, cost of the project and availability of
land. The scheme involves very simple civil
engineering works and can be completed in working
period of 24 months. During Rainy period of four
months work at site may not be possible. Construction
materials like sand, screened gravel and aggregates
etc. are available in the nearby area. The proposed
scheme has minimal interference with the
environment and hence would not affect the ecology
of the area. The scheme will be helpful in generating
employment and improvement of quality of life in the
area. The scheme would result in saving of diesel and
kerosene oil by making available a reliable and stable
electricity supply to the people living in the area.
The cost of generation from the scheme at 75%load
factor is about Rs.2.26 per kWh without subsidy
which is much less than the cost of generation from
alternate source of diesel/solar at such remote sites.
REFERENCES
A.K.Akella, R.P.Saini, M.P.Sharma, Social,
economical and environmental impacts of
renewable energy systems, Renewable Energy
2009, 34:390–396.Einstein, H.H., Choudhury, D.

ISBN 978-81-921121-0-7
A PRAGMATIC APPROACH OF POT HOLES REPAIR UNDER INDIAN

SERVICE CONDITIONS

Assistant Professor, Transportation Engineering Group, Department of Civil Engineering, IIT-Roorkee
Praveen Kumar
Professor, Transportation Engineering Group, Department of Civil Engineering, IIT-Roorkee.
Brnd Kumar
Assistant Professor, Department of Civil Engineering, IT-BHU, Varanasi.
ABSTRACT: Potholes are a form of disintegration of the pavement that may be associated with poorly compacted
material, raveling, cracking, improper longitudinal & transverse joints, moisture intrusion, base failure or ageing of
the bituminous materials. Mostly potholes occur in the wheel paths, particularly the outer path, centerline, near the
edge of the pavement and especially where a pavement has been widened. Potholes are also formed due to improper
resealing of core holes made on the pavement surface during the post construction quality control tests. All over the
world pot holes is a daily challenge for cars and trucks being expensive in their repairing. They often appear after
rain or during thaw periods when pavements are weaker. Causes of potholes, different methods of potholes repairing
and preventive & corrective measures of potholes are discussed in this paper.
INTRODUCTION
Potholes are a small, bowl-shaped depression in the
pavement surface that penetrate all the way through the
bituminous concrete layer down to the bituminous base
course or even goes deeper than these layers due to
untimely repairs. Potholes are a serious problem and
their effective repair is an important issue. For road
users, potholes are not simply a nuisance. They
constitute a dangerous safety hazard that can inflict
substantial damage to vehicles.
US Department of Transportation, Federal Highway

Administration [1] described, potholes are as bowl-
shaped holes of various sizes in the road pavement Fig.2. Moderate severity pothole
surface, with a minimum width of 150 mm. Low
severity potholes are less than 25mm deep (Fig.1),
moderate-severity from 25mm to 50mm deep (Fig.2)
and high-severity potholes are deeper than 50mm
(Fig.3).
Fig.3. High severity pothole
Causes of potholes
The generally accepted mechanisms for pothole

formation are due to cracking, raveling or stripping in
the pavement surface; moisture intrusion which
weakens the underlying pavement layers and
Fig.1. Low severity pothole

subsequently this leads to loss of fines from the surface pavement surface. If pothole occurs within an area of
& underlying pavement layers. fatigue cracking, the area of fatigue cracking is reduced
Pothole formation is mostly associated with bitumen by the area of the pothole [1].
fatigue damage. Potholes on roads and streets of India
are often appearing after the onset of monsoons (2). Disadvantages of potholes
When bitumen road pavement ages, it tend to becomes
more porous and rain water increasingly begins to The following are the major disadvantages of potholes
penetrate through cracks and other flaws (Fig.4). when they are not repaired at all or repaired with the
Additionally, water may enter the road base from the preceding improper techniques [2]:
sides of the road. In colder weather conditions when
sub-zero temperatures occur, the water which has (i) Inconvenience to motoring public
drained through to the road base or foundation layer (ii) User delays which cost money in terms of
freezes to form pockets of ice (Fig.5). When water their time
freezes to become ice, it expands. This expansion, or (iii) Extra fuel consumed when the vehicles have
increase in volume, begins to displace the road to stop or go very slow
pavement surface. When the temperature subsequently (iv) Excessive wear & tear of the vehicles, which
rises above freezing, the pockets of ice melt leaving require frequent repairs
voids beneath the road surface. A succession of freeze- (v) Increased air pollution caused by vehicle
thaw cycles gradually increases the size of these sub emission due to slow or stopped traffic
pavement voids, progressively weakening the road (vi) Accidents resulting from the tendency of the
pavement layer. Traffic loads over the un-supported vehicles to go around the potholes
pavement cause collapse and the formation of a (vii) Increased real cost of repairing potholes in
pothole. However, these kinds of problems are rarely terms of labour because the government
observed in our country. There is no doubt that water is agencies have to keep on repairing the same
the primary cause of potholes, but the ingress of water potholes several times during the monsoon
into the road structure to cause the potholes is mostly Significant environmental impact due to most of the
due to the condition of the road surface [www. csir] reasons mentioned above such as increased fuel usage,
(3). The formation of potholes differs somewhat increased air pollution, increased consumption of
depending on the environmental conditions, traffic energy in repairing the potholes with stopgap
loading, road pavement structure and materials used techniques.
(3). Generally, pothole generated from the top course
initially (Fig.6) then once a pothole has been formed, it
will tend to grow in size because traffic will Methods of pothole patching.
continually remove broken pieces of the road surface
(Fig.7). If the pothole fills with water, its rate of The construction procedures of potholes patching vary
growth will accelerate as the water will wash away with method employed and material selected [4]. There
loose particles under the force of passing traffic. Whilst are three primary patching techniques, throw and roll,
potholes can grow to perhaps half a metre in width semi-permanent patches and injection patching; an
also, if they are left unattended for a longer period then overview of each follows:
ultimately they can cause damage to the pavement
structurally. Fig.8 depicts the stepwise formations of
potholes due to their untimely repairs. Ultimately,
expensive reconstruction of the road is then necessary.
Potholes can also result from improperly filled &
unfilled core holes made on the pavement surface
during post construction quality control tests [Fig.9
(a)]. This formation is prone near the edge of the
pavement especially where a pavement has been
widened [Fig. 9 (b)].
Pot holes measurement

Fig.4. Water penetration through cracks &
Potholes are generally measured in terms of number flaws (www.instarmac.co.uk)
and square meters of affected area at each severity
level. Pothole depth is the maximum depth below

(i) Patching material is placed into the hole, with or
without cleaning and/or drying of the hole.
(ii) The material is compacted using the maintenance
truck tires.
(iii) The finished patch should have about 10 mm
crown to prevent water ponding.
Clean-up is generally not required.
1.5.2. Semi-Permanent Patching
Semi-permanent patching is considered to be an

effective patching method (second only to complete
removal and replacement of the failed area). The steps
of this method are summarized as follows:
(i) Mark the boundaries of the distressed area,

taking care to encompass a slightly larger area
than that reflected by the distress. The repair
boundaries should be as rectangular as
Fig.5. Unsupported pavement fails under
possible and take into consideration the
the traffic loads. (www.instarmac.co.uk.)
dimensions of the equipment that will be used
for removal of the old material and
compaction of the new material.
(ii) Cut the boundaries of the patch square using
either a diamond saw or pneumatic hammer
with a spade bit. In the case of the latter, care
should be taken not to damage the HMA
surface layer in the sound pavement.
(iii) Square up the sides of the hole until the edges
of the hole are sound pavement. This method
Fig.6. Pothole in the wearing course is very simple if the boundaries of the repair
area were cut with a diamond saw or
established with cold milling equipment. It is
usually only required when manual techniques
of material removal are employed. It is
suggested that the depth of the patch be 50%
thicker than the thickness of the failed layer.
(iv) Remove water and debris from the hole and
apply a tack coat of asphalt emulsion to the
sides and bottom of the hole.
(v) Place the patch material in the hole, taking
Fig.7. Pothole in the deeper layer care to avoid segregation. The hole should be
overfilled by 20 to 25 percent of its depth to
provide adequate material for compaction.
1.5.1. Throw and Roll (vi) Compact the patch material with a hand
device or a small vibratory roller. It is
preferably to use compaction equipment
The “throw and roll” method is often used for whose surface is smaller than the size of the
temporary patches. It is only appropriate when weather patch. It is very difficult to achieve
conditions do not allow a semi-permanent patch to be satisfactory compaction with equipment that
placed, or the road is due for rehabilitation in the very bridges the repair area.
near future. It is the least expensive and least labor-
(vii) The patch should be compacted thoroughly
intensive method for patching a pothole and includes
with proper compaction equipment. For
the following steps:

additional compaction by traffic and helps employed for pothole repairing [5]. This kind of
prevent standing water accumulate in the technology could work in quicker ways as well as has
patch area, the finished patch should have a 3 advantages on cost over manual repair. As its initial
to 6 mm crown. investment is very huge, it would be difficult to employ
such costly machinery in every running project.
The edges of the patched area should be sealed with Therefore as on today, we rely more on manual
crack sealant. For areas with significant amount of repairing method than other methods.
rainfall, the entire patch should be fog sealed.
Injection Patching
Injection patching is a rapid and effective method of
patching that requires specialized equipment. This
method is used for low volume traffic roads and is an
alternative to the throw and roll method. The patches
are temporary, but generally have a longer life than
throw and roll patches. The steps for injection patching
are described below: Fig.8. Pot hole damaged the entire pavement layers
(i) Prepare the site for patching by blowing debris and

water from the hole with the application nozzle.
(ii) Spray a tack coat of emulsion on the sides and
bottom of the hole at a rate of approximately 1
2
liter/m .
(iii) Blow asphalt/aggregate mixture into the
hole, filling the hole to the top.
Considering the current practices which are being

adopted by the various construction agencies in India
are briefly discussed here [2]: (i) mostly potholes are Fig. 9a.
being filled with soil, which is usually washed away in
the first rain and this method is not successful (ii)
another way to fill the potholes is dumping bare stones
in the holes. This not only provides a rough riding
surface especially for motorcycles but also the stone
particles start to ravel out of the pothole during
subsequent rains. If the stones survive in the potholes,
these are not usually removed and pothole properly
patched when the road or street is overlaid with hot
mix. This results in reappearance of the same pothole
during next year’s monsoon. (iii) for a very deep Fig. 9b.
potholes, full cement bags of sand is employed to fill Fig.9. Pot hole developed near the centre-line (a) &
the holes which is highly undesirable and (iv) outer wheel path (b).
sometimes a penetration macadam type approach is
used. In this method, a tack coat is applied to the
bottom of the pothole. A layer of large stone is spread
in the pothole. Bituminous binder is applied to the
stone layer and then another layer of stone is
spread and compacted. This penetration macadam is
extremely porous to water
and gets saturated with water when it rains. Even when
the road is resurfaced
with hot mix, the pothole redevelops at the same place
year after year. However, in other developed countries,
sophisticate automated patching technology is being

Fig. 10. Digging out the distressed portion
Fig. 12. Tack coating to the bottom of the

repair area
Fig. 11. Removal of disintegrated materials Fig. 13. Applying tack coat to the bottom as
well as sides of the repair area
Fig. 14. Transportation of patch Fig. 15. Placing of Fig. 16. Hand Fig. 17. Vibratory
material to repair site. patch material & compaction compaction
leveling
A lack of periodic and /or preventive maintenance Table 2. Minimum Residual Bitumen Content by
leads to the development of weaknesses in the road Weight of Mix [2].
pavement structure. Therefore, interval of road Aggregate water Minimum residual bitumen
inspection should be conducted at least once in three absorption, % content, %
months of time. There should be pre-and post-monsoon 1.0 or less 4.5
inspections of the road and the necessary preventive & 1.1 to 1.5 5.0
corrective measures should be carried out to the 1.6 to 2.0 5.5
appropriate standards. Knowledge of the effects of
routine maintenance on pavement performance is Example to explain the residual bitumen content: If the
important to the management of highway pavements at total amount of cutback
both the network and the project level [6]. Rigorous bitumen in the mix is 6.0 % and the cutback has 80 %
repair procedures that involve cutting, cleaning, and bitumen (residual) and 20 % diluent
compacting are the most cost-effective ways to repair (such as kerosene), the residual bitumen content in the
potholes. Throw-and-go procedures cost approximately mix is 6.0 times 80 %, which is
three times more than do the more rigorous procedures equal to 4.8 %. The contractor must obtain the actual
[7]. percentage of residual bitumen in the
MC cutback from the approved supplier.
Preventive & corrective measures
Materials for pothole repairing The mineral aggregate should be clean and surface dry
Materials play an important role in the efficient and before mixing. The temperature of aggregate shall
effective use of maintenance treatments. The two main preferably be in between 50C to 650C and that of
materials comprising flexible pavement maintenance bitumen shall be in between 550C to 950C.
treatments are bitumen and aggregate. Generally, The following two tests should be performed on the
bitumen types like paving bitumen, bitumen emulsion mixture, freshly prepared or taken from a stockpile at
& bitumen cutback are oftently recommended for any time during its storage life (usually 6 months) [2]:
patching activities related to pavement maintenance. 1. Water Resistance Test
Also, suitable modified bitumen grade other than 2. Workability Test
conventional bitumen grade may also be suggested
depending upon the work demand. Under no The water resistance test will indicate whether the
circumstances, cutback bitumen should be allowed to patching mix has a potential for stripping in the pothole
be made by contractors or department by blending in presence of water. If the mix fails this test, it means
kerosene with paving bitumen. It should only be a proper type and/or amount of antis tripping agent has
obtained from approved manufactures; the same should not been used in the mix. If the mix fails in
conform to all requirements of IS 217 specifications. workability, it could be due to improper bitumen type,
The patching mix gradation given in Table 1 may be low bitumen content, excessive fines or improper
adopted for pothole repairing. It should be noted that gradation.
the material passing 0.075 mm sieve is restricted to 2%
maximum (based on washed gradation) for ensuring
quality and durability of the patching mix. A pragmatic approach of pot holes repair
Once the distress areas are located, mark the
Table 1. Gradation of Stockpile Patching Mix [2]. boundaries of the distressed area to be replaced. At the
best possible, one should strive for rectangular areas
Sieve size, mm Percent passing taking into considerations of the equipment that will be
used for removal of the failed material and compaction
9.5 100 of the new material (Figs. 10).
4.75 40-100
2.36 10-40 Cut out the perimeter of the area with suitable tools and
1.18 0-10 remove the disintegrated materials (Fig.11).
0.075 0-2 Clean the dig out area with the help of suitable
As far as possible, aggregate with 1.0 % or less water equipments like cycle pump and air compressor.
absorption should be used. If the water absorption Then tack coat of bitumen emulsion is applied at the
exceeds 1.0 %, the amount of residual bituminous rate of 1liter/m2 approximately to the sides of the repair
material should be increased as shown in Table 2. area as well as along the bottom of the repair area. It is

also important to ensure timely supply of patch 9. Warmer and drier weather should be preferred for
material to the repair site and the same can be long lasting pothole repairs. Under unavoidable
performed with the help of wheel barrow (Fig. 14) for circumstances, temporary repairing may be got
small magnitude of work. The patch material is conducted even during monsoon however, the
typically placed in lifts if the depth of the repair is same shall make a permanent repair as soon as
greater than 100 mm. The thickness of any lift should possible.
not exceed 100 mm.
REFERENCES:
Initially hand compaction may be employed (Fig. 16) 1. US Department of Transportation, Federal Highway
for ensuring proper compaction at the corners & edges Administration, “Distress Identification Manual
of the repair area, Compact each lift using equipment for the Long -Term Pavement Performance
similar to that typically used in hot-mix asphalt Program, Publication Number FWA-RD-03-031,
compaction operations. The width of the compaction June 2003.
equipment should be narrow enough to fit within the 2. Kandhal, P. S. and D. B. Mellott. Rational Approach
repair area. The finished patched area should have a to Design of Bituminous Stockpile Patching
crown of 3 to 6 mm. It is an allowance made for further Mixtures. TRB, Transportation Research Record
compaction under traffic loads. 821, 1981.
3. www.csir.co.za/pothole_guides, non-technical and
CONCLUSIONS & RECOMMENDATIONS technical guides.
1. The effective technique for reducing pothole 4. Caltrans, Maintenance Technical Advisory Guide,
formation on the road surface is timely preventive Volume I, Flexible Pavement Preservation, Second
maintenance to the appropriate standards. Edition, 2008.
2. Interval of road inspection should be frequent. It 5. James R. Blaha “Fabrication and Testing of
should be atleast once in three months. Automated Pothole Patching Machine” Strategic
3. Pothole repair programme should aim to mobilize Highway Research Program (SHRP-H-674),
funds as well create well-trained mobile team for National Research Council Washington, DC, 1993.
each project. 6. Fwa, TF and Sinha, K.C., “A Study of the effects of
4. Holes detected on the pavement surface during routine Pavement Maintenance” Transportation
post construction quality control test should be Research Board, Issue Number: 1102, 1986, p. 6-
made good. The edges in particular holes should 13.
be sealed firmly so as to prevent intrusion of 7. Thomas, HR and Anderson, DA, “Pothole Repair:
surface water. The number of holes in the You can’t afford not to do it right” Transportation
pavement surface should be kept as low as Research Board, Issue Number: 1102, 1986, p. 32-
possible. 40.
5. Joints of pavements are weak planes and are also
one of the vulnerable locations for formations of
potholes hence therefore; extra care should be
taken by ensuring sufficient overlapping distance
& providing requisite degree of compaction at
locations where two lanes are jointing or
terminating.
6. Adequate quantity of tack coat should be sprayed
to the dugout areas before placing premix material
for ensuring strong bonding.
7. Sealing cracks is of the utmost importance. The
presence of open cracks greater than 3 mm allows
the ingress of water into the pavement layers,
thereby decreasing strength (load-bearing
capacity) and making them more susceptible to
formation of potholes.
8. Timely treatment of hungry surfaces on the road
pavement is essential to prevent the formation of
potholes. The same can be treated with
rejuvenator or by spraying suitable grade of
bitumen emulsion.

ISBN 978-81-921121-0-7
ANALYSIS OF THE MEAN MONTHLY TEMPERATURE SEVEN

HOMOGENEOUS REGIONS OF INDIA
R. K. S. Maurya
Department of Geophysics, Banaras Hindu University, Varanasi-221005, Email:rksmaurya@gmail.com
G. P. Singh
Department of Geophysics, Banaras Hindu University, Varanasi-221005, Email: gpsinghbhu@hotmail.com.
ABSTRACT: In the last 100 years or so, the Earth’s surface and lowest part of the atmosphere have warmed
up on average by about 0.6oC. During this period, the amount of greenhouse gases in the atmosphere has
increased, largely because of the burning of fossil fuels for energy and transportation, and land use changes, for
food by mankind. In the last 20 years, concern has grown that these two phenomena are, at least in part,
associated with each other. That is to say, global warming is now considered most probably to be due to the
man-made increases in greenhouse gas emissions. Whilst other natural causes of climate change, including
changes in the amount of energy coming from the Sun and shifting patterns of ocean circulation, can cause
global climate to change over similar periods of time, the balance of evidence now indicates that there is a
discernible human influence on the global climate. Analysis of the mean monthly temperature seven
homogeneous regions, viz., West Himalaya (WH), Northwest India (NWI), North Central India (NCI),
Northeast India (NEI), West Coast (WC), East Coast (EC), and Interior Peninsula (IP) has been studied for the
period of 1901-2000. Detail analysis on annual and seasonal changes in mean monthly temperature (Tavg),
mean maximum temperature (Tmax) and mean minimum temperature (Tmin) have been done. Each year was
divided into four principal seasons, viz. winter, pre-monsoon, monsoon, and post-monsoon. The analysis shows
that the annual mean temperature, mean maximum temperature, and mean minimum temperature are increasing.
As far as different homogeneous region are concerned, the analysis shows that the WH, NWI, NCI, NEI, WC,
EC and IP show a rising trend of approximately 0.53°C, 0.07°C,0.41°C, 0.55°C , 0.67°C , 0.42°C and 0.45°C
per hundred year respectively. The largest temperature rising are found over the WC and minimum over NWI.
INTRODUCTION changes in the Earth’s orbit, fluctuations in energy

In the last 100 years or so, the Earth’s surface and received from the sun, and volcanic eruptions.
lowest part of the atmosphere have warmed up on The main human infl uence on global climate is likely
average by about 0.6oC. During this period, the to be emissions of greenhouse gases (GHG) such as
amount of greenhouse gases in the atmosphere has carbon dioxide (CO2) and methane (CH4).
increased, largely because of the burning of fossil “Global atmospheric concentrations of carbon
fuels for energy and transportation, and land use dioxide, methane and nitrous oxide have increased
changes, for food by mankind. In the last 20 years, markedly as a result of human activities since 1750
concern has grown that these two phenomena are, at and now far exceed pre-industrial values determined
least in part, associated with each other. That is to say, from ice cores spanning many thousands of years. The
global warming is now considered most probably to global increases in carbon dioxide concentration are
be due to the man-made increases in greenhouse gas due primarily to fossil fuel use and land use change,
emissions. Whilst other natural causes of climate while those of methane and nitrous oxide are
change, including changes in the amount of energy primarily due to agriculture”.
coming from the Sun and shifting patterns of ocean They can be defined by the differences between
circulation, can cause global climate to change over average weather conditions at two separate times.
similar periods of time, the balance of evidence now Climate changes differently in different time scales
indicates that there is a discernible human influence and geographical scales. Many studies have reported
on the global climate. that the global mean surface air temperatures have
“Climate” refers to the average weather experienced significantly increased in the last century and
over a long period, typically 30 years. This includes pronounced warming has occurred in the last three
temperature, wind, and rainfall patterns. The Earth’s decades (Jones et al 1999; Houghton et al 2001; Jones
climate is not static, and has changed many times in and Moberg 2003). On the regional scale, Kothawale
response to a variety of natural causes. and Rupa Kumar (2005) have examined trends in
The term “climate change” usually refers to changes surface air temperature over India from 1901–2003
that have been observed since the early 1900s. These and reported that the annual mean, maximum as well
changes in global climate are likely to be due to a as minimum temperatures have significantly increased
combination of both natural and human causes: by 0.20C per decade in last three decades. It is
The Earth’s climate varies naturally as a result of remarkable acceleration of the warming trend
interactions between the ocean and the atmosphere, compared to the trends of past century (1901–2003).
However, manifestation of such changes on shorter

timescale has been recognized to be one of the most RESULTS AND DISCUSSION
crucial factors in terms of socio-economic impacts.
1. Western Himalaya
Increasing atmospheric temperatures and carbon
dioxide along with uncertainties in annual
precipitation have an adverse affect on Indian
agriculture (Aggarwal 2008).
DATA AND ANALYSIS
Mean monthly temperature data for seven
homogeneous regions, viz., Western Himalaya (WH),
Northwest (NW), North Central (NC), Northeast
(NE), West Coast (WC), East Coast (EC) and Interior
Peninsula (IP) are used in the present study over a
network of 121 stations are same as used by Pant and
Rupa Kumar (1997). This is originally from the
monthly weather records of the India Meteorological
Department (IMD). The data have been updated for
the period 1901-2003 from the Indian Daily Weather
Reports (IDWRs) published by the IMD.
We can use the linear regression equation for the Fig. 1 Temperature homogeneous regions of India
trend line . We can apply a statistical The analysis shows change of Average temperature
treatment known as linear regression to the data and over Western Himalaya (WH) has increased by about
determine these constants. 0.53°C and average temperature is 11.86 °C and
standard derivation is 0.49°C (Figure-2). Highest and
Given a set of years and temperature data set lowest temperatures are 13.4 °C (1999) and 10.78 °C
with n is number of years. (1920). The analysis also shows the numbers of cold
and warm cases are 16 and 14 respectively during the
period of study. Table 1 shows the number of cold
and warm years during different decades. An increase
m is slope in maximum temperature is about 0.67°C/decades,
maximum temperature is about 18.03°C/decades, and
standard derivation is about 0.72°C. The highest and
lowest temperatures are 20.09°C (1999) and 16.17°C
(1957) respectively. An increase in minimum
b is y-intercept temperature is about 0.40°C and standard derivation
n
n
about 0.47°C. Highest and lowest minimum
temperatures are 6.78 °C (1956) and 4.52 °C (1975)
n respectively during the period of present study.
σ is standard derivation
Western Himalaya
1.5
1
x is mean
T e m p e r a tu r e (° C )
0.5
n
n
0
n
Coefficient of Variation ( C.V.) = X 100 -0.5

-1
-1.5
A Cold/warm year can be classifies as
1901 1911 1921 1931 1941 1951 1961 1971 1981 1991
Warm Year years
Tavg - σ ≤ +1 Fig. 2 Average surface air temperatures (°C) over
Cold Year Western Himalaya. (anomalies are computed from
mean 1901-2000). Trend is found using 5 points
Tavg - σ ≤ -1 moving average.
Tavg is the average rainfall and σ is standard derivation
for period 1901-2000.

Table 1 Cold (C) and Warm (W) years over 3. North-West India
homogeneous region of the India
Decades WHI NEI NWI NCI ECI WCI IPI The analysis of average temperature over the North-
C WC WC W C WC WC WC W West India (NWI) have shown that the temperature
1900-10 4 0 4 0 1 2 2 1 2 0 7 0 0 1 has increased about 0.07°C during 20th century,
average temperature is 25.24°C, and standard
1911-20 3 0 6 0 1 1 3 0 2 0 2 0 4 0
derivation is 0.38°C (Figure-4). The highest and
1921-30 0 1 1 0 1 1 3 1 2 0 1 0 1 2 lowest temperatures are 26.08 °C (1921) and 24.08 °C
1931-40 0 2 0 0 3 0 3 0 3 0 0 0 3 2 (1917) respectively. The analysis shows the number
1941-50 2 2 0 1 2 2 1 1 1 1 2 1 0 0 of cold and warm years of 14 and 16 for the same
period. Table 1 shows an increase in maximum
1951-60 1 3 0 3 0 3 0 3 2 1 1 0 0 0
temperature is about 0.41°C and average maximum
1961-70 3 0 0 1 1 1 0 2 2 0 0 3 3 1 temperature is about 32.08°C and standard derivation
1971-80 1 1 1 1 2 0 1 1 2 3 2 1 2 3 about 0.49°C. The highest and lowest maximum
1981-90 2 1 0 0 2 2 1 1 0 4 0 3 0 1 temperatures are 33°C (1969) and 30.32°C (1917)
respectively. Decreases in minimum temperature is
1990- 0 4 0 5 1 4 0 4 0 6 0 7 2 5
2000 about 0.27°C, average minimum temperature is about
Total 16 14 12 11 14 16 14 14 16 15 15 15 15 15 18.40 °C and standard derivation about 0.363144 °C .
We have found that High and Low minimum
temperature year for this period 19.28 °C (1915) and
17.71 °C (1945) respectively.
2. North-East India
The analysis shows change in average temperature North West India
over North-East India (NEI) has increased about 0.55 1
°C during 20th century, average temperature is 23.31
°C, and standard derivation is 0.30 °C (Figure-3). We 0.5
have found that High and Low temperature year for

0
this period are 24.17 °C (1958) and 22.35 °C (1905)
respectively. The numbers of cold and warm year are
-0.5
12 and 11 for the same period. Table 1 is shows the
number of cold and warm year in various decades. -1
The average maximum temperature is about 28.63°C
and standard derivation 0.41°C. The highest and -1.5
lowest temperatures are 29.61 °C (1966 & 1979) and 1901 1911 1921 1931 1941 1951 1961 1971 1981 1991
27.38 °C (1905) respectively. An increase of Years
minimum temperature is about 0.09°C and average
minimum temperature is about 17.98°C and standard Fig. 4 Average surface air temperatures (°C) over
derivation is about 0.31°C. The Highest and lowest North West India. (anomalies are computed from
temperatures are 19.98°C (1958) and 17.07°C (1971) mean 1901-2000). Trend is found using 5 points
respectively. moving average.
4. North-Central India
No rth Eas t India
1 The analysis shows a change in average temperature
over North-Central India (NCI) by 0.41°C during 20th
T e m p e r a tur e (°C )
0 .5 century, average temperature is 24.31°C, and standard

derivation is 0.34°C (Figure-5). Highest and Lowest
0
temperature years for this period are 25.17°C (1958)
-0 .5 and 23.18°C (1917) respectively. The numbers of cold
and warm years are 14 and 14 respectively for the
-1 same period.
-1.5 Table 1 shows an increase in maximum temperature
19 0 1 19 11 19 2 1 19 3 1 19 4 1 19 51 19 6 1 19 71 19 8 1 19 9 1 of about 0.68°C, average maximum temperature about
Years 30.68°C and standard derivation about 0.45°C. High
and Low maximum temperature year for this period is
Fig. 3 Average surface air temperatures (°C) over 31.90 °C (1966) and 29.10 °C (1917) respectively.
North East India. (anomalies are computed from mean Increase of minimum temperature is about 0.15°C and
1901-2000). Trend is found using 5 points moving average minimum temperature about 17.95°C and
average. standard derivation is about 0.353391°C. Highest and
Lowest temperature year for this period are 18.98 °C
(1958) and 17.08 °C (1971) respectively.

6.West-Coast India
North Central India
1 The analysis of Average temperature over West-Coast
India (WCI) shows an increase of temperature about
0.5
T em p era t u re (° C ) 0.67°C of during 20th century, average temperature is
0
25.90°C, and standard derivation is 0.29°C (Figure-7).
We have found that High and Low temperature year
-0.5 for this period (1998) 26.73°C and (1917) 25.15°C.
We have seen the number of cold and warm year 15
-1 and 15 for the same period. An increase in maximum
temperature is about 1.13°C and average maximum
-1.5
temperature is about 30.82°C and standard derivation
1901 1911 1921 1931 1941 1951 1961 1971 1981 1991
Years about 0.41°C. Warmest and coldest years are for this
period is 1987 (31.86°C) and 1917 (29.73°C)
Fig. 5 Average surface air temperatures (°C) over respectively. Increase in minimum temperature is
North Central India. (anomalies are computed from
about 0.19°C and average minimum temperature
mean 1901-2000). Trend is found using 5 points
about 20.97°C and standard derivation about
moving average.
0.263816°C. We have found that High and Low
5. East-Coast India minimum temperature year for this period 21.68 °C
(1998) and 20.38 °C (1910) respectively.
The analysis of average temperature over the East-
Coast India (ECI) shows an increase in temperature Wes t Co st India
about 0.42°C during 20th century, average temperature 0.9
is 27.91°C, and standard derivation 0.25°C (Figure-6). 0.6
Highest and Lowest temperatures for this period are T e m pe r a tur e (°C )
28.64°C (1998) and 27.26°C (1917) respectively. The 0.3
number of cold and warm year is 14 and 14 for the
0
same period. An increase in maximum temperature is
about 0.54°C, average maximum temperature is about -0.3
32.34°C and standard derivation about 0.31°C.
Highest and Lowest temperature for this period is -0.6
32.98 °C (1998) and 31.52 °C (1917) respectively. -0.9
Increase in minimum temperature is about 0.29°C and 19 0 1 19 11 192 1 19 3 1 19 4 1 19 51 196 1 19 71 19 8 1 19 9 1
average minimum temperature about 23.49°C and Years
standard derivation 0.28°C. Highest and Lowest
minimum temperature for this period is 24.29 °C Fig. 7 Average surface air temperatures (°C) over
(1998) and 22.91 °C (1971) respectively. West Cost India. (anomalies are computed from mean
1901-2000). Trend is found using 5 points moving
average.
Eas t Cos t India 7. Interior Peninsular India
1
The analysis of Average temperature over Interior
0.5 Peninsular India (IPI) shows an increase of about

0.45°C during 20th century, average temperature is
0 28.57 °C, and standard derivation is 0.659469 °C
(Figure-8). We have found warmest and coldest
-0.5 temperature of 30.16°C (1973) and 26.75°C (1918)
respectively. Table 1 shows the number of cold and
-1 warm year of 15 and 15 for the same period. An
190 1 1911 192 1 193 1 194 1 19 51 19 61 19 71 19 81 19 91 increase in maximum temperature is about 0.42°C.
Years Average maximum temperature is about 32.39 °C and
standard derivation about 0.36°C. Warmest and
Fig. 6 Average surface air temperatures (°C) over
East Cost India. (anomalies are computed from mean coldest years are during study period is 1980
1901-2000). Trend is found using 5 points moving (33.03°C) and 1917 (31.17°C) respectively. An
average. increase in minimum temperature is about 0.39°C and
average minimum temperature about 20.52 °C and
standard derivation about 0.29. While High and Low
minimum temperature year for this period 21.31 °C
(1998) and 19.86 °C (1904 & 1917) respectively.

indices to identify change in climatic extreme;
Interio r Penins ular India Climate Change 42 131–149.
2
Kothawale D R and Rupa Kumar K (2005), On the
1
recent changes in surface temperature trends over
India; Geophys. Res. Lett. 32 L18714,
doi:10.1029/2005GL023528.
0
Maurya R K , G P Singh and P K Singh ( 2010) :
-1 Change in temperature and precipitation over
India and its homogeneous zones during 20th
-2 century; Nava Gavesana Vol 1, Issue 2, Oct-Dec,
page 16-33.
19 0 1 19 11 19 2 1 19 3 1 19 4 1 19 51 19 6 1 19 71 19 8 1 19 9 1 .
Years
Fig. 8 Average surface air temperatures (°C) over

Interior Peninsular India. (anomalies are computed
from mean 1901-2000). Trend is found using 5 points
moving average.
CONCLUSION
Analysis of the mean monthly temperature seven
homogeneous regions, viz., West Himalaya (WH),
Northwest India (NWI), North Central India (NCI),
Northeast India (NEI), West Coast (WC), East Coast
(EC), and Interior Peninsula (IP) has been studied for
the period of 1901-2000. Detail analysis on annual
and seasonal changes in mean monthly temperature
(Tavg), mean maximum temperature (Tmax) and mean
minimum temperature (Tmin) have been done. Each
year was divided into four principal seasons, viz.
winter, pre-monsoon, monsoon, and post-monsoon.
The analysis shows that the annual mean temperature,
mean maximum temperature, and mean minimum
temperature are increasing. As far as different
homogeneous region are concerned, the analysis
shows that the WH, NWI, NCI, NEI, WC, EC and IP
show a rising trend of approximately 0.53°C,
0.07°C,0.41°C, 0.55°C , 0.67°C , 0.42°C and 0.45°C
per hundred year respectively. The largest
temperature rising are found over the WC and
minimum over NWI.
ACKNOWLEDGMENTS
The authors also want to acknowledge to the
Department Science and Technology, Government of
India (No.SR/AS/05/2008) for financial support in the
form of project.
REFERENCES
Aggarwal PK (2008): Global climate change and
Indian agriculture: impacts, adaptation and
mitigation. Indian J Agric Sci 78:911–919.
Houghton J T, Ding Y H, Griggs D J, Noguer M, van
der Linden P J, Dai X, Maskell K and Johson C A
(eds) 2001 The Scientific Basis. Contribution of
Working Group I to the Third Assessment Report
of the Intergovernmental Panel on Climate Change
(Cambridge: Cambridge University Press).
Jones P D, Horton E B, Folland C K, Hulme M,
Parker D E and Basnett T A (1999) The use of

ISBN 978-81-921121-0-7
RELATE AVERAGE CONCENTRATION OF SEDIMENT TO LOCAL

CONCENTRATION FOR STEADY UNIFORM FLOW
N.D.Vernekar
Post Graduate student, Bharati Vidyapeeth University College of Engineering,Pune - 411043, Email: nvgec78@gmail.com
A.R.Bhalerao
Principal & Dean, Bharati Vidyapeeth University College of Engineering,Pune -411043, Email: arbhalerao@bvucoep.edu
ABSTRACT: The measurement of sediment load involves determination of velocity of flow and concentration of
sediment at large number of points in an alluvial channel. This process proves to be time consuming. The existing
theories and empirical equations available for computation of sediment load give approximate values. In practice, the
concentration is measured at 0.6 times the depth below the surface. Sampling at 0.6 times the depth has been worked in
India and some parts of U.S.A., which gives the mean concentration, presumably because the mean velocity occurs
approximately at this level. However this does not hold true at all times because the concentration in the vertical varies
with the flow conditions, size and nature of sediment. Keeping in view these factors, this research involves utilizing the
laboratory data to determine ratio of depth at which the average concentration of the sediment lies under given
conditions of shear velocity, temperature, slope, depth of flow .Also it includes determination of a dimensionless
parameter involving the above variables and thus relates average concentration of sediment to local concentration at a
point under steady uniform flow. This research thus involves analysing the concept of sediment diffusion coefficient
from the available laboratory and field data.
INTRODUCTION C = concentration at depth y from the bed of channel; u =

For sediment transport Einstein assumed that bed load velocity at depth y from bed of channel; D = depth of
transport of a given size range,iBqB occurs in a bed layer flow; d = representative grain size.
of thickness 2d.Here in qB is the rate of bed load transport The average concentration, Cavg ,is obtained by
in weight per unit width per time,iB is the fraction of bed q
load of a given size range and d is the representative s
C  (2)
grain size.Since there will be continuity in the distribution avg q
of the suspended load and the bed load it is assumed that
average concentration of the bed load in the bed layer where q = discharge per unit width.Using the above
must be equal to the concentration of suspended load at equations for computations of average concentration
2d.This theory has been taken into consideration and from proves to be tedious and time consuming.Thus an attempt
the concentration profiles of Laboratory data by James has been made to obtain the dimensionless parameter.
R.Barton and Pin Nam Lin,Emmet M.Laursen and
adopting the method of Integrating curves the total DETERMINATION OF DIMENSIONLESS
suspended load (qs) is computed. This method proves to PARAMETER (P)
be tedious. Hence further an attempt is made to obtain a From the laboratory data (Barton & Lin,1955;Laursen,
dimensionsless parameter by adopting the equation as 1957) the dimensionless parameter is obtained by
suggested by E.W.Lane and A.A.Kalinske and plot curves equation suggested by E.W Lane and A.A.Kalinske
from which we can obtain total suspended load and also (Garde & RangaRaju,2000)
can relate average concentration of sediment to local
concentration at a point.Another aspect of this paper is to
  ω  
o
q  qC Pexp 15 A (3)
  *  

obtain values of sediment diffusion s a u
coefficient,momentum transfer coefficient from the
available laboratory and field data and verify the where Ca = concentration of sediment at depth ‘a’ from
approximate method as suggested by E.W.Lane and the bed;qs = sediment transport rate per unit width;ωo =
A.A.Kalinske. fall velocity of sediment particle;u*=shear
velocity;A=a/D;D=depth of flow;P is a function of ωo/ u*
COMPUTATION OF TOTAL SUSPENDED LOAD and u*/kou.using Mannings equation for u, u*/kou can be
(qS) AND AVERAGE CONCENTRATION (Cavg). expressed in terms of n/D1/6,where n=d1/6/21;d = sediment
The total suspended load is computed by equation size;ko= Karman’s constant;n = Manning’s roughness
D coefficient.
q   Cudy (1) The fall velocity of sediment at a given temperature is
s determined from the graph of sediment size(d) v/s fall
2d
velocity(ωo) (Garde & RangaRaju, 2000).
where qs = sediment transport rate per unit width Utilising the laboratory data (Barton and Lin,1955
;Laursen, 1957) and by adopting the Eq.(3) graph of P

v/s ωo/u* & n/D1/6 in SI units (Fig.2) is plotted for values Lane and Kalinske factor is calculated by equation
of n/D1/6=0.012 to 0.014 & n/D1/6 =0.015 to 0.018.
D gDS
L (7)
15
where L=Lane and Kalinske factor (m2s-1); D=depth of

flow (m); g=gravitational acceleration (ms-2); S=slope.
From the values obtained for εs , εm, L a graph of D v/s εs ,

εm,L is plotted for laboratory data and for field
data.(Fig.3) shows plot of D v/s εs , εm,L for laboratory
data and (Fig.4) shows plot of D v/s εs , εm,L for field
data.
Fig.2 Graph of P v/s ωo/u* & n/D1/6
COMPUTATION OF SEDIMENT DIFFUSION

COEFFICIENT,MOMENTUM TRANSFER
COEFFICIENT,LANE AND KALINSKE FACTOR
In equations involving determination of sediment
concentration,these equations are arrived at based on the
assumption that sediment diffusion coefficient and
momentum transfer coefficient are equal. The
approximate method suggested by Lane and Kalinske
showed that a constant value of sediment diffusion
coefficient and momentum transfer coefficient can be Fig.3 Graph D v/s εs , εm,L for laboratory data
used in the vertical for wide rivers (Garde &
RangaRaju,2000).
The approximate method suggested by Lane and Kalinske
is given by equation
D gDS
ε ε  (4)
s m 15
where εs = sediment diffusion coefficient;εm= momentum
transfer coefficient;g = gravitational acceleration;D=depth
of flow;S = slope.
From the available laboratory data (Barton &
Lin,1955;Laursen,1957) and field data
(Brownlie,1981;Molina,2001) ,the sediment diffusion
coefficient is calculated by equation
c Fig.4 Graph of D v/s εs , εm,L for field data
ω Cε 0 (5)
o s y
where ωo=fall velocity of particle (ms-1); C=time From the values thus obtained we arrive at
averaged concentration (ppm); εs=sediment diffusion New equation
coefficient (m2s-1). Du *
εs  for d ≤ 0.2mm (8)
Momentum transfer coefficient is calculated by equation 15
D  y 0.885
ε  u ky   (6) Du *
m *  D  εs  for d >0.2mm (9)
45
where εm =momentum transfer coefficient (m2s-1); where εs=sediment diffusion coefficient (m2s-1); d=depth
u*=shear velocity (ms-1); k=karman’s constant; of flow (m); u* = shear velocity (ms-1); d = diameter of
y=reference depth (m); D=depth of flow (m). sediment particle (mm).

CONCLUSIONS
From (Fig.2) plot of P v/s (ωo/u*) & n/D1/6 for known
values of (ωo/u*), n/D1/6,q & Ca,qs can be computed and
for various sediment size mixtures we can plot such
curves wherein we can relate average concentration of
sediment to local concentration at a point and the average
concentration depends on the depth of flow,size of
sediment particle,fall velocity of particle,& shear velocity.
From (Fig.3), (Fig.4) plot of D v/s εs,εm,L we can say that

sediment diffusion coefficient,momentum transfer
coefficient,Lane and Kalinske factor are not equal and
from the new equation arrived at we can compute the
values of sediment diffusion coefficient if we know the
slope,shear velocity and depth of flow.
ACKNOWLEDGMENTS
The writers are grateful to late Dr.R.J.Garde for his
guidance during the studies.
REFERENCES
Brownlie, W.R. (1981). Compilation of Alluvial channel
data.
Barton, J.R. and Lin,P.N. (1960). A study of sediment
transport in alluvial channels.
Garde R.J. and RangaRaju K.G. (2000). Mechanics of
Sediment Transportation and Alluvial stream
Problems.
Grenfell,S.E. and Ellery E.N. (2009).
Hydrology,Sediment Transport Dynamics and
geomorphology of a variable flow river:The Mfolozi
River,South Africa.Water SA Vol.35,No.3,271-278.
Laursen, E.M. (1957). An Investigation of Total Sediment
Load.
Molina,A. (2001). Transport of Sediment in large sand
bed rivers.Journal of Hydraulic
Research,Vol.39,No2,135-145.
SYMBOLS
q rate of flow per unit width (m2s-1)
qs rate of sediment transport per unit width
Cavg average concentration of sediment
Ca concentration at reference depth
D Depth of flow (m)
U* shear velocity (ms-1)
n Manning’s roughness coefficient
A elevation of reference concentration
L Lane and Kalinske factor (m2 s-1)
εs sediment diffusion coefficient (m2s-1)
εm momentum transfer coefficient (m2s-1)
ωo fall velocity of sediment particle (ms-1)
d diameter of sediment particle (mm)

ISBN 978-81-921121-0-7
GEOTECHNICAL PROPERTIES OF SOIL CONTAMINATED BY

TANNING INDUSTRY EFFLUENTS
Ashwani Jain
Associate Professor, Department of Civil Engineering, NIT, Kurukshetra-136119, Email: ashwani.jain@rediffmail.com
ABSTRACT: Effluents from industries cause a lot of nuisance to the public. Normally, people are more
concerned about environmental problems; the engineering implications of soil pollution are not appreciated by
them. Discharge of untreated industrial effluents on to the open ground affects not only the index properties of
the underlying soils, but also their engineering behaviour. This paper describes the experimental work that has
been conducted to evaluate the effect of contamination by tanning industry effluents on the properties of soil.
Undisturbed and disturbed soil samples have been collected from the site of a tanning industry, from depths
varying from 0.8 to 1.2 m. The soil samples so collected have been subjected to comprehensive tests, both
chemical and physical. The results indicate that not only the index properties, but also the engineering properties
of soil have been adversely affected due to contamination by tanning industry effluents. It has been observed
that the shear strength of soil decreases, and plasticity index, compressibility and permeability increase with
increase in degree of contamination.
INTRODUCTION imparts a negative charge to the surface of soil

Tanning of hides and skins to convert them into particles and acts as a dispersing agent; the
leather has been an important industrial activity since consequent decrease in inter-particle attractive forces
antiquity. It is a versatile industry adaptable to large has been considered to be responsible for the loss of
and small scales, and to high and low levels of undrained shear strength.
sophistication, and consequently it is found in almost Srivastava et al. (1994) have studied the effect of
every country on earth. The potential environmental waste water from four industries, namely sugar mill,
impact of tannery operations is widely acknowledged. paper mill, fertilizer plant and a power generation
There are fewer consensuses, however, about how and equipment manufacturing industry, on a typical
by whom these impacts should be resolved. Questions alluvial (CL-ML) soil. To study the interaction
of chemical safety, contamination of land and ground effects, soil has been studied with distilled water first
water, inadequate provision for solid waste and sludge and then with the waste waters from the four
disposal, spills and accidents involving chemical industries separately. The study reported engineering
substances, and the general nuisance of odour and and chemical properties of soil, chemical analysis of
noise from installations are taken into account more waste water before and after its interaction with soil
and more often by regulatory and planning authorities. and changes in the soil properties. All the four
Pollution from tanneries has a negative long term effluents were causing overall deterioration in soil
impact on the growth potential of a country, properties and the chemistry of waste water and soil
irrespective of the immediate economic benefits of was thought to be responsible for the interaction
production. Polluted water, air or soil affect people’s behaviour.
health and damages ecological processes that sustain Meegoda et al. (1994) have studied the factors that
the production of food. control the compression index of contaminated soils.
The objective of present study is to determine the The two soils tested are kaolin (clay of low plasticity)
effect of effluents from tanning industry on the and a mixture of bentonite (15% by weight) and
properties of soil in the vicinity of the industry. kaolin (85% by weight) (clay of high plasticity).
Chemicals used in this study are glycerol and 1-
SOME LABORATORY STUDIES propanol. Experimental results indicated that the
Kumpaley et al. (1985) studied the case of accidental compressibility of the two soils used in this study
spillage of a highly concentrated caustic soda solution changed with the type and amount of chemicals in
in to the subsoil. As a result of the spillage and pore fluids.
seepage through cracked drains in an industrial Meegoda et al. (1995) have performed experimental
establishment in Ghana, considerable structural investigation to evaluate the effects of oil
damage has been caused to light industrial buildings contamination on soils and to establish a methodology
in the factory, in addition to localized subsidence. The to identify and classify contaminated soils. Fine-
pH of the uncontaminated soil was 6.5 while that of grained oil contaminated soils are often identified and
the contaminated sub-soil was as high as 10 to 11. classified as granular soils with large particle sizes
Deterioration in soil strength as a result of chemical and may result in selection of improper treatment
reactions in soil was considered as a possible cause of techniques. In this study, four uncontaminated soils
structural distress. Laboratory investigations showed were first identified and classified; then all four soils
general decrease in the undrained strength of soil with were mixed with 3% motor oil. The soils are
increasing caustic soda concentration. Caustic soda classified as clay of high plasticity, clay of low

plasticity, silty clay and silty sand. The oil- parameters in the soil samples increase as the distance
contaminated soils were first tested for identification from the source of contamination decreases.
and classification and then treated by (a) heating in Table 1 Chemical used in the tanning process
ovens at various temperatures (low – temperature (Source UNIDO)
desorption), (b) adding solvents (solvent extraction), Chemicals/Materials Heavy Light Leather
and (c) adding surfactants (treatment with Leather
(kg per 100 kg of raw hides)
surfactants). The processed soils were then
reidentified and reclassified. Test results showed that General Purpose Chemicals
Sodium Sulphide, Na2S 3.0 3.0
treatment with a surfactant produced near virgin soils Calcium Hydroxide, Ca(OH)2 4.5 4.5
for all four soil types. The solvent extraction was
adequate only for the silty sand and the silty clay. The Hydrochloric Acid, HCl 0.3 0.3
low-temperature thermal treatment was inadequate for Ammonium 2.0 2.0
all four soils. To further confirm the above test Sulphate,(NH4 )2SO4
results, the low plasticity clay was mixed with 6% Sodium Bisulphate, NaHSO3 1.5 1.5
motor oil and the above test programme repeated. Sodium Chloride, NaCl 10.0 10.0
This soil with a higher degree of contamination Calcium Formate, Ca(COOH)2 - 2.0
produced same results. To provide an explanation for
the change in soil behavior due to oil contamination, Sulphuric Acid, H2SO4 4.0 4.0
low-plastic soil was mixed with oil up to 70% by Sodium Carbonate, Na2SO3 - 2.0
weight and Atterberg limit tests performed. It was
Sodium Sulphite - 2.0
concluded that fine grained soils, when contaminated
Basic Tanning Materials
with oil, behave as granular soils with particle Chrome Salts Cr2 (SO4)3 - 10.0
aggregation. Since soil in this study was artificially Vegetable tanning materials 12.0 3.0
contaminated, one should exert a caution when
applying the results of this study to soils in contact Performance Chemicals
with oil for many years. Bates 0.8 0.8
Bactericides 0.3 0.3
Syntans - 3.0
CHEMICALS IN TANNING INDUSTRY Fat liquors - 4.0
The major chemicals used in the tanning process are Dyeing auxiliaries - 3.8
shown in the Table 1. Table 2 shows the composition Dyes - 0.6
of a typical combined tannery effluent that has not yet Finishes - 4.0
been treated. It is characterized by a high oxygen Table 2 Composition of typical untreated combined
demand and high salt content and is strongly alkaline. tannery effluent
It also contains high level of suspended solids and Parameter (mg/lit) Chrome Vegetable Tannage
possibly a persistent high land of chrome. Tannage
pH 9 9
SAMPLING DETAILS OF SOIL Total Solids 10000 10000
The industry chosen for this study is tanning industry Total ash 6000 6000
(leather producers) near Karnal. Undisturbed and Suspended solids 2500 1500
disturbed soil samples have been collected from three Ash in suspended solids 1000 500
different locations of this industry. The tubes used for Settled solids (2h) 100 50
the collection of undisturbed soil samples are of 10 BOD5 900 1700
KMnO4 value 1000 2500
cm and 30 cm length. Sample 1 has been withdrawn
COD (K2 Cr2O7) 2500 3000
from a depth of 1.2m at a site far away from the point
Sulphide 160 160
of disposal of effluent and is considered to be an
Total Nitrogen 120 120
uncontaminated soil sample not having any pollution
Ammonia nitrogen 70 70
effect. Sample 2 has been withdrawn from a depth of Chrome (Cr) 70 -
1.2m and sample 3 from a depth of 0.8m at sites near Chloride (Cl-) 2500 2500
to the point of disposal of effluent. Sample 2 and Sulphate (SO4) 2000 2000
sample 3 are partially and highly contaminated soil Phosphorus (P) 1 1
samples respectively. All the soil samples have been Ether extractable 200 200
collected below the storage pond level. Table 3 Chemical properties of the soil samples
Parameter Sample-1 Sample-2 Sample-3
CHEMICAL PROPERTIES OF SOIL SAMPLES pH 7.1 8.5 9.2
Samples collected from the three locations have been Total soluble
11.27 20.15 23.42
Sulphate (%)
subjected to chemical analysis for determination of Chlorides, ppm 192 1860 2942
pH, total soluble sulphates, chlorides and organic Organic matter
matter as per various IS specifications. Chemical 0.72 3.25 4.97
(%)
properties have been summarized in Table 3. In Total soluble
16.92 21.34 25.63
general, it has been observed that the values of these solids (%)

PHYSICAL PROPERTIES OF SOIL SAMPLES Compressibility of polluted samples is greater as is
The samples collected from the three locations have seen from the increase in compression index values.
been subjected to tests for determination of various Shear strength was determined by testing the soil
index and engineering properties as per IS samples in direct shear apparatus. Consolidated
specifications. These properties have been drained tests were conducted on undisturbed samples.
summarized in Table 4. It has been observed that the It has been observed that there is a decrease in
clay size content decreases due to alkaline pollution strength of polluted samples.
from tanning industry. This change may be attributed
to the flocculation occurring in alkaline environment. CONCLUSIONS
All the soil samples have been classified as clayey The study indicates that both chemical and physical
sand (SC) as per IS Classification. In general, the properties of soil are adversely affected from
consistency limits of the soil including its plasticity engineering point of view due to contamination by
index increased with increase in the degree of untreated tanning industry effluents. When industrial
contamination. It is observed that chemically polluted wastewater interacts with soil, complex physico-
soil samples showed higher optimum moisture content chemical process takes place, resulting in significant
values and lower maximum dry density values as change of soil properties. In general, it has been
compared to the uncontaminated soil samples. It has observed that the pH value, total soluble sulphate,
also been observed that chemically polluted soil chlorides, organic matter and total soluble solids in
samples exhibited higher permeability values than the soil samples increase as the distance from the
unpolluted samples which may be due to higher void source of contamination decreases. It has also been
ratios shown by contaminated samples. observed that the shear strength of soil decreases, and
plasticity index, compressibility and permeability
Table 4 Index and engineering properties of soil increase with increase in degree of contamination.
samples REFERENCES
Soil Property Sample Sample Sample IS: 2720 (Part - 21) (1977). Methods of test for soils:
1 2 3 Determination of total soluble solids. Bureau of
Index Properties Indian Standards.
Gravel (%) 5.00 5.5 5.5 IS: 2720 (Part - 22) (1972). Methods of test for soils:
Sand (%) 51.17 52.25 53.56 Determination of organic matter. Bureau of Indian
Silt size (%) 20.12 21.04 21.07 Standards.
Clay size (%) 23.71 21.21 19.87 IS: 2720 (Part - 26) (1987). Methods of test for soils:
Liquid Limit 34.08 41.21 49.35 Determination of pH value. Bureau of Indian
(%) Standards.
Plastic Limit 21.23 23.45 27.26 IS; 2720 (Part - 27) (1977). Methods of test for soils:
(%) Determination of total soluble Sulphate. Bureau of
Plasticity Index 12.85 17.76 22.09 Indian Standards.
Kumapley, N. and Ishola, N.K. (1985). The effect of
IS SC SC SC
chemical contamination on soil strength. Proc. of
Classification
XI International Conference on Soil Mechanics
Activity Ratio 0.54 0.83 1.11
Foundation Engineering, San Francisco, Vol. 3,
Natural Water
21.65 30.83 35.76 1199–1201.
Content (%)
Meegoda, N.J. and Ratnaweera, P. (1994).
Field Dry
1.78 1.59 1.41 Compressibility of contaminated fine grained soils.
Density, (g/cc)
Geotechnical Testing Journal, Vol.17, No.1,
Degree of March, 101-112.
71.0 85.20 91.9
saturation (%) Meegoda, N.J. and Ratnweera, P. (1995). Treatment
Initial void ratio 0.69 0.86 0.95 of oil contaminated soils for identification and
Optimum classification. Geotechnical Testing Journal,
moisture 14.9 15.9 16.5 Vol.18, No.1, March, 41-49.
content (%) Srivastava, R.K. et al. (1994). Laboratory study of
Maximum dry soil-industrial waste water interaction behavior.
1.90 1.83 1.79
density (g/cc) Proceedings XIII ICSMFE, New Delhi, India,
Engineering Properties 1553-1556.
Permeability, Sudhakar Rao, M. and Sridharan, A. (1993).
1.89 3.31 3.92
cm/sec (x 10-6) Environmental Geotechnics - A Review. Indian
Compression Geotechnical Journal, Vol. 23, 235-252.
0.192 0.214 0.271
Index
Shear c 0 0 0
Parameters  37.67 o
33.50 o
30.10o

ISBN 978-81-921121-0-7
WATER MANAGEMENT ON DRIP AND MICRO SPRINKLER

IRRIGATION FOR SUMMER GROUNDNUT CROP UNDER DEFICIT
WATER SUPPLIES
A. B. Rathod
Department of Water Resources Management, Civil Engineering, L. D. College of Engineering, Ahmadabad
anilrathod_222399@yahoo.com
S. A. Trivedi
shaileshatrivedi@yahoo.co.in
ABSTRACT: The experiments were conducted during summer season (Feb. to May) for the consecutive 3
years ( to assess the groundnut crop response to different 6 irrigation levels. (IW/CPE=0.6, 0.7, 0.8, 0.9, 1.0 and
1.2, ratio of irrigation application to cumulative pan evaporation of days from last irrigation) under drip and
micro sprinkler irrigation and its comparisons. Under the micro sprinkler irrigation, the lowest and the highest
pod yield were observed as 2948 kg / ha and 5319 kg/ha under the IWCPE of 0.6 and 1.2 requiring the seasonal
irrigation of 523 ha.mm/.ha and 1047 ha.mm./ha. Through adopting the drip system, the lowest and highest pod
yield were obtained as 1917 kg/ha and 2927 kg/ha in the IW/CPE of 0.6 and 0.9 respectively requiring 502
ha.mm.ha and 757 ha.mm.ha. Under the drip irrigation, the pod yield was increase by 30.32, 23.56, 17.28,
14.78, 15.66 and 17.28% over micro sprinkler irrigation. The seasonal irrigation water as input and pod/ fodder
yield as output could be assembled in quadratic form. The highest possible water use efficiency under micro
sprinkler and drip irrigation fro summer groundnut crop can be 3.40 kg/ha.mm and 4.148 kg/ha.mm
respectively. The highest total net return (including additional net return from water saving due to deficit
application) under micro sprinkler and drip irrigation were found as Rs.16574 in the IW/CPE o 0.9 and
Rs.19,940 in IWCPE of 0.8 by micro sprinkler irrigation and drip irrigation respectively. The ratio of the
additional befits to additional cost of drip irrigation system adoption replacing micro sprinkler irrigation was
increased from 1.69 in IW/CPE of 0.9 to 2.24 in IW/CPE of 0.8
INTRODUCTION frequent droughts during monsoon and early

withdrawal of monsoon, the life saving and
The land availability being limited, leaving little supplement irrigation have to be applied to monsoon
scope for the horizontal expansion, efficient crops. So whatever the groundwater have been
management of the agricultural resources is rapidly recharged, the major part of its are being exploited
becoming a key issue for agricultural development to during monsoon and winter seasons only. Due to
feed ever-growing population from ever-limited adverse topographical conditions and uneven
resources. The judicious management of the resources distribution of the rainfall over monsoon season, little
as a means of increasing production needs, knowledge amount of rainwater is stored as ground water and left
of crop response to water, one of the important inputs. runs away to sea. So there is acute shortage of water
The Saurashtra region of Gujarat state is having 84% of irrigation during summer season. However keeping
of cultivable land under dry arming and only 16% in view the productivity of summer groundnut, some
area under irrigation mainly on dug wells. The total water management strategies should be thought so
annual rainfall if insufficient, scanty and erratic in that we can cultivate the summer groundnut crop from
nature. The coefficient o variation is more than 40%. limited water supplied. This can be achieved by
Out of total annual rainfall 80% of seasonal rainfall adopting the water efficient irrigation methods. Also
received during the month of July and August only. there is question that whether the farmer should grow
Due to intensive rainfall over short period, most of the less areas with more water or more area with deficit
run of water drains away to the sea and left little irrigation. Therefore, the research is needed to study
scope to recharge the aquifer in this region. The of drip irrigation as compared to conventional micro
climate o this region favors to cultivate groundnut sprinkler method or this reason.
crop during monsoon and summer season. During The farmers apply the seasonal irrigation depth of
monsoon season, due to scanty, erratic and uncertain more than 100 cm to her summer groundnut crop
rainfall as well as having more chances of insect and without having any scientific supports. Excessive
pest dieses, the yield is low and uncertain. During irrigation water application coupled with low water
summer season, the yield is more and certain provided availability resulted into low acreage under summer
sufficient irrigation water resources are available. In groundnut crop. Research investigations revealed that
this region, no any major irrigation project exists. The by maintaining certain moisture stress tolerable during
major source fro irrigation is ground water only. Due the crop growing season, higher yield per unit area
to frequent droughts during project exists. The major can be attained that normal and in turn, this saved
source for irrigation is ground water only Due to irrigation water can be utilized to bring additional
land into cultivation under groundnut corps. So to

bring more area under summer groundnut, two 0.7, 0.8, 0.9, 1.0 and 1.2. In the drip irrigation system,
alternates can be thought. One alternate is to increase drip line of 16 mm was fitted at 0.9 m distances so
the water resources availability and second is to that one drop line can serve two rows of the crops.
increase the water use efficiency by optimal deficit The inlet of the drip line was fitted with sub main by
water applications or adopting some advanced water the 16 mm grommet take off the other end of the drip
efficient technologies. There is little cope or line was closed by the end plug of 16 mm. The
increasing the water resources in the low rainfall area operating pressure was 1g/cm2. The length of the drip
and also it needs high investment. Therefore the line was varied to match the plot size of the
efforts were made to study of drip irrigation system replications. The water meter and pressure gauge were
and its relative superiority over micro sprinkler fitted to measure the volumetric flow and pressure of
methods in summer groundnut crop at various water inflow water respectively. Under the micro sprinkler
deficit levels because no firm conclusions had been system, the lateral of 16mm ISI was used to supply
drawn so far on water management strategies for the water to micro sprinkler fitted at 4m distances.
summer groundnut which can be recommended or The lateral to lateral distance was kept as 4.05 m. So
adoption by the farmers. the under one replication, the plot size was 16m X
8.1m. At inlet of each sub main, the pressure gauge
METHODOLOGY was fitted to measure the pressure of the inflow. The
operating pressure was kept as 2 kg/cm2. The
The experiments for all the 3 years were undertaken discharge capacity of the micro sprinkler was
during summer season at College of Agricultural 160LPH. The micro sprinklers were of big swivel
Engineering and Technology, Gujarat Agricultural type.
University, Junagadh. The summer groundnut crop
was shown in February and harvested in May. The RESULTS AND DISCUSSIONS
seeds were treated with the thrum for insects and
fungi control . The agronomic practices like seed rate Micro sprinkler Irrigation
(100 kg/ha), fertilizer does (N:P:K = 25:50:0 kg/ha)
and row spacing (0.45 m) etc. were strictly adopted as (a) Pod Yield
recommended by the university for this region. The The lowest pod yield was observed as 1471 kg/ha
sowing and fertilizer application wad done under the IW/CPE of 0.6 having seasonal irrigation
simultaneously by tractor mounted automatic seed- applications of 523 ha.mm/ha. The additional
cum-fertilizer drill. The interculturing, weeding and irrigation application an increase of pod yield of
plant protection measures were performed as per 388,525,166 and 104 kg/ha from IW/CPE of 0.6 to
requirements. 0.7, 0.7 to 0.8, 0.8 to 0.9 and 0.9 to 1.0 respectively.
The highest pod yield of 2550 kg/ha was found in the
The open pan evaporimeter was installed at IW/CPE of 0.9 having 789 ha.mm/ha irrigation water
experimental site and daily evaporation date at 8.00 requirement. Under the IW/CPE of 0.6, 0.7, 0.8, 0.9
IST were taken. The irrigation frequency was decided and 1.0 and 1.2 the water use efficiency were found as
based upon the cumulative pan evaporation and 2.81,3.01,3.40,3.23,2.81,and 2.10 kg/ha respectively.
IW/CPE ratios. The other climatic data like After the statistical analysis, it could be found that the
temperature, sunshine hours, humidity and wind treatments having IW/CPE of 0.8, 0.9, 1.0 were
velocity were also collected. The soil samples of statistically at par. Also the insignificant difference
different profile depths were taken’ to know the soil was found between the treatmetns of IW/CPE of 0.6
moisture status during irrigation intervals. and 0.7 as well as 0.7 and 0.9. The treatment of
IW/CPE of 0.8 was considered as statistically better
Two parameter namely irrigating methods (surface one. It requires 700 ha.mm/ha seasonal irrigation
and drip) and water application level were studied. water giving the yield of 2384 kg/ha.
The 6 water application levels were based on IW/CPE
vales of 0.6, 0.7, 0.8, 0.9, 1.0 and 1.2. Under the (b) Fodder Yield
different combination of these two parameters, 12 (2 The fodder production was observed as
methods X 6 IW/CPE) treatments were taken. Each 2948,3588,4465,4959,5222 and 5319 kg/ha
treatment was replicated 4 times. The split plot design respectively under the IW/CPE of 0.6, 0.7, 0.8, 0.9
was adopted. The net area of samples taken from the and 1.0 and 1.2 respectively. It could be seen that the
replication plots were 1.8 m x 9m and 2.7 m 6 m odder production was continuously increased at
under surface and drip irrigation method respectively. decreasing rate as the IW/CPE was increased from 0.6
to 1.2. The water use efficiency was obtained as
The controlled amount of water was applied by 5.63,5.81,6.37,6.28,6.01 and 5.08 kg/ha under
respective irrigation method scheduling the irrigation IW/CPE of 0.6, 0.7, 0.8, 0.9, 1.0 and 1.2 respectively.
based on the 60%, 70%, 80%, 90% 100% and 120% The significant difference was found among the
of the cumulative pan evaporation o days since 1st treatments of IW/CPE o 0.6, 0.7, 0.8 and 0.9.
irrigation under the treatment of IW/CPE ratio of 0.6, See Table 1

Drip Irrigation 23545, 29549, 37687, 40589, 39631 and 36644 Rs./ha
(a) Pod Yield respectively. As we increased the IW/CPE from 0.6 to
The pod yield of 1917, 2297, 2796, 2927, 2829 and 1.0, the gross income increased from 23545 to 39631
2586 kg/ha were observed under the IW/CPE o 0.6, Rs/.ha and then it decreased to 36644 at IW/CPE of
0.7, 0.8, 0.9, 1.0 and 1.2 respectively using 502, 589, 1.2. Also the higher benefit cost ratio (gross income to
674, 757, 843 and 1011 ha.mm/ha seasonal irrigation total cost) was found as 1.69 in the IW/CPE of 0.9.
water. The significant differences in the pod yield was The net return (gross income – total cost) in the
found among the treatments having IW/CPE of 0.6, summer groundnut crop under the IW/CPE of 0.6, 0.7,
0.7 and 0.8. However, the treatments having IW/CPE 0.8, 0.9, 1.p and 1.2 were 2722, 7598, 14740, 16574,
of 0.8 and 0.9 were found statistically at par. The pod 14668 and 9533 Rs./ha respectively. The higher ratio
yield was decreased under the treatments having of gross income to total cost (1.69) as well as net
IW/CPE higher that 0.9. The IW/CPE of 0.8 was return was found. So the water saving in order
observed as statically better one. There were no IW/CPE of 0.6 to 1.0 was assessed and the net return
significant differences between the treatments having from this water saving at respective IW/CPE was
IW/CPE 0.8 and 0.9, 0.9 and 1.0 and 1.0 and 1.2. It calculated. The net return including additional return
was found that the water use efficiency could be from water saving (over IW/CPE=1 as base) was
attained as 3.819, 3.899, 4.148, 3.867, 3.356 and 4110, 9725, 16656 and 16574 Rs./ ha in the IW/CPE
2.558 kg/ha.mm respectively under the IW/CPE of of 0.6, 0.7, 0.8 and 0.9 respectively. The IW/CPE of
0.6, 0.7, 0.8, 0.9, 1.0 and 1.2 respectively. 0.9 was found as optimal as the highest net return
(including return from water saving over base
(b) Fodder Yield IW/CPE) of Rs.16656 was found at this IW/CPE.
The fodder production of 3710, 4604, 5443, 5901, See Table 2
6295 and 6640 kg/ha could be obtained using 502,
589, 674, 757, 843 and 1011 ha.mm/ha of irrigation Drip Irrigation
water respectively during the season. The fodder yield The total cost of cultivation under drip irrigation in
could be increased by the tune of 894, 839, 394 and the IW/CPE of 0.6 to 1.2 were varied as 25436 to
345 kg/ha from the IW/CPE 0.6 to 0.7, 0.7 to 0.8, 0.8 30526 Rs/ha. The gross return was 30486, 36767,
to 0.9, 0.9 to 1.0 and 1.0 to 1.2 respectively by 44513, 46903, 46220 and 43578 Rs./ha in the
additional water application o 87, 85, 83, 86 and 168 IW/CPE of 0.6, 0.7, 0.8, 0.9, 1.0 and 1.2 respectively.
ha.mm/ha. The water use efficiency in producing the The highest-benefit cost ratio (gross income to total
fodder could be observed as 7.390, 7.817, 8.076, cost) was 1.68 in the IW/CPE of 0.9. Also the net
7.795 and 7.467 and 6.568 kg/ha.mm under the return was highest of 18917 Rs.ha in the same
IW/CPE of 0.6, 0.7, 0.8, 0.9, 1.0 and 1.2 respectively. IW/CPE. The net return was calculated as 505, 10401,
The fodder yield data under all the six treatments 17357, 18917, 17374 and 13052 Rs/ha in the IW/CEP
were differed significantly. It was found that the of 0.6, 0.7, 0.8, 0.9, 1.0 and 1.2 respectively. The
fodder production was increased at decreasing rate IW.CPE of 0.8 was taken as base because the highest
with increase in the IW/CPE from 0.6 to 1.2. benefit cost ratio as well as highest net return was
found at this level. The total net return including
Comparison additional return from water saving (over base
The crop performance under drip irrigation was found IW/CPE=0.9) was found as Rs.7626, 13495, 19940
far superior that the micro sprinkler irrigation. Under and 18917 in the IW/CPE of 0.6, 0.7, 0.8 and 0.9
the drip irrigation, the pod yield was increased by respectively.
30.32, 23.56, 17.28, 14.78, 15.66 and 17.28% over See Table 3
micro sprinkler irrigation. In comparison to micro
sprinkler irrigation the drip irrigation could produce Comparison
25.85, 28.32, 21.96, 18.99, 20.55 and 24.83 more The additional cost of drip system over order at
fodder under the IW/CPE of 0.6, 0.7, 0.8, 0.9, 1.0 and respective IW/CPE of 0.6 to 1.2 was found as 7188 to
1.2 respectively. 7958 Rs/ha respectively. The additional benefit was
found as 9776, 12535, 16625, 14793, 11839 and
ECONOMICS: 10201 Rs/ha respectively in the IW/CPE of 0.6, 0.7,
0.8, 0.9, 1.0 and 1.2 respectively. The benefit cost
Micro sprinkler Irrigation ratio (additional benefit to additional cost over
The total cost of cultivation o the summer groundnut surface) of the drip system adoption was increased
crop was 20823, 21951, 22947, 24015, 24963 and from 1.36 in IW/CPE of 0.6 to 2.24 in IW/CPE of 0.8
27111 Rs./ha under the treatment having IW/CPE of and then decreased to 1.28 at IW/CPE o 1.2. The
0.6, 0.7, 0.8, 0.9 and 1.0 and 1.2 respectively. At high highest benefit cost ratio of drip system adoption
water application level the total cost of cultivation replacing the surface method was found as 2.24 at
was high due to more expenditure toward labours and IW/CPE of 0.8
diesel/ electricity charges. The gross income under the
IW/CPE from 0.6, 0.7, 0.8, 0.9, 1.0 and 1.2 were

CONCLUSIONS Rank H.D. (1991), Nonlinear Programming Approach
to Agricultural Resources Management. Unpub.
 The statistical and economical analysis of the M. Tech. Thesis. Sub to IIT, Kharagpur.
results data generated through the series of the field Reddi G.H.S. and Hanks R.J. (1978). Efficient use of
experiments yielded into the following conclusion. Irrigation water for Wheat and Groundnut.
Maysore Journal of Agricultural Sciences. 11:22 –
 The optimal water requirements of the summer 27.
groundnut crop under micro sprinkler and drip Ronald L. Elliott, Sam L. Harp, Gerald, D., Grosz and
irrigation can be 700 ha.mm/ha and 674 ha.mm/ha Michael, A., Kizer (1988). Crop Coefficient for
having irrigation scheduling based on the IW/CPE Peanut Evapotranspiration, Agricultural Water
of 0.8 and 0.8 respectively for the South Saurashtra management, 15:155 –164.
Agro Climatic Zone. The drip irrigation can give Saini J.S. and Sandhu R.S. (1973). Yield and Quality
21% more yield with water saving off 11% as of Groundnut as affected by Irrigation and
compared to micro sprinkler irrigation. Fertilizer Levels Journal of Research, Punjab
Agricultural University. 10: 179 – 183.
 The highest possible water use efficiency under Sandhu R.S., Saini, J.S. and Singh T. (1972). The
micro sprinkler and drip irrigation for summer Effect of Irrigation and Fertilizer levels on the
groundnut crop can be 3.40 kg/ha.mm and 4.148 Yield and Quality of Groundnut. Journal of
kg/ha.mm respectively in this zone. Research, Punjab Agricultural University. 9: 535–
540.
 The highest benefit cost ratio of adopting drip Seshadri C.R., Bhavanisankara Rao, M. and Varisai
irrigation system replacing the surface method can Muhammed, S. (1958). Studies on Root
be 2.24 at IW/CPE of 0.8. The drip irrigation Development in Groundnut, Indian Journal of
adoption at low (IW/CPE=0.6) and higher Agricultural Science. 28: 211 – 215.
(IW/CPE=1.2) water application level are not so Stansell, J.R., Shepherd, J.L., Pallas, J.E., Jr., Bruce,
profitable. R.R., Minton N.A., Bell, D.K. and Morgan L.W.
(1976). Peanut Responses to Soil Water Variables
REFERENCES in the Southeast. Peanut Science, 3 : 44 – 48.
Anonymous. (200). Studied on Techno-Economic Su.K.C. and Lu,P.C. (1963). Effect of Time of
Feasibilities of Drip Irrigation for summer Irrigation and Amount of Water Irrigation on
Groundnut Crop and its Comparison with Micro Peanut Yield. Journal of Agricultural Association
Sprinkler and Drip Irrigation Final Report of of China, 41 : 4351.
ICAR Adhoe Project, Sub. To ICAR, New Delhi. Rahmanyam, S.S.D., Sundersingh K.P., Ramaswamy,
Billaz, R. and Ochos R. (1961). The Stage of S.P., Paxkiraj, S.P. and Rajagopalan, K. (1975).
Susceptibility to Drought in Groundnut. Effect of Moisture Stress at Different Growth
Oleasgineux, 16:605. Stages of Groundnut. Madras Agricultural Journal.
Boote, K.J. and Hammond, L.C. (1981). Effect of 62: 587–588.
Drought on Vegetative and Reproductive
Development of Peanut. Processing of American
Peanut Research & Education Society, 13:86
Abstract.
Gorbet D.W. and Rhoads, F.M. (1975) Response to
Peanut Cultivars to Irrigation and Kylar.
Agronomy Journal, 67:373 – 376.
Hammond. L.C. Boote K.J., Varnell, R.J. and
Robertson, W.K. (1978). Water Use and Yield of
Peanuts on a Well Drained Sandy Soil. Proceeding
of American Peanut Research and Education
Association, 10:73, Abstract.
Hewitt.T.D., Gorbet, D.W. and Westberry, G.O.
(1980). Economics of Irrigating Peanut.
Proceeding of Soil and Crop Science Society Fia
39. 135 – 140.
Lenka D. And Misra P.K. (1973) Response of
Groundnut (Arachis hypogaea L.) to irrigation.
Indian Journal of Agronomy. 18:492 – 497.

Table-1: Total Water Applied (ha.mm/ha), Pod and Fodder Yield (kg/ha) under micro sprinkler Irrigation &
Drip Irrigation
IW/CPE Microsprinkler Irrigation (Pooled data or 3 years) Drip Irrigation (Pooled data for 3 years)
Water Pod Yield Fodder Yield Water Applied Pod Yield Fodder Yield
Applied
0.6 523 1471 2948 502 1917 3710
0.7 617 1859 3588 589 2297 4604
0.8 700 2384 4463 674 2796 5443
0.9 789 2550 4959 757 2927 5901
1.0 868 2446 5222 843 2829 6295
1.2 1047 2205 5319 1011 2586 6640
63.26 96.57 63.25 96.58
160.3 275.2 180.3 275.2
10.44 7.35 10.42 7.37
Table-2 Cost Benefits Of The Cultivation Of The Summer Groundnut Crop Under Micro sprinkler Irrigation.
Treatment Total Cost Gross B/C Ratio for Net Water Saving Additional Total Net
(IW/CPE) (Rs./ha) Income Cultivation of Return (mm) Over Net Return Return
(Rs./ha) summer (Rs./ha) Better From (with
Groundnut Treatment Water from
Saving Water
Saving)
0.6 20823 23545 1.13 2722 2835 2575 4110
0.7 21951 29549 1.35 7598 5317 2919 9725
0.8 22947 37687 1.64 14740 9799 3219 16656
0.9 24015 40589 1.69 16574 8479 3727 16574
1.0 24963 39631 1.59 14668 5250 3891 N.A.
1.2 27111 36644 1.35 9533 3266 4543 N.A.
Table-3: Cost Benefits Of the Cultivation Of the Summer Groundnut Corp Under Drip Irrigation
Treatment Total Gross B/C Ratio Net Total Net Additional Addi. B/C ratio
(IW/CPE) Cost Income For Return Return Cost of Benefits of of drip
(Rs./ha) (Rs./ha) Cultivation (Rs./ha) (With System Over Drip Over Adoption
of Summer Return of Surface Surface
Groundnut Water
Saving)
0.6 25436 30486 1.20 5050 7626 7188 9776 1.36
0.7 26306 36767 1.40 10461 13495 7274 12535 1.72
0.8 27156 44513 1.64 17357 19940 7428 16625 2.24
0.9 27986 46903 1.68 18917 18917 7698 14793 1.92
1.0 28846 46220 1.60 17374 17374 7774 11839 1.52
1.2 30526 43578 1.43 13052 13052 7958 10201 1.28

ISBN 978-81-921121-0-7
A REVIEW OF NOISE ANALYSIS FOR ROAD TRANSPORT

SYSTEMS
Brind Kumar
Assistant Professor, Department of Civil Engineering, Institute of Technology, Banaras Hindu University, Varanasi.
E-mail: kumar_brind.civ@itbhu.ac.in
G.D. Ransinchung R.N.
Assistant Professor, Department of Civil Engineering, Indian Institute of Technology, Roorkee, Uttarakhand
Saurabh Gupta, Gajendra Kumar Yadav, Alok Kumar
M.Tech. III Semester, Department of Civil Engineering, Institute of Engineering & Technology,
Sitapur Road, Lucknow
ABSTRACT: Similar to air pollutants, noise is also an acclaimed pollutant which has the potential to degrade
our living standards along roadside. Seemingly, studies pay much attention to air pollution as compared to noise
along road side because the former is visible. However, noise pollution has the capacity to seriously cripple
human beings physiologically and psychologically and in prominence the audio-sensory system. It is ever
increasing its intensity with the growth of population, industry and vehicular traffic. The paper seeks to present a
review of mathematical framework for its analysis under different traffic situations so as to incite studies in this
area.
INTRODUCTION the traffic flow pattern in our country being different

The problem of noise impact from transportation from developed countries who have note worthy
activities is unique among the pollutants in its works to their credit, full scale studies pertinent to our
spontaneity and lack of duration. In any given situation including geometrically prohibitive hill areas
transportation situation, the noise generated by the has to be conducted before this nuisance assumes
sources involved at the moment is not affected by any outrageous dimensions. But, before proceeding with
previous activity, nor does it affect any future activity. such engineering studies, it is necessary to understand
Unlike air or water pollution, noise leaves no residual the various physical mechanisms involved in noise
evidence to serve as a continuing remainder of its analysis, to adequately define the various descriptors
unpleasantness. This is the principal reason that, even used in highway noise and to understand various
though its effects are usually as severe as any underlying concepts which is the objective of this
transportation impact, noise is often the pollutant with paper.
the lowest priority for control. In case of highways,
however, noise can, in certain situations, becomes the PHYSICAL MECHANISMS OF ROAD
critical issue in project development. TRAFFIC NOISE
To fully understand, noise should be viewed as a
In India, no separate regulations are in existence problem of Source-Path-Receiver, because any
exclusively to control road traffic induced noise solution towards its abatement will have to be
pollution. The stipulations made by regulatory presented as solutions of source, path, and receiver
authorities like Central Pollution Control Board, have [1]. Source is a mechanism or device from which
laid down the maximum limits for the ambient air noise is being emitted e.g. the road traffic stream. The
quality in respect of noise, thus, treating road traffic path is the environment through which the emitted
induced noise pollution as a component of noise travels on its way to striking the receiver, which
deterioration in air quality. Contrarily, in developed may be a person, a group of buildings, or some other
countries there are specific legislations and sensitive receptor. Noise is a broad-banded source of
regulations for highway noise control. Due to higher acoustic energy and all its components does not
incidence of highway noise pollution is our cities and impact the ultimate receiver, the human, to the same
certain highway corridors, it is expected that such a degree. So, the goal of noise analysis should be to
regulation will be underway in the near future. identify those parameters, which will affect the
emission, transmission, and reception of the noise by
In developed countries, the regulations, and their root the human. These parameters include:
legislative statutes, have had the effect of forcing (a) Temporal distribution of the sound.
technological advances in the area of prediction (b) Magnitude of the sound.
methodologies for impact analysis. In our country too, (c) Cyclic frequency of the sound.
a few researchers and institutions have endeavoured (d) Coherence or time-variance of the sound.
into the assessment of noise level and formulation of
predictive methodologies with hub of the activities TEMPORAL DISTRIBUTION OF THE SOUND
around metropolitan cities and important highway This is an important consideration, because, the time
corridors in plain areas. The traffic composition and of a day, day of a week, and month of a year affect the

condition of a receiver. For example, a heavy flow of where, dS is the minimum perceptible increase in
truck traffic may not be offensive to an adjacent sensation, dW is the stimulus producing it, and W is
residential area, if flow is concentrated in the the total stimulus originally present. Integration of the
afternoon hours, but will be very offensive if a large above equation yields the relationship known as the
percentage of trucks ply during late night hours. Also, Weber-Fechner Law, where,
a temple located near a heavy traffic zone will not be
bothered ordinarily by regular traffic operation, unless
people gather to listen to sermons of a reputed Saint. S  k log W (2)
Ordinarily, it may be said that noise disturbs activities
requiring concentration and as a consequence it (Note: “log” refers to “log10”)
produces annoyance to the exposed subjects.
Intensity (I) is the energy flow per unit time per unit
MAGNITUDE OF THE SOUND area perpendicular to the direction of flow. Since the
Emitted sound (or noise) travels by vibrating rate at which work is done (or energy emitted) is
individual particles in an elastic medium back and defined as power, the intensity (I) is also the power
forth along the direction of propagation, usually in a transmitted per unit area normal to the direction of
cylindrical or spherical manner. This type of sound propagation. Power produced by travelling sound
wave may be classified as a longitudinal wave, waves is quite small. Interestingly, the average sound
capable of being propagated through solids, liquids, or power developed by a healthy person speaking at
gases. The net displacement for each particle within normal conversational tone is about 10-5 watts [2].
the transmitting medium is zero, even though the
wave pulse moves at a high rate of speed of 343 m/s Human ear is capable of discerning sound of intensity
in air at 20oC. The actual magnitude of sound as as low as 10-12 watts/m2 or 10-16 watts/cm2 and can
perceived by our ears, is caused by short-duration respond without damage to sound one trillion times as
fluctuations in the atmospheric pressure. These intense [2]. Because of such a large range of
fluctuations, called sound pressures, are extremely intensities over which human ear responds, use of a
minute when compared to the atmospheric pressure linear scale will be very difficult, and so, a
which is approximately 1,00,000 N/m2, or Pascals logarithmic scale is used. Additionally, the ear
(1Pa = 10 microbars = 0.000147 psi). For example, a responds not linearly, but logarithmically to stimulus
heavy truck at 7.5 m may generate a sound pressure of as shown in eq. (2).
upto 0.63 Pa, and a jet aircraft on take off may
generate upto 20 Pa. On the other hand, a very quiet Transferring the concept of the Weber-Fechner Law
sound, like the rustling of leaves, may cause a change to the commonly accepted acoustical terminology
in atmospheric pressure in an order of magnitude of yields the widely used ratio of measure, called the
only 0.00002 Pa, which is a million folds smaller than Bel. Simply defined, the Bel is the logarithm of the
that of the jet aircraft. relative intensity of sound intensities being compared,
or, a logarithmic ratio of a measured intensity to
The number of time the sound pressure varies around reference intensity. Mathematically, the measured
its mean value per second is its frequency and, the intensity (I) and the reference intensity ( I o )
maximum displacement from the mean is called corresponding to L Bel is,
amplitude. The amplitude of sound pressures, as
indicated above, are extremely small in comparison to
the atmospheric pressure. These pressure variations I 
L  log   Bel (3)
can be measured by a sound pressure level meter or  Io 
simply, a sound level meter.
But, in practical use the unit Bel was found to be
The most efficient method for analysing and rather too large, and a unit of one-tenth of a Bel called
manipulating the magnitude component of sound is to the deciBel (dB) is now in general use.
examine the relationship between the physical
stimulus and the corresponding physiological
response by the human. E.H. Weber first studied this I 
So, L  10 log  dB (4)
phenomenon in 1825 in connection with human  Io 
estimates of weights of various objects. Weber
noticed that the increase in stimulus required to In other words, the deciBel is a dimensionless unit
produce a given increase in sensation is proportional logarithmically relating two like quantities of different
to the pre-existing stimulus [1]. Expressed magnitudes, one of which is taken as reference
mathematically, magnitude (in this case, sound intensity or pressure).
The most commonly accepted reference sound
dW pressure is ( Po ) is 0.00002 Pa, which corresponds to
dS  k (1)
W

the threshold of human hearing at 1000 Hz. In terms Table1. Representative Sound Pressures and Sound
of the intensity level, the reference intensity is taken Pressure Levels for Transportation Systems
as 10-12 watts/m2 or 10-16 watts/cm2.
Intensity (I) of a plane wave is related to the rms Prms SPL Description
(Pa) (dB)
pressure (P) of a plane wave by [1],
200.00 140 Threshold of pain
2 63.24 130
P
I (5) 20.00 120 Jet take off
oC 6.32 110
2.00 100 Jet landing
0.63 90 Heavy truck at 7.5 m
0.20 80 Adjacent to a major
where,  o C is the characteristic impedance of the
highway
medium. It has a value of 415 rayls (1MKS rayl = 1 0.06 70
kg/m2.sec) for air at 20oC and is a measure of specific 0.02 60 Automobile at 15 m
0.006 50 Typical community
acoustic impedance and named in honour of Lord 0.002 40 Quiet neighbourhood
Rayleigh. So, the equation for sound pressure level 0.0006 30
(SPL) may be expressed as, 0.0002 20
0.0000 10
6
 P2  0.0000 0 Threshold of hearing
SPL  10 log  2  dB 2 (at 1000Hz)
 Po 
Cyclic frequency of the sound
The cyclic frequency ‘f’ of sound or noise is

P determined primarily by the number of times per
SPL  20 log   dB (6) second that the sound pressures fluctuate between
 Po  positive and negative values in a sinusoidal
configuration. This quantitative measure uses the unit
where, P is the measured sound pressure, while both Hertz (Hz) after H.R. Hertz (1857-1894). One Hertz is
P and Po are root-mean-square (rms) pressures. equal to one cycle per second. Frequency is the
reciprocal of the temporal period T, which is the time
When P  Po , SPL  20 log(1) = 0 dB, at the required to complete one cycle on the sine curve.
threshold of hearing at 1000 Hz. When P = 20 Pa for Mathematically, frequency is defined as,
jet aircraft at takeoff, SPL = 20 log (20/0.00002) = 20
log (106) = 120 dB. C 
f   (7)
-12 2  
The reference intensity ( I o ) taken as 10 watts/m
corresponds almost exactly to the reference pressure where, f = cyclic frequency in Hz.
( Po ) of 0.00002 Pa. So, I is separated from SPL by C = speed of sound in the medium (m/s)
only 0.2 dB, and, for all practical purposes of  = wavelength of the sound (m)
transportation noise, it may be stated that L equals
SPL. Table 1 gives examples of transportation related As mentioned earlier, the speed of sound in air is 343
sound pressure levels with corresponding sound m/s at 20oC. Thus a sound with a frequency of 2000
pressures (rms) in Pascals (Pa) [1]. Hz would have a wavelength of 0.17 m; a frequency
of 200 Hz would a wavelength of 1.72 m; and a 20 Hz
From this table it is inferred that, sound levels sound would have a wavelength of 17.2 m.
corresponding to 10, 20 and 100 dB represents 10
times, 100 times and 1010 times as intense the Knowledge of wavelength as well as frequency is
threshold intensity respectively. The area of human very important in transportation noise analysis for a
hearing extends in frequency from about 20 to 20,000 number of reasons. Firstly, the atmosphere has
cps [1, 3], and in intensity from 0 to 140 dB. The first different absorption rates for different frequencies;
sound heard by humans is at a frequency of 20 cps secondly, barriers between the source and receiver are
(threshold of hearing), and, the extreme noise level at less effective against short wavelength dominated
about 20,000 cps becomes painful to human ears sounds; and thirdly, humans simply respond
(threshold of pain). differently to sounds of different frequencies. Because
the frequency spectrum of noise is usually so broad, it
can be very cumbersome to work with. Therefore, a
method is needed to summarise the contributions of
noise in the various bands and correlate the

summaries with human response. This has been done a temporal component, which describes the daily,
quite effectively through the use of frequency weekly, monthly and yearly fluctuations in its impact
weighing networks, which have the added benefit of influence. Another component that needs to be
being easily adaptable to sound measurement understood is the time-variance, which attempts to
instruments. describe and quantify the sporadic and random
manner that transportation noise impinges on the
Weighted sound level acoustic environment. Herein, few questions will have
Human ear does not respond uniformly to sounds of to be addressed before proceeding further, like, does
all frequencies. It is less efficient at low and high one use the average sound level emitted by the source
frequencies and efficient at medium or speech as an indicator of its impact? If so, how is one truck
frequencies at about 600 to 6000 Hz [3], and most per day or two cars per hour averaged in a meaningful
efficient at about 1000-3000 Hz [1], the phenomenon way? If the peak sound level of the source is to be
usually known as Place principle. So, it is necessary used, where is the significance of another slightly
to weight the low and high frequencies with respect to quieter sound to be included? Also, do changes in
medium frequencies. In an attempt to duplicate the human response correlate well with changes in peak
response of human ear, sound level meters are often and averages [1]?
fitted with three internationally defined frequency
weighing filters, ‘A’, ‘B’ and ‘C’. Experience has To provide solutions to these questions effectively,
shown that the ‘A’ weighted deciBel, designated by the concept of Equivalent Sound Pressure Level
dB(A), correlates very well with human response to ( Leq ) is formulated in terms of the equivalent steady-
noise, particularly in estimating the probability of
hearing damage in industry, and in describing state noise level which in a defined period of time
annoyance caused by transportation noise. contains the same noise (acoustic) energy as a time-
varying noise during the same period of time. The
Coherence or time–variance of the sound Leq is an energy summation integration, and as such
do not rely on statistical parameters. In other words,
Noise parameters
Leq is the A-weighted energy mean of the noise level
(i) Percentile levels (L10, L50 and L90) averages over the measurement period. It can also be
When measurements of road traffic noise are made, considered as the continuous steady noise level which
we are dealing with a noise source that is strongly would have the same A-weighted acoustic energy, as
time-dependent. To account for this feature, the real fluctuating noise measured over the same
measurements of noise parameters are important. period of time, and mathematically expressed as,
Their distribution function is determined by analysing
the noise level data with some form of statistical
analysis. The observed noise levels are grouped in  1 t 2 P( t ) 2 
equal incremental range and the percentage of time Leq  10 log 
for which the noise level exists in each group is  2  dt (8)
found. Then the cumulative percentage of time for  (t 2  t1 ) t1 Po 
which the noise level exceeds the average noise level
of the range considered is computed and, the where, P(t) is the time varying sound pressure, Po is
cumulative noise frequency distribution curve is the reference pressure, and t1 and t2 are the limits of
prepared. From this curve, the noise level exceeding a the time period in question. In terms of sound
given percentage of sample time may be determined intensity,
as the percentile noise level in dB(A).
L10 or ten percentile exceeding noise level, is the one
which exceeds 10% of the total observation time. It
 1 t2 I 
Leq  10 log    dt (9)
indicates the peak levels of intruding noise.
L50 or fifty percentile exceeding noise level, is the one
 (t 2  t1 ) t1 I o 
which exceeds 50% of the total observation time. It
indicates the average noise level. Performing integration of the above equation, we
L90 or ninety percentile exceeding noise level, is the have,
one which exceeds 90% of the total observation time.  n SPLi / 10 
It indicates the background noise level.   f i 10 
Leq  10 log  i 1 n  dB(A) (10)
(ii) Equivalent sound pressure level (Leq)  
Up to now, transportation noise is characterised  
i 1
fi 
having an intensity (magnitude) component,
expressed in dB; a frequency component, expressed in where, fi is equal to the number of occurrences at
Hertz (Hz) and often summarised in A-weighting; and SPLi.

For Gaussian distribution of percentile noise levels, an
We know that, L  10 log  I 
alternative expression for Leq may be stated as [4],  
 Io 
 ( L  L90 ) 2  I 
10 L / 10   
Leq  L50   10  dB(A) (11)
56 Io 
 
I  I 010( L /10) (14)
Noise level indices
A major objective of traffic noise study is to relate the
For the 1st vehicle with noise level L1, eq. (14) will be,
human response to certain physical descriptors of
noise. The common descriptors used in traffic noise I  I 010( L1 /10)
studies are – Traffic Noise Index and Noise Pollution
Level. For the 2nd vehicle with noise level L2, eq. (15) will
( L2 /10)
be, I  I 0 10
(i) Traffic noise index (TNI)
It is a traffic noise rating index obtained from a
combination of noise levels, which gives a better Similarly, for the ith vehicle with noise level Li , eq.
correlation with dissatisfaction. It is obtained on the ( Li /10)
(15) will be, I  I 0 10
consideration that, L10 as an average peak level
intrudes into L90 as an average background noise level
The total noise intensity (I) due to ‘n’ number of
when A-weighted noise levels are measured outdoors
vehicles at a constant distance ‘d’ from the source, eq.
[5]. When mathematically stated,
(14) may be written
as,
TNI  [4  ( L10  L90 )  L90  30] dB(A) (12) I  I1  I 2  ........  Ii  I 0 10L1 /10  10L2 /10  .........  10Li /10 
I
The term (L10 – L90) is called “noise climate” by some  (10L1 /10  10 L2 /10  .........  10 Li /10 )
authors and the final arbitrary constant is included to I0
yield more convenient numbers. Traffic noise index is
10 L /10  (10 L1 /10  10 L2 /10  .........  10 Li /10 )
derived on the assumption that, an extensive noise
level fluctuation over time is the dominant factor in Taking log10 on both sides
traffic noise annoyance. TNI attempts to make an L
allowance for the noise variability, since fluctuating  log(10 L1 /10  10 L2 /10  .........  10 Li /10 )
noise is commonly assumed to be more annoying. 10
n 
L  10 log 10 Li / 10  (15)
(ii) Noise pollution level (NPL)  i 1 
According to Robinson, D.W. [6], of the British
National Physical Laboratory, Leq in itself is an Obtained eq. (16) is the equation for deciBel
insufficient descriptor of annoyance caused by “summation” [1]. For e.g., one truck generates a noise
fluctuating noise, and, road traffic noise is a level of about 90 dB at 7.5 m away from the noise
significantly fluctuating noise. The index NPL, was source. Substituting in eq. (15), we get,
developed to estimate the dissatisfaction caused by
road traffic noise comprising of two terms. The first is L  10 log 10 90 / 10  90dB
a measure of the equivalent continuous noise level If the noise source is doubled, i.e. two trucks are
( Leq ) and the second represents the increase of considered, then the resulting noise intensity level will
annoyance caused by fluctuations in that level. For a be,
Gaussian distribution of noise levels, NPL can be L  10 log(1090 / 10  10 90 / 10 )  93dB
expressed as given below:
Three trucks will generate,
NPL  [ Leq  ( L10  L90 )] dB(A) (13) L  10 log(1090 / 10  10 90 / 10  1090 / 10 )  94.77dB
Four trucks will generate,

Some important derivations of road traffic noise L  10 log(10 90 / 10
 10 90 / 10
 10 90 / 10
 10 90 / 10 )  96dB
Noise intensity level for more than one vehicle

So, an important observation made is that, the
doubling of noise source strength will result in a 3 dB I  I 
increase in noise intensity, and halving the source L1  L2  10 log  1   10 log  2 
strength will result in a 3 dB decrease. This concept Io   Io 
can be generalised to add any number of noise sources
of different strengths to obtain the resultant noise
intensity level.
 10 log I1  10 log I o  10 log I 2  10 log I o
Relationship between noise intensity level and
distance from the source ‘d’
I 
The source of road transportation noises is the vehicle  10 log  1  (16)
or vehicles emitting noise. These sources as a whole I2 
can be considered as aggregate or ‘macro’ sources.
But, the motor vehicle, is actually an aggregate of But, at d2=2d1, A2=4A1, and I1=4I2
several ‘micro’ sources, including engine, exhaust,
and tyre-roadway interaction amongst others. In turn, This implied that,
each of these ‘micro’ sources can be classified as  4I 
monopole, dipole or quadrupole radiators of noise. A L1  L2  10 log  2   10 log 4  6 dB
monopole source is one that produces noise waves  I2 
through a change in mass outflow (or volume). A
dipole source, on the other hand, has no volume Therefore, for point sources, there is a decrease of 6
change, because it includes two equal and opposite dB in noise intensity for every doubling of distance,
components. A quadrupole noise radiator is in effect and an increase of 6 dB for every halving of distance.
two equal and opposite dipole noise radiators, with
total dipole strength equal to zero. Thus a quadrupole The nature of line source is to conform to cylindrical
has neither a volume change nor a change in the force spreading and to obey the inverse first power law.
on the medium. It may be presumed that motor According to this principle, the power flux (intensity)
vehicle noise is predominantly monopole in nature. A is reduced by one-half as the distance increases from
truck exhaust is an example of monopole source, a d to 2d, because, the size of the area of impact of the
rotating fan blade is a dipole source, and a turbulent noise signal only doubles as the distance is doubled.
jet is a quadrupole source [1]. Mathematically,
There are three basic types of noise sources found in At d2 =2d1, A2=2A1, and I1 =2I2
transportation noise. The first is the point source,
which would be represented by an individual vehicle Substituting in eq. (16) we have,
e.g. a lone truck on a highway. The second is the pure
line source, which may be an extremely heavy and
uniform traffic flow. The third type is the one most
often found in typical highway traffic, the modified  2I 
L1  L2  10 log  2   10 log 2  3dB
line source, which has characteristics of both point  I2 
and line sources [1].
So, for a line source, we observe that, there is a
Point sources are subjected to a phenomenon called
decrease of 3 dB in noise intensity for every doubling
spherical spreading, which results from the inverse
of distance, and an increase of 3 dB for every halving
square law. Basically this means that the noise signal
of distance.
emanating from a source strikes an area A, four times
as large at a distance 2d as it does at a distance d from
When considering traffic noise, a careful distinction
a source. Thus, the power per unit area, or Intensity, I
should be drawn between emission from individual
at 2d has reduced by a factor of four, from its value at
vehicles and emission from traffic stream [7]. This is
d. Mathematically,
because the peak noise level defined by L10 are due to
the passage of individual vehicles in the stream acting
I  as a point sources where the spatial decay rate is 6 dB
At d1, L1  10 log  1  per doubling of distance, as one moves away from the
 Io  road. However, the background noise level, defined
by L90 , contains significant contributions from a
I  greater number of vehicles in the traffic stream, and
At d2, L2  10 log  2  therefore acts as a line source, attenuating by 3 dB per
 Io  doubling of distance. It should be noted that these
relationships only apply to increase or decrease of

noise level with respect to distance. They do not Chhatwal, G.R., Satake, M., Katyal, M., Mehra, M.C.,
consider factors such as atmosphere, attenuation, Katyal, T., and Nagahiro, T., “Noise Pollution”,
absorption, and so on. Encyclopaedia of Environmental Pollution and its
Control, Anmol Publications Pvt. Ltd., Daryaganj,
Thus, an individual vehicle is a point source, with a New Delhi, Vol. 3, 1995.
spatial decay rate of 6 dB per distance doubling. As Purdom, Walton P., “Noise Evaluation and Control”,
the flow rate increase, the traffic stream becomes a Environmental Health, Drexel University,
line source, with a spatial decay rate of 3 dB per Academic Press, New York, pp. 146-429, 1972.
distance doubling. In practice, most traffic stream Hansi, Lt. Col. J.S., Gupta, A.K., and Nigam, P.S., “A
densities are somewhere between point and line Study on Traffic Noise for Various Land Uses for
sources, with resultant decay rates of 6 to 3 dB, [1]. Mixed Traffic Flow”, Indian Highways, Vol. 14,
Studies also show that, the spatial decay rate for Leq No. 2, pp. 30-47, February 1986.
Griffiths, I.D., and Langdon, F.J. (1968). “Subjective
(or energy mean emission level), is always 3 dB per response to road traffic noise.” Journal of Sound
distance doubling at all traffic densities, [1]. & Vibration, 8(1):16-32.
Conversely, doubling of source strength would result Robinson, D.W. (1971). “Towards a unified system of
in 3 dB increase in Leq for all traffic densities. noise assessment.” Journal of Sound and
Vibration, 14(3):279-298.
Again, if speed is doubled, everything else held Hassall, J.R., and Zaveri, K. (1988). “Acoustic Noise
Measurement”, Bruel & Kjaer:1-289.
constant, the result would be a 3 dB decrease in Leq .
The reason for this decrease is that, with volume held
constant, doubling the vehicular speed actually
reduces the amount of time that each vehicle
acoustically influences a given receiver of sound. This
reduction would of course be more than offset by the
increase in reference energy mean emission levels of
corresponding vehicle types which would be 11.5 dB
per speed doubling for automobiles, 10.2 dB for
medium trucks, and 7.4 dB for heavy trucks [1].
CONCLUSIONS
Studies on traffic noise induced pollution are in
preliminary stages in our country. The dangers
emanating from this pollutant are eminent. Its role as
a polluter to human health is well recognised.
Concerted efforts towards its quantification suitable to
the Indian context are to be made, and abatement
strategies are to be developed through research and
development. Government and public funding should
be mobilised and more number of engineers and
scientists are to be involved in this assignment. There
is a need to form a national level coordination
committee under the ages of the Indian Roads
Congress in line with the Federal Highway
Administration of U.S.A. for coordinated research at
the national level on the subject.
The paper highlights the review of technical and

engineering aspects of quantifying road traffic noise
pollution at the basic level so as to be useful for
further work in this area in our country.
REFERENCES
Cohn, Louis F., and Mcvoy, Gary R., “Environmental

Analysis of TransportationSystems”, John Wiley
& Sons Inc., New York, U.S.A., pp. 153-186,
1982.

ISBN 978-81-921121-0-7
BEARING CAPACITY OF POND ASH REINFORCED WITH

A COIR GEOTEXTILE
M.V.S. Sreedhar
Research Scholar & Asst. Professor, Osmania University, email: mvs_sreedhar@yahoo.com
G. Venkatappa Rao
Former Professor, IIT Delhi & Hon. Professor, Osmania University, email: gvrao.19@google.com
R. Ramesh Reddy
Professor, Osmania University, Hyderabad-500007, email: rameshreddy@hotmail.com
ABSTRACT: The need for utilisation of coal ashes is well recognised. However, the utilisation of these ashes
can be increased by understanding their engineering behaviour. If necessary, their performance can be improved
by reinforcing them with materials that possess better tensile strength. India being a largest producer of coir, it
will be mutually beneficial if pond ash is reinforced with coir products. In line with this, efforts are made in this
project to investigate the behaviour of Pond Ash reinforced with a coir Geotextile by conducting load tests in a
model tank. The load bearing mechanism is investigated in terms of effect of depth of placement of the
reinforcement and the effect of overburden. The results indicated that, the maximum improvement in bearing
capacity was occurred when the coir geotextile was placed at a depth of 0.5 B, where B is the width of model
footing. The effect of overburden was to increase the bearing capacity.
INTRODUCTION studied at macroscopic level in terms of variation in

In India, about 67% of the electric power needs are settlement as the bearing stress is increased.
met by coal based Thermal Power Plants. With vast
coal reserves, the dependance on thermal power REVIEW OF LITERATURE
generation will continue in future as well. However, Several studies are carried out on the bearing capacity
the thermal power plants are emerging coal ashes to of reinforced soils. However, there are very limited
the extent of 170 milling metric tons per year. It is number of studies on the characteristics of
expected that, by 2031, the ash generation in India unreinforced and reinforced pond ash (Kim et al.
will be around 300 million metric tons per year. Huge 2005), particularly pertaining to their strength
chunk of land is used in disposal of the un-utilised ash characteristics.
and substantial amounts are being spent on the R.S.Jakka et.al. (2010) have carried out CD and CU
maintenance of ash ponds. The better option will be to bar tests on pond ash and reported that, the shear
increase the utilisation of coal ashes, more so that of behavior of Pond Ash in general is comparable with
the Pond Ash. that of sands; however variation is found from Pond
Geotechnical applications such as construction of to Pond and within a given pond from point to point.
pavements, embankments and retaining walls may Goutham Kumar Pothal and G.V.Rao (2008), have
provide bulk utilisation of ash,provided the carried out triaxial and load tests on reinforced pond
mechanisms are thoroughly understood. Though few ash and reported improvement in bearing capacity.
reinforced pond ash retaining walls are constructed A.K.Bera et.al.(2008), have reported conducting
and functional in the country, there is need for laboratory load tests on Pond Ash at its MDD
increasing the confidence level. Further, India being reinforced with one type of Jute geotextile and three
the largest producer of coir and related products, there types of Polypropylene geotextiles. They reported
is great necessity for its utilization domestically. It is that, ideal depth of placement was 0.255 B, where B
therefore appropriate to reinforce the pond ash with is width of the model footing, that resulted in a
coir products to increase their bulk utilization. maximum improvement of 34% in the bearing
The primary objective of this research is to investigate capacity of reinforced pond ash.
the behaviour of Pond Ash reinforced with a coir
geotextile. The objective include studying the effect METHODOLOGY
of overburden on the bearingcapacity of reinforced The methodology includes collection of Pond Ash and
Pond Ash. its characterization. Preparation of sample for the
The scope of this research includes carrying out load laboratory plate load test at a relative density of 70%
tests on one type of pond ash only at its relative and at a moisture content of 26% without and with
density of 70%, in partially saturated state resembling placement of reinforcement and conducting strain
just compacted state. One type of coir geotextile is controlled load tests as per the scheme of
used as reinforcement material. The mechanism is experiments.

Characterisation of pond ash
The Pond Ash was collected from National Thermal
Power station, Ramagundam in Andhra Pradesh. The
Pond Ash was collected from a place close to the
discharge point that is about 18m from the bank, at a
depth of 3m below the then ash level in
February,2008. The index and engineering properties
of the Pond Ash used in this project are as given in
Table 1. The photograph of the Pond Ash is shown in
Fig. 1.
Table 1 Geotechnical characteristics of Pond Ash

Parameter Value units
Specific Gravity 1.93

Particle Size distribution
Gravel Size particles 3.98 % Fig. 1 The test set up
Sand Size particles 87.31 %
Silt size particles 8.71 % Model test tank
Consistency Non-Plastic A model test tank of size 300mmx750mmx600mm is
Coefficient of uniformity, Cu 4.29 fabricated using 12mm thick Perspex sheets placed in
Coefficient of curvature, Cc 0.65 a mild steel angle frame. The bottom plate is provided
IS Heavy Compaction test results with a locking arrangement. The sizes of the tank are
Maximum Dry Density 1.17 g/cc governed by the allowance in the loading frame of the
Optimum Moisture content 28.10 % PC controlled tri-axial facility.
Classification as per IS:1498-1970 SP
Model footings
Characteristics of Reinforcement element
The tests are carried out using model footings of size
The characteristics of the coir geotextile (CGT) used
50mm such that the width of the tank 300mm and
in this research are as shown in Table2. It was
depth of the pond ash bed 250mm, will be more than
manufactured by M/s Charankattu Coir Mfg.Co.(P)
or equal to 5B, such that the boundary effects are
Ltd.,Shertallay, Kerala.
eliminated. The model footings are made of 25mm
thick aluminium plates with a rough base.
Table 2 Properties of CGT
Property Value
Scheme of experiments
Name of the product CCM-00H2M2
The investigations are carried out systematically as
Weight per unit area (g/sqm) 773
per the scheme of experiments, which include
Thickness under 2kPa (mm) 8.184
determining the bearing capacity of unreinforced pond
Tensile strength 9.19 / 9.95
ash followed by the pond ash provided with
(ASTMD4595)(kN/m)
reinforcement at 0.50,1.00,2.00 B below the top level
Elongation (MD/CD) (%) 38.29/37.01
(u), where ‘B’ is width of the model footing. The
Mesh opening size (mm) 9.03 x 8.33
experiments are repeated with a sand bed of dry unit
weight 16.50 kN/cum laid to a thickness equal to 1.0
Test set up
B, above the model footing level, such that the
The experimental test set up is shown in Fig 1. The
overburden acts like a surcharge as assumed in
PC controlled Tri-axial test facility is utilised to
Terazaghi’s theory of bearing capacity.
conduct the model plate load tests. The application of
load is by hydraulic control system and the load is
measured by an electronic load cell with a sensitivity RESULTS
of 1 kg. The settlement is measured by electronic The load Vs settlement curve for Pond Ash reinforced
Linear Voltage Differential Transducer (LVDT) of ± with coir geotextile is shown in Fig. 2.
50 mm ranges. The PC controlled facility is run by
software that enables to give the operating conditions
as input. The facility logs the load and settlement
observations continuously and provide online display
of the progress of the mechanism.

Fig. 4 (u/B) ratio Vs BCR plot
Fig. 2 Pond Ash at RD=70% with CGT
The behaviour of pond ash reinforced with coir

geotextile and provided with 1.0B thick surcharge is
shown in Fig. 3.
Fig. 5 (u/B) ratio Vs Axial strain at failure
Fig. 3 Pond Ash at RD=70% with CGT + 1.0 B thick

surcharge of sand at 16.50 kN/cum.
OBSERVATIONS
From Fig. 2, the bearing pressure and the axial strain
at failure for un-reinforced Pond Ash were found to
be 275 kPa and 10.97% respectively. The effects of
reinforcement and overburden on the Bearing
Capacity Ratio (BCR), Axial strain at failure, Tangent
Modulus, Post-failure Modulus are shown in Figs.4-7,
respectively.
Fig. 6 (u/B) ratio Vs Tangent Modulus

6. As seen from Fig.4-7, the overburden improved the
BCR, Tangent Modulus and post-failure modulus
while it decreased the axial strain at failure,
emphasizing the fact that, the overburden has
significant influence on the modification of plastic
zone formation and results in considerable
improvement in bearing capacity.
CONCLUSIONS
Based on the investigations carried out in this project,
the following conclusions are made
1 The Pond Ash reinforced with a coir
geotextile has shown phenomenal
improvement in bearing capacity.
2 The load bearing mechanism of reinforced
Pond Ash is similar to that of cohesionless
soils.
3. The ideal depth of placement for the
Fig. 7 (u/B) ratio Vs Post-failure Modulus materials used in this research was found to
be 0.50 B, where B is width of the footing.
Based on the results, the following observations can 4. The overburden in the form of surcharge as
be made assumed in Terazaghi’s theory of bearing
1 As seen from Fig. 4, the maximum improvement in capacity, showed improvement in load
bearing capacity with surcharge was found when bearing mechanism both on un-reinforced
the reinforcement was laid at a depth of 0.50 B and and reinforced Pond Ash.
the improvement decreased as the depth of
placement is increased. This emphasizes the fact ACKNOWLEDGEMENTS
that, the reinforcement function is maximum when The first author acknowledge the support provided
it is placed in the zone of shear failure such that it under TEQIP phase-I and Research Infrastructure
modifies the plastic zone formation and thus fund of UGC.
contributes to improvement in bearing capacity.
However, in without surcharge case, the maximum REFERENCES
BCR was observed at a (u/B) ratio of 1.00. The Bera A. K., Ghosh A. and Ghosh A. (2005),Regression
reason as to why the maximum was not at model for bearing capacity of a square footing on
(u/B)=0.5, may be due to lack of adequate reinforced pond ash, Geotextiles and Geomembranes,
23, No. 3,261–285.
anchorage due to the presence of only 0.50 B thick Bera A. K., Ghosh A. and Ghosh A. (2008), Bearing
overburden and due to model constraints. Capacity of square footing on reinforced pond ash,
2. The Fig. 5 shows that the axial strain at failure is Ground Improvement,161,GI 1,pp 17-22.
minimum at which the BCR was maximum. This Goutham Kumar Pothal & G.V.Rao (2008),Model studies
indicates the improvement in stiffness of the on Geosynthetic reinforced Pond Ash, JGE,Vol. 13,
reinforced pond ash. Bund. H
Jakka RS, Datta M, Ramana GV (2010) Shear Behaviour of
3. As seen from Fig. 6, the tangent modulus is
Loose and Compacted Pond Ash, Geotech Geol Eng
maximum when the BCR was maximum. This (2010) 28:763–778
signifies improvement in stiffness of the reinforced Kim B, Prezzi M, Salgado R (2005) Geotechnical properties
pond ash. of fly and bottom ash mixtures for use in highway
4. The Fig. 7 shows the improvement in post-failure embankments, J Geotech Geoenviron Eng-ASCE
modulus when the u/B ratio was 0.50 irrespective of 131(7):914–924
where the BCR was maximum. This emphasizes the Pandian NS, Sridharan A, Chittibabu G (2001) Shear
strengthof coal ashes for geotechnical applications. In:
fact that, the reinforcement function in without Proceedingsof the Indian geotechnical conference,
surcharge case was mobilized in post failure stage Indore, India, pp 466–469
on par with “with surcharge” case overcoming the Sridharan A, Pandian NS, Rajasekhar C (1996)
lack of anchorage in the early stages as stated in 1. Geotechnical characterization of pond ash, Narosa
5. The Fig. 2-3 indicates the fact that, the overburden Publishers, Delhi, pp. 97–110.
in the form of surcharge has significant effect on
the load bearing mechanism, particularly in the post
failure stage. When total failure was observed in
without surcharge cases, the medium with
surcharge was continued to bear the loads even up
to an axial strain of 20%.

ISBN 978-81-921121-0-7
PERFORMANCE OF CONCRETE FILLED STEEL (CFT) HOLLOW BEAM

SECTIONS IN FLEXURE
Sunil Kute
Professor, K. K. Wagh Institute of Engg. Education and Research, Nasik (MS) 422 003, sunil_kute@rediffmail.com
T.Jayashree
Lecturer, Guru Gobind Singh Polytechnic, Nasik, (MS) 422003
Shantanu Pande
Asst. Professor, K. K. Wagh Institute of Engg. Education and Research, Nasik (MS) 422 003, pande_shantanu@yahoo.com
ABSTRACT:Significant research work is ongoing worldwide to understand the behavior of Concrete Filled Steel
(CFT) hollow sections. The recent research works over CFT hollow sections have illustrated that their behavior in
terms of structural advantage and economy is much better than conventional reinforced concrete woks. The different
properties of CFT hollow sections like high strength, ductility and large energy absorbing capacity have made their
applications popular in zones of high seismicity. Most of the research work has been done in understanding the axial
behavior of CFT hollow sections.The present paper attempts to understand the behavior of CFT hollow beams under
flexure. The high strength rectangular steel hollow section specimen with fixed depth, width and thickness were tested
to failure under pure bending. The different grades of concrete like M15, M20 and M25 were filled in hollow tube
sections and some sections were provided with reinforcement at top and bottom to find influence of different grades of
concrete and effect of reinforcement in CFTs under flexure.It has been observed that the entire specimen demonstrated
favorable post yield behavior with good ductility performance. The increase in grade of concrete enhances the flexural
strength of CFT beam sections significantly. Also, the failure mechanism of beam sections result in an excessive
deflection with no significant lateral disturbances or any form of instability.
INTRODUCTION
CFT consists of a steel tubular encasing of any shape Much of the computational research conducted on CFTs
such as square, rectangular and circular with concrete to date has focused on computing the ultimate axial and
(generally high strength) filled into it. The most flexural capacity of members. Relatively little research
commonly adopted cross sections are square, has been reported on the flexural behavior of CFT
rectangular and circular as shown in Fig1. beams. A number of experimental and theoretical
Concrete filled tubes are known for their excellent studies on the related subject have been carried out. M.
earthquake resistant properties such as high strength, Elchalakani, X.L. Zhao, R.H. Grzebieta(2001) have
high ductility and large energy absorption capacity. Due conducted study on ‘Concrete filled circular steel tubes
to bond between steel tube and concrete, the local subjected to pure bending’. Their paper presents an
buckling of steel tube is delayed by the restraint of experimental investigation of the flexural behavior of
concrete and the strength of concrete is enhanced by circular CFT subjected to large deformation pure
confining effect of the steel tube. bending where d/t =12 to 110. Wie-Min Gho, Dalin
Liu(2004) have conducted study on ‘Flexural behavior
of high-strength rectangular concrete-filled steel hollow
sections’ . Their paper presents a study of the flexural
behavior of 12, 1600-mm-long high-strength
rectangular concrete-filled steel hollow section
(CFSHS) specimens tested to failure under pure
bending. Dalin Liu, (2004) in his work on ‘Behavior of
Fig. 1 Various cross-sections of CFTs high strength rectangular concrete-filled steel hollow
CU => Circular Unstiffened SU => Square Unstiffened section columns under eccentric loading’, has presented
SS => Square Stiffened using reinforcing ties an experimental study on the behavior of 12 high
Concrete filled steel tubes have been used increasingly strength rectangular concrete-filled steel hollow section
as columns and beam- columns in braced and unbraced columns subjected to eccentric loading. The slenderness
frame structures. They are used as compression ratios of the specimens were 20 and 50, while the load
members in low rise and open floor plan construction. eccentricity ratios (e/B; e is the load eccentricity, B is
Use of Concrete filled tubes as primary lateral the breadth of cross-section) varied between 0.17 and
resistance columns in multi storied braced and unbraced 0.40. Favorable ductility performance was observed for
frames has also, gained popularity.

all specimens during the test. F.W. Lu, S.P. Li, D.W. Li, Table 1 Geometric Properties of the specimens
and Guojun Sun, (2007) have worked on ‘An Dimensions Design Grade
Remarks
experimental study on the mechanical behavior of non- Specimen (mm) of RHS of Material(Mpa)
uni-thickness walled rectangular concrete-filled steel Designat
ion B D t Steel Concrete
tube (CFT) beams subjected to pure bending’. Their
experimental and analytical results suggested that a
non-uni-thickness walled rectangular CFT beam
Steel hollow
exhibits adequate ductility, and the flexural strength of a SP 01 48 96 4.8 310 -
section
non-uni-thickness walled CFT beam is superior to that
of a uni-thickness walled CFT beam. Arivalagan
Soundararajan; Kandasamy Shanmugasundaram, (2008)
SP 02 48 96 4.8 310 19 CFT Beam
conducted study on ‘The Flexural behavior of concrete
filled steel hollow section beams’. They have
experimented the study of normal mix, fly ash, quarry SP 03A 48 96 4.8 310 24 CFT Beam
waste and low strength concrete (brick-bat lime
concrete) contribution to the ultimate moment capacity
Reinforced
of square steel hollow sections. Their experimental with 1.22mm
investigation results showed that normal mix, fly ash, SP 03B 48 96 4.8 310 24
MS wire 4
quarry waste and low-strength concrete enhance the nos.
moment carrying capacity of steel hollow sections.
SP 4A 48 96 4.8 310 27 CFT Beam
EXPERIMENTAL INVESTIGATION:
Reinforced
Test Setup: SP 4B 48 96 4.8 310 27
wtih1.22mm
6 types of specimens were prepared to investigate the MS wires 4
nos.
flexural behavior of rectangular CFT beams. The
specimens were filled by three different grades of
concrete (Grade M15, M20, M25) and Mild steel outer cement conforming to IS: 8112-1989 and tested as per
casing of 4.8mm thickness. The specimens were tested IS: 4031-1988. The results of various tests conducted on
on 200 ton capacity modified compression testing cement used are reported in table
machine with a central loading mechanism. All beam
specimens of span length 1000mm were tested to Table 2 Physical properties of the cement
investigate the flexural behavior of rectangular CFT Specified value
Results
beams. Characteristics Units
obtained
as per IS:8112-
Deflection of the specimens was measured by magnetic 1989
base dial gauges which were placed at the centre and l/4
distances from the edge of the specimen. The Fineness
instruments were controlled and calibrated according to 1. Dry sieving cm2/gm 5 < 10
2. Air permeability % 2518 < 3500
the standard specifications.
Specimens: -
The steel hollow specimens were filled with concrete of Specific gravity - 3.14
three different grades in many layers and carefully
compacted by a steel rod to avoid any gaps that occur Soundness
mm 3 < 10
Le Chatelier Expansion
inside the specimen. The overall geometric properties of
the specimens are summarized in a table given below. Specified value
Results
Characteristics Units as per IS:8112-
obtained
Material properties 1989
Standard consistency % 28 -
Setting time
Cement: 1. Initial Minutes 90 > 30
Ordinary Portland cement 53 grade from a single lot 2. Final Minutes 210 < 600
was used throughout the course of the investigation. Compressive strength
The cement was carefully stored in airtight silos to 1. 3-days MPa 23.2 > 23
2. 7-days Mpa 34.6 > 33
prevent deterioration in its properties due to 3. 28-days Mpa 45.8 > 43
atmospheric effects. The physical properties of the

Fine aggregates: measured by three dial gauges. One was placed at the
Locally available river sand was used as fine mid span of the specimen and other two were placed at
aggregates. The physical properties of the fine L/4 distance from the edge of the specimen.
aggregate are listed in table.
Table 3 Physical properties of fine aggregates

Characteristics Results obtained
Grading Zone III (IS:383-1970)
Fineness modulus 2.51
Specific gravity 2.64
Density (Loose) (kN/m3) 15.23
Water absorption (%) 1.20
Moisture content (%) 0.35
Coarse aggregates:
Locally available crushed stone aggregates of maximum
nominal size of 10 mm were used as course aggregate.
Physical properties of the aggregate are listed in table
Fig. 2 Test set up
Table 4 Physical properties of coarse aggregates
Characteristics Results obtained
Results and discussions:
Fineness modulus 6.45
The tests were conducted till the failure of beam shows
Specific gravity 2.67
deflection up to 50 mm on the Compression testing
Density (Loose) (kN/m3) 16.19
machine. The ductile behavior can be studied from the
Water absorption (%) 0.60
below given Load vs. Deflection at center-span and L/4
Moisture content (%) Nil
distance from the edges of all the specimens.
Load Vs Deflection SP-01
Water:
The water to be used both for mixing and curing of 100
90
80
concrete, should be free from deleterious materials. In 70
Load(KN)
60
present investigation, potable tap water was used. 50 L/4
40
Steel: 30
20
Steel hollows used were factory-made products. They 10
0
are produced by Tata Steel Industries, India. The 0 10 20 30 40 50
length of a specimen was 1.00m. The size and thickness Deflection(mm)
of rectangular specimen were 96x48x4.8mm. Grade of
steel was Fe 310Mpa as per IS 4923:1997 “Indian Load Vs Deflection SP -01
Standard Hollow Steel Sections for structural use 100
90
specifications”. 80
Reinforced mild steel wires of 1.22mm diameter were 70
Load(KN)
60
placed two each at top and bottom of the specimen. 50 Central
40
Test procedure: 30
The specimens were tested under bending about the 20
10
major axis. In the test, the deflections of the specimen 0
were recorded at each loading interval of 5 KN. For all 0 10 20 30 40 50
Deflection(mm)
levels, the load was applied continuously. The local
bucking was observed near the compression side for all
the specimens. Deflections of the beam specimens were
Fig. 3 Load vs. Deflection curve for Specimen 01

Load Vs Deflection SP-02
Load Vs Defle ction SP-02
100
90 100
80 90
70 80
Load (KN)
60 70
Load (KN)
50 Central 60
40 50 L/4
30 40
30
20
20
10
10
0 0
0 10 20 30 40 50 0 10 20 30 40 50
Deflection(mm) Deflection(mm)

A comparison of Load vs. Deflection curves plotted for specimen 1(hollow section without any filler) and specimen 2
(with M15 grade concrete filler) shows that
there is a definite increase in flexural strength.
Load Vs Deflection SP-3A

Load Vs Deflection SP-3A
100
100
90
90
80
80
70
70
Load(KN)
Load(KN)
60 60
50 Central 50 L/4
40 40
30 30
20 20
10 10
0 0
0 10 20 30 40 50 0 10 20 30 40 50
De fle ction(mm) Deflection(mm)
In comparison of Load vs. Deflection curves plotted for specimen 2 (with M15 grade concrete filler) and specimen 3A
(with M 20 grade concrete filler), it is found that the deflection at a given load has reduced by 3.26% indicating the
increase in flexural strength.
Load Vs De fle ction SP-3B
Load Vs De fle ction SP-3B
100
100 90
90 80
80 70
Load(KN)
70 60
Load(KN)
60
50 L/4
50 Central
40
40
30
30
20 20
10 10
0 0
0 10 20 30 40 50 0 10 20 30 40 50
De fle ction(mm) De fle ction(mm)
Load Vs Deflection SP- 4A

Load Vs Deflection SP- 4A
100
100 90
90 80
Load (KN)
80 70
70 60
Load (KN)
60 50 L/4
50 Central 40
40 30
30 20
20 10
10
0
0
0 10 20 30 40 50
0 10 20 30 40 50
De fle ction(mm) De fle ction(mm)
A comparison of Load Vs. Deflection curves plotted for specimen 3A (with M20 grade concrete filler) and specimen
4A (with M 25 grade concrete filler) shows that the deflection at a given load has been reduced by 10.56%

Load Vs Deflection SP - 4B
Load Vs Defle ction SP- 4B
100
100 90
90 80
80 70
Load(KN)
70 60
Load(KN)
60
50 L/4
50 Central
40
40
30
30
20 20
10 10
0 0
0 10 20 30 40 50 0 10 20 30 40 50
Defle ction(mm) De fle ction(mm)
Behavior of all beams was predicted during the test. REFERENCES

The beam specimens developed the full flexural
strength of the section. The experimental results of this Arivalagan Soundararajan, Kandasamy
study demonstrated the predominant failure mechanism Shanmugasundaram (2008). “Flexural behavior of
of the beam specimens to be excessive deflection concrete-filled steel hollow sections beams”,
accompanied with some local distortions near the point Journal of Civil Engineering and Management,
of load application at a stage very close to maximum 14(2): 107–114.
loading.The measured load Vs deflection at mid-span Dalin Liu (2004). “Behavior of high strength
showed all beam specimens established similar rectangular concrete-filled steel hollow section
behavior with respect to flexural strength with the columns under eccentric loading”, Thin-Walled
increasing grade of concrete filler.The tested beams Structures 42, 1631–1644.
failed in a very ductile manner. No tension fracture was Elchalakani, M., Zhao,X.L., Grzebieta, R.H.
observed on the tension flange. (2001).“Concrete-filled circular steel tubes
subjected to pure bending”, Journal of
CONCLUDING REMARKS: Constructional Steel Research, 57
The tests conducted in this study on concrete filled steel F.W. Lu, S.P. Li, D.W. Li, Guojun Sun (2007).
hollow section beams with M15, M20 and M25 grade “Flexural behavior of concrete filled non-uni-
concrete filling and with MS wire reinforcement two thickness walled rectangular steel tube”, Journal of
each at top and bottom in case of M20 and M25 lead to Constructional Steel Research, 63, 1051–1057.4.
the following conclusions. I.S. 4923:1997, “ Indian Standard Hollow Steel Sections
 Beams filled with M15, M20 and M25 grade for Structural Use Specifications”.
concrete were capable of improving flexural Wie-Min Gho, Dalin Liu (2004). “Flexural behavior of
strength of their sections. The concrete filling high-strength rectangular concrete-filled steel
enhanced the flexural strength of the steel hollow sections”, Journal of Constructional Steel
hollow sections. All specimens demonstrated Research 60, 1681–1696.
favorable post yield behavior with good
ductility performance.
 From the experimental results of the current
investigation, it can be concluded that the
failure mechanism of the beam sections result
in an excessive deflection with no significant
lateral disturbances or any form of instability.
 Specimens with M20, M25 concrete filling and
MS wire reinforcement of two each at top and
bottom have shown marginally higher flexural
strength.
 Finally tests conducted in this investigation
confirm that with the increase in the grade of
concrete, considerable increase in flexural
strength is observed. Hence for the composite
sections, better economy can be achieved with
higher grade of concrete.

ISBN 978-81-921121-0-7
FORECASTING OF GROUND WATER LEVEL OF SIMGATEHSIL,

RAIPUR DISTRICT, CHHATTISGARH USING ARTIFICIAL NEURAL
NETWORK
Umank Mishra
Research Scholar, Dept. Of Civil Engineering, IT, BHU, Varanasi (U P)
K K Pandey
Associate Professor, Dept. Of Civil Engineering, IT, BHU, Varanasi (U P)
S B Dwiwedi
Associate Professor, Dept. Of Civil Engineering, IT, BHU, Varanasi (U P)
ABSTRACT: The performance of Artificial Neural Network (ANN) is examined in groundwater level forecasting.
The standard feed-forward neural network trained with Levenberg-Marquardt, has been used for forecasting
groundwater level of Simga Tehsil (Raipur) ,in order to identify an optimal ANN architecture that can simulate the
decreasing trend of the groundwater level and provide acceptable predictions. Annual rainfall, temperature,
sunshine, evaporation and relative humidity has been chosen as inputs. The model efficiency and accuracy were
measured based on the root mean square error (RMSE) and regression coefficient (R2). The model provided the best
fit and the predicted trend, followed the observed data closely (RMSE = 2.250 and R2 = 0.96). Thus, for accurate
groundwater level forecasting, ANN with Levenberg-Marquardt algorithm, appears to be a promising tool.
Keywords: Artificial Neural Networks, back-propagation, feed-forward, forecasting, groundwater level fluctuation,
MATLAB.
INTRODUCTION groundwater levels well in advance is the need of the

Groundwater level is an indicator of groundwater hour to devise sustainable water management
availability, groundwater flow, and the physical protocols.
characteristics of an aquifer or groundwater system. Physical models (Upadhyaya and Chauhan, 2001; and
Groundwater contributes greatly to the amount of fresh Song et al., 2007), water balance models (McCarthy
water in the world. In many areas, groundwater is and Skaggs, 1991; and Roulet, 1991), and statistical
considered a major water source for both drinking and regression models (Yakowitz, 1976; and Joginder et
irrigation purposes. The India uses more than 83.3 al., 1981) have been developed in the past to simulate
billion gallons of fresh groundwater each day for water table variation in different areas. However, all of
private and public water supplies, irrigation, livestock, these models need extensive observations to perform
manufacturing, mining, and other purposes (CGWB, the modeling. The physically based model also requires
2008). Earlier, it was considered to be a limitless or at an explicit relationship between the input and output
least fully renewable natural resource, but during the parameters. The presence of errors or uncertainties in
past 20 years or so, there has been a tremendous the observations will result in errors or deviations in
pressure on this precious natural resource mainly due model output (Shirmohammadi et al., 2006). Artificial
to rapid industrialization and growth of human Neural Network (ANN) modeling provides another
population, though, the advancement in agricultural approach to predict water table fluctuation. ANN
technology has been impressive. Poor ground water technology was developed around 65 years ago
monitoring and management has resulted in (McCulloch and Pitts, 1943). Since then, it has been
considerable depletion of the groundwater table, land widely used on pattern/speech recognition and
subsidence and deterioration in the water quality, image/signal processing in a variety of fields (Widrow
making the availability of water in the future highly and Lehr, 1992). The application of ANN in hydrology
uncertain. Keeping in mind the scarcity of available started in the early 1990s. In the late nineties, ANN
water resources in the near future and it impending modeling began to be used in the simulation of water
threats, it has become imperative on the part of water table fluctuations at different locations (Yang et al,
scientists as well as planners to quantify the available 1997; Yang et al, 2000; Coulibaly et al., 2001; and
water resources for its judicial use. Thus, a ready Affandi et al., 2007). These studies indicate that ANN
reckoner to monitor the fluctuations in modeling is a convenient tool for predicting water table
fluctuation, especially in areas where the aquifer
system information is not available or where the

available records are relatively short. The purpose of The carbonate rock includes the dolomite and
this study was to develop a quick, easy, and convenient stromatolitic limestone. Cherty shale’s are the
water level forecasting model using weather dominant rock formation in the study area and serve as
parameters through ANN technology. highly potential source of ground water in the area.
Both dolomite and limestone are less potential
MATERIALS AND METHODS geological formation for the ground water.
Study Area
The input data for the present study were collected for DATA COLLECTION
Simga tehsil of Raipur district of Chhattisgarh state, Monthly water levels have been collected from 19
India. Study area is located in the western part of the wells distributed in the area (Fig.1) Groundwater level
Raipur district, bounded in the north by Bilaspur data collected from Central Ground Water Board
district, in the west by Durg district in a continuous Raipur. Data on weather / hydrological parameter i.e.
pattern of land and people. The state of Orissa with its rainfall, temperature, sunshine and humidity were
district area of Sambalpur and Mahasamund district of collected from agricultural department (Meteorological
Chhattisgarh state defines the eastern boundary of the Division). Both the set of data belong to the period
Raipur di st ri ct , where as its south east and southern from year 2000 to 2007.
boundary is delimited by the district ar ea of Balangir,
Kalahandi of Orissa state and Ranker district of
Chhattisgarh State. Sigma tehsil lies between latitudes
( N 21063’to N 21075’) and longitudes (E810 79’ to
E83087’). It covers an area about 91.73 km2. The
district falls under survey of India toposheet No. 64G.
The topographic elevation is about 261-276 m amsl.
The winter season lasts till February. During winter,
the night temperatures may sometimes drop below
10°C. From about mid-February both day and night
temperatures increase rapidly till May which is the
hottest month of the year with daily maximum
temperature at 480C characterize the area, with a
distinct rainy season from June to September. The
temperature ranges from 220 C to 440C, with an Fig. 1 Location of study area
average annual rainfall of Raipur district is 1290 mm.
The maximum rainfall of the district recorded in the METHODOLOGY
past is 2324 mm and minimum rainfall is 826 mm. Artificial Neural Network (ANN)
From the analysis of long-range records it becomes Artificial neural networks (ANNs) are information
clear that the district receives 87.0% of the total rainfall processing system based on the present understanding
from southwest monsoon during June to September. of the biological nervous systems. The processing units
The winter rainfall accounts for 9 % of the total rainfall of an artificial neural network are called neurons,
and summer rainfall accounts for 4% of the total annual which are arranged in layers. Neurons between layers
rainfall. The summer season humidity is the lowest, are connected by links of variable weights. The most
about 36 %, while it is higher during the monsoon commonly used architecture of neural network (NN) is
period about 86%. The humidity again decreases from the feed forward neural network. Feed forward network
October onwards due to rise in temperature and also has no feedback connections (Zurada, 1999). Feed-
due to receding monsoon. forward neural network (FFNN) along with
The surface water bodies are rivers and streams, lakes, Levenberg–Marquardt back propagation (LMB)
springs, ponds and tanks etc. Drainage pattern of the algorithm is used in MATLAB programming.
area is well defined by Fig.1in which different types of Feed-Forward Neural Network
drainage are observed in the unutilized area. The Here, the source nodes in the input layer of the
pattern obtained is namely first order, second order, network, supply the respective elements of the
third order and fourth order. Fourth order drainage is activation pattern that constitute the input signals to the
more prominent in southern part of the study area and neurons in the second layer. The output signals of the
first and second orders are observed in the northern second layer are used as inputs in the third layer, and
part of the area. The different order of the drainage so on for the rest of the network. Thus, typically the
pattern in the area indicates the scarcity of the ground neurons in each layer of the network have their inputs
water table in the area. Geologically the watershed of from the output signals of the preceding layer only.
the area consists of carbonate rocks and cherty strata. The set of output signals of the neurons in the output

layer of the network constitutes the overall response of Training and Testing of the Data
the network supplied by the source nodes in the input The total identified nodes include the monthly
layer (Haykin, 2006) as shown in Fig.re 2, which observed water level in 19 sampled well points with
explains a typical FFNN with one hidden layer. local weather parameters, viz. Rainfall, temperature
(minimum and maximum), evaporation, and relative
humidity and sunshine. Weather parameters were
chosen as input because they have a direct influence on
groundwater levels. Several neural network models and
their architecture were optimized through rigorous
experimentation for each of the nineteen wells to
predict the variation of ground water level in a
particular month of the year. The input data vector
consists for 32 samples for each well. 75% of the data
were used for the training while rest 25% were used for
testing purposes. The target data were made up of the
ground water level corresponding to vector of local
weather parameter. For a particular well, a feed-
forward neural network is designed with 5 neurons in
Fig.2. A typical three-layer feed forward ANN the input layer and 1 neuron in the output layer. The
Levenberg–Marquardt Back Propagation (LMB) number of neurons in the hidden layer was optimized
Algorithm experimentally .Thus, the prediction above ground
The LMB algorithm was designed to approach second- water level was made with different neural network
order training speed without having to compute the model for different wells.
Hessian matrix. When the performance function has The efficiency of selected network for accurate output
the form of a sum of squares (is typical in training was measured using root mean square error (RMSE)
FFNNs), then the Hessian matrix can be approximated and regression coefficient ( ) using following
as equations.
= ,
= ∑ ( − ) / (1)
and the gradient can be computed as
= , ∑( )
=1− ∑
(2)
∑
where J is the Jacobian matrix that contains first
derivatives of the network errors with respect to the
weights and biases, and e is a vector of network errors. where is the observed data, the calculated data
The Jacobian matrix can be computed through a and n the number of observations. RMSE indicates the
standard back propagation technique that is much less discrepancy between the observed and calculated
complex than computing the Hessian matrix. values. Lower the RMSE, the more accurate is the
Therefore, the Levenberg-Marquardt algorithm uses prediction. The best fit between observed and
this approximation to the Hessian matrix in the calculated values would have as 1 and RMSE as 0.
following Newton-like update
= −( + ) Training of the network has been carried out using
Levenberg-Marquardt algorithm. The values of
When the scalar is zero, this is Newton's method learning rate η and the momentum factor α were
using the approximate Hessian matrix. When is large, decided after examining different combinations.
this becomes gradient descent with a small step size. Learning rate η = 0.1 and momentum factor α = 0.9
Newton’s method is faster and more accurate near an have been found to be the best combinations.
error minimum. So the aim is to shift towards
Newton’s method as quickly as possible. Thus, is
decreased after each successful step (reduction in
performance function) and is increased only when a
tentative step would increase the performance function.
In this way, the performance function is always
reduced at each iteration of the algorithm.

Table1. Summary of the experimentally determined
number of neurons in the hidden layer of neural
network designed for each well.
Well Hidden Well Hidden

No. Neuron No. Neuron
Layers Layers
1 2 12 2
2 3 13 4
3 3 14 2
4 4 15 3
5 3 16 2
6 3 17 3
7 2 18 3
8 3 19 2
9 2 Fig.4. Regression Analysis of simulated and Target
10 2 values for G.W.L
11 3
Number of hidden layers in the wells varies from 2 to
The number of neurons in the hidden layer of the 4.Table 1 shows the number of hidden layers at each
neural network has been finalized after a trial and error well point.
procedure using different combinations of learning rate Further, the data were analyzed separately for each
and momentum factor. Each combination of learning independent well point to have a clear comparison of
rate and momentum factor has been tested for different the mean observed and estimated water levels. Here
numbers of hidden neurons. The network has been also, the results exhibited that FFNN–LMB was found
trained for 5000 epochs (iterations), as there is very to be the best fit for predicting groundwater levels at
negligible improvement of the RMSE value and R2 the Simga Tehsil. The training performance of the
value after 5000 iterations. Fig. 3 and 4 show reduction FFNN-LMB for different well is given in Table 2.
of MSE and R for a network having 2 neurons in the The performance trend of the model, individually for
hidden layer and for some of the combinations of each of 19 wells along with actual and predicted values
learning rate η and momentum factor α. It was is indicated in the figure 5.
observed that for a network having 2 neurons in the
hidden layer, the good value of η and α is 0.9 and 0.1 Here, the predicted water level trend followed the
respectively. observed trend closely, showing that the difference
between actual and predicted values is close to zero for
each of the19 wells.
Fig.3. Decrease of MSE of well no.1 during training

Table2. RMSE and R2 values for training and testing
of network for different wells.
Well TRAINING TESTING
No. RMSE R2 RMSE R2
1 1.7958 0.9659 2.251 0.8363
2 1.9062 0.9249 1.2546 0.6679
3 1.7503 0.9976 0.554 0.8973
4 2.2361 0.9372 1.4902 0.94904
5 0.552 0.9903 0.469 0.96051
6 2.508 0.9772 0.8721 0.864
7 0.2535 0.8547 0.2496 0.7639
8 0.8986 0.96024 0.4512 0.8505
9 2.8474 0.8771 1.0622 0.3818
10 0.6823 0.8906 1.2859 0.90607
11 1.4576 0.87502 2.3086 0.65096
12 1.824 0.897 3.0584 0.8765
13 2.0319 0.75009 2.1688 0.6664
14 1.074 0.9616 2.2333 0.6794
15 0.4737 0.873 2.8495 0.855
16 1.415 0.856 1.5466 0.759
17 0.8711 0.8767 2.0933 0.8467
18 0.5724 0.9175 1.5049 0.729
19 1.1616 0.919 0.9013 0.9477

Level Fluctuation, Journal of Spatial Hydrology, v.7
pp. 17-32
Ahmed,Sarma, A.K.,(2007):Artificial neural network
model for synthetic stream flow generation, Water
Resource Management v. 21, pp.015–1029.
Ajmera1, K and Rastogi, A.K., (2008):Artificial Neural
Network Application on Estimation of Aquifer
TransmissivityJournal of Spatial Hydrology. v, 2 pp.
15-31.
Azhar, K., Watanabe, K., (2007): Application of an
artificial neural network to estimate groundwater
level fluctuation, Journal of. Spatial Hydrology. v.,17
pp. 23-46.
Balkhair, K.S. (2002) Aquifer parameters
determination for large diameter wells using neural
network approach, Journal of Hydrology, v.265,pp.
118–128.
Central Ground Water Board (2008): Ground water
resources and development of Raipur, .pp. l-80
Chauhan,S., Shrivastava.R.K.,(2009): Performance
Evaluation of Reference Evapotranspiration
Estimation Using Climate Based Methods and
Artificial Neural Networks Water Resource
Management, v. 23,pp.825–837..
Chung,Y.W.,(2008):Unpublished Ph.D. thesis
“Predicting water table fluctuations using Artificial
Neural Network” Professor. Adel Shirmohummali
Fischell Department of Bioengineering 192p.
Coulibaly, P., Anctil, F. (2001): Artificial neural
nrtwork modeling of water table depth fluctuations,
Water Resource Research, v.,37,pp. 885-896.
Daliakopoulos, L.N., Coulibaly, P., (2005):
Groundwater level forecasting using artificial neural
networks, Journal of Hydrology,v. 309, pp. 229-240
Doğan, E., Koklu, R., (2007): Estimation of Biological
Oxygen Demand using ANN Network earthquake
symposium, Kocaeli pp.3345-3351.
Hamzaçebi, C., (2008): Improving artificial neural
networks performance in seasonal time series
forecasting, Journal of Science v.,pp. 4550–4559.
Hamed, M., Khalafallah, G., Hassanie, E. A., (2004):
Prediction of wastewater treatment plant
performance using artificial neural networks, Journal
of Science,v. 19 pp. 919–928.
Fig.5.Performance trend of the model Haykin, S., Neural Networks, A Comprehensive
Foundation, Pearson Education, 2006, 2nd edn.
Holger, R., Maier, Graeme C. Dandy., (2000): Neural
ACKNOWLEDGEMENT networks for the prediction and forecasting of water
Authors are thankful to the Head, Department of Civil resources variables: a review of modelling issues and
Engineering for providing the infrastructure to applications, Journal .Science Direct v. 15,pp.101–
complete this work. 124
Joginder, S., Kumar, R and S. Rajinder.(1981): Ground
REFERENCES water regression model for prediction of water table
Affandi, A. K., Watanabe, and India, Journal of Research. Haryana. Agric Univ.
Tirtomihardjo.H.,(2007): Application of An Hissar, India, v.11, pp.528- 533.
Artificial Neural Network to Estimate Groundwater

Joorabchi, A., Zhang, Hand Blumenstein, M., (2007): Optimum Structural Analysis and Design”, under the
Application of artificial neural networks in flow supervision of Dr.M.K.Gupta (Prof. & Head –
discharge prediction for the Fitzroy River, Australia Department of Civil Engineering, B.I.T, Durg).93p
Journal.Coastal Research v.51,pp. 287 – 291. Shirmohammadi, A., I. Chaubey, R. D. Harmel, (2006)
Lallahema, S., Maniaa, J., Hania, A., Najjarb, Y. : Uncertainity in TMDL Models. Transaction.
(2005): On the use of neural networks to evaluate ASABE, v.494,pp. 1033-1049
groundwater levels in fractured media, Journal of Song , Z.,Kong,L. Li, J and Zhang.H. ( 2007): A new
Hydrology,v. 307,pp. 92-111. analytical solution of tidal water table fluctuations in
McCulloch, W. S., and Pitts.W.,(1943): A Logical a coastal unconfined aquifer, Journal of Hydrology,
Calculus of the Ideas Immanent in Nervous Activity, v.340,pp. 256- 263.
Bulletin of Mathematical Biophysics,v. 5:,pp. 115- Sreekanth,P.D., Geethanjali, N., Sreedevi,P.D.,
33. Ahmad,S., Kumar,N.R.and Kamala
McCarthy, E. J., and. Skaggs. R. W ,(1991): A jayanti,P.D.,(2009): Forecasting groundwater level
simplified model for predicting drainage rates for using artificialneural networks.Current science, v.
changing boundary conditions, Transaction. 96, pp.933-938..
ASAE,v. 34,pp. 443-448 Upadhyaya, A., and Chauhan,H. S. (2001): Water table
Morshed, J.,Jagath, Kaluarachchi.J., (1998): fluctuations due to canal seepage and time varying
Application of artificial neural network and genetic recharge, Journal of Hydrology, v.244, pp. 1-8.
algorithm in flow and transport .Widrow, B., and Lehr.M. A., (1992): 30 Years of
simulations,Advances in Water Resources,v.l. 22, Adaptive Neural Networks: Perceptron, Madaline,
pp. 145–158. and Backpropagation. In Neural Networks –
Minyoung, K., Gilley, J.E.,(2008): Artificial Neural Theoretical Foundations and Analysis edited by
Network estimation of soil erosion and nutrient Clifford Lau, pp. 27-53
concentrations in runoff from land application Yakowitz, S. (1976): Model free statistical methods for
area,JournalScienceDirect,v. 6 4 ,pp. 268–275. water table prediction, Water Resources Research,
Nayak, P.C., Rao, Y.R.S., and Sudheer, K.P.(2006): v.12,pp. 836-844
Groundwater level forecasting in a shallow aquifer Yang, C. C., Prasher,S.O., Lacroix, R., Sreekanth, S.,
using artificial neural network approach, Water Patni, N. K. and Masse, L. (1997): Artificial neural
Resources Management, v.20, pp.77-90. network model for subsurface-drained farmlands,
Ranjithan, S., Eheart, J.W. and Garrett Jr.J.H. (1993): Journal of Irrigation and Drainage Engineering –
Neural network-based screening for groundwater ASCE, v.123,pp. 285-292.
reclamation under uncertainty, Water Resources Yang, C. C., Tan,C. S. and Prasher, S.O.,(2000):
Research, v.29, pp.563-574 Artificial neural networks for subsurface drainage
RosminaBustami,NabilBessaih, et.al(2007):Artificial and subirrigation systems in Ontario, Canada,
Neural Network for Precipitation andWater Level Journal of the American Water Resources
Predictions of Bedup River, International Journal of Association, v.,36,pp. 609-618.
Computer Science, v.34,pp.1-10.. ZengYuhong ,HuaiWenxin (2009): Application of
Roulet, N. T. (1991): Surface level and water table artificial neural network to predict the friction factor
fluctuations in a subarctic fen, Arctic Alpine of open channel flow, Journal .Science Direct,v. 14
Research,v.23,pp. 303-310 ,pp. 2373–2378.
Sarkar, K., (2009): Unpublished Ph.D thesis on
“Application of Artificial Neural Networks for

ISBN 978-81-921121-0-7
REDUCTION OF FLUORIDE WITH FREE JETS FOR INDUSTRIAL

APPLICATIONS
Suresh Kumar. N
Professor, Dept.of Civil Engg, College of Engg.(A), Osmania University, Hyderabad-500007 (A.P).
Email: nskdr@yahoo.co.in
Ravi Kumar, S.,
Post Graduate Student, Dept.of Civil Engg, College of Engg.(A), Osmania University, Hyderabad-500007 (A.P).
ABSTRACT: Fluoride is an essential element in water with certain concentration limit. Excess of fluoride
causes the seizing of the human body growth and leads to fluorosis. The population exposure to levels of above
1.5 ppm of fluoride had increased and led to several incidences of skeletal deformation and permanent
disability. A simple method called Nalgonda technique, not only helps in removal of fluorides, but also helps in
removing colour, odour, turbidity, bacteria and organic contaminations from raw waters. The present
investigation has been taken-up to investigate the mechanism of fluoride removal with help of hydraulic action
without any mechanical devices. The study has revealed that, the fluoride reduction through jets is as high as
90%. Hence, the usage of jet clariflocculators for fluoride reduction in public water supply systems would save
considerable time and result in smaller size of tanks due to smaller detention times. Moreover, the absence of
moving or mechanical components can eliminate most of the problems associated with corrosion, and
breakdown of mechanical components. As a consequence, the need for stand-by units is not obligatory.
INTRODUCTION contents are retained by this, through

Water is one of the most important natural resources. physical/chemical adsorption
Water is required for the survival of mankind. Water
is vital to life and development. In the past, a wide processes and fluoride free/reduced concentration
range of defluoridation materials and methods have water is obtained.
been tested and evaluated. Nevertheless, Nalgonda technique is available for Community
implementation is sparse and most often restricted to Defluoridation and Domestic Defluoridation.
limited areas in developing countries. Most common However, this system was not successful as the
defluoridation methods implemented in rural areas of community and individuals are not familiar with
developing countries are adsorption on activated correct dosages of chemicals, stirring speeds for
alumina, bone char and/or clay, co-precipitation with flocculation, and setting time. Hence, safe, efficient
aluminum hydroxide (known as Nalgonda and cost effective defluoridation
Technique) and contact precipitation with calcium techniques/processes need to be developed in order
and phosphate. Other methods like reverse osmosis to prevent the occurrence of fluorosis.
or electro dialysis though efficient are too The aim of the present study is to investigate the
sophisticated for implementation in rural areas of mechanism of jet clariflocculation for the reduction
developing countries. of fluoride concentration.
Natural concentration of fluoride in ground water DESCRIPTION OF EXPERIMENTAL SETUP

depends on many factors such as geological, Details of the experimental set-up used in the present
chemical and physical characteristics of the area investigation are shown in Fig.1and Fig.2. The
(nature of soil, porosity of rocks, pH, temperature clariflocculator was fabricated out of M.S. sheets and
and depth of wells). It is possible to map the pattern an anti corrosive paint was applied on all internal
of fluorine path by taking the exact location of each walls to avoid the corrosive action.
monitoring well and analyzing ground water flow
paths. The levels of fluoride in the water are Sodium fluoride was used to prepare the artificial
alarming in Nellore, Nalgonda and Vijayanagaram fluoride concentrations of required levels. It was
Districts of Andhra Pradesh. Further, high fluoride stored in a tank and pumped to the clariflocculator
concentrations occur as shock loads occasionally in with the help of a 0.5 h. p. pump. The required flow
smelter units of aluminum manufacturing industries. rate was set by adjusting the valve number 1, while
the excess flow was diverted back to the raw water
REVIEW OF LITERATURE tank through valve number 2. The inlet to the
Considerable work has been done all over the world flocculator was a submerged jet issued through a
on treatment of Fluorosis. Unfortunately the results pipe and the outlet from the clarifier was a
indicated that the effects of fluorosis are irreversible. suppressed sharp crested weir.
Defluoridation is a process of removal of fluoride ion
in drinking water. When the fluoride rich water The rate of flow entering the clariflocculator was
passes through the bed of said media, the fluoride measured with a calibrated portable dual frequency

Doppler flow meter, a product of Thermo Poly enter into a settling chamber through 4 numbers of
Sonics. The circular clariflocculator used in the rectangular outlets of size 19.6 mm x 12.5 mm. The
present investigation has a plan area of 1963.5 sq. dimensions of these outlets were computed according
cm. and a liquid depth of 450 mm. Earlier, it had to mass balance equation by assuming a flow
been found that for a proper mixing by a velocity of 0.15 m/s. The center line of these outlets
was provided at 86% of the liquid depth in the
horizontal jet, the optimum depth-wise location of flocculator measured from the free surface. This is to
the nozzle could be at three quarters of liquid depth ensure the same liquid level in the flocculator and the
when the liquid depth is equal to the tank diameter. settling tank (Suresh Kumar, N, 2005).
However, the optimum location is at the mid depth of According to field experience, the fluoride
the liquid, when the depth available is smaller than concentration in industrial waters normally ranges
the tank diameter. For horizontal position of the up-to 20 mg/l. Hence, in the present investigation
nozzle the mixing time was less than that obtained by fluoride reductions from a concentration of 20 mg/l
tilting the nozzle at different angles of inclinations to 1.0 mg/l are targeted. Sodium fluoride was used
(Maruyama Toshiro, 1986). to prepare the artificial fluoride concentrations of
required level. A calibration chart was used to predict
Failure to establish the desired flow pattern could the required concentrations which will yield selected
result in a tangential swirl flow, with very poor top- Sodium fluoride values. Fluoride concentration in the
to-bottom turnover in the vessel. Although tangential water sample was measured in a spectrophotometer
swirl may be capable of producing a good mixing, it using SPADNS solution. The decrease in colour is
also greatly increases the blend time for a given input directly proportional to the fluoride concentration.
power. Further, an incorrectly mounted nozzle may
only be satisfactory for providing uniformity of
mixing in the lower portion of the vessel (Oldshue, 3Al 2 ( SO 4 ) 3 18 H 2 O + NaF + 17NaHCO 3 
James Y., 1983). It has been recommended that, the
liquid level in the tank should be less than 1.25 times 5 Al(OH) 3 Al(OH) 2 F + 9Na 2 SO 4 + 17CO 2 +
the diameter of the tank or the dimension of the
54 H 2 O (1)
length of the chamber as the case may be (Holland,
F. A., and Chapman, F. S., 1966). Otherwise it may
require either multiple impellers or excessive power From the above equation
(as much as twice the amount necessary for the same (http://www.bibliotecapleyadas.net/salud/salud_fluor
degree of mixing). 23.htm. ) it is clear that, 42 parts of fluoride requires
17 x 84 parts of alkalinity as sodium bicarbonate.
In the present investigation the liquid depth in the Further, 42 parts of fluoride requires 3 x 666 parts of
flocculator was kept less than one diameter of the aluminum sulphate. Therefore, the alkalinity of the
chamber and the jet was injected from the middle of raw water will have to be measured during each run
water depth in the flocculator (Suresh Kumar, N, to assess the extra alkalinity to be created by artificial
2005). Aluminium sulphate solution of required means for each run.
strength was stored in a tank. Flow rate of coagulant
entering the raw water pipe line and leading to the jet Fluoride reduction studies were conducted using 5%
clariflocculator was adjusted according to the strength (Bhole, A. G., 1993) of aluminum sulphate
influent fluoride concentration. It is desirable that solution (5 parts of chemical to 95 parts of water to
coagulant solution be injected radially in the plane give 100 parts by weight of solution). After properly
perpendicular to the receiving stream as a counter- mixing the aluminum sulphate in water, its dosing
current axial jet. This is because the turbulent was started. Aluminum sulphate solution dosing rate
velocity intensities in the longitudinal, radial and was controlled by a peristaltic pump. Aluminum
circumferential directions generally increase from the sulphate flow rates for the selected fluoride
centre of the pipe towards the wall and would result concentrations are shown in Table 1 This following
in more efficient mixing (Pratte, B. D. and Baines, equation was used in this regard.
W. D., 1967). However, due to practical difficulties
in the present case, coagulant solution was injected Q1C1  Q2 C2  Q3C3 (2)
as a co-current axial jet through a 0.7 mm diameter
seamless stainless steel tube with the help of a Q 3C 3
peristaltic pump, a product of Miclins Pvt Ltd. Q1  (3)
C1
The jet diameter was selected as 12.5 mm. The flow where
rate was 0.1473 lps at detention time of 10 minutes in Q1 = Aluminum sulphate flow rate through the
the flocculation chamber (Suresh Kumar, N, 2005). peristaltic pump (lps)
The detention time chosen for the settling chamber C1 = Aluminum sulphate solution strength = 50,000
was 45.0 minutes. Flocculated water was allowed to mg/l

Q2= Flow rate in the influent line of the It is seen from the Table3that, the fluoride rate of
clariflocculator (lps) reduction is marginal or in other words it fairly
C2 = Concentration of the raw water remained constant.
Q3 = Flow rate in the flocculator = 0.1473 lps The allowable sulphate concentration in the potable
C3 = Concentration of aluminum sulphate needed for supplies is 200 mg/l (Expert Committee of Central
the fluoride removal (mg/l) Public Health and Envr. Engg. Org., 1991). Hence,
the effluent sample water (in case 20 mg/l influent
Table 1 Peristaltic Pump Flow Rates for Various fluoride concentration) was analyzed for sulphate
concentration. It was noticed that the sulphate
Fluoride Concentrations
Fluoride C3 Q1 Pump
concentration was about 200 mg/l. Therefore, the use
concentrationin (mg/l) (ml/min) setting of jet clariflocculators for fluoride reduction can be
water (mg/l) used up-to 20.0 mg/l. This is because of the fact that
12.0 570.85 100.80 223 the effluent sulphate concentration has reached the
14.0 666.0 117.72 260 maximum permissible limit of 200 mg/l.
16.0 761.14 134.53 296
18.0 856.28 151.35 334 CONCLUSIONS
20.0 951.43 168.17 371
The Bureau of Indian Standard recommends the
fluoride concentration to be 1.0 ppm for drinking
The alkalinity required for the removal of fluoride
water supplies. The allowable settling time ranges
was computed and alkalinity available in the raw
from 1 to 2 hours (Expert Committee of Central
water was measured during each run. Additional
Public Health and Envr. Engg. Org., 1991). As the
alkalinity to be created was computed (in sodium
effluent fluoride concentration could be brought
bicarbonate scale). It was converted into an
down to about 1.20 mg/l in 45 minutes of settling
equivalent sodium carbonate scale. This is because
time in jet clariflocculators, their use in water
creation of extra alkalinity by the addition of sodium
treatment works is recommended. Hence, they can be
carbonate is cheaper. (Table2)
retrofitted even in the existing systems without
modifying the units significantly
Table 2 Results of Alkalinity Tests
Influent Alkalini Alkalinit Extra Extra
Fluoride ty y of raw alkalinity alkalin Since the fluoride reduction is as effective as the
concent needed water in created in ity conventional flocculating systems, the usage of jet
-ration for NaHCO3 NaHCO3 create clariflocculators in public water supply systems
(mg/l) fluoride scale scale d in would save considerable time and result in smaller
removal (mg/l) (mg/l) Na2C size of tanks due to smaller detention times.
in O3 Moreover, the absence of moving or mechanical
NaHCO scale components can eliminate most of the problems
3 (mg/l) associated with corrosion, and breakdown of
scale
mechanical components. As a consequence, the need
(mg/l)
for stand-by units is not obligatory. Hence, the jet
12.0 408.0 232.84 175.16 87.58
14.0 476.0 232.84 243.16 121.58 clariflocculators can be cost effective for fluoride
16.0 544.0 188.16 355.84 177.92 reduction.
18.0 612.0 188.16 423.84 211.92
20.0 680.0 238.56 441.44 220.72 In addition, the usage of free jets for fluoride
reduction can be used in smelter units of aluminium
The water samples were collected at the outlet of the manufacturing industries wherein it comes as a shock
clariflocculator after the selected detention time. The load occasionally.
fluoride concentration in the effluent water was
measured by a spectrophotometer and an average of REFERENCES
the three samples is reported. Each run was repeated Bhole, A. G. (1993), ‘Performance of static
to ensure the repeatability of the performance of the flocculators’, Water Science and Technology,
tank. The results of fluoride reduction studies are IAWQ, 27(10) : 181-194
given in Table 3. Expert Committee of Central Public Health and
Envr. Engg. Org. (1991), ‘Manual on public
Table 3 Results of Fluoride Reduction Studies water supply and treatment’, Ministry of Urban
Influent Fluoride Effluent Fluoride Fluoride removal development, New Delhi : 290-298.
concentration concentration percentage (%)
(mg/l) (mg/l) http://www.bibliotecapleyadas.net/salud/salud_fluor2
12.0 0.92 92.33 3.htm. Drinking water by Anne marie,
14.0 0.95 93.21 helmemhine, A compilation of fluoride treatment
16.0 0.98 93.88 methods
18.0 1.00 94.44
Holland, F. A., and Chapman, F. S. (1966), ‘Liquid
20.0 1.20 94.00
mixing and processing in stirred tanks’, Reinhold

Publishing Corporation, Chapman and Hall Ltd., Pratte, B. D. and Baines, W. D. (1967), ‘Profiles of
London : 1-65 the round turbulent jet in a cross flow’, J. Hydr.
Maruyama Toshiro (1986), ‘Jet mixing of fluids in Div., ASCE, 93 (6) : 53-64
vessels’, Encyclopedia of Fluid Mechanics vol. 2 Suresh Kumar, N (2005), ‘Low-Cost Jet
edited by Cheremisinoff, N. P., Gulf Publishing Clariflocculators’, R&D Report, Dept. of Civil
company book division, Houston, Texas : 544- Engg., College of Engg., Osmania University,
556 Hyderabad
Oldshue, James Y. (1983), ‘Fluid mixing
technology’, McGraw-Hill Publications
Company, New York : 176-178
Fig. 1 Elevation of Jet Clariflocculator Set Up
Fig. 2 Plan of Jet Clariflocculator Set Up

ISBN 978-81-921121-0-7
COMPARATIVE BEHAVIOUR OF TWO STOREY BRICK MASONRY

BUILDING MODELS OF CONFINED, UN-CONFINED AND MODEL
AS PER IS CODE
Vikash Khatri
Research Scholar, Department of Civil Engineering, Institute of Technology, Banaras Hindu University, Varanasi-221005,
India, Email: khatri.hbtik@gmail.com
P. K. Singh
Professor, Department of Civil Engineering, Institute of Technology, Banaras Hindu University, Varanasi-221005, India,
Email: prof_pks@yahoo.co.in
P.R. Maiti
Assistant Professor, Department of Civil Engineering, Institute of Technology, Banaras Hindu University, Varanasi-221005,
India, Email: prmaiti.civ@itbhu.ac.in
ABSTRACT: During earthquake structural failure of brick masonry buildings take place mainly due to
following two reasons. (i) Weak buildings may collapse under high seismic load and ii) Due to resonance
amplitude buildup may cause collapse. As per Euro Code 8, a construction system where plain masonry walls
are confined on all four sides by reinforced concrete members is called Confined Brick Masonry (CBM). CBM
buildings are very strong as well as ductile, and are ideally suited for building construction in areas where soil
cover is more than 75m, for the following reasons (i) Deep alluvial soil strata eliminates chances of resonance
and ii) Good quality bricks and construction facility are economically available. Dynamic test results of 2-storey
building models on 1/5th scale of unconfined brick masonry (UBM), model constructed as per BIS provision,
and CBM are reported. Due to scale effect g-level of the prototype buildings shall be 1/5th of the models. It has
been found that CBM buildings are very strong and no damage occurred to the CBM building model even at
1.04g-1.88g level, in comparison to the unconfined model which failed at 0.32g-0.39g, and BIS code model
which failed at 0.65g-1.04g. Thus, CBM buildings are found to be approximately 5.9 times stronger in
comparison to UBM and 2.9 times stronger in comparison to buildings constructed as per BIS Code.
Keywords: Earthquake, Confined, Masonry, Buildings.
INTRODUCTION masonry buildings with reinforced concrete

As per Euro Code 81, a construction system where floors/roof have been used more widely in China.
plain masonry walls are confined on all four sides by
reinforced concrete members or reinforced masonry is
called confined brick masonry (CBM). In case of
CBM buildings the design philosophy adopted is that
neither the brick masonry nor reinforced concrete gets
damaged during earthquake condition.
Euro Code 8 limits the construction of CBM houses
located in seismic zones having design ground
acceleration (ag) greater than or equal to 0.3g to four
storeys (Table 1).
Table 1. Recommended maximum height of
building (H) and number of storeys (n). Fig. 1 Brick Masonry buildings in China
Design ground acceleration ag < 0.2 0.2 -0.3 ≥ 0.3 Buildings of confined brick masonry type (Fig. 2) are
[g] [g] [g] found in all regions of Chile.
Unreinforced H[m] 12 9 6
masonry
n 4 3 2
Confined H [m] 18 15 12
Masonry
n 6 5 4
Reinforced H [m] 24 21 18
masonry
n 8 7 6
Significance of Brick Masonry Buildings in India

As being followed in China and Chile brick masonry
apartment buildings can be the future of the apartment
buildings in India also. Since 1990, base isolated brick
Fig. 2 Brick Masonry buildings in Chile.

230 mm
4x12 dia
Socio-Economic Impact bars
6 dia stirrups @
Seismic zones II, III & IV of the Gangetic plain 150 c/c
are shown in Fig.3, where alluvial soil deposit is

having a depth of more than 75m and goes up to
few kilometers in some areas. Total population
M20
residing in the area is approximately 32.91 crores.
250 mm
Concrete
Therefore, earthquake resistant confined brick 0
Lean Concrete
masonry building for the area will have very high
100 mm
1200 mm
6x10 dia bars
socio-economic impact. either way
Fig. 5 Details of foundation for a four storey CBM

building
Site Effect
Seismic effect of local soil conditions on peak ground
acceleration are shown in Fig.6.
Fig. 3 Seismic zones II, III & IV of the Gangetic

plain having more than 75m soil cover
Technical Details Fig. 6 Effect of local soil conditions on peak ground

According to Euro Code 8 the cross-sectional area of acceleration.
rebars for tie-columns can be selected in dependence From the figure, it is seen that maximum seismic
of seismicity of the location and number of storeys in acceleration is considerably lower in the alluvium
the house. In composite confined brick masonry deposit in comparison to the rock mass.
buildings the column shall be of 230 mm x 230 mm Fig.7 gives a relationship between the natural period
having 4 bars of 12 mm diameter as longitudinal of soil and alluvium depth.
reinforcement and 6 mm diameter stirrups at the
spacing of 150 mm centre to centre. The details of
column are shown in the Fig. 4.
230 m m
6 d ia stir r u p s @
8 5 c /c
4 x 1 2 d ia b a r s
M20
C o n c r e te
230 mm
4 x1 2 d ia
150 mm
b ars
230 m m
6 d ia stir ru p s @ 1 5 0 c /c
b ) B a n d d e ta ils
a ) C o lu m n d e ta ils
Fig. 7 Relationship between the natural period of

soil and alluvium depth
Fig. 4 Details of composite column and Lintel level As the depth of soil deposit increases, fundamental
band period of the deposit also increases. Due to plastic
The foundation details corresponding to allowable deformation and cracking of the soil, high frequency
bearing capacity of 100 kN/m2 is given in Fig.5. The content of the earthquake waves can not be supported
width of strip footing for brick masonry shall be 1200 by the soil, and it quickly dies out in the soil.
mm and the dimensions for column footing shall be Therefore, in deep alluvial soil deposit area only low
1200mm x 1200 mm. The column footing shall be frequency and high amplitude earthquake waves are
reinforced with 6 bars of 10 mm diameter in both the experienced at the ground level.
directions. Fig. 8 shows relationship between damage and the
fundamental period of the soil in the 1967 Caracus
earthquake.

connection. The Load Deflection Curves for the tested
models are given in Figure 10.
According to them, in-plane strength of CBM wall
may go up approximately 10 times in comparison to
unconfined brick masonry.
300
250
200 Frame without infill
Infilled frame without

shear connection
150 Infilled frame with
Load (kN)
shear connection
100
Fig. 8 Damage and the natural period of the soil in

the 1967 Caracus earthquake. 50
As reported for the 1967 Caracus earthquake, 0

0 2 4 6 8 10 12 13
buildings up to 3 to 5 storeys constructed at places
Deflection (mm)
having soil cover more than 75 meters suffered
minimal damage, and suffered considerable damage at Fig. 10 Load deflection curves for static loading
places having soil depth less than 75 meters. Continuous Lintel Band Action
Similarly, buildings up to 10 to 14 storeys suffered Effects of continuous lintel band on out of plain
considerable damage at places where soil cover was vibration of the wall, and in plane strength of the wall
more than 75 meters and suffered minimal damage at are discussed below.
places where soil cover was less than 75 meters. a) Out of plane effect
In case of CBM building continuous lintel band is
Structural Action of CBM provided all around the building. This lintel band
Some structural actions of CBM are presented here breaks the wall height and thereby increases stiffness
for its clear structural understanding. of the wall and results in its reduced deflection to
Load Sharing about one fifth (Fig. 11).
In the CBM building, flexible nature of the slab and
slab slab
the lintel level band, helps the brick masonry wall and
w
the reinforced concrete column to act together to
support all the vertical loads in direct compression. lintel band
Load redistribution between reinforced concrete
column and brick masonry wall mainly at the offsets, w
w
ensures equal strain in the reinforced concrete and the
brick masonry at their interface (Fig. 9).
slab slab
A p p r o x im a te ly 1 m m
R C C o lu m n
Fig. 11 Deflection Fig.2.5

of BM(b) withwith
wall linteland
band without
B ric k M a so n ry
lintel band
a ) F r o m c o l u m n to w a l l b ) F r o m w a ll t o c o l u m n
Deflections of the wall for the two cases are compared
below.
Deflection without lintel band action,
Fig. 9 Load redistribution between reinforced
column and brick masonry wall wh 4

384 EI
CBM Action under In-Plane Static Loading
Singh et al.2 conducted experiment on three models, Deflection with lintel band action,
namely; (i) Reinforced concrete frame without infill
(ii) Brick masonry infilled reinforced concrete frame 4
having no shear connection, and (iii) Brick masonry wh 4  2  16 4 w 
    h 
infilled reinforced concrete frame with shear 384 EI  3  81 384 5

Thus, out of plane deflection of the wall reduces to the load bearing walls were increased by prestressed
about 1/5th due to the continuous lintel band where the to a desired level without affecting the similarity in
wall is assumed to be supported. mass distribution. The attachment of sufficient
b) In-plane effect additional masses is a difficult task for small scale
Singh et al.3 have reported experimental results of in- models in which the ratio of additional mass density
plane effect of continuous lintel band. They have to structural mass density becomes very large.
tested models of infilled frame without opening, Ignoring Gravity Model:
infilled frame with opening having continuous lintel In this model gravity acceleration may be ignored
band, and infilled frame with opening having isolated when the stress created by gravity force is rather
lintel band. smaller than that created by earthquake force. For
The ultimate load carrying capacity of infilled frame small height buildings, the stresses induced by gravity
with opening having continuous lintel band was loads are small compared to the stresses generated by
reported to be 1.7 times that of the infilled frame with seismic motions. Since gravity has a negligible effect
opening having isolated lintel. therefore there is no need to simulate gravitational
Separation of Orthogonal Walls at the Corner a
In the CBM buildings, the corner column, which is effect i.e. the ratio   need not be equal to one,
tied at the lintel and floor level, provides flexural  g r
support to the two orthogonal walls at the corner. This which allows more freedom in selecting model
action prevents wall separation at the corner during materials and scaling parameters. The advantage of
earthquake. this model is that they are simply scaled down
versions of the prototypes without the need of
MODEL TESTS additional masses. However in this modeling the input
Scale effect accelerations and frequency is increased by the factor
For small scale dynamic tests the model and prototype
must perform the same physics phenomenon,
1
and the time be compressed by the factor lr when
therefore the model must simulate the prototype in lr
size, material, boundary conditions and force effect. using prototype material in the models. Therefore for
There are several methods, with their advantages and testing such models on a shake table, the shake table
drawback for founding the similarities between model must be capable of reproducing high accelerations at
and prototype. These methods are derived from high frequencies. The similarity conditions of all the
dimensional homogeneity using Buckingham’s three types of models were compared in Table-2.
  theorem. Three main type dynamic models are: Table-2. Similarity Conditions
True Replica Model:
This model requires a material whose specific Similarity Conditions
E Model True Replica Artificial Ignoring Gravity
stiffness follows the same scaling law as the Types Models Mass Models
 Scaling Models
Parameters
length dimension. For model test on earth, the gravity
Length r l lr lr lr
acceleration is normally equal to 1g. So the model’s Time
elastic modulus should be smaller than prototype’s or lr1/2 lr1/2 E
1/ 2
tr lr  
the density or model material should be larger than
prototype’s, i.e. materials also have to simulate the   r
Velocity
stress-strain properties in the inelastic range. This lr1/2 lr1/2 E
1/2
condition limits the selective field of materials and for vr

 
most practical purpose these type of material cannot   r
be found easily. Frequency
Artificial Mass Model: lr 1/2 lr 1/2 E
1/2
r 1
lr  
In this model prototype materials which do not
influence the strength and stiffness of the model are   r
Gravitational
used for the construction of the models. Similarity in Acceleration 1 1 neglected
dynamic behaviour requires similar distribution of
masses and stiffness along the height of the prototype gr
and the model. Similarity in failure mechanism Acceleration
1 1 E
requires similarity of working stress level in the load ar lr 1  
bearing and structural walls of the prototype and the   r
model. Additional masses should be added to the Displacement
model and fixed to the floor slabs in order to fulfill lr lr lr
the requirements of similarity of mass distribution.
r
For similarity of working stress level the stresses in

Mass density Test Results
Er r r
r The building model after failure is seen in Fig. 14.
lr
Strain
1 1 1
r
Stress
Er Er Er
r
Modulus of
Er Er Er
Elasticity Er
Force
Er lr 2 Er lr 2 Er lr 2
Fr
Specific Stiffness
E
E lr r  
    r
  r
Energy
Er lr 3 Er lr 3 Er lr 3 Fig. 14 Test results viewed from East Face
 EN r
r is the ratio between model and prototype. Amplitude Measurement
lr is the ratio of length factor between model and prototype. Detailed measurements are taken using Laser sensors
and CATMAN Easy software. The amplitude at the
The models have been prepared and tested on 1/5th roof slab level was also measured by using a scale
scale. Since the tested models are only of two storeys, mounted on the stand and a pointer fixed to the
gravitational effect is neglected. In the direct model model, with the help of video recording.
study, response of the prototype is directly determined
from measurement of response of the model4. Frequency and amplitude
According to modeling theory5 input frequencies and Amplitude of the shake table was fixed at ± 10mm. In
accelerations for the models have to be increased by this model rate of change of frequency was 0.033Hz
five times. Elastic behavior of this type of model has /s. Total number of cycles subjected to the model was
been reported to be quite accurate when compared to 122 cycles in 87sec. The time interval was kept as
prototype structure. 3sec for each frequency step.
UBM Building model Test
An UBM 2-storey model was prepared on
geometrically reduced scale of 1/5th which is seen as
mounted on the shake table in Fig. 13.
Materials used for the brick masonry and RC works in
the experiment are; (i) Portland pozzolana cement, (ii)
1st class country bricks of size 46×23×14mm, having
average water absorption of 10.7% and compressive
strength of 35MPa, (iii) Coarse sand of size 4.0mm Fig. 15 Time Vs amplitude plot of UBM model (0-
downgraded to 1mm having FM of 6.29 used as 27sec)
coarse aggregate, and (iv) washed locally available The plots between time and amplitude at top of the
Ganga river sand used as fine aggregate having FM of model are shown in figures 15 and 16, which
2.81. represent plot for the first 27sec, and last 54 to 84sec,
Concrete mix of 1:1.5:3 by weight with water cement respectively. From figure 16, it is clear that model
ratio of 0.5 was adopted for all RC works, and cement vibrated with maximum amplitude of ± 12.5mm at the
mortar of ratio 1:3 by weight was used for the brick top, with a storey drift of 2.5mm.
masonry.
Fig. 16 Time Vs amplitude plot of UBM model

Fig. 13 UBM model Mounted on shake table (54-84sec)

g- level at failure of the UBM model
If the displacement / amplitude is given by:
y = a sin ωt
Then, ÿ = - a ω2 sin ωt
And, ÿ max = - a ω2
Maximum acceleration at the base level

= -a ω2
= 0.010  17.582
= 3.09 m/ sec2
= 0.32g Fig. 18 Building model as per IS 4326 with Laser

Maximum acceleration at the top slab level sensors.
= -a ω2
Test Results
= 0.0125  17.582 The building model after failure is seen in Fig. 19.
The failure took place at 0.65g level at base level and
= 3.86 m/ sec2 1.04g at top. The mode of failure was failure of the
corner steel in tension.
= 0.39g Amplitude of the shake table was fixed at ± 5mm. In
The main conclusion drawn from the above this model rate of change of frequency was 0.037Hz
experimentation is that the UBM building model /s. The amplitude with respect to time at roof slab
failed at 0.32g level in bending. Therefore, level, middle floor slab level and base level are
reinforcement at the corners was found to be measured with laser sensors.
necessary to enhance the g-level of the building
before failure.
Building Model as per IS 4326-1993.

A building model geometrically similar to UBM
model was prepared as per IS 4326-19936 provisions,
except confinement of openings (Fig. 17 and Fig. 18).
Fig.17 shows the model where masonry up to window
sill level with corner reinforcement welded to the base
plate is completed.
Fig. 19 Test results viewed after failure.

Frequency and amplitude
The plots between time and amplitude at top, middle
and bottom of the model are shown in figures 20, 21
and 22, which represent plot for 14.9-16.08sec,
75.82-76.8sec and 98.8 to 100.0sec, respectively.
Observed maximum amplitude at the top is ± 8.0mm.
Fig. 17 Masonry up to window sill level with

corner reinforcement.
The complete building model mounted on the shake
table is seen in Fig.18.
(14.9-16.08sec)

shake table which became unstable at 7.2Hz
frequency.
Test Results
Frequency and amplitude

Amplitude of the shake table was fixed at ± 5mm.
Time vs. amplitude plot for the CBM model are
shown in figures 25 and 26. The CBM model was

(75.82-76.8sec)

(98.8-100.0sec) Fig.23. Complete CBM model after mounting on
shake table
The main conclusion drawn from the above
experimentation is that the building model failed at
5.7Hz frequency i.e. at 0.65g level by the way of
rupture of vertical reinforcement at base level. No
other failure mode was noticed.
CBM Building Model

A building model on 1/5th scale and geometrically
similar to UBM model was prepared (Fig. 23).
Reinforcement details adopted in the CBM model (at
1/5th scale) are given in the table 3.
Table 3. Reinforcement details of the model

S. Particulars Dia of Nos./ Dia of wire
No rebars spacing used in the
As per model
Euro Fig. 24 CBM model after test
code 8 subjected to amplitude at the top level of model of
1. Column rein- 12mm 4 2.4 mm ±9mm and base amplitude of ±5mm.
forcement
2. Slab 8 mm 22mm 1.6 mm
reinforcemen c/c
t
3. Beam rein- 12mm 2.4 mm
forcement 6
5
amplitude in
4. Stirrups 6mm 2legged 1.2 mm

4
3
@ 17
2
1
0
-1
mm
mm c/c -2
5. Lateral ties 6mm 2legged 1.2mm -3
-4 012345678910
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
@ 30 -5
-6
mm c/c
6. Binding wire 22 26 gauge Time in sec
gauge
The model after test is shown in Fig. 24. As seen from Fig. 25 Time vs amplitude plot for CBM model ( 0-
here, there is no damage to the model at all after the 27sec)
test. The test had to be stopped due to limitation of the

level. Therefore, it is concluded that for the CBM
buildings, there is no necessity to provide offsets, as
given in Euro
10
9 In the CBM Building model there is no failure or
8
amplitude in mm
7
6
5 crack observed at openings and at junction of concrete
4
3 and masonry even at 1.88 g-level. Hence, it is
2
1
0
-1 concluded that there is no need to confine the
-2
-3
-4
-5 11 11 11 11 11 11 11 11 11 12 12 openings as given in the IS 4326-1993.
-6
-7 The CBM building model, tested without bond beams,
-8
-9 1 2 3 4 5 6 7 8 9 0 1
-10 exhibited no deficiency during the test even at 1.88g
Time in sec level. Therefore, it is concluded that in CBM
buildings provision of bond beam below the slab
level, as given in Euro code 8, is not necessary.
Fig. 26 Time vs amplitude plot for CBM model The CBM buildings are found to be approximately 5.9
(111- 120sec) times stronger in comparison to UBM and 2.9 times
stronger in comparison to buildings constructed as per
The CBM model did not fail even at 7.2 Hz BIS Code.
frequency. The maximum g-level of CBM at the base
level was 1.04g and at the top slab level it was 1.88g. ACKNOWLEDGEMENT
CONCLUSIONS The research work was carried out under Special

On the basis of experimental study, the following Assistance Program of the University Grants
salient conclusions are drawn. Commission, New Delhi in the Department of Civil
The frequencies and accelerations of the models is Engineering, Institute of Technology, Banaras Hindu
five times that of the prototype due to the scale effect. University.
In the UBM model initial cracks due to flexure
appeared at 0.32g – 0.39g at the base in the horizontal REFERENCES
direction at a frequency of 2.82Hz, which lead to final
failure. Therefore, there is need for sufficient vertical Euro Code 8: ‘Design provisions for earthquake
reinforcement at the corners to prevent this type of resistance of structures.’ Part 1-2: General rules
failure. for buildings’. ENV 1998-1-2:1995 (CEN,
Building model as per IS 4326-1993 failed at the Brussels, 1995).
maximum shake table frequency of 5.7Hz. Fixed P.K.Singh, S.Saxena and Prof (Dr) B N Roy.
amplitude at the base level was +5mm, and the 'Behavior Of Brick Masonry Infilled Reinforced
observed maximum amplitude at the roof slab level Concrete Frames Subjected to Static Loading’
was +8.0mm. The corresponding g-level at the base Journal of the Institutions of Engineers (India), vol
was 0.65g and at the roof slab level was 1.04g. Model 82, no 01, June 2001, pp 23-29.
failed by the way of corner reinforcement rupture at P.K.Singh, V. Singh and S.Yadav ‘Effect of Opening
the base level. Hence, it is concluded that single bar as on Behavior of the Infilled Frame with and
the vertical corner reinforcement is insufficient. without Continuous Lintel Band’ Journal of the
The CBM model was subjected to maximum Institutions of Engineers (India), vol 87, August
practically possible frequency of the shake table of 2006, pp 33-37.
7.2Hz (± 5mm base amplitude) in 486 cycles. No ‘Model Analysis of Structures’ by T.P. Ganeshan,
damage to the model was observed, and the model University Press (India) Limited.
remained intact after the test. Krawinkler H. and Moncarz P. D. “Similitude
In case of CBM, the shake table amplitude was fixed requirements for dynamic models.” Dynamic
at ±5mm, and maximum roof slab amplitude of the Modeling of Concrete Structure,SP-73, American
model was observed to be ±9mm. The corresponding Concrete Institute,1982,pp.1-22
g-level at maximum possible frequency and amplitude IS 4326-1993, ‘Earthquake resistant design and
was 1.04g at the base, and it was 1.88g at the roof slab construction of buildings – Code of Practice’,
level. Bureau of Indian Standards, New Delhi.
In the CBM model no separation of the brick masonry
and RC at the interface was observed even at 1.88g

ISBN 978-81-921121-0-7
DAMAGE ESTIMATION OF R.C.C STRUCTURAL MEMBER BY

DISPLACEMENT BASED ANALYSIS TECHNICS
A.Vimala
Ph.D(Computer Aided Structural Engg.), IIIT-Hyderabad, A.P, India, e-mail: vimlareddy@gmail.com
Ramancharla Pradeep Kumar
Associate Professor & Head, Earthquake Engg. Research Center, IIIT- Hyderabad, A.P, India
ABSTRACT: Reinforced concrete buildings constitute the dominant type of construction in the earthquake
prone countries. In this study a R.C.C building of four storey is considered in the seismic environment and
response of the structure under five ground motions are investigated using MAT LAB, SEISMOSIGNAL and
SAP packages and the damage is estimated for given ground motions.
The investigation is roughly divided into three parts, first part includes investigation of the dynamic
characteristics of ground excitations, second part includes evaluation of the nonlinear seismic behavior of
building subjected to the given ground excitations and third part includes evaluation of damage of the building
for the given ground motions Earthquake can cause damage to the structural and non structural components of
building. This document addresses structural damage. The present study is based on analytical investigation of
seismic performance and potential seismic damage of a member of a typical reinforced concrete framed building
due to earthquakes in India, using rational nonlinear modelling and displacement-based analysis techniques.
Since the seismic damage is directly correlated to the displacement (deformation) of the structure, the yield
displacement of the structure is calculated based on the material properties and geometry of the structure. From
the displacement time history of the structure response, the peak values which exceeding yield displacement of
the structure have been identified to calculate the damage of the structure. The estimation of structural damage
is calculated as per the model of Park and Ang following the modifications introduced by Kunnath et al(1992)
for the given ground motions
Keywords: time history, yield displacement, ductility, frequency, damage.
INTRODUCTION:
Details of the structure:
Earthquake ground motion is highly uncertain and
difficult to be predicted. The records of strong ground
motions from past earthquakes can serve to
provide a wealth of information that will be
indispensable in the damage analysis and design of
earthquake resistant structures. Earthquake is the
worst among all the natural disasters. It is not possible
to prevent earthquakes from occurring The disastrous
effects of these, however, can be minimised
considerably through scientific understanding of their
nature, causes, frequency, and magnitude. The seismic Fig.1
hazard analysis refers to the estimation of some Type of building : R.C framed structure
measure of the strong earthquake ground motion Plot area of the building : 369.5 sq.m
expected to occur at a selected site. This is necessary Built up area : 217.75 sq.m
for the purpose of evolving earthquake resistant No. of storeys : Ground+ 3 floors
design of a new structure or for estimating the safety Height of each floor : 3m
of an existing structure of importance, like dams, No. of columns : 24
nuclear power plants, long-span bridges, high-rise Size of column : 0.3m x 0.4m
buildings, etc. at that site. A fundamental objective in Size of beam : 0.23m x 0.3m
seismic evaluation of building structures is the Grade of concrete : M20
prediction of displacement imposed by the earthquake Grade of steel : Fe415
on the structure both at global and elemental levels.
The present study proposes a new approach for the List of earthquakes used in the analysis:
estimation of damage for each earthquake event by 1. India, Sept10,1986
analyzing the response of the structure under each 2. chamba, Mar24,1995
ground motion. In this study a R.C.C building is 3. chamoli, Mar29,1999
considered in the seismic environment and the 4. dharmasala,Apr26,1986
damage is estimated for given ground motions. 5. India Bangladesh border, Feb6,1988

Displacement time history of the structure: Yield displacement:
It is uneconomical to design structures to withstand
The displacement time history of the structure is
major earthquakes elastically. Therefore, the trend of
performed for five ground motions. The results are
design is that the structure should have sufficient
presented in figure2 to figure6.
strength and ductility to withstand large tremors
0.002
inelastically. For this the interconnections of the
0.001
members must be designed particularly to ensure
D is p la c e m e n t [ m ]
0.001
sufficient ductility.
0
Ductility (μ) of a member or structure is defined as
-0.001
the ratio of absolute maximum deformation
-0.001
(elongation, curvature or rotation) to the
corresponding yield deformation. Thus:
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Time [s ec]
µ= with respect to displacement
Fig.2 (India, Sept10,1986) = with respect to curvature
0.01 = with respect to rotation

D i s p la c e m e n t [ m ]
0.005
0
Ductility of a beam is normally determined for a
-0.0 05
particular cross section. A beam of width = b,
-0.0 1 effective depth = d, depth of the neutral axis = kd (in
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
elastic range) and = x (in the limit state of collapse),
u
Time [s ec]
yield strain of tensile steel = ε , yield curvature = Φy,
y
Fig.3 (chamba, Mar24,1995) we know:

Φy =
0.01
Where εy = = = 2.07 x 10-5
0.005
0
k = -mp +
-0.0 05
m = Es / Ec
-0.0 1
Φu =
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Time [s ec]
Where εu = ultimate strain of concrete = 0.0035.
Fig.4 (chamoli, Mar29,1999) = = ≤ ( )
0.015 Yield displacement = ∆y =

0.01
0.005
Where l = length of the column
0
Percentage of steel : 1.3
-0.005
Diameter of the bar : 16 mm
-0.01
-0.015 Size of the column : 0.3m x 0.4m
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Time [s ec]
Effective depth d : 352 mm
Fig.5 (dharmasala, Apr26,1986) m= = 2.07x10-3
0.015 k=-(8.95x1.3)+ √(8.952x1.32 + 2x8.95x1.3)=-1.385

0.01
Фy=2.07x10-3/(352-(-1.385x352))=2.465x10-6
D i s p la c e m e n t [ m ]
0.005
0 ∆y=2.465x10-6x30002/3=7.39 mm = 0.0073m
-0.005
-0.01 Calculation of peak displacements:
-0.015
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8 8.5 9

The peak displacements of the structure for five
Time [sec]
ground motions are presented in figure7 to figure11
Fig.6 (India Bangladesh border, Feb6,1988) Earthquake-1 (India, Sept10,1986)

Damage estimation
Structural Damage in a structure due to earthquake

loading may be due to excessive deformation, or it
may be due to accumulated damage sustained under
repeated load reversals. For the present study, the
damage is proposed to be calculated as per Park and
Ang Model (1985). This model consists of simple
X-axis -Time in sec, Y-axis - Displacement in mts linear combination of normalized deformation and
Fig.7 hysteretic dissipated energy as
Damage Based Life of Historical Constructions in
Earthquake-2 (chamba, Mar24,1995) Seismic Environment
Dlk = +β
Where xm is the maximum displacement that the

equivalent linear SDOF system would be subjected to
during the base excitation, xu = (µxy where µ is the
available ductility) is the ultimate displacement of the
system under monotonic loading, β represents the
X-axis -Time in sec, Y-axis - Displacement in mts
effect of cyclic loading on structural damage, Eh
Fig.8 represents the total energy dissipation in the structure
Earthquake-3 (chamoli, Mar29,1999) during the excitation and Qy is the yield strength of
the structure. The first term accounts for pseudo-static
displacement and the second term accounts for the
cumulative damage. Eh is the hysteretic energy
dissipation of the system, which is assumed as given
by Takeda et al. model.
Following the modifications introduced in the model
by Kunnath et al. (1992), the modified form has been
considered in the study
Fig.9 Dlk = +β
Earthquake-4 (dharmasala, Apr26,1986)
where Qy= xy K
Where K is the stiffness of the structure
xy is the yield displacement
N
Eh =  4 Q (E(xi)-xy)
i 1
y
Where N= No.of peak displacements exceeding the xy

th
E(xi) is the i peak displacement
Fig.10
Peak displacements
Earthquake-5 (India Bangladesh border)
The peak displacements of structure which exceeding
yield displacement for 5 earthquakes are presented in
the table1.
Table 1.
eq1 eq2 eq3 eq4 eq5
-
0 0.0089187 0.0077364 0.0077364
X-axis -Time in sec, Y-axis - Displacement in mts 0.0095967
-
Fig.11 -0.0135861 0.0087756 -0.015793
0.0131332
-
0.0119002 0.0155328 0.0150918
0.0110905

- damage for the structure due to this earthquake and
-0.0107308 0.0092988 -0.0166859
0.0155067 the structure is safe under such type of ground
-
0.0112507
0.0149149 0.0155328 motions.
- A fundamental objective in seismic evaluation of
0.0109201 -0.011072
0.0156684 building structures is the prediction of displacement
- demand imposed by the earthquake on the structure
0.0088844 0.0149149
0.0123076
both at global and elemental levels, since the seismic
0.0131308 -0.0077135 -0.011072 damage is directly correlated to the displacement of
- the structure or structural element.
0.0084683 0.0088844
0.0128437
This article presents an analytical investigation of the
-
0.0135256 -0.0088198 seismic performance and potential seismic damage of
0.0088158
- the frame using rational displacement-based analytical
0.0084111 0.0084683
0.0134138 method such as nonlinear dynamic time-history
- analysis.
0.0114905 -0.0074056
0.0077135
0.0084111
REFERENCES
-
0.0074056
A.K.Jain, “Quality Assurance and Control for
Structures Subjected to Earthquakes”, Indian
Concrete Journal, Vol. 70, No.11, pp. 601 – 608,
Damage calculation for earthquake 2:
November 1996.
Arlekar, J. N., Jain, S.K., Murty, C.V.R., “Seismic
K = 15.26 x 106 N/m
Response of RCFrame Buildings with Soft First
µ=3 Storeys” Proceedings of the CBRI
Chopra A.K., Dynamics of Structures: Theory and
β = 0.1
Applications to Earthquake Engineering. Pearson
xy=0.0073 m Prentice Hall, Third edition, 2007.
Golden Jubilee Conference on Natural Hazards in
xm= 0.0135
Urban Habitat, 1997,New Delhi.
Qy= 0.0073 x 15.26 x 106 = 11139800 N P.C.Varghese, “Advanced Reinforced Concrete
Design”, Prentice-Hall of India Private Limited,
N=4 (0.0089, 0.0135, 0.0119, 0.0073)
New Delhi, 2001.
Eh =4 x11139800 [(0.0089-.0073)+(0.0135- Pradeep Kumar, R. (1997). Damage Based Life of
Structure in Seismic Environment. Dept. of Civil
.0073)+(0.0119-.0073)+(.0107-.0073)]
Eng., Indian Institute of Technology – Kanpur
= 704035.36 J India.
Dlk = 0.0135 + 0.1 x 704035.36 = 0.489

0.0405 11139800 x 0.0405
Damage calculations for other earthquakes are
presented in the following table 2.
Table 2.
eq.
xy xm µ xu Qy N Eh D lk
no.
0.0073 0.00138 3 0.00414 11139800 0 0 0

1
2 0.0073 0.013586 3 0.040758 11139800 4 704035.4 0.489

3 0.0073 0.013526 3 0.040578 11139800 12 2227381 0.679
4 0.0073 -0.01669 3 0.05 11139800 12 1952673 0.56
5 0.0073 -0.01567 3 0.047 11139800 14 2497133 0.687
CONCLUSIONS
The uncertainty quantification of the ground motions
is important in reliability analysis and performance
and optimal design of structures
From the peak displacements of the structure response
for the earthquake1, it is observed that there is no

ISBN 978-81-921121-0-7
LONG-TERM STATISTICAL EXPRESSION OF TEMPERATURE

OVER INDIA
R. K. S. Maurya
Department of Geophysics, Banaras Hindu University, Varanasi-221005, Email:rksmaurya@gmail.com
G. P. Singh
Department of Geophysics, Banaras Hindu University, Varanasi-221005, Email: gpsinghbhu@hotmail.com.
ABSTRACT: Climate is the long-term statistical expression of short-term weather. Climate can be defined as
"expected weather". When changes in the expected weather occur, we call these climate changes. They can be
defined by the differences between average weather conditions at two separate times. Climate may change in
different ways, over different time scales and at different geographical scales. In recent times, scientists have
become interested in global warming. Mean monthly temperature over all-India has been studies that for the
period of 1901-2000. Detail analysis on annual and seasonal changes in mean monthly temperature (Tavg), mean
maximum temperature (Tmax) and mean minimum temperature (Tmin) have been done. Each year was divided
into four principal seasons, viz. winter, pre-monsoon, monsoon, and post-monsoon. The analysis shows that the
annual mean temperature, mean maximum temperature, and mean minimum temperature are increasing
approximately at the rate of 0.40°C, 0.65°C, and 0.40°C per 100 years respectively. On the regional scale,
Kothawale and Rupa Kumar (2005) have examined trends in surface air temperature over India from 1901–2003
and reported that the annual mean, maximum as well as minimum temperatures have significantly increased by
0.20C per decade in last three decades. It is remarkable acceleration of the warming trend compared to the trends
of past century (1901–2003). However, manifestation of such changes on shorter timescale has been recognized
to be one of the most crucial factors in terms of socio-economic impacts. Increasing atmospheric temperatures
and carbon dioxide along with uncertainties in annual precipitation have an adverse affect on Indian agriculture
(Aggarwal 2008).
INTRODUCTION energy balance. How this energy balance is regulated

Climate is the long-term statistical expression of depends upon the flows of energy within the global
short-term weather. Climate can be defined as climate system. Major causes of climate change
"expected weather.” When changes in the expected involve any process that can alter the global energy
weather occur, we call these climate changes. They balance, and the energy flows within the climate
can be defined by the differences between average system. Causes of climate change include changes in
weather conditions at two separate times. Climate the Earth’s orbit around the Sun, changes in the
changes differently in different time scales and amount of energy coming from the Sun, changes in
geographical scales. Many studies have reported that ocean circulation or changes in the composition of
the global mean surface air temperatures have the atmosphere. Large volcanic eruptions can affect
significantly increased in the last century and the global climate over only a few years. By contrast,
pronounced warming has occurred in the last three the movement of continents around the world over
decades (Jones et al 1999; Houghton et al 2001; hundreds of millions of years can also affect global
Jones and Moberg 2003). On the regional scale, climate, but only over these much longer time scales.
Kothawale and Rupa Kumar (2005) have examined In the last 100 years or so, the Earth’s surface and
trends in surface air temperature over India from lowest part of the atmosphere have warmed up on
1901–2003 and reported that the annual mean, average by about 0.6oC. During this period, the
maximum as well as minimum temperatures have amount of greenhouse gases in the atmosphere has
significantly increased by 0.20C per decade in last increased, largely because of the burning of fossil
three decades. It is remarkable acceleration of the fuels for energy and transportation, and land use
warming trend compared to the trends of past century changes, for food by mankind. In the last 20 years,
(1901–2003). However, manifestation of such concern has grown that these two phenomena are, at
changes on shorter timescale has been recognized to least in part, associated with each other. That is to
be one of the most crucial factors in terms of socio- say, global warming is now considered most
economic impacts. Increasing atmospheric probably to be due to the man-made increases in
temperatures and carbon dioxide along with greenhouse gas emissions. Whilst other natural
uncertainties in annual precipitation have an adverse causes of climate change, including changes in the
affect on Indian agriculture (Aggarwal 2008). amount of energy coming from the Sun and shifting
The overall state of the global climate is determined patterns of ocean circulation, can cause global
by the amount of energy stored by the climate climate to change over similar periods of time, the
system, and in particular the balance between energy balance of evidence now indicates that there is a
the Earth receives from the Sun and the energy which discernible human influence on the global climate.
the Earth releases back to space, called the global

DATA AND ANALYSIS RESULTS AND DISCUSSIONS
Monthly temperature data for all-India are uses in
the present study over a network of 121 stations are 1. Average Temperature
same as used by Pant and Rupa Kumar (1997). This The analysis shows that the change in average
is originally from the monthly weather records of the temperature over All-India (AI) has been by about
India Meteorological Department (IMD). The data 0.4 °C in 20th century and average temperature is
have been updated for the period 1901-2003 from 24.25 °C and standard derivation is 0.2655 0C for the
the Indian Daily Weather Reports (IDWRs) same period (figure-2). Highest and lowest
published by the IMD. temperatures are 24.896 0C (1941) and 23.363 °C
We can use the linear regression equation for the (1917) respectively. Numbers of cold and warm
trend line . We can apply a statistical years are 14 and 15 respectively for the same period.
treatment known as linear regression to the data and Table 1 clearly shows the decadal variation in
determine these constants. number of cold and warm years and an increase in
Given a set of years and temperature data set temperature during last decade. Figure-3 shows
depicts a seasonal variation in temperature for winter
with n is number of years.
season, pre-monsoon, monsoon season, and post-
monsoon season. It is noticed that a increase in
temperatures in all seasons by about 0.51°C, 0.31°C,
m is slope 0.12°C and 0.79 °C respectively.
Mean annual average air Temperature

0.8
b is y-intercept 0.6
n
n
A n o m a lie s o f T e m p e r a tu r e ( °C ) 0.4
0.2
n
σ is standard derivation 0
-0.2
-0.4
-0.6
x is mean
-0.8
n
n
-1
n
X 100 1901 1911 1921 1931 1941 1951 1961 1971 1981 1991
Coefficient of Variation ( C.V.) = Years
A Cold/warm year can be classifies as: Fig. 2 Surface air temperature anomalies (°C) over
Warm Year all India (anomalies are computed to their respective
Tavg - σ ≤ +1 mean 1901-2000). Trend is obtained using 5 points
Cold Year moving average.
Tavg - σ ≤ -1
Tavg is the average temperature and Table 1 Cold (C) and Warm (W) years over all-India
σ is standard derivation for period 1901-2000. Decades WI Cumulative sum
C W C W
1901-1910 5 0 5 0
1911-1920 2 0 7 0
1921-1930 1 0 8 0
1931-1940 2 0 10 0
1941-1950 2 1 12 1
1951-1960 0 3 12 4
1961-1970 0 2 12 6
1971-1980 2 2 14 8
1981-1990 0 2 14 10
1991-2000 0 5 14 15
Fig. 1 Observed the all India meteorological station. Total 14 15

Winter Season
1.5 2. Maximum Temperature
A n o m a l i es o f T e m p e r a tu r e (° C )
1 The analysis of maximum temperature over all-India
0.5 (AI) shows an increase in temperature about 0.65°C
0
during 20th century and average maximum
temperature is 30.2°C and standard derivation about
-0.5
0.34°C for the same period (figure-4). Highest and
-1 Lowest temperatures during the period of study are
-1.5 30.95 °C (1966) and 28.95 °C (1917) respectively.
-2 Mean annual maximum Temperature
1
-2.5
1901 1911 1921 1931 1941 1951 1961 1971 1981 1991
Years
A n o m a l ie s o f T e m p e ra tur e (° C )
0.5
Pre-Monsoon Season
2 0
1.5
-0.5
A n o m a lie s o f T e m p e r a t u r e ( °C )
0.5
-1
-0.5 -1.5
1901 1911 1921 1931 1941 1951 1961 1971 1981 1991
Years
-1
-1.5 Fig 4. Surface air maximum temperature anomalies

-2
(°C) over all India (anomalies are computed to their
1901 1911 1921 1931 1941 1951 1961 1971 1981 1991 respective mean 1901-2000). Trend is obtained using
Year
5 points moving average.
Post-Monsoon Season
1.5
3. Minimum Temperature
The analysis shows a change of Minimum
A n o m a lie s o f te m p e r a tu r e (°C )
1
temperature over All-India (AI) has increased about
0.5 0.14°C and average minimum temperature about
18.3°C and standard derivation about 0.27 °C
0
(Figure-5). Highest and Lowest minimum
temperatures are 19.14 °C (1958) and 17.61 °C
-0.5
(1971) respectively.
-1
Mean annual minimum Temperature
0.9
-1.5
1901 1911 1921 1931 1941 1951 1961 1971 1981 1991
0.6
Anomalies of Temperature (°C)
Years
0.3
Summer monsoon Season
0.9
0
0.6
A n o m a lie s o f te m p e r a t u r e ( °C )
-0.3
0.3
-0.6
0
-0.9
1901 1911 1921 1931 1941 1951 1961 1971 1981 1991
-0.3 Years
-0.6 Fig 5. Surface air minimum temperature anomalies

(°C) over all India (anomalies are computed to their
-0.9 respective mean 1901-2000). Trend is obtained using
1901 1911 1921 1931 1941 1951 1961 1971 1981 1991
Years
5 points moving average. .
CONCLUSION
Fig 3. Seasonal surface air temperature anomalies Discuses of the temperature over India for the 20th
(°C) over all India (anomalies are computed to their century and found are increase. Table-1 shows that
respective mean 1901-2000). Trend is obtained using increase of temperature for increase the decade year.
5 points moving average. The temperature is increasing 4 to 6 decade and

decreasing 6 to 9 decade. Detail analysis on annual
and seasonal changes in mean monthly temperature
(Tavg), mean maximum temperature (Tmax) and mean
minimum temperature (Tmin) have been done. Each
year was divided into four principal seasons, viz.
winter, pre-monsoon, monsoon, and post-monsoon.
The analysis shows that the annual mean
temperature, mean maximum temperature, and mean
minimum temperature are increasing approximately
at the rate of 0.40°C, 0.65°C, and 0.40°C per 100
years respectively.
ACKNOWLEDGMENTS
The authors also want to acknowledge to the
Department Science and Technology, Government of
India (No.SR/AS/05/2008) for financial support in
the form of project.
REFERENCES
Aggarwal PK (2008): Global climate change and
Indian agriculture: impacts, adaptation and
mitigation. Indian J Agric Sci 78:911–919.
Houghton J T, Ding Y H, Griggs D J, Noguer M, van
der Linden P J, Dai X, Maskell K and Johson C A
(eds) 2001 The Scientific Basis. Contribution of
Working Group I to the Third Assessment Report
of the Intergovernmental Panel on Climate
Change (Cambridge: Cambridge University
Press).
Jones P D, Horton E B, Folland C K, Hulme M,
Parker D E and Basnett T A (1999) The use of
indices to identify change in climatic extreme;
Climate Change 42 131–149.
Kothawale D R and Rupa Kumar K (2005), On the
recent changes in surface temperature trends over
India; Geophys. Res. Lett. 32 L18714,
doi:10.1029/2005GL023528.
Maurya R K , G P Singh and P K Singh ( 2010) :
Change in temperature and precipitation over
India and its homogeneous zones during 20th
century; Nava Gavesana Vol 1, Issue 2, Oct-
Dec, page 16-33.
.

ISBN 978-81-921121-0-7
EFFECTS OF CORROSION INHIBITOR ON THE PERFORMANCE OF

MILD STEEL AND PRESTRESSING STEEL
Vasugi Jegan
Undergraduate Student, P.S.G. College of Technology, Coimbatore
Vandhana Mary Jacob
Undergraduate Student, M.N.M. Jain Engineering College, Chennai
Radhakrishna G. Pillai
Assistant Professor, Department of Civil Engineering, IIT Madras, Chennai
ABSTRACT: Chloride-induced corrosion is a major deterioration mechanism in concrete structures. A method

to delay corrosion initiation and minimise corrosion rate is the addition of corrosion inhibitors in fresh concrete
(i.e., as an admixture). Calcium nitrite [Ca(NO2)2] inhibitors are widely used for this purpose. Literature
provides some information on the performance of mild steel reinforcement embedded in mortar with corrosion
inhibitor (Obi-Egbedi 2011, Ilayaraja et al. 2011, Rastogi et al. 2010). However, the information on the
corrosion performance of prestressing steel in such environment is scarce. Such information is necessary in
assessing the corrosion propagation period and can be helpful in enhancing durability of prestressed concrete
structures. This paper presents a comparative study on the performance of mild steel wires and prestressing steel
wires embedded in mortar cylinders (i.e., lollipop samples) with different dosages of corrosion inhibitors. The
mortar cylinders were 12 cm in length and 3 cm in diameter. Mortar was prepared with a water-cement ratio of
0.55 and calcium nitrite corrosion inhibitor dosages of 0, 2, and 4% by weight of cement respectively. The
samples were subjected to accelerated curing (as per IS 9013:1978) and then cyclic wet-dry exposure to chloride
solution. This was done to accelerate chloride migration towards the steel wire at the centre of the lollipop
sample. Just before the end of each wet period, the electrical potential and corrosion rate were measured. The
Linear Polarisation Resistance (LPR) technique and a 3-electrode corrosion cell setup were used to measure the
corrosion rate. The percent reduction in the capacity of the reinforcing bars was estimated using the corrosion
rates determined using the LPR technique. Corrosion propagation periods for a capacity reduction of 20% are
estimated for both mild steel and prestressing steel.
Keywords: Concrete, corrosion, chloride, corrosion inhibitor, calcium nitrite, mild steel, prestressing steel
INTRODUCTION
Service life of concrete structures The first phase from construction time to the
Chloride induced corrosion is a major deterioration
corrosion initiation time is termed “corrosion initiation
mechanism influencing the service life of concrete
structures. Fig. 1 shows a schematic of the overall phase”. When embedded in uncontaminated concrete,
service life of a structure. Typically the overall
the steel reinforcement is well-protected against
service life can be divided into three phases as
follows: 1) corrosion initiation phase, 2) corrosion corrosion. This is because, when steel comes in
propagation phase and 3) repair phase.
Damage
contact with the highly alkaline uncontaminated
level First Last concrete, an oxide film (known as the passive layer) is
repair repair
formed over the steel surface. Although this passive
Maximum Allowable Damage Level
film is very thin, dense, and invisible, it protects the
steel from further corrosion. However, chlorides from
external sources like sea water and/or ground water
can penetrate through the concrete cover, reach the
Time
Propagation Repair steel reinforcement surface, and break the passive film.
Initiation Phase Phase Phase
Once the passive film is broken and sufficient oxygen
Overall Service Life and moisture are present, corrosion initiation occurs.
Fig. 1 Schematic showing various phases during The amount of chloride ions required to break the
the service life of a concrete structure exposed to passive film is known as critical chloride threshold
corrosive environments (Pillai 2011)

ISBN 978-81-921121-0-7
(Clthreshold). Pillai (2003) estimated that the 0.5 to 1.0 Moderate to

high rate
Clthreshold of typical mild steel reinforcement can be
> 1.0 High rate
approximately 0.5 kg/m3 of concrete (0.09% by
weight of cement). It should be noted that there Corrosion inhibitors for concrete applications
exists significant scatter in the Clthreshold values A corrosion inhibitor is a chemical compound that
when mixed with concrete can inhibit or decrease the
published in the literature. This is because of the corrosion rate of the embedded steel. Calcium nitrite
complex electrochemical mechanisms of chloride- [Ca(NO2)2] has been used as corrosion inhibitor for
steel-cementitious systems. El-Jazairi and Berke
induced corrosion. Once actively initiated the (1990), Berke and Weil (1992) suggested that a more
chloride-induced corrosion is difficult to stop; coherent, homogeneous passive film forms due to the
catalytic action of nitrite in forming the passive film.
unless the chlorides are extracted or removed from In general, the Ca(NO2)2 -based corrosion inhibitor can
the steel-cementitious interface. The second increase the Clthreshold and reduce the corrosion rate of
mild steel in chloride contaminated mortar [Ann et al.
phase between the corrosion initiation time and the (2006)]. However, information on the corrosion
first repair time is termed “corrosion propagation performance of prestressing steel could not be found.
The information on the Ca(NO2)2 dosage and
phase”. The corrosion rate depends on the steel corresponding corrosion rates of prestressing steel can
microstructure, the cementitious environment, and be used to estimate the corrosion propagation period,
and hence, to design durable structures. This paper
the amount of chloride at the steel-cementitious provides such information on both mild steel and
interface. Significant amount of research has been prestressing steel available in India.
1.1 Corrosion rate measurement - Linear Polarization
done on the corrosion performance of mild steel Resistance (LPR) Technique
with the addition of various mineral and chemical Among various electrochemical techniques, the Linear
Polarization Resistance (LPR) technique can be used
admixtures (Bleszynki et al. 2002, Malhotra et al. for repeated measurements of corrosion rate of steel
2000, Li et al. 1999). Haleem and Wanees (2002) embedded in mortar. Fig. 2 shows the schematic of a
typical LPR curve. The electrical potential of steel
have studied the environmental factors affecting when the current, I, is zero (i.e., at origin; with an open
the corrosion behaviour of reinforcement steel. electrical circuit) is termed open circuit potential
(OCP). A positive or negative current flow will occur
Based on information from McDonald et al. (1996), when the potential is forced to change (i.e., polarized)
Pillai (2003) estimated that ASTM A615 mild steel and vice versa. The extent of this potential change
caused by the net current flow to or from an electrode
reinforcement in severe corrosive environments can is called polarization. Typically, the LPR technique
exhibit a corrosion rate of 24 mils/year (1 mil = uses a polarization or an overvoltage, E, of
approximately +20 mV only (versus OCP), which
0.001 inches). Gu et al. (1996) reported that the cannot cause significant electrochemical damage to the
corrosion rate of stainless steel is at least 50 times steel surface. Therefore, LPR technique can be
considered non-destructive and used for repeated
lower than that of mild steel in chloride corrosion rate measurements on same steel specimen.
contaminated environment.
Table 1 shows typical ranges of corrosion rates and

corresponding qualitative categories.
Table 1 Qualitative classification of corrosion

current density (Broomfield et al. 1994)
Corrosion current density, Corrosion
Icorr (µA/cm2) rate category
< 0.1 No corrosion
expected
0.1 to 0.5 No to
moderate rate

ISBN 978-81-921121-0-7
RESEARCH SIGNIFICANCE AND OBJECTIVES

Prestressed concrete elements are increasingly used in
major infrastructure projects such as metros, highway
bridges, ports etc. and housing projects. Information
on the corrosion rate of high-strength prestressing steel
strand is important to design durable prestressed
concrete structures. However, very limited
information is available on the corrosion rate of these
strands in concrete. In addition, the use of corrosion
inhibiting admixtures to enhance the durability of
concrete structures is increasing. However, there is a
lack of sufficient information on suitable dosages and
corresponding effectiveness in reducing the corrosion
rate of steel. This article fills these knowledge gaps to
Fig. 2 A conceptual linear polarization curve
some extent by providing information on the corrosion
(adapted from Pillai 2003)
rates of prestressing steel and mild steel in mortar with
various dosages of Ca(NO2)2 inhibitor. In addition,
The slope of the polarization curve, 1/Rp, at zero
the estimates of the time taken for a particular loss in
current (i.e., as the curve crosses the vertical
the cross-sectional area of the steel are provided. Such
current axis) can be used to calculate corrosion
information is helpful for designing durable and
current as follows:
cost-effective prestressed concrete structures.
B
I corr  (2)
Rp MATERIALS AND EXPERIMENTAL
PROCEDURES
where, Icorr is the instantaneous corrosion current; Experimental design
Rp is the polarization resistance; and B is 26 mV for Table 2 shows the experimental design of the testing
passive corrosion conditions. The instantaneous program and the number of samples tested in each
corrosion current density (or instantaneous variable combination. Mortar with a
corrosion rate) can be calculated by dividing the water:cement:sand ratio of 0.55:1:2.75 was used.
Icorr by the exposed surface area, A. As the Ordinary Portland Cement was used for mortar
corrosion activity near the OCP region increases, preparation. The Grade I, II and III sand (as per IS
the slope of the polarization curve will increase. In 383:1970) were mixed in equal proportions and used
other words, the higher the slope, the larger will be for making the mortar. Different dosages of calcium
the corrosion rate. It should also be noted that the
nitrite (Ca(NO2)2) corrosion inhibitor were added to
measured corrosion rate is not an average quantity.
the mix, per Table 2. Corrosion performance of mild
The two objectives of this research are: 1) to steel and prestressing steel samples were tested. Three
compare the corrosion rate of mild steel and samples were tested for each variable combination – a
prestressing steel embedded in mortar with total of 18 lollipop samples.
different dosages of Ca(NO2)2 corrosion inhibitor Table 2. Experimental design of the corrosion test
and 2) to predict the corrosion propagation period program
of mild steel and prestressing steel in mortar with Dosage of
different dosages of calcium nitrite corrosion Number
corrosion
Mortar of
inhibitor. type
inhibitor Steel type
samples
The organisation of the remaining paper is as (% by weight of
tested
cement)
follows. A brief discussion on the significance of Mild steel 3
this research is provided next. Then an overview 0 Prestressing
3
about the materials and experimental procedures steel
used in the test program is provided. It includes the 0.55 w/c Mild steel 3
experimental design followed by sample OPC 2 Prestressing
3
mortar steel
preparation, casting, curing, corrosion testing Mild steel 3
procedures and calculation of corrosion 4 Prestressing
3
propagation time. The paper then presents the steel
experimental results and the estimated time taken Total Number of samples  18
for 20% loss in tension capacity. Finally, the
conclusions drawn from this research are provided.1.2 Sample preparation, casting, and curing procedures
Fig. 3 shows the schematic of a lollipop sample (15 cm
long steel wire embedded in a 120 mm length and
30 mm diameter mortar cylinder). All the mild steel
wires were obtained from same stock. Similarly, all

ISBN 978-81-921121-0-7
the prestressing steel wires were obtained from The 50 mm long region at the centre of the steel wire is
same stock. Prior to embedding in mortar, the ends considered as the working electrode (WE). Two
of these wires were coated (as shown in Fig. 3) platinum plates (20 mm × 7 mm) kept at radially
with a 2-part epoxy. This was done to ensure that opposite sides of the WE were used as counter
only the centre 50 mm region of the steel wire electrode (CE1 and CE2). Saturated calomel electrode
(unshaded region in Fig. 3) was exposed to (SCE) was taken as the reference electrode (RE). It
corrosive environment and the other regions were was positioned such that the tip of the salt bridge
free of corrosion. connecting the electrode tip remained in close
The required quantities of cement, sand, water, and proximity with the working electrode. In other words,
corrosion inhibitor were batched. The three grades this was kept very near to the surface of the mortar
of standard sand were mixed for about 2 minutes. covering the exposed steel surface. All the electrodes
The cement was added to the sand and dry mixed were placed into a specially fabricated glass container
for another 2 minutes. The batched corrosion with 3.5% NaCl solution (i.e., the corrosion cell). This
inhibitor was added to mixing water. Water with setup was then connected to a computer and
inhibitor was then mixed with the cement-sand mix potentiostat (i.e., Solartron 1287 Electrochemical
for 3 minutes. After a rest period of 3 minutes, the Workstation) for performing LPR tests discussed in
mortar was mixed again for 2 minutes. The mortar Section 1.1.
was then placed into the plastic moulds (i.e.,
centrifuge tubes shown in Fig. 3) and tamped using
a 5mm diameter acrylic rod. Then the steel wire
was penetrated or embedded into the mortar and
tapped around the mould surface for proper
consolidation. The plastic moulds were removed
after drying the samples in laboratory room
conditions (27 o C and 60% RH approx.) for
23±1 hours. After this, the samples were immersed
in boiling 3.5% sodium chloride (3.5% NaCl)
solution (as per IS 9013:1978) for three and a half
hours. This was done to accelerate the curing of
mortar and the migration of chloride ions towards
the steel surface. It was found that sufficient
amount of chlorides required for active corrosion
was available at the steel surface. The samples Fig. 4 Schematic of the 3-electrode corrosion cell setup
were then taken out of the boiling NaCl solution used for LPR test
and placed in 3.5% NaCl solution at room Following is a demonstration of how the corrosion rate
temperature (27 o C) for another one day. is calculated using the LPR tests. In the LPR tests, a
small potential, say ±20 mV is applied between the
electrodes and the resulting current is measured. The
polarization resistance [Rp = (∆E/∆I)|∆E 0] is the
ratio of the applied potential, E (volts), and the
resulting current, I (µA/cm2). The measured Rp is
inversely related to the corrosion rate.
Fig. 5 shows a representative LPR curve obtained as
part of this program. Based on the slope of the best-
fitted line near the zero-current region and assuming
the Stern-Geary constant “B” equal to 26 mV, the rate
of corrosion is calculated to be 12.8 µA/cm2 (detailed
calculations are shown in
Fig. 5). The corrosion rates calculated this way can be
considered as the corrosion rate of the steel during the
corrosion propagation period.
Fig. 3 Schematic of lollipop sample. Corrosion

testing procedures
The cured lollipop specimens were subjected to a

cyclic wet-dry chloride environment using the 3-
electrode corrosion cell test setup shown in Fig. 4.

ISBN 978-81-921121-0-7
(hollow circular marker) is less than that of mild steel

(solid circular marker). The average corrosion rate in
samples with 0% inhibitor is approximately 50% and
66% higher than that in samples with 2% and 4%
inhibitor, respectively. This clearly indicates that the
corrosion rate can be decreased by the addition of 2%
1
Slope   4.93 10-4
Rp corrosion inhibitor. In Fig. 6(a), no statistically
B
I   0.026  4.93 10-4  12.8 A / cm 2
significant change in the corrosion rate is observed for
prestressing steel. Similar behaviour is observed for
mild steel samples shown in
Fig. 5 LPR curve of test sample with a corrosion
rate=12.8 µA/cm2 Fig. 6(a). Similar behaviour is exhibited by the data
1.3 Corrosion propagation time shown in Fig. 6(b) and (c). Therefore, the exposure
Fig. 1 shows the three different phases (including was stopped after 6 wet-dry cycles. At this time, the
the corrosion propagation phase) of a structure in chloride concentration in the mortar adjacent to the
service. The corrosion rates calculated using the steel in a representative specimen was found to be
LPR tests can be used to estimate the duration of 5.5% by weight of cement. This value is more than the
the corrosion propagation period. The following chloride threshold values reported in literature for mild
Faraday’s law is applied in determining the mass steel reinforcement. Therefore, it can be assumed that
loss of the sample, the active corrosion is ongoing at a constant rate. Fig.
Ita 7 compares the corrosion rates of mild steel and
m prestressing steel samples after 6th wet-dry cycle. It
nF should also be noted that, in the case of prestressing
(2) steel, no significant reduction in the corrosion rate is
where I is the current in amperes; t is the time in observed by adding more than 2% corrosion inhibitor.
seconds; a is the atomic number of iron; n is the The effect of dosage on mild steel is also similar in
number of electrons transferred; F is Faraday’s nature. This bar chart clearly shows that the
constant (96400 Coulombs per equivalent). For prestressing steel is having better corrosion resistance
mass, m, of steel or iron corroded can be calculated than the mild steel reinforcement.
using the following: I obtained from the LPR tests,
t = 31536000 seconds per year, a = 26, n = 2, and
F = 96400 Coulombs per equivalent. This
calculated value of m is equal to the mass loss per
year and can be assumed constant - resulting in a
constant rate of reduction in the area of cross-
section and, hence, tension capacity. Using this
information, deterministic estimates of the time
taken for a particular reduction in the cross-
sectional area can be calculated – giving an
indication of the corrosion propagation period.
2 Corrosion Rates Obtained From LPR Tests

Fig. 6(a), (b) and (c) show the corrosion rates of the
mild steel and prestressing steel samples embedded
in mortar with 0, 2, and 4% (by weight of cement)
of corrosion inhibitor, respectively. The number of
wet-dry cycles is plotted on abscissa and the
corrosion rate on ordinate. The plots show that, in
general, the corrosion rate of prestressing steel

ISBN 978-81-921121-0-7
account in this study. Therefore, these results on

prestressing steel may be non-conservative in nature.
Further experimental studies are required in this area.
Fig. 6 Corrosion rates of mild and prestressing steel

in mortar with 0, 2, and 4% Ca(NO2)2 inhibitor.
Fig. 7 Comparison of corrosion rates of mild steel F

and prestressing steel with respect to corrosion ig. 8 Time taken for 20% capacity loss of steel when
inhibitor dosage (Corrosion rates after 6th exposure embedded in mortar with various dosages: (a) mild
cycle) steel and (b) prestressing steel.
2.1
2.2 Corrosion propagation time
Fig. 8(a) and (b) show the estimates of the time CONCLUSIONS
taken for 20% capacity loss of mild steel and The following conclusions were obtained from the
prestressing steel when embedded in mortar with results:
various dosages. A comparison between Fig. 8(a) 1. In general, the prestressing steel corrodes at a
and (b) show that the corrosion propagation periods lower rate mild steel, when exposed to
for prestressed concrete systems can be more than uncontaminated and chloride-contaminated
that of a conventional concrete system with mild environments.
steel. For example, when no corrosion inhibitor is 2. The addition of at least 2% (by weight of cement)
added, the mild and prestressing steel systems will Ca(NO2)2 -based corrosion inhibitor to the mortar
take 11 and 18 years, respectively, for a tension can decrease the corrosion rate; and hence,
capacity loss of 20%. Similarly, when 2% increase the corrosion propagation period of both
corrosion inhibitor is added, the mild and mild and prestressing steel systems.
prestressing steel systems can take approximately 3. Addition of more than 2% corrosion inhibitor may
16 and 38 years, respectively, for a tension capacity not significantly decrease the corrosion rates and
loss of 20%. Interestingly, the dotted and dashed therefore, may not significantly increase the
curves in Fig. 8(a) is very close to each other, corrosion propagation periods.
indicating that the addition of more than 2%
corrosion inhibitor may not significantly enhancing ACKNOWLEDGEMENT
the durability of a mild steel system. Similar is the The authors acknowledge the Department of Civil
behaviour of prestressing steel system, see Fig. Engineering and the Central workshop at the Indian
8(b). In general, the presence of corrosion Institute of Technology Madras, Chennai for providing
inhibitors can increase the service life of the the laboratory facilities and other support for this
concrete systems with mild steel and prestressing research. We specially thank Mr. Jayachandran,
steel. Mr. Murali, Ms. Malarrvizhi, Mr. Soundarapandian,
However, it should be noted that the effect of the Mr. Senthilnathan, Mr. Subramaniam, Mr. Murthy, and
prolonged prestressing forces, which exist on actual Mr. Krishnan of the Dept. of Civil Engineering for
prestressed concrete structures, is not taken into their technical support.

ISBN 978-81-921121-0-7
Beaudoin J J, Zhang, M H, Malhotra V M., 2000,

REFERENCES “Long-term performance of steel reinforcing bars
Gu, P., Elliott, S., Beaudoin, J. J., and Arsenault, in Portland cement concrete incorporating
B., 1996, "Corrosion Resistance of Stainless moderate and high volumes of ASTM Class F Fly
Steel in Chloride Contaminated Concrete," ash”, Material Journal Publication, Vol No.97, pp
Cement and Concrete Research, V. 26, No. 8, 409-417.
August, pp. 1151-1156 Li Z, Peng J, Ma B., 1999, “Investigation of Chloride
McDonald, D. B., Pfeifer, D. W., and Blake, G. T., Diffusion for High Performance concrete
1996, The Corrosion Performance of containing fly ash, microsilica and chemical
Inorganic-, Ceramic-, and Metallic-Clad admixtures” Material Journal Publication, Vol
Reinforcing Bars and Solid Metallic No.96, pp 391-396.
Reinforcing Bars in Accelerated Screening Haleem and Wanees., 2002, “Environmental factors
Tests," FHWA Report No. FHWA-RD-96- affecting the corrosion behaviour of reinforcement
085, October, Federal Highway steel II. Role of some anions in the initiation and
Administration, McLean, VA. inhibition of pitting corrosion of steel in Ca(OH)2
Broomfield J.P, Rodriguez J, Ortega L.M, Garcia solutions”, Elsevier, Corrosion Science
A.M., 1994, “Corrosion rate measurement in IS 9013:1978, 1978, "Method of making, curing and
reinforced concrete structures by a linear determining compressive strength of
polarization device”. In: R.E. Weyers (ed.), accelerated-cured concrete test specimens",
Philip D Cady Symposium on corrosion of Bureau of Indian Standards, New Delhi.
steel in concrete. ACI Special Publication, Vol ASTM G 109, 2007, “Standard test method for
No.151-9. 1994, pp 163. determining effects of chemical admixtures on
Obi-Egbedi N O, Essien K E, Obot I B., 2011, corrosion of embedded steel reinforcement in
“Computational simulation and corrosion concrete exposed to chloride environments,”
inhibitive potential of alloxazine for mild steel American Society of Testing and Materials,
in 1M HCl” Scholars Research Library J. Conshohocken, PA.
Comput. Method. Mol. Design, 2011, 1 (1): 26- Pillai, R.G., 2011, “Corrosion and service life
43 prediction of reinforced concrete structures,”
Ilayaraja G, Sasieekhumar A R, Dhanakodi P, Proceedings of the National Conference on
2011, “Inhibition of Mild Steel Corrosion in Advances in Materials and Structures (AMAS
Acidic Medium by Aqueous Extract of Tridax 2011), Pondicherry, India.
procumbens L” ISSN: 0973-4945; CODEN El-Jazairi, B, and Berke, N.S., 1990, “The use of
ECJHAO, E-Journal of Chemistry, 8(2), 685- calcium nitrite as a corrosion inhibiting admixture
688 to steel reinforcement in concrete”, in: C.L. Page,
Rastogi R B, Singh M M, Singh K, Maurya J L., K.W.J. Treadaway, P.B. Bamforth (Eds.),
2010, “Electrochemical Behavior of Mild Steel Corrosion of Reinforcement in Concrete, Elsevier
in Dimethyl Sulfoxide Containing Science Publishers Ltd., Wishaw, Warwickshire,
Hydrochloric Acid” Portugaliae UK, pp. 571–585.
Electrochimica Acta 2010, 28(6), 359-371, Berke, N.S., and Weil, T.G., 1992, “World-wide
DOI: 10.4152/pea.201006359 review of corrosion inhibitors in concrete,
Hope B B, Ihekwaba N M, Hansson C M., Advances in Concrete Technology, pp. 899–924.
1995, “Influence of Multiple Rebar Mats on
Electrochemical Removal of Chloride from
Concrete”. Material Science Forum, 1995.
192-194: p. 883-890
Tait W.S., 1994, “An introduction to
electrochemical corrosion testing for practicing
engineers and scientists”,
http://www.amazon.com/introduction-
electrochemical-corrosion-practicing-
scientists/dp/B0006PFYBO.
Pillai, R.G. and Trejo, D., 2005, “Surface
conditions effect on critical chloride threshold
of steel reinforcement”, ACI Materials Journal,
Vol.102, No.2, pp 103-109.
Bleszynki R, Hooten R D, Thomas M D A and
Roger C A., 2002, “Durability of ternary blend
concrete with silica fume and slag”, ACI
Materials Journal, Title no. 99-M51.

ISBN 978-81-921121-0-7
REMEDIAL MEASURES FOR RESTORATION OF DISTRESSES IN

PART PARALLEL TAXIWAY AND APRON – A CASE STUDY OF LAL
BAHADUR SHASTRI AIRPORT AT BABATPUR, VARANASI
Brind Kumar
Assistant Professor, Department of Civil Engineering, Institute of Technology, Banaras Hindu University,Varanasi-221005
Pramod Kumar Singh
Professor, Department of Civil Engineering, Institute of Technology, Banaras Hindu University,Varanasi-221005
INTRODUCTION distress was made and analysed vis-à-vis their structural

The Airports Authority of India (AAI) has executed the design, construction and operating conditions. This
work of the extension of runway, part parallel taxiway paper presents the details of field observations made and
and the apron for upgradation of the Lal Bahadur the summary of remedial measures suggested.
Shastri Airport at Babatpur, Varanasi to cater to
international flights. While the runway is of flexible PAVEMENT HISTORY
pavement, the taxiway and apron are rigid pavements. As per information provided by AAI, the details of crust
After execution of the concrete work in 2010, the apron thickness of the rigid pavements for apron and part
was opened to traffic in November 2010. The part parallel taxiway are shown in Fig. 1. The pavement
parallel taxiway is not under traffic operation as yet. history reveal that, 41 cm Pavement Quality Concrete
Soon after their completion, distresses were observed by (PQC) was laid over 10 cm thick Dry Lean Concrete
the executing agency in form of cracks, popouts, (DLC) separated by 400 µm thick polythene sheet.
spalling, corner breaks and faulting. Realising the need Below the DLC layer, the pavement consists of 20 cm
to restore the distresses at the right earnest for thick Wet Mix Macadam (WMM) laid over 25 cm thick
maintaining their operational ability, the AAI Granular Sub-Base (GSB). The sub-grade was 30 cm
approached the experts for their advice to suggest thick cement stabilized layer, laid over the compacted
suitable remedial measures. natural soil.
A couple of site visits were made to understand the
nature and extent of distresses. Detailed study of each
400 µ thick polythene

PQC = 41 cm thick sheet
DRLC = 10 cm thick
WMM = 20 cm thick
GSBC = 25 cm thick
Soil cement stabilized subgrade = 30 cm thick
Fig. 1. Crust thickness of Apron and part parallel Taxiway

PRELIMINARY FIELD OBSERVATIONS loading, cracking in these panels is clearly due to
Apron area shrinkage only, and they are not structural cracks.In the
Distresses in the panels of the apron area were carefully apron area, cracking is similar to the taxiway panels.
studied and the following observations were made: Average crack width of the cracks in the taxiway panels
(i) Cracks were available in the panels, mostly is 1.30 mm, and is 1.71 mm in the apron panels.
extending from one construction joint to the Therefore, comparing the two it can be concluded that
other. cracks in the apron panels are due to shrinkage and
(ii) Crack followed the pattern of concreting. these are not structural crack due to failure of PQC and
(iii) Some cracks were found along the joints. underlying layers.
(iv) Some cracks extended into the adjacent panel. Comparing the cracking patterns in the panels for the
(v) Popouts were observed in some panels. taxiway and apron areas, it is observed that in both the
(vi) Corner breaks were observed in some panels. cases, the shrinkage cracks have formed along the
Generally, the cracks were confined to single panels shorter span only, and these are not propagating into the
running parallel to the shorter edge. adjacent panels unless there is some construction defect.
Taxiway area Therefore, once again it is indicated that, the cracks are
Cracking patterns in the part parallel taxiway area were due to shrinkage alone.
of similar nature to those of the apron area. Some of the
cracks in the taxiway panels were repaired by REASONS FOR CRACKING
“trenching and refilling” about one year back. Typically all the cracks appeared during the following
Apparently, no fresh cracks were observed at these winter season after the casting during the summer
locations. This suggests that the same methods may be season. The cracks being parallel to the shorter side of
successful at other locations where distresses were the panel in approximately straight line and near the
available. middle of the panel are indicative of shrinkage and
thermal cracks. Panels cast during winter were
DETAILED FIELD OBSERVATIONS apparently having no cracks.
The layout plan of the apron and part parallel taxiway is The reason of cracking in the panels includes shrinkage
shown in Figs. 2 and 3, respectively. The apron is strain and thermal strain. Whereas, shrinkage strain had
segmented into 11 parts from “A” to “K”, while the part been normal, thermal strain was higher than normal in
parallel taxiway has 8 segments from “A” to “H”. Every this case.
distress was numbered and its salient details were As per Clause 5.2.4.1 of IS: 1343-1980 [2], shrinkage
recorded corresponding to their respective panel strain is of the order of 0.0003. Therefore, shortening of
numbers. Crack width in affected panels was measured the 4.2 m long panel of M40 grade concrete is as
at 5 equidistant locations using vernier calipers having calculated below:
least count of 0.002 mm as shown in Fig. 4. Average of Shrinkage Shortening = εc x L
these crack widths were computed and recorded for = 0.0003 x 4200
each crack as shown in Table 1. = 1.26 mm
Spalling of joints, popouts and faulting of joints were As the concreting of the taxiway and apron panels was
measured in accordance with the “Distress Identification done during May-June when the temperature peaks to
Manual” of the Federal Highway Administration about 45οC, thermal shortening during winter season
Publication No. FHWA-RD-03-031, June 2003, U.S.A. when temperature is about 5οC, may be calculated as
[1]. given below:
Thermal shortening = αt x L x ΔT
= 12 x 10-6 x 4200 x 40
PHOTOGRAPHS OF THE DISTRESSES = 2.02 mm
For each distress, one photograph showing its relevant Therefore, it is observed that due to casting of concrete
details was taken in the apron and part parallel taxiway. at higher temperature (45°C), there has been excessive
contraction of concrete resulting in thermal cracks in the
DISCUSSION OF THE DISTRESSES panels.
Since the taxiway is yet to be opened for traffic, the The total contraction due to shrinkage and thermal
panels therein are not subjected to the design loading. shortening may of the order as given below:
Whereas, the apron area is in use for approximately one Due to shrinkage = 1.26 mm
year, the panels are subjected to the design loading. Due to thermal shortening = 2.02 mm
Panel cracking has been observed in both the areas i.e. Total = 3.28 mm
taxiway and apron. Since taxiway is not subjected to

Supposing that the entire 3.28 mm shortening produces resulting crack would be parallel to the shorter side and
the tensile stress in the concrete, then tensile stress in nearly at the middle of the longer side unless there is
concrete will be: some construction joint or other defects in the concrete.
If we assume that the concrete is free to slip on the
3.28 polythene sheet in a total of its half panel length from
  4200  40 3.28
4200 the edges, then maximum crack width of ( =) 1.64
2
= 20.74 N/mm 2
This is a very high tensile stress and this cannot be mm may be expected. This compares well with the
sustained by the concrete. As such cracking of concrete average observed crack width of 1.30mm in taxiway
particularly during extreme winter is obvious. The and 1.71mm in the apron area.
70/71 15 69 68 67 66 65 30 45
14 75 7629 63 44
C
13 7728 62 43
12 27 64 42
11 7826 41 61
10 25 5540 56 57 58 59 60
A
B
9 24 39 54 53 52
8 79 23 38 51
7 22 90 9137 92 95 94 50
6 8021 36 93 49 48
5 20 35 47
4 8119 34 46
3 82 18 88 89 96 33 45
2 72 73 74 83 84 17 86 87 98 97 32 44 43 42
1 16 85 31
4.1 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2
1 3
2 4 SHOULDER
106 121
105 120
104 119 15
I
103 118
102 117 16
H
101 116
100 115 14 13
99 114
98 113 17
97 112
96 111
95 110 20 18
94 109 23 19
93 108 24 21 25
92 107 41 22 26
60 75
59 74
58 73
57 72
56 27 71
E
55 30 70
54 28 29 69
53 31 68
52 32 67
51 66
D
50 3365
49 64 37
48 63
47 62 35 34 36
46 61
4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2

5 7 3.75
6 8 3.75
136 151 4.0
135 6 5 150 3 3 3 3 3 3 3 3 4.0
134 7 149 4.0
133 148 4.0
K
J
132 9 8 147 4.0
131 10 146 4.0
130 145 4.0
129 144 4.0
128 143 4.0
127 142 4.0
126 12 141 4.0
125 140 4.0
124 139 4.0
123 138 4 4 4 4 4 4 4 4 4 4.0
122 137 4.0
90 105 4.0
89 104 4.0
88 103 4.0
G
87 102 4.0
F
86 11 101 4.0
85 100 4.0
84 99 4.0
83 98 4.0
82 97 4.0
81 96 2 4.0
80 95 4.0
79 94 4.0
78 38 93 1 4.0
77 40 39 92 4.0
76 91 4.0
4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.5 4.5 4.5 4.5 4.5 4.5 4.5 4.5 4.5 4.5 4.5 4.5
Fig. 2. Layout plan of APRON showing distress numbers

57 SHOULDER 58 59
6 12 6 12 18
5 11 5 17
A 4
3
2
B 1
2
3 4 10
9
8 C 8
7
10
9
16
15
14
1 7 11 13
49 SHOULDER 50 51
4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2
60 61 62
24 30 36
23 29 35
D 22
21
20
E 28
27
26
F 14 13
34
33
32
19 25 31
52 53 54
4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2
63 64 3.5
42 48 3.83
41 47 3.83
G
40
39
38
H 46
45
44
3.83
3.83
3.83
37 43 3.83
55 56 3.5
4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2
Fig. 3. Layout plan of part parallel TAXIWAY showing distress numbers

Table 1. Details of field observations of distresses on Apron and part parallel Taxiway
Distress Panel size Description of distress / Crack width (mm)

Panel No. Description of distress Remarks
No. (m)
I II III IV V Av.
APRON:
Transverse cracking in Available in the full panel width. No
1 G-93 4.5 x 4 1 1.2 1 1.1 1.1 1.08
panel spalling was observed.
No spalling, no faulting, no loss of

2 G-96 4.5 x 4 Corner break 1.2 1.1 1.5 1.2 1 1.2
material and no patching was observed.
Cracking along Construction joint between panels is

3 K-150 4.5 x 4 1.2 1.5 1.6 1.5 1 1.36
longitudinal joint exposed.
4 G-138 4.5 x 4 2 2.5 2 2 2.5 2.2
Available in the full panel width.
Transverse cracking in
5 J-135 4.2 x 4 0.6 0.8 1.1 1.2 1.1 0.96 Spalling for average width of 15mm was
panel
observed.
panel
observed.
7 J-134 4.2 x 4 1.1 1.2 1 1.5 1.1 1.18 Spalling for average width of 13mm was
panel
observed.
panel
observed.
9 J-132 4.2 x 4 1 1 1.5 1.5 1 1.2 Spalling for average width of 14mm was
panel
observed.
Available in the full panel width in two
10 J-131 4.2 x 4 1.2 1.1 1.3 1.5 1.5 1.32 separate stretches. Spalling for average
panel
width of 11mm was observed.
Transverse cracking in Available in the full panel width. No
11 F-86 4.2 x 4 1 0.9 0.8 1.2 1.1 1
panel spalling was observed.
panel
observed.
13 I-115 4.2 x 4 3.5 3.5 3.6 3.7 2.8 3.42 Spalling for average width of 15mm was
panel
observed.
14 I-115 4.2 x 4 1.1 1.2 0.9 0.8 1 1 separate stretches. Spalling for average
panel
15 I-119 4.2 x 4 1.5 2 1 1.5 1.1 1.42 Spalling for average width of 17mm was
panel
observed.
Cracking cuts across transverse joint in
Transverse cracking
16 I-117 4.2 x 4 1 1.5 1.3 1.5 1.1 1.28 full panel width. Spalling for average
across panels
17 I-113 4.2 x 4 0.9 0.9 1.5 1.5 1 1.16 Spalling for average width of 14mm was
panel
observed.
panel
observed.
panel
observed.

Transverse cracking in Spalling for average width of 19mm was
20 I-110 4.2 x 4 0.8 0.5 0.5 0.6 1.9 0.86
panel observed. One end has low severity D-
cracking.
panel
observed.
panel
observed.
panel
observed.
panel
observed.
panel
observed.
panel
observed.
27 D-56 4.2 x 4 2.1 1.5 2.7 1.5 1.4 1.84 separate stretches. Spalling for average
panel
28 D-54 4.2 x 4 0.4 0.5 0.6 0.7 0.6 0.56 Spalling for average width of 13mm was
panel
observed.
Transverse cracking in Available in about half panel width. No
29 D-54 4.2 x 4 panel 0.4 0.5 0.4 0.5 0.4 0.44 spalling was observed.

Transverse cracking
30 D-55 4.2 x 4 2 1.5 1.5 1 0.9 1.38 full panel width. Spalling for average
across panels

31 D-53 4.2 x 4 2.1 1.5 2.5 1.6 1.8 1.9 Spalling for average width of 21mm was
panel
observed.
Multiple cracks with low severity

spalling was observed. No faulting was
32 D-52 4.2 x 4 Corner break 1 0.5 0.3 0.25 0.7 0.55
noticed. Average spalling width was
34mm.
Damaged plaster over

33 E-65 4.2 x 4 Plaster is fully damaged.
PQC
Cracking under No spalling and no faulting was

34 E-62 4.2 x 4 9 8.5 7.5 6.5 1.2 6.54
aerobridge pillar observed.
Cracking under No faulting was observed.Average

35 E-62 4.2 x 4 4.5 4.3 3.5 3.7 3.5 3.9
aerobridge pillar spalling width was 7mm

36 E-62 4.2 x 4 0.2 0.25 0.28 0.21 0.31 0.25
longitudinal joint exposed. No joint cutting was done.
Four numbers joints were not sealed and

37 E-64 4.2 x 4 No sealing of joints 0.6 0.46 0.52 0.72 0.65 0.59
are exposed.

Low intensity corner break has no
Corner break and
38 F-78 4.2 x 4 0.5 0.2 0.5 0.3 0.45 0.39 spalling or faulting. Low intensity
transverse cracking
transverse crackhas no spalling.
Cracking under No faulting. Average spalling width was
39 F-77 4.2 x 4 5.5 4.5 4 2.5 2.1 3.72
aerobridge pillar 11mm.
Cracking under
No faulting and no apparent spalling.
40 F-77 4.2 x 4 aerobridge pillar 7.5 6 5.1 1.5 1 4.22
41 I-107 4.2 x 4 1.1 1.2 0.9 1.1 1 1.06 Spalling for average width of 11mm was
panel
observed.
42 C-32 4.2 x 4 4.2 4.7 3.5 2.8 2.5 3.54 Spalling for average width of 23mm was
panel
observed.
panel
observed.
panel
observed.
panel
observed.
panel
observed.
panel
observed.
panel
observed.
Available in mid region of the panel
49 C-36 4.2 x 4 0.5 0.7 0.6 0.4 0.35 0.51 width. Spalling for average width of
panel
4mm was observed.
panel
observed.
panel
observed.
panel
observed.
panel
observed.
panel
observed.
Spalling of expansion No faulting. Average spalling width was
55 C-40
joint 7mm.
panel
observed.

panel
observed. No faulting.
panel

Transverse cracking
59 C-40 4.2 x 4 2.5 2.5 2.3 2.9 2.2 2.48 full panel width. Spalling for average
across panels

panel
observed.
panel
observed.
Corner broken into multiple pieces and a
Corner break and
62 B-28 hollow has formed. Average spalling
spalling of joint
width was 16mm.
63 B-29 4.2 x 4 0.6 0.8 1.1 0.4 0.3 0.64 Spalling for average width of 7mm was
panel
observed.
64 B-27 4.2 x 4 0.5 0.9 0.8 0.7 0.6 0.7 Spalling for average width of 5mm was
panel
observed.
Average spalling width was 27mm with
65 A-15 Spalling of joint
loss of material.
loss of material.
67 A-15 Spalling of joint Average spalling width was 17mm.
loss of material.
loss of material.
loss of material.
75 0.4 0.4 0.7 0.6 0.5 0.52
A-14 longitudinal joint exposed.
76 B-29 Spalling of joint Average spalling width was 9mm.
79 A-08 4.2 x 4 Popouts Multiple low severity popouts.
85 B-16 4.2 x 4 1.6 1.5 2.1 1.8 1.5 1.7

89 B-18 4.2 x 4 Popout 80 mm diameter Single popout.
90 B-22 4.2 x 4 Popouts Multiple low severity popouts.
91 C-37 Spalling of joint Average spalling width was 11mm.
Average crack width = 1.71
PART PARALLEL TAXIWAY:
Repaired transverse
cracking in panel with
T-1 B-10 4.2 x 3.8 1.5 1.2 1.3 1.5 1.5 1.4 Spalling for average width of 18mm was
new joints spaced at
300mm.
Repaired transverse
300mm.
Repaired transverse
400mm.
Repaired transverse
Available for about two-third panel
T-4 B-10 4.2 x 3.8 1.5 1.2 1.1 1.4 1.5 1.34 width. Spalling for average width of
14mm was observed. No faulting.
300mm.
Repaired transverse
T-5 C-17 4.2 x 3.8 1.7 1.5 1.5 1.7 1.1 1.5 Spalling for average width of 21mm was
300mm.
Repaired transverse
400mm.
Repaired transverse
400mm.
Repaired transverse
400mm.

Repaired transverse
400mm.
Repaired transverse
cracking in panel with Available in the full panel width. No
T-10 C-15 4.2 x 3.8 0.4 0.4 0.3 0.5 0.4 0.4
new joints spaced at spalling and no faulting was observed.
700mm.
Repaired transverse
600mm.
Repaired transverse
400mm.

T-13 F-32 4.2 x 3.8 1.6 1.4 1.3 1.1 1.5 1.38 Spalling for average width of 24mm was
panel.
Repaired transverse
T-14 F-32 4.2 x 3.8 1.1 1.2 1.1 1.4 1.1 1.18 Spalling for average width of 9mm was
700mm.
Average crack width = 1.30
Fig. 4. Recording of crack width using Vernier

Calipers

ISBN 978-81-921121-0-7
AN EXACT SOLUTION OF MHD FLOW DUE TO ACCELERATED

MOTION OF INFINITE POROUS PLATE
J. Singh
Dept. of Civil Engineering, Institute of Technology, Banaras Hindu University, Varanasi-221005
Munna Lal
Dheeraj Agrawal
Mona Goyal
ABSTRACT: An exact solution of MHD flow due to accelerated motion of infinite porous plate is considered in
this paper. Expressions for the velocity field and induced magnetic field have been obtained using Laplace
transformation technique when the magnetic Prandtl number is unity. The variations of velocity profile and induced
magnetic field are presented in graphical format, whereas numerical values of skin friction and current density are
given in tabular form.
Keywords: Magneto-hydrodynamics; Porous Media; Induced Magnetic Field; Laplace Transformation; Skin
Friction
INTRODUCTION MHD flow past an accelerated surface embedded in a

Magneto-hydrodynamics (MHD) is a branch of porous plate. Singh et al. (2010) analyzed the free
continuum mechanics which deals with the flow of convection effect on the flow of elactico-viscous fluid
electrically conducting fluids in electric and magnetic past an accelerated plate with constant heat flux.
fields. Many natural phenomena and engineering
problems are worth being investigated using MHD The present paper deals with the flow of an
analysis. Such types of flow find huge relevance in incompressible viscous and electrically conducting
engineering problems involving movement of natural fluid due to accelerated motion of an infinite porous
gas, oil and water through the oil reservoir. In chemical plate with a constant suction rate.
engineering, filtration and water purification processes
also involve MHD flows. This problem is also of great MATHEMATICAL FORMULATION &
interest in industrial technology and geothermal ANALYSIS
application such as liquid metal fluid and MHD power Consider the flow of an incompressible, viscous and
generation system etc. due to its frequent occurrence. electrically conducting fluid past an accelerated infinite
Magneto-hydrodynamics equations are ordinary porous plate. At time t’<0, both the fluid and the plat
electromagnetic and hydrodynamic equations modified are assumed to be at rest. At time t’>0, the infinite plate
in a form to account the interaction between the motion starts moving with a velocity u’= ct’ (where c is a
of the fluid and electromagnetic field. The formulation positive constant). Taking the x’ axis along the plate
of electromagnetic theory in the mathematical form is and y’axis perpendicular to it, the basic governing
known as Maxwell’s equation. The flow of an equations for the flow are:
′
incompressible, viscous, electrically conductivity fluid ′ = 0, (1)
past an infinite flat plate in the presence of a uniform
transverse magnetic field when a plate execute simple ′ ′ ′ ′
harmonic motion parallel to itself has been discussed ′ +v ′ =v ′
+ ′ (2)
by Ong and Nicholls (1959). They solved the problem
neglecting the induced magnetic field. Singh and Singh ′ ′ ′ ′
+v = + (3)
(1983a) have considered MHD through porous medium ′ ′ ′ ′
bounded by an impulsively started porous plate. Singh

and Singh (1983b) analysed MHD flow of an elastic Where u’ and v’ are the velocity component along and
viscous fluid past an accelerated plate. Ferdows and perpendicular to the porous plate, h’, the induced
Ota (2005) considered the similarity solution for MHD magnetic field, the density of fluid , the kinetic
flow through vertical porous plate with suction. viscosity , the electrical conductivity of the fluid and,
Choudhary and Jain (2010) gave an exact solution for the magnetic permeability of fluid.

The inertial and the boundary condition are: + +M - sh =0 (11)
′
For <0
Subject to boundary conditions
′ ′ ′ ′ ′ ′
( , ) = 0, ℎ( , )=0 (4)
= , ℎ = 0 at y = 0
′
(0 , ′ ) = c ′ , ℎ′ ( ′
, ′) = 0
= 0, ℎ = 0 at y→ ∞ (12)
For ′ > 0
Solution of equations (10) and (11) with boundary
′ ′
(∞, ) = 0, ℎ′ ( ∞ , ′ ) = 0 condition (12) and assumption =1 is given by:
The continuity equation (1) gives = [exp [-{(M+ ) + ( + ) + 4 } /2 ]
’
V= - (5)
+ exp [-{( + )+ ( − ) + 4 } /2 (13)
Were (>0) is the suction velocity of the fluid
through the porous plate. ℎ= [exp [-{(M+ ) + ( + ) + 4 } /2 ]
Now, in order to transform eqs. (2) and (3) in non-

- exp [-{( + )+ ( − ) + 4 } /2 ] (14)
dimensional form we introduce the following non-
dimensional quantities.
After the inverse transformation of equation (13) and
′ 1/3, ′ 1/3 (14), the velocity field and the induced magnetic field
y= (c / ) t= ( / )
are given by
u = ′ /( )1/3, u = ′ /( )1/3 ( )
u= [( +t) erf c{ }
√
1/2
pm = ,M=( ) /
( )
–( -t) exp{ -( + ) } erf { }
√
′
1/2
h= ( ) /
(6) ( )
+ [( +t)erf c{ }
√
In view of expressions (6), (2) and (3) can be put as:
( )
–( -t) exp{ -( − ) } erf { } (15)
√
- = +M (7)
( )
h= [( +t) erf c{ }
√
- = +M (8)
( ± )
–( -t) exp{ -( + ) } x erf { }
√
And the corresponding boundary condition assume the
form for t<0 ( )
− [( +t)erfc{ }
√
u ( , ) = 0, h( , ) = 0
( )
–( -t) exp{ -( − ) } erf { } (16)
√
u ( 0, ) = 0, h( 0, ) = 0 (9)
The velocity field being known the skin friction in
For t > 0
dimensionless form in given by
( ∞, ) = 0 ℎ( ∞, ) = 0
=[ + Mh]y=0
Applying the Laplace transform
( )√
= [{ +( + )erf c{ }
f (y, s)= ∫ ( , ) dt (s>0)
( )√
In eqs. (7) and (8) , we have –{ +2 }{ + ( − )erf c{ }
+ +M - su=0 (10)

(18)
- {exp[- t/4] –exp[- (17)
Results and Discussion:
Similarly the current density in dimensionless form is
given by The Variation of non-dimensional velocity and induced
magnetic field for different parameters values have
J=- - [ erf c{ } been shown in Figs. 1 and 2. The figures reveal that the
Figure 1: Variation of Velocity profile with y
–( -y+t( )} exp{ - Figure 2: Variation of induced magnetic field with λ

} + [( +t)erf c{ velocity increases with increase in M, t and decreases
}– } with increasing λ, whereas, induced magnetic field
shows contrary trend. Tables 1 and 2 present the
numerical values of skin-friction and current density. A
+{ – y+t( )y} } closer look at tables indicates that skin-friction and
current density decrease with M, and t.
- exp(
TABLE 1. The value of skin-friction
-( +t)exp-{ Values for

M λ
t = 0.2 t = 0.4
1.0 0 0.1000000 0.2000000
-( -t)exp{-( }exp{ - } 1.0 0.5 1.1083147 0.223233
2.0 0 0.2000000 0.4000000
2.0 0.5 0.2161528 0.4443926
TABLE 2. The value of current density
M λ t y
0 0.5 1.0 1.5 2.0
1 0 0.2 0.769195 0.769195 0.769195 0.769195 0.769195
1 0 0.4 0.683922 0.683922 .683922 0.683922 0.683922
- - - - -
1 0.5 0.2
0.667084 0.667084 0.667084 0.667084 0.667084
- - - - -
1 0.5 0.4
0.565320 0.565320 0.565320 0.565320 0.565320
2 0 0.2 0.490625 0.490625 0.490625 0.490625 0.490625
2 0 0.4 0.382637 0.382637 0.382637 0.382637 0.382637
2 0.5 0.2 0.456897 0.456897 0.456897 0.456897 0.456897
2 0.5 0.4 0.518938 0.518938 0.518938 0.518938 0.518938
+( )exp{-( ]
REFERENCE:
Choudhary and Jain (2010), International Journal of
Heat Transfer 53, 1609-1611.
Ferdows and Ota (2005), Journal of Computational and
Applied Mechanics 6 (1), 15-25.
Ong, R.S. and Nicholls, J. A. (1959), Journal of
Astrophysics and Space Science 26, 313.
Singh and Singh (1983a), Indian Journal of Theoretical
Physics 31, 219-223.
Singh and Singh (1983b), National Academy of
Science Letters 6 (7), 233-238.

ISBN 978-81-921121-0-7
GPS DERIVED STRAIN ANALYSIS AND CRUSTAL DEFORMATION IN

INDIA
Abhay P. Singh
PhD Student, NCEMP, University of Allahabaad, Allahabad 211002, Uttar Pradesh, , Email abhaysingh2000@gmail.com
Abhishek Rai
Scientist-B, NCEMP, University of Allahabaad, Allahabad 211002, Uttar Pradesh, , Email abhaysingh2000@gmail.com
ABSTRACT: The geodetic data at 18 sites in India are continuously recorded by several universities and research
institute using Global Positioning System (GPS). For study of Crustal deformation National Center of Mineralogy and
Petrol- ogy, University of Allahabad, Allahabad installed highly efficient and accurate LEICA GRX1200GG Pro
receiver at Ghoorpur near to Allahabad. This instrument is also equipped with MET3A sensor to record pressure,
temperature, humidity at regular interval of 30 second for study of precipitable water vapor content in troposphere.
The Latitude and longitude of the GPS sites is 25.21N, 81.30E. GPS sites installed on the Indian plate to measure rate of
displacement of indan plate. The velocity solutions of the measurement sites found after processing and error
corrections using GAMIT/GLOBK software. The GPS velocity data after error corrections and coordinate stabilization
in ITRF2000 reference frame were used for the determination of strain. First, the amounts of site velocity given as
northward velocity component and eastward velocity component were converted to the resultant velocity. The resultant
velocity data of two sites were resolved into components along the vector joining the sites and along a direction
perpendicular to the site vector. The ratio of the difference of the along site component and the distance between the
sites gives the longitudinal strain rate. The ratio of the differences of the components perpendicular to the site vectors
and the distance between them provides the magnitude of the shear strain rate. As the strain is a direction dependent
quantity, the orientation of the vector joining the two sites with respect to the geographical north direction was
determined. This angle gives the azimuth of the longitudinal strain. The azimuth of the shear strain was determined by
adding 90 degree to the azimuth of the site vector. The earthquakes of the Indian plate interior are interpreted to be
results of periodic accumulation of stress/strain due to shortening and release of strain along the same directions during
the extension. The tropospheric and ionospheric regions of the atmosphere affect the propagation of GPS signals. The
neutral atmosphere, which is the non-ionized part of atmosphere, can Normally be divided into two components, the
hydrostatic (dry) and wet portions of the troposphere. The hydrostatic component consists of mostly dry gases
(normally referred to the dry part) ,whereas the wet component is a result of water vapor. Objective of this paper to find
integrated water vapor over Allahabad.
keywords: Crustal deformation, Strain accumulation, Indian plate dynamics, PWV.
INTRODUCTION GPS
A high precision dual frequency Geodetic type GPS
receiver, choke-ring Antenna along with Met-Package is
installed at India Meteorological Department Campus,
Ghurpur, Allahabad. GPS satellites broadcast at two
frequencies, 1.57542 GHz (L1 signal) and 1.2276 GHz
(L2 signal). Most of the receivers, including the one
installed at Allahabad, produce all types of measurement
data (L1, L2, code and phase). The satellite network uses
a CDMA spread-spectrum technique where the low-
bitrate message data is encoded on high-rate pseudo-
random (PRN) sequence that is different for each
satellite. The PRN code for each satellite is hard-wired
into the receiver so that it can reconstruct the actual
signal from the satellites. Fig 2. The visibility of different satellites as a function of time of the
day (24 hrs). Atleast 8-9 satellites are visible at any particular time of
the day
The above figure shows the path and visibility of the

satellites at a particular day. The distribution of satellites
are homoginious above the GPS station. At any
particular day, at least 8 to 9 satellites are visible to the
antenna, which can provide sufficient accuracy to the
Fig 1. Satillite visibilitiy at Allahabad. Upper figure shows the path of GPS data analysis.
various satellites as they pass above the antenna.

GPS Data Quality Analysis The raw data in rinex format was provided to the
National GPS-Geodetic data bank at Survey of India,
Dehradun and was also processed locally at NCEMP,
Allahabad University. The data from Indian GPS-
Geodetic network, including Allahabad data is made
available to the scientific community in India for
research Fig. 5. GPS stations used for studying
Crustal deformation. purpose, from the SOI data bank
at Dehradun. At NCEMP, Allahabad Univerity, the GPS
data recorded has also been analyzed locally, along with
data recorded at neighboring GPS stations. For this, the
GPS data from neighbouring GPS stations such as
Lucknow (GSIL), Hyderabad (HYDE), Bangalore
(IISC), Dehradoon (DEHR), Wadia Institute (WIHG),
Bhopal (2191), Jabalpur (BARG), Tezpur (TZPR),
Gawhati (GHTU), were obtained from GPS-Geodetic
data bank of Survey of India, Dehradun. Adjacent figure
shows the location of GPS stations in the Indian
Fig. 3. Quality control parameters: Signal to noise ratio,
subcontinent, which were used in our analysis.
and multi-path effect on two frequencies, as a function of
Simultaneous processing of data from many station
azimuth and elevation. Various colors show the effect for
removes many common mode errors and consequently
different GPS satellites.
improves the overall accuracy of analysis.
Fig. 5 GPS sites location deployed in India
The following table shows the location, type of receiver

and data availability period for the station used in our
study. An a-priori requirement of a tightly-constrained
coordinate repetability solutions, is good azimuthal
distribution of GPS stations.
Table 1 (GPS sites location used in Analysis)
Fig. 4. Quality control parameters: (Same as in above

fugure, but as a function of time of the day).
The accuracy of GPS data processing to a large extent

depends on the quality of GPS data. Therefore, quality
control and pre-processing of data set is an important
consideration while studying millimeter rage accuracy.
Above figure illustrates the data quality for Allahabad
GPS station. As obvious from the above figure, singal to
noise ratio increases the elevation of the satellite
increases, whearas the multi-path effect decreases when
the elevation of the satellite increases. Multipathing is
minimum when satellites are nearly above the receiver.
We did regular quality control analysis of the data by
post-processing the GPS data, at regular intervals.

Fig: 6. Coordinate repeatability plots: HYDE (top left),
GPS Data Processing IISC (top right), ALDU (bottom left), and BARG
processing of the GPS data various softwares are (bottom right) time series.
available in private and public domain. We used the
public domain processing software developed at MIT,
with minor changes required to suit our needs. For pre-
processing and quality control of GPS data we used the
various other softwares available in public domain and
locally developed at NCEMP for various purposes. The
major steps in GPS data processing requires processing
of phase and code data recorded during the period of
experiment. The basic steps in data processing requires
obtaining the accurate satellite navigational path which
becomes available for download, in few weeks time
from, by various international agencies. The other major
steps in data processing requires, an iterative cleaning of
GPS data for cycle slips, modeling of theoretical phase
and code data, computing partial derivatives of phase
with respect to various parameters, estimation of
atmospheric and ionospheric delay parameters and
finally least – square estimation of parameters and
goodness of fit. The goodness of fit is upgraded by an
iterative procedure. Other minor steps require removing
effects of ocean-tidal effects, clock drifts and visual and
quantitative inspection of coordinate repeatability time-
series and sky-plots.
Fig. 7. GPS velocity vectors: GPS site velocity in Indian frame of

reference (left panel), relative GPS velocity vectors with ALDU
velocity fixed.
For the frame stabilization HYDE and IISC is used. The

stations for which data availability is only for a limited
period, is loosely coupled, however sites having
adequate data available, the solutions are tightly coupled.
The coordinate repeatability time series for various
stations are shown in above figure.The velocity of IGS
Hyderabad site (HYDE) is 32.78±0.07, 40.14±0.16,
−0.61±0.42 in the three directions. The result depends on
analysis of 665 days of data and is very near to results
obtained by International agencies such as SOPAC. The
velocity offset of IISC, Bangalore is 36.16±0.07,
41.69±0.16, −2.97±0.42 and is based on analysis of 660
days of data, whereas POL2 velocity offsets are
6.28±0.07, 27.22±0.13, 42.15±0.35 based on based on
data of 534 days. Velocity offset of DEHR is
32.70±0.36, 33.02±0.56, 3.03±1.27 on the basis of
analysis of 384 days. Velocity offset of BARG(
44.02±1.26, 37.70±2.50,15.87±6.15 result is based on
analysis of only 184 days of data and therefore results
are loosely constrained as data availability is only for
limited period. The velocity offset of Allahabad station

(ALDU) is 40.48±0.41, 39.07±0.73, −6.98±1.71. This triangle. The magnitude of strain in the study region of
conclusion is based on analysis of 237 days of data even order of micro-strain/yr.
the total time span of data availability is for two years,
therefore this solution is tightly constrained solution. The We intend to use the data generated at Allahabad during
frame of reference has been stabilized using IGS sites this project, along with data from other stations to
IISC, HYDE. The solution could be improved by adding understanding the dynamics of deformation in Indian
more IGS sites on same cratonic blocks. In our study subcontinent.
region only two IGS station IISC, HYDE are available.
The velocity offset has been calculated using ITRF2005
frame of reference. Velocity vector in ITRF2005 has
been shown in Figure 5.3. As we go from south to north
India the velocity of plate motion changes rapidly and
rotating in anti clock wise direction. The preliminary
estimate of velocity at NCEM is about 50 mm/yr and
angle is 43.98N . The circle at the top of velocity vector
represent the error bar. The error bar is direction of
velocity is sufficiently low. KUNM and NTUS is going
in SW direction. The velocities of all sites have been
calculated in Indian frame of reference (Fig 5). The
relative velocity of all the sites have been calculated
fixing IISC, and NCEM and is shown in Fig. 7.
Fig. 9
Estimation of Water Vapor Content in the
Atmosphere
Another important outcome of the GPS data processing
is the determination of precipitable water vapor content
in the atmosphere above the receiver. This parameter is
of immense important in short and medium term weather
forecasting as well as studying the atmospheric cycle.
The meteorological parameters were recorded at a 30s
interval by Paroscientific Met3A sensor (Fig. 1).
Pressure, temperature and humidity data collected during
the experiment was used to compute water vapor content
and its seasonal variation above the GPS station. The
neutral atmosphere, which is the non-ionized part of
atmosphere, can normally be divided into two
components, the hydrostatic (dry) and wet portions of the
Fig 8: GPS site Velocity in ITRF2005 frame of troposphere. The hydrostatic component consists of
reference. Scale 20mm/yr. mostly dry gases (normally referred to the dry part),
whereas the wet component mainly comprises of water
To find change in strain field and deformation pattern of vapor present in the atmosphere. The troposphere causes
the Indian plate besides measuring the plate-velocity at radio signal delay. The hydrostatic fraction contributes
the station and perceptible water-vapor content in the approximately 90% of the delay. The pressure,
atmosphere. For anaysis of the crustal deformation temperature, relative humidity profiles for Allahabad
reference frame realization site Bangalore (IISC), GPS station are shown in Figure 11.
Hyderabad (HYDE) has been used. In the analysis data
from IGS station KUNM, SAMP, POL2, KIT3, BAHR,
NTUS has been used from SOPAC data archive.
Allahabad station is located in a quiet environment at
local IMD campus at Ghoorpur Allahabad. The data
from this station provides valuable information about the
Indian plate dynamics and important parameters of water
vapor content in the atmosphere, for weather forecasting
models. All GPS sites have triangulate to show the strain
accumulation in the GPS network. The direction of
contraction and extension are plotted at the centre of

Fig. 10 ZWD variation for one year data set. Monthly Integrated water vapor variation in the data
range July 2007 to June 2009
For Allahabad and adjoin regions the GPS wet delay
when compared with theoretical models, it is observed REFERENCES
that GPS calculated wet-delay closely resembles with Welch, G. and G. Bishop, An Introduction to Kalman Filter.
Hopfield and Saastamonien models, during the summer Department of Computer Science,Univ. North Carolina at Chapel
Hill. TR-95-041.http://www.cs.unc.edu/ welch/publications.html.
season. However, during the winter season the wet delay Webb, F. H., and J. F. Zumlierge, An Intoduction to GIPSY-OASIS-II
matched closely with Ifadis model. The wet delay can be JPL Publ., D-11088,300pp.
easily converted into integrated water vapor. For wet www.sopac.ucsd.edu/dataArchive/
delay calculations, we used empirically formula to Shuanggen Jin and Pil-Ho Park , Tectonic activities and deformation
in South Korea constrained by GPS observation Issue 2, Vol. 1,
determine Tm (Bevis et. al.). The integrated water vapor 2007, international journal of geology.
has been calculated at 00:00 hrs, 06:00 hrs, 12 hrs, 18:00 E. Calais,1 L. Dong,1 M. Wang, Z. Shen, and M. Vergnolle
hrs and plotted in the Figure 12. The smoothed integrated Continental deformation in Asia from a combined GPS solution
water vapor has been shown by blue curve (Figure 12a). Geophysical Research Letters, VOL. 33, L24319.
Brunner, F.K., Coleman, R. and Hirsch, B., A Comparison of
The mean of IWV for each month has calculated Computation Methods for Crustal Strains from
separately and plotted in the Figure 12b. GeodeticMeasurementsTectonophysics, 71: 281-298.1981.
Brunner, F.K., On the Analysis of Geodetic Networks for the
It is observed that for Allahabad and its neighboring Determination of the Incremental Strain Tensor Survey Review,
XXV (192): 56-67.1979
region the water vapor content varies from maximum Feigl, K., R. W. King, and T. H. Jordan, Geodetic measurement of
70Kg/m2 to a minimum of 15 Kg/m2 (Figure 12a). The tectonic deformation in the Santa Maria fold and Thrust Belt
monthly distribution of water vapor is shown in Figure ,California, Journal of Geophysical Research, 95(B3), 26792699,
12b. Water vapor content reaches its maximum during 1990.
King, R. W. and Y. Bock, Documentation for the GAMIT GPS
the rainy season and minimum during winter, as software analysis version 9.9, Massachusetts Institute of
expected. Thus, atmospheric water vapor variation could Technology, Cambridge,1999.
be used as a proxy for short-term weather prediction, if IERS96. 2000. The 1996 IERS Conventions Homepage: the Chapters,
water vapor estimation is done in real time. The long Subroutines, and Associated Information,
http://maia.usno.navy.mil/conventions.html.
term study of water vapor can be used for to develop a .http://igscb.jpl.nasa.gov or
model for the region which can then be used for http://igscb.jpl.nasa.gov/overview/viewindex.html. Accessed July
prediction of short- and medium term climate changes, 2008,
very useful for agriculture based rural India. JPL. 2000a. JPL Navigation Server.,
ftp://navigator.jpl.nasa.gov/pub/ephem/export
JPL. 2000b. GPS Time Series,
http://sideshow.jpl.nasa.gov/mbh/series.html.
JPL. 2000c. JPL’s High-Rate Ephemerides Database,
ftp://sideshow.jpl.nasa.gov/pub/jpligsac/hirate
Kouba, J. 2000., IGS Orbit/Clock Navigation Satellite Clock
Interpolation With No SA.IGSMAIL-2824.
http://igscb.jpl.nasa.gov/mail/igsmail/2000/msg00163.html
Lanyi, G., Tropospheric Delay Effects in Radio Interferometry.
TheTelecommunications and Data Acquisition (TDA),Progress
Report 42-78, Imprint:JPL, California Inst. Tech., 152-159.(1994)
Laurila, S. H., Electronic Surveying and Navigation,545p-(1976).
LAREG. 2000a. ITRF. , http://lareg.ensg.ign.fr/ITRF
LAREG, 2000b. ITRF96, http://lareg.ensg.ign.fr/ITRF/itrf96.html
LAREG, 2000c. ITRF97, http://lareg.ensg.ign.fr/ITRF/ITRF97.
LAREG, 2000d. ITRF2000, A Call for Participation.
http://lareg.ensg.ign.fr/ITRF/ITRF2000.
NGS. 2000. The NGS GPS Orbital Formats,
http://cddisa.gsfc.nasa.gov/dgps/orbits.txt.
NIMA. 2000. SP3 Format Description,
http://164.214.2.59/GandG/sathtml/sp3format.htm
Fig. 11: The integrated water has been calculated at

00:00 hrs, 06:00 hrs, 12 hrs, 18:00 hrs and plot.(b)

ISBN 978-81-921121-0-7
MODELLING HETEROGENEOUS TRAFFIC USING VISSIM
A.K. Maurya
Assistant Professor, Department of Civil Engineering, Indian Institute of technology Guwahati, ASSAM – 781039,
Email:maurya@iitg.ernet.in
S. Sharma
Former U.G. Student, Department of Civil Engineering, Indian Institute of technology Guwahati, ASSAM – 781039
S. Biswas
P.G. Student, Department of Civil Engineering, Indian Institute of technology Guwahati, ASSAM – 781039
ABSTRACT: In India, road traffic is characterized to be heterogeneous. A wide variation in static and
dynamic characteristics of vehicles is observed on the roadway. Vehicles do not follow lane discipline here;
their lateral movement is only restricted by the width of the road. This paper evaluates the potential of
microscopic simulation model VISSIM in modelling Indian traffic. The inherent model of VISSIM incorporates
longitudinal as well as lateral interactions resulted from both roadway and prevailing traffic characteristics. This
paper focuses on uninterrupted traffic flow modelling under Indian traffic scenario. Successful validation
extends this research in: (a) developing macroscopic relationships for various road widths, and (b) effects of
road geometry on driver behaviour, under Indian traffic conditions.
INTRODUCTION
The traffic pattern in developing countries like India RELATED WORK
can be characterized as heterogeneous traffic: a wide This section describes the previous studies regarding
variation in speed, size, operating characteristics, and uninterrupted traffic flow modelling for Indian traffic.
lateral spacing of vehicles is observed. Moreover, Khan and Maini (2000) have recollected previous
motorized and non-motorized modes share the same modelling developments related to heterogeneous
road space. On the other hand, lane discipline is traffic prevailing in developing countries. They note
generally discarded by vehicles on Indian road. Non- that several studies have provided only few details
lane based lateral movement implies a continuous regarding modelling approach used for mixed flow
lateral movement within and between the lanes, and conditions.
overtaking within the same lane if lane width allows
that possibility (Vortisch and Gopalakrishanan, 2008). Chakroborty et al. (2004) have developed a potential
This lateral movement is only restricted by the road field model of driver behaviour in uninterrupted
width. Maurya and Chakroborty (2008) note that lack traffic flow. This model includes certain aspects: the
of lane discipline forces longitudinal as well as lateral driver’s concern for safety and the driver’s urge to
interactions. The above mentioned complexity of reach destination as early as possible. However, this
Indian traffic has become a motivation to conduct this model is computationally intensive which narrows
study. Heterogeneity arising from modes of various down the applicability of this model in simulating
static dynamic characteristics and non-lane based large traffic stream. To overcome the aforesaid
discipline, without clear lane markings on the drawback, Maurya (2007) developed CUTSiM
roadways, poses challenges for modellers. In this (comprehensive unidirectional traffic simulation
regard, a variety of simulation tools are available for model) to model unidirectional and uninterrupted
modelling homogeneous traffic. Homogeneity with traffic streams in computationally efficient and
lane discipline simplifies the modelling and executes realistic manner. Many other researchers also has
it in a well structured fashion. On the contrary, these developed models to simulated Indian heterogeneous
approaches developed for homogeneous traffic do not traffic like (Arasan and Koshy, 2005; Dey et al., 2008;
prove to be effective for heterogeneous traffic flow. Mathew and Radhakrishnan, 2010). VISSIM has been
a potential microscopic simulation tool in developed
For this paper, uninterrupted traffic in India is countries. Literature review yields various traffic
modelled using microscopic simulation model simulation studies in context to developing world.
Verkehr in Staedten Simulation (VISSIM). The Hossain (2004) was able to calibrate VISSIM to satu-
inherent model in VISSIM incorporates longitudinal ration flows, measured at intersection in Dhaka,
as well as lateral interactions resulted from both Bangladesh, with good results. Ahmed (2005)
roadway and prevailing traffic characteristics, which conducted VISSIM calibration study for Khobar and
allows the modelling of heterogeneity with non-lane Dammam, Saudi Arabia, by adjusting the default
based discipline. An understanding of complexity values of the number of observed vehicles, additive
behind modelling is crucial to replicating the and multiplicative parts of desired safety distance,
prevailing traffic situations effectively. The aim of amber signal decision, and distance required in
this paper is to determine how effective VISSIM is in changing lanes. Mathew and Radhakrishnan (2010)
modelling Indian traffic at macroscopic level. Next have calibrated VISSIM for signalized intersection in
section deals with the related work done in the past. India. Results of above simulation studies developed

our faith to conduct Indian traffic stream simulation lane is better than on the current. The final decision is
study using VISSIM. about the possibility to go to the desired lane using
gap acceptance modelling: a driver is willing to accept
The approach presented in this paper builds upon a certain deceleration for a vehicle approaching from
prior works and expands into topic areas where past behind on the desired lane. The value of this
research has suggested it most useful. Full road width deceleration is subjected to user’s calibration
use for lateral movement of vehicles yields new approach and driver’s aggressiveness.
insight into psycho-physical car-following model of
VISSIM. With great understanding of how vehicles Lateral Behavioural Model
move laterally, the car-following parameters may be Lateral behavioural model has incorporated space-
fine-tuned more precisely. oriented movement behaviour. This extension allows
drivers to choose any lateral position within the lane
SIMULATION MODEL OVERVIEW while longitudinal behaviour incorporates more than
VISSIM is a microscopic, time-step based, one leading vehicle on the same lane. The choice of
behavioural traffic flow simulation model. The model the lateral position in a lane follows this simple
contains a psycho-physical car following model for principle: the driver chooses the lateral position (i.e.
longitudinal vehicle movement and a rule-based desired position) where he has maximum longitudinal
algorithm for lateral movements (Fellendorf and time to collision. To move to this position from his
Vortisch, 2001). The following description of inherent current position, driver applies a constant lateral speed
model of VISSIM is entirely credited to Vortisch and depending on vehicles type and on his longitudinal
Gopalakrishanan (2008), as VISSIM manual (PTV, speed.
2009) does not provide much evidence of this nature.
Therefore this paper presents this description as an aid The car-following model is modified to take into
to fellow modellers and to improve the understanding account lateral safety distances when deciding which
of the modelling approach behind VISSIM. of the preceding vehicles are relevant for the
acceleration behaviour. In this model, the lateral
Longitudinal car following model safety distance that a driver wants to keep while
VISSIM uses Weidemann’s psycho-physical car passing another vehicle depends on both vehicle types
following model, first published in 1974 (Wiedemann, and on speed of overtaking vehicle, i.e., on maximum
1974) and further developed in 1991 (Weidemann, of both speeds. To facilitate this concern, a linear
1991). The psycho-physical car following model relationship between lateral safety distance and speed
assumes that the driver behaviour is dependent on the is assumed.
four modes of driving: free driving, where driver
seeks to maintain a constant individual desired speed; Based on current lateral position, the longitudinal
approaching mode, when a driver reduces his speed to model looks at all preceding vehicles, within a user
the speed of a preceding slower car; following mode, defined ‘look ahead distance’ and determines relevant
when a driver follows the preceding car without any leading vehicles. These are all vehicles which the
conscious acceleration or deceleration and he seeks to driver cannot pass at the moment and with its current
maintain a constant desired safety distance; and speed and lateral position. Then for each of the
breaking mode, when the minimum desired safety relevant leading vehicle, an interaction according to
distance is violated and the driver applies medium to psycho-physical car following model is computed.
high deceleration. For each mode, acceleration results The final longitudinal acceleration applied is the
from speed of the vehicle, speed difference, and driver minimum of all acceleration resulted from the
and vehicle characteristics. On reaching a certain individual interactions.
threshold, the driver switches from one mode to
another. The thresholds take into account the human
perception capability and this is reason why the model DEVELOPING UNIDIRECTIONAL UN-
is called psycho-physical. INTERRUPTED TRAFFIC FLOW MODEL
Lane Change Logic Base Model

The lane change logic in VISSIM is based on a series Unidirectional uninterrupted traffic flow is considered
of three decisions. In the first decision, the driver on modelling grounds. A road network, 4 km long, is
decides if he wants to leave the current lane. This is created in VISSIM. Road geometry is assumed to be
the case either for mandatory lane change, when the plane. Table 1 lists the static characteristics of the
driver moves to another lane to make a turn on vehicles used in study. As noted in Table 1, a wide
intersection or in situations when the driver cannot variation in lengths and widths among different
drive at his desired speed on current lane. The second modes of traffic is common in India. Lateral gap
decision compares the situation of the current lane function considers total lateral spacing between
with the situation of the neighbouring lane. This neighbouring vehicles on the road.
decision is positive if the situation on neighbouring

TABLE 1 Static characteristic of vehicles. MODEL CALIBRATION AND MACROSCOPIC
Vehicle type Lengt Widt Lateral gap VALIDATION
Truck 7.3 2.5 (0.76  0.04  v ) Simulation model needs to be calibrated and validated
Car 4.3 1.8
to replicate reality. Model calibration is defined as the
(0.60  0.04  v ) process by which the individual components of the
Motorized 3 wheeler 3.0 1.4 (0.46  0.06  v ) simulation model are adjusted or tuned so the model
Motorized 2 wheeler 1.8 0.6 (0.30  0.06  v) will accurately represent field-measured or observed
Source: Maurya (2007) traffic conditions. The improved base model is
calibrated against the fundamental speed-flow
Table 2 shows the speed characteristics of different relationship observed on a 14.6 m wide highway with
vehicles. Desired speed of vehicles is assumed to be unidirectional and uninterrupted traffic. Field
normally distributed: N(µ, σ2). Depending on vehicle observed speed-flow data is taken from PhD Thesis of
type, the range of desired speed, mean value, and Maurya (2007); this data have been collected on
standard deviation are incorporated in VISSIM Delhi-Gurgaon highway (NH-8) during March, 2007.
accordingly. In absence of field-observed acceleration Traffic at this location was uninterrupted and mainly
and deceleration profile of different vehicles for consists of passenger-cars (≈ 70%) and motorized two
Indian conditions, default values, provided by wheelers (≈ 30%). The highway was 14.6 m wide. As
VISSIM, are used. noted in Maurya (2007), desired speed distribution for
calibration includes: mean speed (65 km/hr);
TABLE 2 Speed characteristics of vehicles. minimum speed (50 km/hr); maximum speed (80
Vehicle type Minimum Maximum Mean Standardkm/hr). Lowens and Machemehal (2006) have
(km/hr) (km/hr) (km/hr) deviationinvestigated the sensitivity of Weidemann’s 99 car-
(km/hr) following model parameters on roadway capacity.
Truck 50 70 60 5 Standstill distance (CC0) and headway time (CC1) are
Car 70 90 80 5 observed to have predominant effect and thus are
more germane in the context of our approach. Only
Motorized 3 wheeler 40 60 50 5
free flow traffic regime is considered from the point
Motorized 2 wheeler 70 90 80 5 of view of calibration. A trial and error analysis
Source: Maurya (2007) followed by sensitivity check yields only a change in

the value of CC1 from 0.9 (default) to 1.5. Keeping
Wiedemann’s 99 car following model is used in this CC0 unchanged, a comparison between field-
study for its potential to model highway traffic observed and simulated speed-flow profile is
together with the use of numerous driver behaviour presented in Figure 1. A close match between field
parameters. Lateral behavioural model is adjusted for observed and simulated speed-flow relationship can
non-lane based lateral movements. “Any” option is be observed from Figure 1.
selected in desired position at free flow. Observe
vehicles on next lane(s) option, along with diamond
shaped queuing, is checked; besides, overtaking on
the same lane is allowed from both sides – left and
right. In minimum lateral distance option, lateral gap
function (see Table 1) is used to calculate the safe
lateral gap for various vehicle types at 0 km/hr and at
50 km/hr.
The time step influences the resulting driver

behaviour, because smaller time steps allow an earlier
and therefore smoother reaction to the driving action
of the preceding vehicle (Fellendorf and Vortisch, Fig. 1 Simulated speed-flow relationships for 14.6 m
2001). Thus we have used time step of 0.1 seconds to wide road (70% passenger cars and 30% motorized
investigate the finer details of lateral behavioural two wheelers) using VISSIM.
model as well as to depict the speed-flow
relationships on aggregate level. SIMULATION RESULTS
Unrestricted lateral movement Fundamental Speed- Flow Relationships

As mentioned before, previous studies access full road As for assessing the impact of road width on capacity
width to implement an unrestricted lateral movement of roadway, we generated speed-flow relationships for
(Arasan and Koshy, 2005; Dey et al., 2008). various road widths: 3.5m, 7.0m, 10.5 m, and 14.0 m
Therefore in VISSIM, roads are assumed as one lane (Figure 2). These relationships are obtained for
but of width equal to entire road width. homogenous traffic stream consisting of passenger-
cars only. Chiefly theses diagrams are obtained for

free-flow regime. Static characteristics of vehicles and The results can be explained by the lack of lane
speed distribution data are taken from Table 1 and 2. discipline on the roadway. It is clear from the figure
that the vehicles’ lateral positions selection is random,
Table 3 presents the estimated impacts of roadway and the vehicles do not appear to follow each other
width on capacity. In general results are intuitive: As within a well defined lane. Vehicles’ lateral positions
road width increases, capacity also increases, but not within a lane appear to be randomly distributed.
at a proportional rate as it does in case of homogenous Therefore it is reflected that macroscopic
traffic with lane discipline. This non linear increment characteristics of heterogeneous traffic stream can be
in our case can be attributed to the existing non-lane modelled using VISSIM by calibrating its parameters
based lateral movement. suitably.
Zoomed view of
circled portion
Fig. 3 Lateral profiles of passenger cars on a 14.6 m

wide road obtained from simulation.
PERSPECTIVE
This paper introduced to the integration of
unrestricted lateral movement approach with car-
following theory. Both are part of the existing psycho-
physical car-following microscopic simulation model
Fig. 2 Simulated Speed-flow relationships for VISSIM. Our empirical findings underpin the
different road widths with passenger cars only. potential of VISSIM in modelling non-lane based
lateral movements. Overall, results suggest that
TABLE 3 Roadway capacity estimates for various VISSIM could produce realistic measurements.
road widths obtained from VISSIM simulation for
traffic comprising passenger-cars only. In future, the first implementation effort will be
directed to the calibration and validation of VISSIM
Road width (m) Capacity at microscopic level with more data collection.
Considering genetic algorithm based optimization will
3.5 1880 – 2080
enhance the realistic measurements. Afterwards,
7.0 2820 – 3310 inclusion of non-motorized modes in our model will
10.5 4470 – 5040 lead us to realistic Indian traffic modelling.
14.0 5810 – 6430
REFERENCES
Ahmed, S.A. (2005) Calibration of VISSIM to the
Investigation of Lateral Behavioural Model of
Traffic Conditions of Khobar and Dammam,
VISSIM
Saudi Arabia, M.S. Thesis, King Fahd University
In Figure 3, finer details of lateral behavioural model
of Petroleum Minerals, Saudi Arabia.
are shown. The two-dimensional reference positions
Arasan,V.T. and Koshy, R.Z. (2005). Methodology
of vehicles, taken at time step of 0.1 sec, are plotted
for modeling highly heterogeneous traffic flow.
against each other. Two-dimensional referencing
includes longitudinal as well as lateral coordinates.

Journal of Transportation Engineering, ASCE, Road Transport. edited by the Commission of the
131(7), 544-551. European Community, DG XIII, Brussels.
Brackstone, M. and McDonald, M. (2000). Car Wiedemannn, R. (1974). Simulation des
following: a historical review. Transportation Straβenverkehrsflusses, (In German)
Research Part F: Traffic Psychology and Schtiftenreihe des Instituts fur Verkehrswesen der
Behaviour, 2, 181–196. Universitat Karlsruhe, Heft 8.
Chakroborty, P., Agrawal, S., and Vasishtha, K.
(2004). Microscopic modeling of driver behavior
in uninterrupted traffic flow. Journal of
Transportation Engineering, ASCE, 130 (4),
438-451.
Dey, P.P., Chandra, S., and Gangopadhyay, S. (2008).
Simulation of mixed traffic flow on two-lane
roads. Journal of Transportation Engineering,
ASCE, 134 (9), 361-369.
Elefteriadou, L., Torbic, D. and Webster, N. (1997).
Development of passenger car equivalents (PCE)
for freeways, two-lane highways and arterials.
Transportation Research Record, 1572, pp. 51-
58.
Fellendorf, M. and Vortisch, P. (2001) Validation of
the microscopic traffic flow model VISSIM in
different real-world situations, Available in the
80th annual meeting of the Transportation
Research Board Compendium of Papers.
Hossain, M. J. (2004). Calibration of microscopic
traffic flow simulation model VISSIM for urban
conditions in Dhaka city. Master thesis,
University of Karlshruhe, Germany.
Khan, S.I. and Maini, P. (2000). Modelling
heterogeneous traffic flow. Transportation
Research Record, Transportation Research
Board, Washington, D.C., 1678, 234–241.
Lownes, N.E. and Machemehl, R.B. (2006). VISSIM:
a multi-parameter sensitivity analysis, In
Proceedings of the 38th Conference on Winter
Simulation (Monterey, California, December 03 -
06, 2006. L. F. Perrone, B. G. Lawson, J. Liu, and
F. P. Wieland, Eds. Winter Simulation
Conference. Winter Simulation Conference,
1406-1413.
Mathew, T.V. and Radhakrishnan, P. (2010).
Calibration of microsimulation models for
nonlane-based heterogeneous traffic at signalized
intersections. Journal of Urban Planning and
Development. ASCE, 136 (1), 59-66.
Maurya, A.K. (2007). Development of a
comprehensive microscopic model for simulation
of large uninterrupted traffic streams without
lane discipline. PhD Thesis, Department of Civil
Engineering, Indian Institute of Technology
Kanpur, India.
PTV (2009). User’s manual, VISSIM 5.20, Planning
Transport Verkehr AG, Karlsruhe, Germany.
Tiwari, G. (2001). Traffic flow and safety: need for
new models in heterogeneous traffic, Injury
Prevention and Control, Eds. D. Mohan & G.
Tiwari, London: Taylor & Francis, 71-88.
Wiedemann, R. (1991). Modelling of RTI-Elements on
multi-lane roads. In: Advanced Telematics in

ISBN 978-81-921121-0-7
RECENT ADVANCES IN HYDROLOGICAL ANALYSIS, MODELING,

AND DECISION-MAKING
R. D. Singh
Director, National Institute of Hydrology, Roorkee
M. K. Goel
Scientist “F”, National Institute of Hydrology, Roorkee
ABSTRACT: Water is essential for sustaining life on Earth. However, in view of the increasing demands for
various competing sectors such as urbanization, industrialization, growing population, and increasing living
standards, the finite freshwater resources are getting vulnerable to deficit in terms of quantity and quality.
Further, because of the high time and space variability in rainfall in this country, optimum conservation and
management of the available water resources for different purposes is the need of the hour.
In the past few decades, the science of hydrology has made significant progress in various dimensions.
Hydrological instrumentation, data collection, processing, storage and retrieval have been computerized to a
considerable extent. Spatial database provided through remote sensing satellites and spatial analyses through
Geographic Information System (GIS) are finding applications in a number of modeling studies. Nuclear
techniques are finding applications in a number of hydrological applications related to identification of recharge
areas, groundwater seepage and its movement and identification of flow components from different sources in a
stream. Further, Decision Support Systems (DSS) are being developed at the scale of river basins which can
help the decision makers in analyzing various scenarios for rational surface and groundwater development and
management without going deep into the modeling intricacies. This paper discusses some of these recent
developments in various facets of hydrology and water resources.
INTRODUCTION
Of all renewable Earth resources, water has a accounts for about 80% of domestic water requirement
unique place. It is essential for sustaining all forms and more than 45% of the total irrigation in the
of life, food production, economic development, country.
and for general well-being of the society. It is truly Hydrology is the science of occurrence, distribution,
a unique gift of nature to mankind as it is difficult and movement of water in the atmosphere, at the
to de-pollute, expensive to transport and impossible surface of Earth and below the surface. With time, the
to substitute for most of its uses. Water is also one advances in hydrological techniques have grown
of the most manageable of the natural resources as manifold. Automation in Hydrological and Hydro-
it is capable of diversion, transport, storage, and meteorological instrumentation provides adequate and
recycling. All these properties impart great utility accurate data at desired time interval for the
of water for human beings. application of advanced modeling tools to address
The water resources of our country (both, surface various water resources problems. User interactive
and groundwater) play a major role in agriculture, Software for Digital hydrological data storage,
hydropower generation, livestock production, processing and retrieval are being developed and
industrial activities, forestry, fisheries, navigation, extensively used for data base management. Spatial
recreational activities etc. In the planning and database provided through remote sensing satellites
operation of water resources, our National Water and spatial analyses through Remote Sensing and
Policy broadly allocates priorities as: (i) drinking Geographic Information System (GIS) are important
water, (ii) irrigation, (iii) hydropower, (iv) ecology, tools for the development of spatial decision support
(v) agro-industries and non-agricultural industries, system (SDSS) for spatial data base management and
and (vi) navigation. India receives annual analysis. These tools provide important input for the
precipitation of about 4000 km3, including modeling studies. The outcomes of such studies are
snowfall, out of which monsoon rainfall is of the being utilized for addressing the number of water
order of 3000 km3. India is gifted with a river resources management issues. There have been
system comprising more than 20 major rivers with significant developments and applications of Soft
several tributaries, most of which are perennial. computing techniques, such as Artificial Neural
The rivers like Ganges, Brahmaputra and Indus Network (ANN) and Fuzzy Logic etc, in the field of
originate from Himalayas and carry water hydrology and water resources. Nuclear techniques are
throughout the year. More than 50% of water finding applications in a number of hydrological
resources of India are located in various tributaries applications which include identification of recharge
of these river systems [Lal (2001)]. Apart from the areas, groundwater seepage and its movement and
water available in the various rivers of the country, identification of flow components from different
the groundwater is also an important source of sources in a stream. Further, Decision Support Systems
water for drinking, irrigation, industrial uses, etc. It (DSS) are being developed at the scale of river basins

which can help the decision makers in analyzing required for the regional stations and early warning
various scenarios for rational surface and stations are transmitted through satellites. These
groundwater development and management instruments are solar power-based
based system and runs on
without going deep into the modeling intricacies. the battery which is charged by the sunlight through a
The present paper
aper gives a brief overview of recent small solar panel. In India, India Meteorological
developments in hydrological modeling and Department
partment (IMD) has installed a number of state-of-
state
analysis. art AWS all over the country and efforts are
continuing. To improve meteorological services, it is
ADVANCES IN HYDROLOGIC also planned to set up about 1000 AWS through the
INSTRUMENTATION expansion programmes of Indian an Space Research
The planning, development and operation of water Organisation (ISRO) and Snow and Avalanche Study
resources projects is solely dependent upon the Establishment (SASE).
hydrological data. For economic ic and optimum Stream discharge is the volume of water passing a
utilization, planning, design and development of cross-section
section per unit of time and is generally
water resources, determination of the extent and expressed as cubic meter per second (cumec).
availability of surface and groundwater is the first Discharge is simply velocity times the cro
cross-sectional
requisite and this in turn requires readily area. Cross-sectional
sectional area is determined by stretching a
accessible, reliable and adequate observation data tape across the channel to measure distance at the
on the elements of hydrological cycle and related cross-section
section locations where depth is measured with a
factors. Hydrometry is concerned with measuring calibrated rod. In large rivers, the cross-section
cross is
and observing elements of the hydrological cycle. covered on a moving boat and nd water depth is measured
Generally, the main meteorological variables of with echo-depth recorder. Automatic and digital water
interest include maximum and minimum level recorders are getting due recognition for the
temperature, dry and wet bulb temperature, rainfall, measurement of river stages. A radar adar water level
wind velocity and direction, sunshine, albedo, measuring instrument (Fig. 2)
evaporation, soil moisture etc. Main hydrological
variables of interest include water level in the
rivers and corresponding discharge. These
variables are generally measured with conventional
instruments like thermometers, raingauges, wind
vanes and anemometers, sunshine duration
recorders and pyranometers, pan evaporimeters etc.
WMO (1998) has listed various instruments and
their observation methods. Fig. 2 View of radar water level recorde
recorder
In the recent past, automatic weather stations
(AWS) have been introduced for the repetitive using radar pulses to measure levels without direct
recording of hydro-meteorological variables at the contact with water. It can bee located on a bridge, pier
desired frequency and store their values in a data or any structure over the water’s surface which, if
logger for downloading and use. These instruments connected to a logging sensor, can record data at user
user-
(shown in Fig. 1) specified interval. Flow velocity is generally measured
either with a float or current meter which again are
available in a number of varieties. Digital velocity
meters have also been developed for measuring
average flow velocities in rivers and streams.
SOFTWARE FOR HYDROLOGICAL DATA

ENTRY & PROCESSING
Advances in hydrology are dependent on good, reliable
and continuous measurements of hydrological
Fig. 1: A view of AWS
variables. Hydro-meteorological
meteorological data collected from
various sources/instruments are generally raw and
record measurements like wind speed and
cannot as such be used
sed for most hydrological analysis.
direction, humidity, temperature, sunshine,
Such data are first computerized and then subjected to
atmospheric pressure, solar radiation, rain gauge,
various kinds of tests for ensuring the quality. Data
soil moisture and soil temperature. These AWS are
processing is a broad term covering all activities from
of paramount importance in uninhabited, remote,
receiving records of observed raw data to making them
inaccessible areas for data related to weather
available in a usable form. The raw data are in a
forecasting and cyclone forecasting etc. Automatic
variety of formats such as hand-written
written records, charts
weather stations send weather data through
and digital records. Raw data as observed
served and recorded
satellites to the base stations and processed data
may contain many gaps and inconsistencies. These raw
RECENT ADVANCES IN CIVIL ENGINEERING

ENGINEERING-2011 649
data are passed through a series of operations, entry personnel for entering and preliminarily
typically: data entry, making necessary validation validating all types of meteorological and surface
checks, filling missing values in a data series, water quality and quantity data. The main screen of
processing of raw data to estimate required SWDES software is presented in Fig. 3.
variables, compilation of data in different forms
and analysis of data for commonly required HYMOS is an information system for storage,
statistics etc. The rapid advances in computer processing and analysis of hydrological and
technology, in hardware speed of operation and environmental data. It combines an efficient database
data storage capacity as well as the capability of structure with powerful tools for data entry, validation,
hydrological software has greatly simplified the completion, analysis, retrieval and presentation. The
management of large quantities of hydrological wide variety of data processing and analysis features
data and has rendered obsolete those time- make HYMOS very suitable for typical project
consuming manual methods which were formerly application in water-related studies. HYMOS has been
the norm. Computer-based hydrological developed into a flexible and multi-functional
information systems offer various advantages viz., hydrological information system. A typical window of
to permit and promote the standardization of HYMOS software is presented in Fig. 4.
processing, validation and reporting procedures; to HYMOS makes use of a data structure, part of which is
handle a very large amounts of data and users can open, which ensures convenient interaction and data
be provided with data in the required tabular or transfer with other existing databases (such as SWDES)
graphical format. and modeling systems. Being part of DELFT TOOLS,
Surface Water Data Entry System (SWDES) has HYMOS is fully integrated with other software
been developed under Hydrology Project - I and packages of WL | DELFT HYDRAULICS like
introduced in 1999. The prime objective of this SOBEK, DELWAQ and RIBASIM. The HYMOS
data entry software is to provide sufficient facilities database is time series oriented with common facilities
for entering all types of meteorological and for spatial analysis. In combination with a
hydrological quality and quantity data. The Geographical Information System (GIS) for
software provides suitable facilities for entry of comprehensive geographical data analysis, it covers all
rainfall, water level, stage-discharge, water quality, data storage and processing requirements for planning,
sediment and climatic data as observed by surface design and operation of water management systems.
water agencies. The data so entered using this Exchange of data between HYMOS and most common
software is subjected to essential data entry checks GIS systems is easy and user friendly.
so that the possibility of committing mistakes is
minimized. The graphical interface so provided SOFT-COMPUTING TECHNIQUES FOR
gives added advantage of visual means of HYDROLOGICAL MODELING
validating the entered data. The software is used for Soft computing techniques are based on the
entering the historical and current data, available information processing in biological systems. The
with various agencies in huge volumes. complex biological information processing system
enables the human beings to survive with
accomplishing tasks like recognition of surrounding,
making prediction, planning, and acting accordingly.
Human type information processing involves both
logical and intuitive information processing.
Conventional computer systems are good for the
former, but their capability for the later is far behind
that of human beings. For a computing system to have
Fig. 3: A window showing main switchboard of human like information processing facility, it should
SWDES be flexible enough to support three features: openness,
robustness, and real-time processing. Openness of a
system is its ability to adapt or extend on its own to
cope with changes encountered in the real world.
Robustness of a system means its stability and
tolerability when confronted with distorted,
incomplete, or imprecise information. The real-time
characteristic implies the ability of the system to react
within a reasonable time in response to an event.
Information processing systems with all these three
Fig. 4: A typical window of HYMOS data characteristics are known as real world computing
processing software systems (Chaudhuri and Chattopadhyay, 2005a). A
real-world system should be capable of distributed
This software provides a front-end for the data representation of information, massively parallel

processing, learning, and self-organization in order handle the concept of partial truth which implies that
to achieve flexibility in information processing. the truth takes a value between a completely true value
Thus, Soft Computing can be viewed as the key and a completely false value.
ingredient of real systems. Several authors (Zhang Soft computing techniques have been
and Scofield, 1994; Zhang and Knoll, 2001; recognized as attractive alternatives to the standard,
Chaudhuri and Chattopadhyay, 2005 a, b) have well established “hard computing” paradigms.
discussed the potential of soft computing Traditional hard computing methods are often too
techniques in solving real-world problems. cumbersome for today’s problems. They always
Soft Computing has three basic components, require a precisely stated analytical model and often a
namely, Artificial Neural Network, Fuzzy Logic, lot of computational time. Soft computing techniques,
and Genetic Algorithm. In recent times, two other which emphasize gains in understanding system
methods have been incorporated. They are Rough behaviour in exchange for unnecessary precision, have
Set Theory and Ampliative Reasoning. Artificial proved to be important practical tools for many
Neural Network (ANN) is useful in the situations, contemporary problems. ANNs and FLMs are
where underlying processes / relationships may universal approximators of any multivariate function
display chaotic properties. ANN does not require because they can be used for modeling highly
any prior knowledge of the system under nonlinear, unknown, or partially known complex
consideration and are well suited to model dynamic systems or processes. Genetic Algorithm and Particle
systems on a real-time basis (Maqsood et al., Swarm Optimization Techniques have emerged as
2002). Fuzzy Logic is another Soft Computing potential and robust optimization tools in recent years.
technique necessary for analyzing complex Ever increasing demand for the supply of fresh water
systems, especially where the data structure is to the various sectors of human needs has given rise to
characterized by several linguistic parameters. the problem of optimal management of water resources
Another component of soft computing techniques on all parts of the world, more so, in a developing
is genetic algorithm. This algorithm is based upon country like India, where the distribution of the water
the concept of evolution and “survival of the resource is highly uneven both in space and time. For
fittest”. Representation of neurons as artificial efficient management of water resources, prediction of
neural network structure for solving real-life various hydrological events such as rainfall-runoff
problems is shown in Fig. 5. correlation, forecasting of inflow into a reservoir,
forecasting of rainfall, forecasting of evaporation,
forecasting of maximum flood and optimum reservoir
operation policy etc. are required. Soft computing
techniques are very effective for forecast of complex
system in water resources engineering. A number of
studies based on ANN have been reported in
hydrological literature. Modeling in hydrology and
Fig. 5: Representation of neurons as artificial water resources using fuzzy logic is a relatively new
neural network area of research although last decade has witnessed
Scientific and engineering community has acquired quite a few studies mainly in reservoir operation and
extensive experience in developing and using soft water resources management. In other important areas
computing techniques during last few decades. Due like stage-discharge-sediment relationship, rainfall-
to their strong learning, cognitive ability and good runoff modeling and hydrological forecasting etc., the
tolerance of uncertainty and imprecision, soft fuzzy logic applications have been proved [Lohani et
computing techniques have emerged as potential al. (2005, 2007a)].
computing techniques. These methods offer several
advantages over conventional methods. Neural APPLICATIONS OF ISOTOPE TECHNOLOGY
network technology has offered many promising IN HYDROLOGY
results in the field of hydrology and water Isotope methods were introduced into catchment
resources simulation. Fuzzy logic is another soft hydrology research in the 1960s as complementary
computing technique, which has very recently tools to conventional hydrologic methods for
received attention in hydrology. It is a paradigm for addressing questions of where water goes when it
an alternative design methodology which can be rains, what pathways it takes to the stream and how
applied in developing both linear and non-linear long water resides in the catchment (McDonnell,
systems. The discipline of fuzzy logic, fuzzy 2003). Despite slow incorporation into routine research
systems, and fuzzy modeling has witnessed its applications, the last decade has seen a rapid increase
greatest success in real-world automatic control in isotope-based catchment studies. These have been
applications, including subway control, mainly carried out in small well-instrumented
autonomous robot navigation, auto-focus cameras, experimental catchments, of the order of 0.01 to 100
image analysis, and diagnosis systems. In the fuzzy km2 and located typically in headwater areas (Buttle,
logic approach the Boolean logic is extended to 1998). In contrast, little has been done in terms of

application and transfer of these concepts and understand and identify the problem for suggesting the
methodologies to large (100s to 1000s km2 river possible remedial solutions. Major applications of
basins) and less instrumented basins. However, isotopes in hydrology are:
much potential is yet to be realized in terms of how 1. dating of waters and sediments,
isotope information may be used to calibrate and 2. estimation of soil erosion in watersheds/
test rainfall-runoff models and to aid in the catchments,
sustainable water resources management. 3. estimation of groundwater flow velocity and
Isotope hydrology deals with the application of direction,
isotopes as tracers in water resources development 4. discharge measurement of mountainous rivers,
and management. Applications of isotopes in 5. seepage/leakage from water bodies,
hydrology and water resources has increased in 6. hydrograph separation (shown in Fig. 6),
recent years due to increase of problems in water 7. sources of salinity and sea water intrusion,
sector, particularly depleting groundwater quantity, 8. identification of recharge sources and recharge
deterioration in water quality and many other zones for shallow and deeper aquifers,
unpredictable natural events that affect the 9. groundwater surface water interaction,
hydrological cycle. The environmental isotopes are 10. snow and glacial studies, and
the naturally occurring isotopes of elements found 11. climate change studies.
in abundance in our environment such as H, C, N,
O and S. These are principal elements of
hydrological, geological and biological systems.
The stable isotopes of these elements serve as
tracers of water, carbon, nutrient and solute
cycling. Radioactive environmental isotopes such
as C-14 and H-3 can be used to estimate
age/circulation of groundwater.
A large variety of environmental stable and Snow melt water = 56%, Glacier Melt water=41%,
radioactive isotopes (e.g., 2H, 3H, 3He, 4He, 6Li, Rainfall contribution=3%
11
B, 13C, 14C, 15N, 18O, 34S, 36Cl, 37Cl, 81Br, 81Kr,
87
Sr, 129I, 137Cs 210Pb etc.) are employed for Fig. 6: Discharge separation at Gomukh using nuclear
hydrological studies. However, the stable isotopes techniques
have the distinct advantage over injected (artificial)
tracers (3H, 46Sc, 60Co, 82Br, 131I, 198Au, etc.) in that The National Institute of Hydrology, Roorkee has a
they facilitate the study of various hydrological well equipped Nuclear Hydrology Laboratory for
processes on a much larger temporal and spatial measurement of stable and radio isotopes. The Institute
scale through their natural distribution in a system. has carried out various studies using the isotopic
Thus, environmental isotope methodologies are techniques in the recent past for example, discharge
unique in regional studies of water resources to measurement of Teesta River, hydrograph separation
obtain time and space integrated characteristics of Bhagirathi River at Devprayag, identification of
whereas the artificial tracers generally are effective leakage from Tehri dams, estimation of water balance
for site–specific, local applications. Earlier, of Nainital Lake, estimation of sedimentation rate and
artificially produced radioactive isotopes were pattern of major lakes of North India, river -
being used with a very limited scope of their use as groundwater interaction along Yamuna river at Delhi,
tracers only, but now environmental isotopes, both estimation of monsoon contribution to river Bhagirathi
radioactive and stable isotopes, are widely used for at Gaumukh [Rai et al. (2009)], identification of
a variety of applications with no fear of health recharge zones and source of springs in mountainous
hazards. Environmental isotopes are freely region and groundwater in plain region, and
available in the atmosphere and automatically characterization of Indian precipitation [Kumar et al.
injected in to the hydrologic cycle. Therefore, the (2010)] etc.
users have neither to purchase these isotopes nor to
inject them in hydrological system. REMOTE SENSING & GIS APPLICATIONS IN
With the development of sophisticated and HYDROLOGY
automated instrumentation for very precise isotopic Availability of the right information at the right time to
measurements, new approaches are being the right person and at the right cost is a crucial factor
developed and new applications and tools are being in decision making. For any water resources
added to the isotope toolbox. Now Isotope development project, information on existing land
techniques can effectively be used in carrying out use/cover, cropping pattern, soil characteristics and
various hydrological studies. Specifically, the their spatial distribution is a prerequisite. These data
Isotope can be used for tracing the source, and information need to be carefully, effectively and
movement, and pollution of ground water. Isotopes efficiently analyzed, stored and retrieved whenever
can be used for various investigations in order to required. The conventional procedure of storing,

handling and updating records is slow, un- [Govardhan (1993), Bastiaanssen (1998, 2000), and
systematic, occupies large space and requires huge Menenti (2000)]. Advances in remote sensing
manpower. Further, such records are rarely technology have led to considerable saving in time and
updated. With the advent of remote sensing and money spent in data collection and input. There are
GIS tools, it has become possible to prepare the variety of available packages for image processing of
dynamic resource maps at various levels of satellite data (Bos et al. 2001), ranging from low-cost
confidence. Remote sensing provides multi-spectral and simple (MapInfo, IDRISI, ER-Mapper, PCI) to
and multi-temporal synoptic coverage for the area complete and simple (ILWIS, GRASS) and
of interest, while the GIS provide the facilities to professional and expensive (Erdas-Imagine, IDL-
integrate and analyze multi-disciplinary data. For Envi). In last few years, remote sensing technology has
natural resource management to be socially been increasingly used for evolving an objective,
acceptable, decision-making must be a public and cheaper and faster methodology for identification of
participatory process. The capability of GIS to landuse, soil type, snow cover areas, irrigated areas
express the information in maps and pictures, can and cropping pattern, crop water stress and crop yield,
make the general public more informed and can crop evapo-transpiration, soil moisture and salinity.
involve them in decision making activities. Remote sensing data is being increasingly used for
Remote sensing implies sensing from a distance. rainfall-runoff modelling, reservoir sedimentation
Each satellite image is considered as representing a assessment, watershed prioritization, flood forecasting,
spatial distribution of energy coming from the flood risk zoning and flood management. The scope of
earth's surface in one or several wavelength ranges remotely sensed data can be significantly enhanced
of the electromagnetic spectrum. By using multi- through integration with other types and sources of
spectral satellite data suitably, different ground data in a GIS environment. Potential applications of
features can be differentiated and thematic maps GIS in hydrological studies include topographical
can be prepared. Basic principles of remote sensing analysis (creation of DEM, slope, aspect, drainage
are depicted in Fig. 7. network, and contributing basin maps), interpolation of
point information (rainfall or groundwater level) to
generate spatially distributed maps of groundwater
depth and rainfall, and integration of maps of various
variables and different sources for water logging
assessment, soil/land suitability classification, water
resources management and participatory decision-
making (by using the GIS to express information as
maps and pictures to make the general public more
informed and involve them in decision making.
DEVELOPMENT OF DECISION SUPPORT

SYSTEMS (DSS)
With time, we are becoming increasingly aware of the
fact that our water supplies are limited both in quantity
and quality. Water has multiple and often competing
uses and water systems are interrelated with other
physical and socio-economic systems. This precious
resource has competitors that need water in one way or
the other as a result of which it often becomes
challenging in space and time to fully satisfy the needs
of all competitors. Therefore, effective management of
Fig. 7: Illustration of principles of remote sensing water and related environment in a river basin requires
Satellite remote sensing measurements have been an integrated and coordinated planning within the
used to provide regular information on agricultural basin. Further, introduction of water resources projects
and hydrological conditions of the land surface for (reservoirs, irrigation canals etc.) in a river basin sets
vast areas and have played vital role in developing in new hydrological regime in the basin with revised
and monitoring water management plans. Vastness conditions of surface and groundwater availability and
of the river basins, time and manpower constraints use. In addition, a number of interacting water-related
in data collection and seasonal changes in the variables are involved in a river basin such as rainfall,
information require fast inventory of study areas. In evapo-transpiration, land use and land cover,
all these circumstances, remote sensing can be landscape, soil types and their water holding and
looked upon as an aid in planning and decision- transmission characteristics, prevailing cropping
making (Vidal, 2000). The usefulness of remote pattern and their water demands, irrigated/rain fed
sensing techniques in various hydrological studies area, groundwater potential/development and its
have been demonstrated in various studies temporal availability, urban/rural area and water

demands, industrial development and water (developed by Delft Hydraulics, The Netherlands),
demands, water resources structures and their MIKE BASIN (developed by DHI, Denmark), and
characteristics, hydro-meteorological network and SWAT (Arnold et al., 1994; Neitsch et al., 2002) are
spatial and temporal data availability etc. For some other DSSs that have been used world-wide to
preparing any integrated river basin plan, it is develop basin hydrologic plans and manage water
imperative to gather the details about these resources efficiently.
interacting variables and to establish their linkages Under the Hydrology Project – II, a DSS (Planning) is
for evaluation of their impact on the basin water being developed at NIH for integrated water resources
resources. planning and management. Five major components of
A Decision Support System (DSS) as an interactive the DSS are: surface water planning, integrated
computer-based support system that helps decision operation of reservoirs, conjunctive use of surface and
makers utilize data and models to solve ground water, drought monitoring, assessment, and
unstructured problems. DSS for integrated water management, management of surface water and
resources planning and management can be very groundwater quality. The DSS applications are being
important tools that are helpful for fast developed for the nine Peninsular States of India. In
computations, easy data management and drawing addition, a DSS (Real-Time) is being developed at
conclusions about alternative water policies. Bhakhra-Beas Management Board (BBMB) to assist
General architecture of a DSS is shown in Fig. 8. the decision makers in real-time flood forecasting and
management of floods from the reservoirs. This DSS is
being customized for the Satluj river basin. Both the
DSS are being developed with the assistance of DHI,
Denmark as consultants.
CONCLUSION
With time, we are becoming more increasingly aware
of the fact that our water supplies are limited, both in
quantity and quality. In the present age, hydrology and
water-related studies are given due recognition as all
developmental activities depend on water. However,
Fig. 8: General architecture of a DSS
with technological innovations and emergence of high
speed digital techniques, hydrological observation,
As computing speed and ease is becoming more
analysis, and management is rapidly advancing with
powerful, more complex yet more comprehensive
time. Hydrological instrumentation has improved
DSS are being developed. Such computer models
substantially with the advent of AWS and GPRS
as MODSIM, RIBASIM, AQUATOOL, TERRA,
technology while software have been developed for
MIKE BASIN, RiverWare, WaterWare, and
storage, analysis, processing, and dissemination of
BHIWA etc. have been used as Decision Support
hydrological data. More rigorous hydrological models
System (DSS) for integrated river basin planning
are being used in conjunction with remote sensing and
and management. MODSIM is widely referred
GIS techniques for detailed representation of real
DSS for conjunctive management of water
systems. Soft computing techniques are being
resources under the prior appropriation water right
integrated with the models for simulating the
(Fredericks et al., 1998). It is designed as a DSS for
hydrological response of water systems. Nuclear
developing improved basin-wise and regional
techniques are finding applications in a number of
strategies for river administration, drought
hydrological applications. Decision Support Systems
contingency planning, evaluating groundwater
(DSS) are being developed at the scale of river basins
exchange programs, managing recharge and
to help the decision makers in analyzing various
augmentation projects and resolving conflicts
scenarios for integrated surface and groundwater
between urban, agricultural and environmental
development and management without going deep into
concerns. TERRA (TVA Environment and River
the modeling intricacies. All these developments are
Resource Aid) is a DSS developed for the
intended to help water resources professionals in
management of river, reservoir and power
formulating adaptive strategies to cope with future
resources of Tennessee Valley Authority (TVA)
water stress situations in view of rising demands and
and the Electric Power Research Institute (EPRI)
altered water supply pattern as a result of climate
(Reitsma, et al., 1996). RIVER WARE is a DSS,
change impact.
developed at the University of Colorado, for a
general river basin modeling for wide range of
applications (Zagona, 1998) using three REFERENCES
Andreu, J., J. Capilla, and E. Sanchis (1996).
fundamental solution methods: simple simulation,
“AQUATOOL: a generalized decision support
rule-based simulation and optimization. Similarly,
system for water-resources planning and
AQUATOOL (Andreu, et al., 1996), WATER
management”, Journal of Hydrology, 177, pp. 269-
WARE (Jamieson and Fedra, 1996), RIBASIM

291. characteristics of Indian precipitation”, WRR, Vol.
Arnold, J. G., J. R. Williams, R. Srinivasan, K. W. 46, W12548, doi:10.1029/2009 WR008532.
King, R. H. Griggs (1994). “SWAT, Soil and Lal, M. (2001). “Climate change – Implications for
Water Assessment Tool”, USDA, Agriculture India’s water resources”, J. India Water Res. Soc.,
Research Service, Temple, TX 76502. 2001, 21, 101–119.
Bastiaanssen, W.G.M. (1998). “Remote sensing in Lohani, A. K., N. K. Goel and K. K. S. Bhatia (2005).
water resources management: The state of the “Real time flood forecasting using fuzzy logic”, In:
art”, International Water Management Institute, Hydrological Perspectives for Sustainable
Colombo, Sri Lanka. Development, Volume I, Allied Publishers Pvt. Ltd.,
Bastiaanssen, W.G.M., D. J. Molden and I. W. New Delhi, 168-176.
Makin (2000). “Remote sensing in irrigated Lohani A. K., N. K. Goel and K. K. S. Bhatia (2007a).
agriculture: examples from research and possible “Deriving stage-discharge-sediment concentration
applications”, Agricultural Water Management, relationships using fuzzy logic”, Journal of
46: 137-155. Hydrological Sciences, 52(4): 793–807.
Bos, M.G., S. A. Dayem, W. G. M. Bastiaanssen, Maqsood, I., R. K. Muhammad and A. Abraham
and A. Vidal (2001). “Remote sensing for water (2002). “Neurocomputing Based Canadian Weather
management: The drainage component”, Report Analysis”, Computational Intelligence and
of expert consultation organised by ICID, Applications, Dynamic Publishers Inc., USA, 39-44.
IPTRID, ILRI, World Watch and the World McDonnell, I. J. (2003). “Where does water go when it
bank, May 2001, Ede-Wageningen. rains? Moving beyond the variable source area
Buttle, J. M. (1998). “Fundamentals of small concept of rainfall-runoff response”, Hydrological
catchment hydrology”, In: Isotope Tracers in Processes, 1869- 1875.
Catchment Hydrology (I. McDonnell, C. Menenti, M. (2000). “Irrigation and Drainage”, In:
Kendall, Eds) Elsevier, Amsterdam, 1-49. Remote Sensing in Hydrology and Water Resources,
Chaudhuri, S. and S. Chattopadhyay (2005a). Edited by G.A. Shultz and E.T. Engman, Springer,
“Neuro-Computing Based Short Range New York.
Prediction of Some Meteorological Parameters Rai, S. P., Bhishm Kumar and Pratap Singh (2009).
during Premonsoon Season”, Soft Computing - A “Estimation of contribution of southwest monsoon
Fusion of Foundations, Methodologies and rain to Bhagirathi River near Gaumukh, Western
Applications, 9, 5, pp. 349 – 354. Himalayas, India, using oxygen-18 isotope”, Current
Chaudhuri, S. and S. Chattopadhyay (2005b). science, VOL. 97, NO. 2, 240-145.
“Prediction of Severe Thunderstorm with Reitsma, R. F., Jr. P. Ostrowski and S. C. Wehrend
Minimal A-Priori Knowledge”, Advances in (1994). "Geographically distributed decision
Complex Systems, 8, 1, pp. 75 – 86. support: the Tennessee Valley Authority (TVA)
Delft Hydraulics (August 2004). “RIBASIM”, TERRA system", In: Water Policy and Management:
http://www.wldelft.nl/soft/ribasim/int/index.html Solving the Problems. Proceedings of the 21st
DHI (August 2004). “MIKE BASIN”, Annual Conference, ASCE, Fontane, D. G. and
http://www.dhisoftware.com/mikebasin Tuvel, H. N. editors. Denver, CO. 311-314.
Fredericks, J. W., J. W. Labadie, and J. M. Vidal, A. and A. Baqri (1995). “Management of large
Altenhofen (1998). “Decision support system for irrigation projects in Morocco”, In: Use of remote
conjunctive stream-aquifer management”, sensing techniques in irrigation and drainage, Vidal,
Journal of Water Resources Planning and A. and Sagardoy, J.A. (Eds.), FAO, Water Report 4.
Management, ASCE, 124(2), pp. 69-78. Rome, Italy.
Govardhan, V. (1993). “Remote sensing and water WMO (1998). “Instruments and Observing Methods”,
management in command areas”, International Report No. 68.
Book Distributing Co., Lucknow, India. Zagona, E. A., T. J. Fulp, H. M. Goranflo and R. M.
International Commission on Irrigation and Shane (1998). "RiverWare: a general river and
Drainage (2004). “Basin Wide Holistic reservoir modeling environment.", In: Proceedings
Integrated Water Assessment (BHIWA) Model – of the First Federal Interagency Hydrologic
User Manual”, ICID, New Delhi. Modeling Conference. Las Vegas, NV. 5- 113-120.
ITC. (2007). “ILWIS Open”, 52 North, Zhang, J. and A. Knoll (2001). “Neuro Fuzzy
International Institute for Aerospace Survey and Modeling of Time Series”, In: Soft Computing for
Earth Sciences, Enschede, The Netherlands. Risk Evaluation and Management, PhysicaVerlag,
Jamieson, D. G. and K. Fedra (1996). "The Heidelberg, New York, 140-154.
'WaterWare' decision-support system for river- Zhang, M. and A. R. Scofield (1994). “Artificial
basin planning. 1. Conceptual design", Journal Neural Network techniques for estimating rainfall
of Hydrology, 177, 163-175. and recognizing cloud merger from satellite data”,
Kumar, Bhishm, S. P. Rai, U. Saravana Kumar, S. International Journal of Remote Sensing, 16, 3241 –
K. Verma and Pankaj Garg (2010). “Isotopic 3262.

ISBN 978-81-921121-0-7
STATIC AND DYNAMIC STABILITY OF PLATES AND SHELLS
L.S. Ramachandra
Professor, Department of Civil Engineering, IIT Kharagpur-721302
ABSTRACT: The buckling of different elastic structures is the rich source of interesting and
important problems in nonlinear solid mechanics. In this study, linear and nonlinear static and
dynamic buckling of plates and shells are discussed. Historical development of the subject is
outlined. Various methods adopted to determine the critical condition for the elastic
structures subjected to external loading are discussed. Few results on the static and dynamic
buckling of plates and cylindrical shell panels are given. The characteristics of nonlinear
dynamic instability of plates are clearly brought out by dynamic response analysis.
INTRODUCTION
Due to neglect of stability in design, many the stable equilibrium configuration. Hence, if were to
structures have collapsed in the past. The following disturb the ball, it will return to its original position.
collapses illustrate the importance of considering There is a third possibility. Take for an example a ball
structural stability in design: the collapse of lying on a plane. If disturbed, it will tend to remain in
Tacoma Narrows Bridge (1940) due to the disturbed configuration. This represents a neutral
aerodynamic instability, collapse of Quebec bridge equilibrium configuration. Hence, the equilibrium
(1907) due to buckling of one latticed diagonal of a configuration of the system tells us nothing about the
truss having insufficient shear stiffness, failure of stability of those equilibrium configurations.
space frame at Hartford civic center as the main
chords had insufficient buckling strength, collapse
of Ferrybridge cooling tower, failure of aircraft
wings, failure of numerous bridges, tanks, silos, (a) Unstable equilibrium
processing containments, chimneys etc. Depending
on the type of loading, the elastic structure may
respond in linear/nonlinear static bending, stability, (b) Stable equilibrium
vibration, dynamic stability. There are many books
dealing with each of this topic. Here an attempt has (c) Neutral equilibrium
been made to introduce linear and nonlinear static Fig. 1 Equilibrium positions
and dynamic stability of plates and shells. The
scope of this paper is limited and only a few salient METHODS ADOPTED IN STABILITY
features of buckling are covered here. ANALYSIS
Failure of structural elements is broadly classified
The word ‘stability’ means the capability of elastic
in to two categories: material failure and structural
structure to regain its original state of equilibrium,
instability. The material failure can be predicted by
when it is being subjected to small external
solving the equilibrium equations of structures. To
disturbances. The structural elements can become
model this type of phenomenon, the equilibrium
unstable in many ways depending on the structural
equations are written based on the initial
geometry and load characteristics. Numbers of
undeformed configuration of the structure. As
different approaches have been adopted to determine
against this, the stability failures are predicted from
the critical condition for the elastic structures subjected
the equations written based on the deformed
to external load. In the oldest method of determining
configuration of the structure. Since the deformed
the stability of the structure one tries to find out
configurations are not known in advance, even the
whether the structure admits two or more different but
linear stability problem becomes nonlinear.
infinitesimally close equilibrium states (Ziegler 1968,
An example often given to make stability
Simitses and Hodges, 2006). Consider for example a
concept clear is shown in Fig. 1. The system
slender straight hinged-hinged column subjected to an
consists of ball of weight W resting at different
axial load P. Here the column is assumed to be
positions. Points of zero slopes denote static
perfectly straight and the load, P, is applied along the
equilibrium positions. The ball on a hilltop (Fig.
centre line of the column. As the load is increased from
1a) represents an unstable equilibrium
zero quasistatically, the column remains straight but
configuration. If we were to give an infinitesimal
compressed. Now if small external disturbances
small disturbance, it will roll down the hill. In
(deformations or velocities) are applied in addition to
contrast, the ball in a trough or a valley represents
axial load P and the structure oscillates about the

deformed equilibrium state, the equilibrium said to given by q i (t ) and q i (t ) . If the response is also
be stable. Here, by small it is meant that as small
as desired. However, there exists a load denoted bounded, i.e, qi (t )   and q i (t )   , then static
by Pcr (critical load) for which the column tends to equilibrium state is considered as stable. This is shown
deviate from the original equilibrium state and this in Fig.3, schematically. This definition of stability is
equilibrium is said to be unstable. It does not return generally used in all fields (structural mechanics,
to its straight configuration on removing the biology and economics). In simple words, the system
disturbance. The column continues to be in the is stable if for small disturbances in initial conditions
deformed adjacent equilibrium configuration. leads to small change in the response. For conservative
Hence at this level of load there exists two different system, this may be stated as the equilibrium is stable
equilibrium states infinitesimally close to each provided potential energy is positive definite (van der
other. In other words a bifurcation point exists for Heijden, 2009).
this load.. This method is generally known as the The energy criterion of stability for elastic systems
equilibrium approach or the bifurcation approach. subject to conservative loading is almost universally
The lowest axial load for which this behaviour is adopted by researchers in the field of structural
observed is defined as the critical load or the Euler stability (Hutchinson and Koiter, 1970) to determine
buckling load of the column as this problem was the critical condition. A positive definite second
first solved by Euler in 1759. The important feature variation of the total potential energy about a static
of the problem just discussed is that there are equilibrium state is accepted as a sufficient condition
multiple equilibrium states, among which an for the stability of that state. In this connection Koiter
unbuckled or trivial state in which the column (Dym, 1974) has proved the following two theorems.
remains straight as shown in Fig.2. Theorem 1 (Koiter): A proper minimum of the
potential energy in the equilibrium configuration, in
P P P comparison with the energy in all other possible
configurations characterized by a sufficiently small L2-
Fig. 2 Two equilibrium configurations of the norm, is sufficient condition for stability.
column under the same loading. Theorem 2 (Koiter): The existence of a proper
minimum of the potential energy in the equilibrium
The bifurcation analysis can only be applied to constitutes, for all practical purposes, both necessary
perfect structures. In real world, structures being and sufficient condition for the stability of this
imperfect ones, the bifurcation load obtained configuration in the dynamical sense of Lyapunov.
through classical eigen value problem can serve
only as the guideline. For imperfect structures, the
buckling occurs when the deflection becomes
excessive or even infinite. Also, from the above sta
unst
discussion, it is clear that concept of motion be qi ( 
introduced to discuss the stability. In strict sense,
adjacent equilibrium state can only be discussed by 
considering the dynamic process of movement
from one stable state to another (Dym, 1974).
Hence, in general the concept of stability is a

dynamic phenomenon. The dynamic definition of
q 0 (t )
stability is applicable to all structural stability qi (
problems. In this case, we study the response of t
elastic structures subjected to external loads. Here
it is required to determine at what level of load the
response ceases to be bounded. This approach is Fig. 3 Response of disturbed system
known as the kinetic or dynamic approach. Hence for all practical purposes, a minimum of the
potential energy or the positive definite second
The kinetic criterion of stability was first
variation of the potential energy implies the stability in
introduced by Lagrange for a system with finite
the sense of Lyapunov.
degrees of freedom. A more general definition of The structural element (thin plate or shell) can store
dynamic stability was given by Lyapunov (Simitses large amount of membrane strain energy under
and Hodges 2006). Consider the n-degree freedom
compressive loads without deforming much. However
system with generalized coordinates. The static
it must deform much more to store equivalent amount
equilibrium state is denoted by q i  0 and the of bending strain energy. If the structure stores most of
system be subjected to bounded disturbances its energy as membrane strain energy when subjected
to compression and if there is a way that the stored up
qi0   and q i0   at t = 0. Here . stands for membrane energy can be converted to bending energy,
L2 norm. At any instant t > 0, let the response be then the structure will fail in the process called

buckling, where membrane strain energy is point is defined as the point where a perfect structure
exchanged into bending energy (Bushnell, 1985). switches from an initial equilibrium path (primary
path) to another equilibrium path i.e. secondary path.
HISTORICAL DEVELOPMENT OF THE Characterization of perfect structure can be done by
bifurcation point and imperfect structure by limit point.
SUBJECT Limit point is the point on equilibrium path, which
The theory of stability of structures is an important corresponds to zero structural stiffness. Nonlinear
part of structural mechanics. It is a basis for analysis is used to predict the nonlinear collapse and
rational design of new structures and examination limit points. Snap-through and snap-back buckling is
of the safety of existing structures. The theory of the phenomenon of nonlinear structural analysis. This
stability of structures started about 250 years back, is characterized by a sudden jump from one
with the work of Euler on buckling of columns in equilibrium configuration to another equilibrium state.
1759. However, the results of Euler could not be At the collapse load the load deflection curve has zero
verified experimentally. The reason for this slopes and if the load is maintained as the structure
discrepancy was explained by Young in 1807, in deforms, failure of the structure is almost
terms of imperfections such as initial curvature, instantaneous. This type of buckling occurs in hinged
initial bending moments or load eccentricity. A arches and clamped shallow spherical shells subjected
rigorous mathematical definition of stability was to uniform external pressure loading. The postbuckled
given by Liapunov in 1892 (Dym, 1974). The deflection curves of an axially compressed flat and
beginning of the modern theory of structural curved panel, with and without initial geometric
stability can be traced to the Doctoral Thesis of imperfections are shown in Figs. 4. (a) and (b). The
Koiter (Koiter, 1945). The thesis initiated curves in the figures indicate that the postbuckled
asymptotic analysis of postbuckling behaviour of equilibrium paths for flat and curved panels are
structures in the vicinity of critical states. It started significantly different. Flat panels exhibit a stable-
with the sensitivity analysis of structures to initial symmetric point of bifurcation. The equilibrium paths
imperfections. The study of stability of for flat plates are hardening in nature.
conservative and non-conservative systems was Curved panels (Fig. 4b), however, exhibit an unstable
made systematically by Ziegler (1956). Koter’s postbuckling response and the load-carrying capacity
work was further developed by Budiansky and of the curved panels is reduced to a value below the
Hutchinson (1964). Thompson (1963) initiated the bifurcation load when the response jumps to a new
study of nonlinear elastic stability in terms of stable equilibrium configuration at a limit point.Snap-
generalized coordinates. through and snap-back buckling phenomena (Fig. 5)
With the advent finite elements, the are some of the most difficult problems in nonlinear
computation load of many problems has become a structural analysis. The Newton-type methods are often
routine. However, correct results for the buckling used to solve nonlinear structural stability problems.
of statically indeterminate arches were obtained as The usual Newton-Raphson method or its modified
recently as 1970. Both experimental and theoretical version, self-correcting or standard incremental
investigations on the instability of shells revealed method, belong to this category. It is necessary to
shortcomings of the linear theory. It was observed modify the standard algorithm of these methods if
that however precise the linear theory be, its results these are employed to trace the postbuckling
were much less suitable for the practical design of configurations including snap-through and snap-back
shells. This was in contrast to linear theory of solution paths. For instance, referring to Fig. 5, if the
columns and plates. The problem for this displacement w were to be prescribed, the limit point B
discrepancy was identified by Koiter and he could be passed and the
attributed them to imperfections. It was pointed out
by Koiter that moderate reductions due to
imperfections occur in frames but in cylindrical
shells and spherical domes the reduction is up to
1/8 to 1/3 of the critical load. Efforts on removing
the discrepancies between the experimental and
theoretical predictions for buckling of cylindrical
shells significantly influenced the development of
nonlinear theories for shell buckling.
TYPES OF STABILITY
In the bifurcation buckling, a perfect structure
looses its stability by moving from one equilibrium
state to another equilibrium state which is Fig. 4 Postbuckled deflection curves showing
infinitesimally close to the first equilibrium state. bifurcation and limit points under uniaxial
The load associated with the bifurcation point is compression: (a) Flat panel, (b) Curved panel.
known as the critical buckling load. Bifurcation

load-shedding curve BC could be traced. However, boundaries which separate stable and unstable regions
the displacement control method would fail at, or in the parametric space defined above. Unlike in the
just before, the limit point F. A great number of case of resonances occurring under transverse forced
procedures have been proposed to overcome these vibration, parametric resonance will occur over a wide
problems. The most widely used scheme is the arc- range of excitation frequencies, particularly when the
length method proposed by Riks (1979, 1984) and amplitude of the dynamic load is large (Bolotin, 1964).
Crisfield (1981, 1983).It establishes the loading The dynamic instability results in transverse vibration
parameter as a variable. Since the loading level is of structures with ever increasing amplitude. Linear
treated as a variable one more equation is required strain-displacement relations are erroneous at higher
in addition to the usual equilibrium equations. This amplitude of vibration and hence geometric
additional equation uses the arc-length of the load- nonlinearities need to be considered. Nonlinear theory
displacement curve as a controlling parameter. results in new phenomena (Shivamoggi, 1977), which
have no place in the corresponding linear problem. The
new phenomena are: (i) existence of solution for all
values of the frequency rather than only for a set of
characteristic values (ii) dependence of amplitude on
frequency (iii) removal of resonance infinities and (iv)
appearance of jump phenomena.
FORMULATION
Here, nonlinear static stability analysis of shell panels
subjected to non-uniform in-plane loads and nonlinear
dynamic buckling analysis of plates are presented. The
static component and the dynamic component of the
Fig. 5 Load displacement curve showing snap- applied periodic non-uniform in-plane loading are
through and snap-back buckling assumed to vary according to either parabolic or linear
distribution. Initially, the plate membrane problem is
Often, plates are part of complex structural system solved using Ritz method to evaluate the plate in-plane
and hence load coming on it may not be always stress distributions within the prebuckling range.
uniform. For example, in the case of I-beam or Subsequently, using the evaluated prebuckling stress
wide flanged beam subjected to bending moment at components, the equations of motion of the cylindrical
the ends or lateral loads on the flange, the web of shell panels subjected to dynamic non-uniform in-
the beam is subjected to non-uniform in-plane plane loading are obtained using Hamilton’s principle
loads. The load exerted on the aircraft wings, or on in terms of force and moment resultants.
the stiffened plate in the ship structures or on the Expressing stress and moment resultants in terms of
slabs of a multi-storey building by the adjoining displacements the governing partial differential
structures usually is non uniform. The type of equations may be written in terms of displacement
distribution in an actual structure depends on the components as,
relative stiffnesses of the adjoining elements. The
non-uniform edge loading introduces all 3- Li (u 0 , v 0 , w0 , 10 , 20 )  0 i  1, 2...,5 (1)
components of stress (  xx ,  yy and  xy ) in the Where, Li are the nonlinear differential operator, uo,
plate which may considerably influence the static vo,wo are displacements of a generic points on the
stability behaviour of plate and shell panel. middle surface in three orthogonal coordinate
There are two types of dynamic instability. The directions respectively and 1 and 2 are rotations of
flutter is a type of dynamic instability which is normal to mid-plane about y-axis and x-axis
important for the design of aircraft wings, respectively. Adopting Galerkin’s approximation, the
suspension bridges, tall chimneys and similar governing nonlinear partial differential equations are
structures. Another type of instability which is converted into a set of nonlinear ordinary differential
important for foundations of rotating machinery, equations describing the plate dynamic instability. This
bridge columns and other structures is the may be written as,
parametric resonance. Plates and shell panels
subjected to dynamic in-plane loading, for certain MXKL KNL NsKG NtKGcosptX 0 (2)
combinations of the load amplitude, disturbing
frequency of transverse vibration, may undergo where [ M ] , [ K L ] , [ K NL ] and [ K G ] are
unstable transverse vibrations. The dynamic in-
respectively the mass, linear stiffness, non-linear
plane loading is known as parametric excitation
stiffness and geometric stiffness matrices and ‘p’ is the
and the range of the values of parameters causing
excitation frequency. Initially only the linear dynamic
dynamic instability is called the dynamic instability
instability of rectangular composite plate/shell panel is
region or parametric resonance region. Parametric
instability zones are generally defined by the

studied and hence the nonlinear terms in the Eq. (2) figure. The tensile load in y-direction makes the panel
are not considered. deflects downward and subsequent application of
compressive load in x-direction increases the deflection
MX KL NsKG NtKGcosptX 0 (3)
in downward direction. For zero and small positive
values of load ratio (N0) the equilibrium path shows
limit load and snap-through behavior.
It may be noted that, the effect of non-uniform in-
plane load is reflected in the computation of [ K G ] E1 / E2  25,G12  G13  0.5E2,G23  0.2E2,12  0.25
matrix. The Eq. (3) is a second order differential
equation with periodic coefficients. In the above
equation Ns and Nt are varied as N s   N cr and
N t   N cr , where  and  are static and
dynamic load factors respectively and N cr is the
static buckling load. The regions of instability are
located by boundaries of instability having periodic
solutions with period T and 2T to Eq. (3). These
are determined by the method suggested by Bolotin
(1964) in the present investigation. The solutions to
Eq. (3) with periods T and 2T are assumed in the
form of Fourier series respectively as, Fig. 6 The influence of the load ratio (N0) on the

kpt kpt (4a) nonlinear behavior of symmetric cross-ply [0/90/0]
x(t )  b0   (ak sin bk cos )
k 2,4,6 2 2 cylindrical shell panel subjected to parabolic uniaxial
 compression.
kpt kpt
x(t )   (a
k 1,3,5
k sin
2
bk cos
2
) (4b)
Figure 7 shows the principal zone of instability and
secondary zone of instability for symmetric 3-layered
where, ak and bk are arbitrary constants. Here, it cross-ply [0/90/0] laminated plate subjected to
parabolic edge loading for a static load factor
may be noted, two solutions of identical periods
Ns
bound the region of instability, two solutions of  ( )  0, by considering geometric
different periods bounds the region of stability. E22 h
Substituting Eq. (4a) or (4b) into Eq. (3) and nonlinearity. For parabolic in-plane load the static and
kpt dynamic component of compressive load at the edge
equating the coefficients of identical sin and
2 y2
y=0 are denoted as N s   Ncr (1  ) and
cos
kpt
leads to a system of homogeneous b2
2 y2
algebraic equations in ak and bk . For nontrivial Nt   N cr (1  ) respectively. It is observed from
b2
solution, the determinant of the coefficient matrix the figures that as the nonlinearity (i.e. w/h) increases,
of ak and bk must vanish. The size of the above the principal instability zones as well as secondary
determinant is infinite as we have assumed the zones of instability narrows down and shifts to higher
solution in the form of infinite series. The frequencies. This is because at higher (w/h) ratio, the
plate becomes stiffer due to stretching of its middle
determinants are shown to be belonging to a class
plane.
of converging determinant known as normal
determinant (Bolotin, 1964).

Fig. 6 (Sarat Panda and Ramachandra 2009) shows
the postbuckled equilibrium path for parabolically
varying biaxial in-plane edge loading for simply
supported symmetric crossply [0/90/0] cylindrical
shell panel for various load ratios. The load ratio N0
(=Ny/Nx) is the ratio of compressive edge load in Fig. 7. Non-liniear instability zones of simply
the x-direction (Nx) to compressive (positive) or supported composite cross-ply plate (a/b=1, a/h=100,
tensile (negative) edge load in the y-direction ( and Ns
  0.
E 22 h
Ny). For all negative load ratio N0, the panel
deflects in the inward direction as shown in the
E1/E2 40, G12 G13 0.5E2, G23 0.6E2, 12 0.25

Next, the principal dynamic instability regions are
evaluated for the 3 layered symmetric cross ply
[0/90/0] square composite plates (a/b = 1, a/h =
100 and  = 0.5) subjected to time dependent non-
uniform in-plane loads. They are shown as the plot
of dimensionless excitation frequency against
dimensionless dynamic load factor (  ) in Figure 8.
It may be observed that the width of the dynamic
P =275
instability region is the maximum for the uniform
(b) Non-linear responcse in the stable region
in-plane load and minimum for the pure bending
case. This is due to the fact that the buckling load is
minimum for uniform loaded plate and maximum
for the pure in-plane bending case. It is instructive
to plot the responses in stable and unstable zones,
as this is expected to clearly bring out the
difference between linear and non-linear analysis.
For this purpose the dynamic instability plot for the
three layered cross-ply composite plate shown in
Fig. 8 is considered.
For this case, the responses are plotted first for a
frequency of p=275, which lies above the unstable P= 230
zone. In this zone both linear and non-linear (c) Linear – response in the unstable region
analysis gives the stable response (Figs.9a and 9b).
However, the amplitude of nonlinear response is
smaller than amplitude of linear response. One can
also observe beats from these figures as reported in
[1, 9]. In the unstable region, for the frequency
p=230, the amplitude of linear response (9c)
increase continuously and the non-linear response
(9d) amplitude remain within bounds. Beats is
observed in Figure 9(d).
P=230
(d) Non-linear-response in the unstable region
Fig. 9 Dynamic response of composite plate in stable

and unstable regions.
CONCLUSIONS
This paper deals with the linear and nonlinear static
stability and dynamic instability of composite plates
and cylindrical shell panels subjected to dynamic non-
uniform in-plane loading. The static buckling load of
Fig. 8 Principal instability zones for SSSS cross- non-uniformly loaded cylindrical panels is evaluated in
ply composite plate subjected to parabolic in- two steps as discussed above. The dynamic instability
planeload behavior depends on the in-plane stress distribution,
the natural frequency of unloaded structure and
frequency of excitation. The plate in-plane stress
distribution depends on the type of applied in-plane
edge loading. The primary instability regions that
correspond to solution with period 2T has greater
practical importance as the width of the instability
regions are more in this case compared with period T.
At higher (w/h) ratio, the plate becomes stiffer due to
stretching of its middle plane and hence instability
zone narrows down and shifts to higher frequency.
Also response in different regions clearly brings out
the different nature of linear and nonlinear analysis of
P =275
(a) Linear response in the region

plate, i.e., existence of beats, their dependency on Shivamoggi, B.K., 1977. Dynamic buckling of a thin
the forcing frequency. elastic plate: Non-linear theory, Journal of Sound
and Vibration, 54(1), 75-82.
ACKNOWLEDGEMENTS Simitses, G.J. and Hodges, D.H.,2006. Fundamentals
Author would like to thank his students, Dr. Sarat of Structural Stability, Butterworth-Heinmann,
Kumar Panda and Mr. Tanish Dey, for the results New York.
reported in this article. Thompson, J.M.T., 1963. Basic principles in the
general theory of elastic stability, J. Mech. Phys.
REFERENCES Solids, 11, 13.
Bazant, Z.P., Cedolin, L., 1991. Stability of Timoshenko, S.P., and Goodier, J.N., 1970. Theory of
Structures, Oxford University Press,Inc. elastic stability, McGraw-Hill Publication, New
Bazant,Z.P., 200. Structural Stability, International York.
Journal of Solids and Structures, 37, 55-67. van der Heijden, A.M.A., W.T. Koiter’s Elastic
Bolotin, V.V., 1964. The dynamic stability of Stability of Solids and Structures, Cambridge
elastic systems, Holden-Day Publication, San University Press, 2009.
Francisco. Young,T., 1807, A Course of Lectures on Natural
Budiansky, B., and Hutchinson, J.W., Dynamic Philosophy and the Mechanical Arts, London.
buckling of imperfect sensitive structures, Ziegler, H., 1956. On the concept of elastic stability,
Proc. XI Internat. Cong. Appl. Mech., Munich, Advances in Applied Mechanics, 4, 351,
1964 (Springer-Verlag). Academic Press, New York.
Bushnell, D., 1985. Computerized Buckling Zeigler, H., 1968. Principles of Structural Stability,
Analysis of Shells. Martinus Nijhoff Blaisdell Publishing Co. (A division of Ginn and
Publishers, Dordrecht. Co.), Waltham,Mass.
Crisfield, M.A., 1981. A fast incremental/iterative
solution procedure that handles snap-through.
Computers and Structures, 13, 55-62.
Crisfield, M.A., 1983. An arc-length method
including line searches and accelerations. Int.
J. for Numerical Methods in Engineering, 19,
1269-1289.
Dym, C.L., 1974. Stability theory and its
applications to Structural Mechanics, Dover
Books, New York.
Girish, J., 2005. Postbuckling and Postbuckled
Vibrations of Composite Plates and Shells,
Ph.D. thesis submitted to IIT, Kharagpur.
Hutchinson, J.W. and Koiter, W.T., 1970.
Postbuckling theory. ASME Applied
Mechanics Reviews 23(12), 1353-1366.
Koiter, W.T., 1945. On the stability of elastic
equilibrium. Dissertation, Delft, Holland, 1945
(English translation issued as NASA
Tech.Trans., F 10, 833, 1967).
Riks, E., 1979. An incremental approach to the
solution of snapping and buckling problems.
Int. Journal Solids and Structures, 15, 529-
551.
Riks, E., 1984. Some computational aspects of the
stability analysis of nonlinear structures.
Comp. Meth. Appl. Mech. Engng, 47, 219-
259.
Sarat Kumar Panda and Ramachandra, L.S., 2010.
Postbuckling Analysis of Cross-ply Laminated
Cylindrical Shell Panels under Parabolic
Mechanical Edge Loading, Thin walled
Structures, 48(8), 660-667.
Sarat Kumar Panda and Ramachandra, L.S., 2011.
Dynamic Instability of Composite Plates
Subjected to Non-uniform In-plane Loads,
Journal of Sound and Vibration (In press).

ISBN 978-81-921121-0-7
EFFECT OF BRIDGE PIER’S SHAPE ON SCOURING
Manvendra Singh Chauhan

Post Graduate Student, Civil Engineering/IT BHU, Varanasi – 221005, Email: Manvendra.hbti@gmail.com
U. K. Choudhary
Professor, Civil Engineering/IT BHU, Varanasi – 221005, Email: ukc@bhu.ac.in
P. K. Singh Dikshit
Professor, Civil Engineering/IT BHU, Varanasi – 221005, Email: pk_sd@yahoo.com
ABSTRACT: In the present age of fast communication, bridges occupy a place of pride in the society as they
facilitate rapid movement of material and personal safety and quickly across the rivers and the other drainages.
The disruption of traffic due to the failure of bridge therefore results in a disaster. The main cause of concern in
stability of bridges founded in river bed is the local lowering of river bed level caused by the river flow around
the bridge elements such as piers, abutments and spurs dikes and is termed as ‘scour’ and this phenomena is
called scouring. An accurate estimation of scour depth below stream bed during design is important since this
determines the foundation levels of the bridge elements such as piers, abutments, guide bank, spur, groyene, etc.
Usually, a structural engineer checks the safety of the bridge components while an architect designs the bridge
shape. The hydraulic engineer predicts the scour depth at the bridge site for a design flood before the
construction and evaluates the need for scour protection at the bridge site after the construction.
INTRODUCTION In every bridge failure many people lost their life.

Scour is defined as the erosion of streambed Besides their human toll, bridge failures cost millions
sediment around an obstruction in a flow field. The of rupees each year in direct expenditure for
mechanism has the potential to threaten the structural replacement and restoration in addition to the indirect
integrity of bridges and hydraulic structures, expenditure related to the disruption of transportation
ultimately causing failure when the foundation of the facilities.
structures is undermined. It is the consequence of the
erosive action of flowing water, which removes and LOCAL SCOUR MECHANISM
erodes material from the bed and banks of streams The basic mechanism causing local scour at piers or
and also from the vicinity of bridge piers and abutments is the formation of vortices at their bases
abutments. (known as horseshoe vortex at a pier and horizontal
Scour can either be caused by the normal flow or vortex at an abutment).The horseshoe vortex results
flood events. Normal flow can lower the channel bed from the pileup of water on the upstream face of the
but scouring is most assisted during a peak flow in obstruction and subsequent acceleration of the flow
which the flow velocity is higher. In other words, around the nose of the pier or embankment. The
scour can occur under any flow condition that makes action of the vortex removes bed material from
the bed mobile within the vicinity of the obstruction around the base of obstruction. The transport rate of
but the rate of scouring is much higher with larger sediment away from the base region is greater than
flow events. The amount of the reduction below an the transport rate into the region and consequently a
assumed natural level (generally the level of the river scour hole develops. As the depth of scour increases,
bed prior to the commencement of the scour) is the strength of the horseshoe vortex is reduced,
termed the scour depth. A scour hole is defined as the thereby reducing the transport rate from the base
void or depression left behind when sediment is region. Eventually for live bed scour, equilibrium is
washed away from a stream or river bed. established and scouring ceases. For clear water
Any structure placed in a river, whether of natural or scour, scouring ceases when the shear stress caused
human origin, will tend to promote scouring and by horseshoe vortex equals the critical shear stress of
deposition due to sudden change in the flow direction sediment particles at the bottom of scour hole. In
or high velocity flow. Such obstacles can include a addition to the horseshoe vortex around the base of
boulder, an island, a sharp bend, or a pipe or road pier, there are vertical vortices downstream of pier,
bridge pier. Local scour causes holes to be dug called the wake vortex; both the horseshoe and wake
around structures such as bridge piers and abutments vortices remove particles from the pier base region.
which are built on or below the river bed. These However, the intensity of wake vortices diminishes
holes can be exceptionally deep, e.g. they can easily rapidly as the distance downstream of pier increases.
run to 20 to 30m. Scouring can also result in the sand Therefore, immediately downstream of a long pier
around the pier being removed to such a depth that there is often deposition of material.
the pier ends up being supported on nothing. Because
of this the pier will fail and the structure will
collapse.

Melville and Chicw (1997) studied the temporal
development of clear water local scour depth at
cylindrical bridge piers in uniform sand beds.
Experiments were conducted in four different
flumes. The results showed that the scour depth after
10% of the time to equilibrium is between about
50% and 80% of the equilibrium scour depth,
depending on the approach flow velocity. Totapelly
et al. (1999) examined the temporal variations of
local scour under steady flow and using stepped
hydrographs to calculated scour depth & concluded
that also logarithmic equation represented the
variation of scour with time better than a power
equation and questioned the existence of an
Figure: Flow pattern around bridge piers.
equilibrium depth. They found that for steps with
duration on the order of 2.5 hours, the results of the
FACTORS AFFECTING LOCAL SCOUR super positioning were comparable to experiments.
Graf et al. (2001) investigated the flow patterns in
Factors which affect the magnitude of the local scour
planes upstream and downstream of a cylindrical
depth at piers are as follows:
pier embedded and vertically in the scour hole using
 Approach flow velocity
an acoustic-Doppler velocity -profiler (ADVP). They
 Flow depth found that the shear stress was reduced in the scour
 Pier width hole as compared to the approach flow but that the
 Gravitational acceleration turbulent kinetic energy was very strong at the foot
 Pier shape of the cylinder on the upstream side. The kinetic
 Size and gradation of the bed material energy was also very Strong in the wake behind the
 Angle of attack of the approach flow to the cylinder.. Faruque and Nago (2003) have calculated
pier the scour hole on the basis of a sediment transport
 Bed configuration equation. A reliable prediction of local scour depth
related to hydrological characteristics and proposed a
design method to predict the local scour depth with
LITERATURE REVIEW
time. Wen-Yi Chang et al. (2004) conducted
The process of local scour around bridge piers is
experiments of bridge pier scour under steady and
fundamentally complex due to the three dimensional
unsteady clear water scour conditions with uniform
flow patterns interacting with the bed materials.
and non uniform sediments and proposed a scheme
Many researchers around the world have studied the
for computing the scour-depth evolution under
phenomenon of local scouring. Few of them
unsteady flow.. Sheppard et al. (2004) performed
presented their view as from their practical
local clear water scour tests with different diameter
experience and others as per their experimental
circular piles, different uniform cohesionless
results. A large number of literatures have been
published so far. Few of the literature are presented sediment diameters and a range of water depths and
flow velocities and concludes equilibrium scour
in this section to give an overview of the scouring
depths were f o u n d to depend on t h e wash load
effect around the piers. Dey et al. (1995) derived
concentration. Dey and Barbhuiya (2005) studied the
quasi-analytical equations for the flow field (in the
time variation of scour at abutments, they have
scour hole, adjacent to the pier above the flat bed and
compared the time variation of scour depth at short
in the wake region) by satisfying the continuity
abutments (abutment length/flow depth< 1) for the
equation and determining empirical coefficients (by
vertical wall.45° wing wall and semi circular wall.
curve fitting experimental data). They analysed a
Clear water scour was studied for uniform and non
model and the model may be useful for stimulating
uniform sediment. They compared their results with
the flow field under prototype conditions. Gosselin
other investigators and found good agreement.
(1997) conducted a series of clear water scour
Sheppard and Miller (2006) conducted local clear-
experiments to measure the velocity field (at various
water and live-bed scour tests for a range of water
times during the scour process). He modelled the
depths and flow velocities with different uniform
velocities in the scour hole with a commercial three
cohesionless sediment diameters. They used a
dimensional flow model. Bruce W. Melville (1997)
circular pier for analysis and tests were performed in
used results from an extensive program of bridge
a tilting flume located in the Hydraulics Laboratory
scour and determines empirical relations, termed K-
at the University of Auckland in Auckland. Scour
factors which were evaluated by fitting envelope
depth predictions were made with four different
curves to existing data for piers and abutments and a
local scour equations for the conditions of the tests
new extensive data set for abutments.
and the results were compared with the measured

values. Ettema et al. (2006) did their experiments present experimental data and their performances
on local scour at vertical cylinder, placed in a sand were evaluated.
bed, shows that the similitude of large scale
turbulence is an important consideration influencing EXPERIMENTAL WORK AND SETUP
equilibrium depth of local scour. Ashtiani and The experiment was conducted in the Hydraulics
Beheshti (2006) carried out experiments of local Laboratory of Department of Civil Engineering,
scour around pile groups under steady clear water Institute of Technology, Banaras Hindu University,
scour conditions. A variety of conditions including Varanasi. The laboratory channel considered was
different pile group arrangements, spacing, flow rectangular having two consecutive bends of reverse
rates, and sediment grain sizes were considered. It order, 1.0m wide, 22.5 cm deep, central radius (rc)
was observed that the scour-hole depth for some 1.0 m and central angles of 1800 each. The bends are
cases of pile groups increases as much as two times connected with a stilling basin, an upstream straight
more than its magnitude for the cases of single piles. reach 4.5 m long, a downstream straight reach 6.5 m
Dey and Raikar (2007) conducted an experimental long and a side channel conveying.
study on the turbulent horseshoe vortex flow within Experiment consists of three parts, in first part no
the developing intermediate stages and equilibrium pier was embedded, in second part circular piers
scour holes at cylindrical piers measured by an were embedded and in last part same periphery (as
acoustic Doppler velocity meter (ADV). The circular piers) elliptical piers were embedded. During
imperative observation is that the flow and the experiment erosion / sedimentation data were
turbulence intensities in the horseshoe vortex flow collected at the different sections which give a useful
in a developing scour hole are reasonably similar. conclusion.
Subhasish Dey et al. (2008) observed that in a
degrading bed, the bed forms cause edge failure of CONCLUSIONS
the riprap at an abutment when the dunes propagate The following important conclusions are drawn on
over the riprap layer. Also the damaged riprap layer
the basis of data tables, figures and places mentioned
was significantly vulnerable against a subsequent
flood accompanied by large dunes. above
Dey and Singh (2008) conducted experiments on 1. In case of flow without pier sedimentation occurs
clear-water scour below underwater pipelines near the convex bank and it is increase as increase in
initially laid on the sediment bed in uniform and θ, and the maximum depth of sedimentation occurs at
nonuniform sediments under steady flow. θ= 120°.
Equilibrium scour profiles were modelled by a cubic 2. As the location of width increases from the convex
polynomial. The experimental results were examined to concave sedimentation decreases and the pocket of
to describe the influence of various parameters on erosion starts near the concave bank at θ= 120°.
equilibrium scour depth and concluded that the scour
depth with an armor layer was less than that without 3. As θ increases from 120° erosion zone increases
armor layer for the same sand bed, when the scour and its depth is also increased.
hole was shielded by the secondary armor layer. 4. Sedimentation depth is maximum in case of
Carmelo Grimaldi et al. (2009) performed laboratory circular pier and the location occurs at convex side at
experiments to investigate the effectiveness of bed θ= 120°.
sills as countermeasures against local scouring at a
smooth circular bridge pier, for flow conditions near 5. With increase in θ depth of sedimentation starts
the threshold of uniform sediment motion. The reducing on the convex side and width of erosion
results showed that a bed sill placed at a short starts developing on the concave side.
distance downstream of the pier reduces the scour 6. The maximum depth of erosion occurs in case of
depth, area, and volume. circular pier at θ= 150°.
Francesco Ballio et al. (2010) performed live-bed
scour experiments at a vertical wall abutment with 7. After θ= 210° (second bend θ= 30°) the concave
specific attention paid to the conceptual issues side of first bend is converted into convex side of
concerning the temporal development of local scour second where zone of erosion exist.
phenomenon. A conceptual model was constructed to 8. The maximum erosion depth in second bend at θ=
explain some previous experimental results which do 270° when accounted in continuously which is θ=
not confirm the explanation for scour fluctuations. 90° of the second bend, The maximum erosion in
B. Ataie Ashtiani et al. (2010) performed local scour case of circular pier as compared to elliptical pier.
experiments around complex piers under steady
clear-water condition for a variety of configuration, 9. The erosion pocket in second bend extends more
including different sizes and shapes of complex in case of circular pier and this pattern continuous till
piers. The available procedures for estimating scour θ= 360°.
depth around the complex pier were applied to the

REFERENCES Melville B W, Sutherland A J (1988) “Design
Ansari S A, Kothyari U C, Ranga Raju K G (2002) methods for local scour at bridge piers”. J.
“Influence of cohesion on scour around bridge Hydraul. Eng. 114:
piers”. J. Hydraul. Res. 40: 717–729 Melville, B.W. and Chiew, Y.M. (1999). “Time scale
Chang, W.Y., Lai, J.S., and Yen, C.L. (2004). “ for local scour at bridge piers”. Journal of
Evolution of scour depth at circular bridge Hydraulic Engineering, ASCE, 125(1): 59-65.
piers.”J. Hydraul. Eng., 130(9), 905-913. Raudkivi, A.J. and Ettema, R. (1983). “Clear-water
Cheremisinoff, P.N., Cheremisinoff, N.P. and Cheng, scour at cylindrical piers”. Journal of Hydraulic
S.L. (1987). Hydraulic mechanics 2. Civil Engineering, ASCE, 109(3): 339-350.
Engineering Practice, Technomic Publishing Richardson, E.V. and Davies, S.R. (2001).
Company, Inc., Lancaster, Pennsylvania, U.S.A. Evaluating scour at bridges. Rep. No. FHWAIP-
780 p. 90-017 (HEC 18), Federal Administration, U.S.
Chitale S V (1988) “Estimation of scour at bridge Department of Transportation, Washington, D.C.
piers”. J. Irrigation and Power (CBIP) 45:
Chitale S V (1993) Indian practice of scour
estimation—limitations. J. Irrigation and Power
50:
Dey, S., Bose, S. K., and Sastry, G. L. N. (1995).
“Clear water scour at circular piers: A model.” J.
Hydraul. Eng., 121 (12), 869-876.
Dey, S. (1999). Time-variation of scour in the
vicinity of circular piers. Water & Maritime
Engineering journal, proceedings of the Institution
of Civil Engineers, Thomas Telford Journals,
Londan, 136 (2), 67-65, paper 11426, June.
Dey, S., and barbhuiya, A. K. (2005). “Turbulent
flow field in a scour hole at a semicircular
abutment.” Can. J. Civ. Eng., 32 (1), 213-232.
Dey, S., rajkumar V.R. (2007). “Clear water scour at
piers in sand beds with an armor layer of gravels”.
J. Hydraul. Eng., 133 (6), 703-711.
Garde R J, Kothyari U C (1998) “Scour around
bridge piers”. J. Ind. Nat. Acad. New Delhi, 64:
569–580
Garde R J, Ranga Raju K G (2000) “Mechanics of
sediment transportation and alluvial stream
problems”. (New Delhi: New Age)
Indian Railway Standards, IRS 1985 Code of
practice for the design of sub-structures and
foundations of bridges
IRC 1998 Standard specifications and code of
practice for road bridges, section 1, IRC: 5-1998
IRC 2000 Standard specifications and code of
practice for road bridges. section-VII, IRC: 78–
2000
Kothyari U C, Garde R J, Ranga Raju K G (1992a)
“Temporal variation of scour around circular
bridge piers”. J. Hydraul. Eng. 118: 1091–1106
Kothyari U C, Garde R J, Ranga Raju K G (1992b)
“Live-bed scour around cylindrical bridge piers”.
J. Hydraul. Res. 30: 701–715
Kothyari U C (2003) Bridge pier scour in gravel –
cobble and cohesive bed rivers, Report submitted
to the IRC – sub-committee to review the aspects
of scour around bridge foundations
Kothyari U C (2007) “Indian practice on estimation
of scour around bridgePiers-A comment” Sadhana
Vol. 32, Part 3, June 2007, pp. 187–197.

AUTHOR INDEX
Abd Elbasit, M. A. M. 1 Divya, P. S. 58

Agrawal, D. 635 Domke, P. V. 239
Ahluwalia, R. S. 286 Dubey, O. P. 69
Arora, M. K. 336 Dutta, A. 215
Pradhan, B 112 Dutta, A. 525
Bahirat, E.M. 181 Dutta, S. 507
Bansal, V. 450 Dwivedi, S. B 294, 349, 595
Barman, P.J. 112 Gao, E. 392
Basak, S. 525 Ghadiyali, G.A. 250
Bhalerao, A.R. 563 Ghosh, K. K. 195
Bhamare, P.R. 205 Goel, A. 103
Bharath, G. 31 Goyal, M. 635
Bhardwaj, A.K. 122 Goel, M. 648
Bhaskar, R. 299 Gupta, A. 461
Bhatt, R. 548 Gupta, S. 577
Bhattacharjee, B. 234 Himanshu, S. K. 421
Bhattacharya, A. 336 Ingle, V. 480
Bhide, D.D. 205 Jacob, V.M. 618
Bhunia, D. 450 Janbandhu, M.S. 538
Bhushan, S 492 Jain, A. 566
Bisht, R.S. 144 Jain, H. 294,502
Biswas, S. 643 Jain, S. K. 286
Buradkar, P.S. 151 Jayashree, T. 591
Burman, A. 492 Jegan, V. 618
Chadaga, M. 440 Jha, M. 27,100
Chaithanya, K. 227 Jindal, A. 445
Chandra, S. 276 Jindal, S. 256
Chattopadhyay, S. 525 Johri, A. 405
Chaudhuri, S. R. 74 Joshi, S. V. 323
Chauhan, M.S. 663 Julaganti, A. 392
Chauhan, P. A. 210 Kadian, V. 440
Chitkeshwar, A.K. 426 Kadu, M. S. 323
Choudhary, A. 450 Kalantari, B. 317
Choudhary, R. 392 Kalpana.V.G. 280
Choudhary, U. K. 663 Kalvit, S.K. 205
Chowdhury, P.R 256 Kanakabandi, S. 417
Chowdhury, S. R. 74, 195 Kartha, S.A. 112
Dalvi, R. E. 170 Katti, B.K. 250
Das, S. 276 Kene, S. D. 239
Das, T. 175 Khati, B.S. 472
Deb, S.K. 215 Khatri, V. 602
Deepu, S.P. 74 Koppineni, A. 227
Deotale, R.S. 239 Krishan, B. 306
Deshmukh, S. D. 239 Krishan, G. 400
Deshpande, P.K. 271 Krishnan, J. M. 58
Dikshit, P. K. S. 408, 663 Kujur, J. 234

Kulkarni, S.N. 205 Pande, P. P. 283
Kumar, A. 492 Pande, S. 591
Kumar, A. 577 Pandhar, V.B. 205
Kumar, B 377, 400 Pandey, K.K. 131, 595
Kumar, B. 417, 551, 577, 625 Paramkusam, B.R. 48
Kumar, B. 507, 519 Parauha, A. 450
Kumar, D. 48 Pardeshi, A.K. 157
Kumar, D. 360 Pardeshi, D.K. 63
Kumar, D. 421, 472 Parekh, B.J. 328
Kumar, D. 461 Patel, S.C.R 108
Kumar, J.P. 164 Patel, S.K. 533
Kumar, M. 1 Patra, B.K. 365
Kumar, P. 210, 445, 551 Petkar, A.S 181
Kumar, P. 497 Pillai, R. G. 618
Kumar, R. 95, 294, 365, 405 Porwal, S. K. 497
Kumar, R. S. 598 Poudel, S. S. 440
Kumar, R.P. 610 Pradhan, B 112
Kumar, S. 48 Prakash, B.S.N. 370
Kumar, S. 164 Prakash, D. 81
Kumar, S.N. 598 Prasad, A. 317, 433
Kumar, S.V. 52 Prasad, K. 88,137
Kumar, V. 331, 365, 405 Prasad, V.R 299
Kumari, S. 311 Prathima, G. 513
Kumthekar, M.B. 271 Purushothaman, P. 377, 400
Kute, S. 591 Quraishi, M.A. 331
Ladhane, K.B. 311 Rai, A. 638
Lakshminarasimhan, S. 201 Rai, P.K. 262
Lal, M 635 Rai, S. P 286, 377
Lelitha, V. 52, 227 Ramachandra, L.S. 656
Lodha, P P 438 Ramakrishna, C. 370
Maiti, P. R. 343 Ransinchung, R.N. G.D. 210, 445, 551, 577
Maity, D. 492 Rao, C. B. K. 299
Malaya, C. 290 Rao, G. V. 388, 584
Mallik, N. 355 Rao, M.S. 400
Mandal, S. 360 Rao, S. 1
Marthong, C. 215 Rao, S.V. 457
Mathew, T. V. 489 Rao, T. D. G. 469
Maurya, A.K. 643 Rathod, A. B. 328, 569
Maurya, R. K. S. 558, 614 Rathor, A.O. 63
Meshram, V.G. 426, 538 Ravindranath 440
Mishra, K.R. 187 Rawat, Y.S. 377, 400
Mishra, U. 595 Reddy, M. A. 31
Mohan, D. 502 Reddy, R.R. 388, 584
Mukhopadhyay, S.K. 525 Regulwar, D. G. 63, 157, 480
Narayan, S. 7 Roy, D. 7
Nema, A.K. 311 Roy, L.B. 164
Niyogi, A. G. 276 Sahu, N.K. 548
Ohri, A. 122 Samrajyam, K. 469
Ojha, C.S.P. 1,88,137 Saran, S. 472
Padhye, R.D. 271 Shah, M.K. 250

Shah, N.B 108 Singla, V. 95
Shankar, K.V. R. R. 489 Sivapullaiah, P.V. 13, 370
Shankar, Y.S. 502 Sobha, B. 469
Sharma, I.N. 81 Sreedeep, S. 290
Sharma, M. L. 336 Sreedhar, M.V.S. 388, 584
Sharma, R. K. 396 Srimani, P. 195
Sharma, S. 643 Srivastav, A. L. 37
Sharma, Y.C. 37 Srivastava, A. 343
Sherasia, N.K. 187 Subramanian, K. 280
Shrivastava, N. 450 Sujan, M. 246
Sidharth, K. 227 Talukdar. S. 507
Kazemian, S. 433 Tewari, S. 81
Singh, A.P. 638 Thalange, S.B. 271
Singh, B. 144, 175, 533 Theunuo, K. 349
Singh, B.N. 331 Thiyam, T.D. 519
Singh, C.S. 396 Tignath, S. 27,100
Singh, G. P. 558, 614 Trivedi, S. A. 569
Singh, J. 635 Upadhyay, M.P. 548
Singh, K. M. 88,137 Vashisth, A. 131
Singh, N.P. 190 Vasu, R. M. 7
Singh, P. C. 81 Venkatesh, S. 221
Singh, P. K. 37, 122 Vernekar, N.D. 563
Singh, P.K. 602, 625 Vimala, A. 95, 610
Singh, R. 408 Vishal, V. 22,
Singh, R.D. 648 Yadav, G.K. 577
Singh, S.K. 331 Yadav, H. L. 421
Singh, S.M. 116 Yasuda, H. 1
Singh, S.P. 382 Yuvaraaj, K.S. 201
Singh, T. N. 22 Zuned, S.M. 438
Singh, Y. 382
Singhal, D. 306

Race 2011

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Race 2011

Uploaded by

Copyright:

Available Formats

RECENT ADVANCES IN

Department of Civil Engineering,

Institute of Technology, Banaras Hindu University,

Proceeding of the National Conference on Recent Advances in Civil Engineering

October 14-16, 2011, Varanasi

Department of Civil Engineering,

Institute of Technology, Banaras Hindu University,

RECENT ADVANCES IN CIVIL ENGINEERING-2011 i

1 VARIABILITY OF MASS TRANSFER COEFFICIENT IN CONTROLLED CHAMBER 1-6

RECENT ADVANCES IN CIVIL ENGINEERING-2011 ii

RECENT ADVANCES IN CIVIL ENGINEERING-2011 iii

RECENT ADVANCES IN CIVIL ENGINEERING-2011 iv

RECENT ADVANCES IN CIVIL ENGINEERING-2011 v

RECENT ADVANCES IN CIVIL ENGINEERING-2011 vi

RECENT ADVANCES IN CIVIL ENGINEERING-2011 vii

VARIABILITY OF MASS TRANSFER COEFFICIENT IN

RECENT ADVANCES IN CIVIL ENGINEERING-2011 1

RECENT ADVANCES IN CIVIL ENGINEERING-2011 2

RECENT ADVANCES IN CIVIL ENGINEERING-2011 3

200 Thus, to screen out the next best performing

It is interesting to see that strategies 3, 4 and 5

3, 4 and 5 of controlled experiments. Day one is used to 400

calibrate mass coefficient and this calibrated mass 200

transfer coefficient is used to compute evaporation for 0

RECENT ADVANCES IN CIVIL ENGINEERING-2011 4

Evaporation rate (g/m hr)

complete data in calibration range, is found to

Fig. 9 Comparison of observed and computed

RECENT ADVANCES IN CIVIL ENGINEERING-2011 5

RECENT ADVANCES IN CIVIL ENGINEERING-2011 6

A SMOOTH DMS-FEM FORMULATION FOR MINDLIN PLATE

RECENT ADVANCES IN CIVIL ENGINEERING-2011 7

splines (Sunilkumar and Roy 2010a, 2010b; ( n  1)( n  2)

RECENT ADVANCES IN CIVIL ENGINEERING-2011 8

connecting the knots located close to the vertices of the H T

where N nd represents the number of nodes in Ω,

RECENT ADVANCES IN CIVIL ENGINEERING-2011 9

Fig. 4 Triangulation of a Quadrant of square

RECENT ADVANCES IN CIVIL ENGINEERING-2011 10

For the second set of examples, again consider a

thin plate series solution of Morley (1963) and

Fig. 7 Morley’s skew plate

RECENT ADVANCES IN CIVIL ENGINEERING-2011 11

RECENT ADVANCES IN CIVIL ENGINEERING-2011 12

ENHANCING THE REACTIVITY OF FLY ASH FOR

PRODUCTION OF FLY ASH

RECENT ADVANCES IN CIVIL ENGINEERING-2011 13

Table 1 Range of chemical composition of Indian Morphology

RECENT ADVANCES IN CIVIL ENGINEERING-2011 14

RECENT ADVANCES IN CIVIL ENGINEERING-2011 15

RECENT ADVANCES IN CIVIL ENGINEERING-2011 16

To minimize the potential for environmental problems

granular lift berm should b provided. The berm and

wind. Preventing ground water intrusion is equally BC soil + 2N NaOH

To minimize deeper percolation, fly ash is used along

RECENT ADVANCES IN CIVIL ENGINEERING-2011 17

RECENT ADVANCES IN CIVIL ENGINEERING-2011 18

RECENT ADVANCES IN CIVIL ENGINEERING-2011 19

Metal Extraction (%)

Fig. 3 Effect of Lime and Gypsum on the Hydraulic

Addition of gypsum is found to be much more

RECENT ADVANCES IN CIVIL ENGINEERING-2011 20

Prashanth JP, Sivapullaiah PV, Sridharan A (2000)