You are on page 1of 25

LANDSLIDE HAZARD POTENTIAL ANALYSIS

USING GIS, PATALGANGA VALLEY,


GARHWAL, WESTERN HIMALAYAN
REGION OF INDIA

Dr.Kumar Kishor
Scientist F, Geotechnical Engineering Division, Central Road Research
Institute
New Delhi, India
Rahul Devrani
Department of Geology,University of Delhi, India
rahuldevrani18@gmail.com
Sudhir Mathur
Scientist F, Geotechnical Engineering Division, Central Road Research
Institute
New Delhi, India

Abstract

Patalganga drainage basin, a mountainous valley, is located in North western Himalayan


region of Garhwal between longitude 790 29’ 15” E - 790 29’ 30” E and Latitude 300 29’ 25”
N - 300 29’ 15” N. An important National Highway No. 58 goes across the valley and is
the only connecting route between this region and rest of the country. The valley as well
as the National Highway had suffered from bewildering variety of Landslides and other
mass movements during an unprecedented rainfall of July 1970. Since 1970, the
population and the associated developmental activities in the valley got increased by
many folds without any disaster mitigation and management planning. The vulnerability
of the valley to landslide disasters also increased relatively to the developmental
activities. To study the current status as well as the future probability of landslide
development and their consequences, a project was initiated in the year 2002. Under this
project Landslide Hazard Potential (LHP) analysis of the basin was carried out to prepare
a Landslide Susceptibility Potential Map (LSPM) for guiding future development within
the basin. To do such analysis and prepare LSPM a variety of thematic information maps
pertaining to the different factors were generated in 1:12500 scale using latest satellite
data and extensive field investigations. All the maps with their analysis and outputs are
prepared in GIS platform. In this paper a stepwise approach applied for landslide hazard
potential analysis is described with a final output of Landslide Susceptibility Potential
Map of the area which will be helpful in guiding the development of the valley in future.

Keywords: Landslide Susceptibility Potential Map, Thematic Information Layers,


Factors Rating, Probability Analysis
1. Introduction
Hazards related to landslides are major societal and environmental concern. These are
responsible for considerable socio – economic losses than is generally recognized (Khire,
2004) and because of their socio economic implications as well as the scientific interest,
these can be considered as a problem of greater relevance (L. Cascini et al., 1992). The
study of landslides has drawn global attention mainly due to the increasing awareness of
its socio-economic impacts and increasing pressure of urbanization on the mountain
environment (Aleotti and Chowdhury, 1999). Landslides have represented 4.89% of the
total natural disasters that occurred worldwide during the years 1990 - 2005 (www.em-
dat.net). These are rapidly becoming the focus of major scientific research, engineering
study and practices, and landuse policy making throughout the world (Mihail Popescu,
1996). The human activities and necessities accentuate slope failures, accelerate soil
erosion and landslides (Valdiya, 1992).The development of human activity has demanded
the utilization of unstable slopes which often results in severe damage to the
constructions and residential areas (Hermosa et al., 1996).The increasing trend of
landslide occurrences will continue in future with the increased unplanned urbanization
and development, continued deforestation and increased regional precipitation due to
changing climatic patterns (Schuster,1996) and Ercanoglu and Gokceoglu, 2004).

There is hardly any kilometer of the road left without landslides. It is estimated that every
square kilometer in the fragile Himalaya shows up at least two landslide scars and one
more is added every 6 kilometer square (Bhandari, et al., 1984). Severity of the problem,
as measured by the losses or the damages caused or by the engineering complexity
involved, varies widely from location to location as well as time to time (Rao et al.,
1996). For the last three decades frequency of landslides all over hilly areas of the
country, particularly in and around developmental sites, has increased. It is therefore
required to orient the studies in much larger perspective to identify the areas vulnerable to
landslide hazards, for sustainable and risk free development. In order to cope with the
increasing demand of modern industrialized India a large number of the developmental
projects for communication, highways, dams, reservoirs etc has already started and are in
their different stages of development out of which some are being executed in and around
study area. In view of this, landslide susceptibility analysis of the area was initiated to
explore the past, present position and future projection in respect of landslide
development in the area.

2. Area of Interest and it strategic importance


Patalganga valley, the study area, between latitudes 30o 25’ & 30o 29’ and corresponding
longitudes 79o 28’ & 79o 35’ in North western Himalayan region of Garhwal, was once
heavily damaged due to devastating landslides in 1970 (Fig. 1). The heavy rain followed
by the cloudburst at the higher reaches of Patalganga generated a series of landslides along
the two tributaries of Patalganga. A huge pile of debris discharged by the river Patalganga
blocked the main river Alaknanda and built a 20m high dam (Kishor Kumar et al., 2005).
With the breaching of dam whole Belakuchi village (Fig.2) hat was located at less than a
kilometer away from the dam along with many other settlements in the downstream was
flushed within a few hours. The impact of this flashflood was so high that the entire
Srinagar city, located about 100 km downstream along Alaknanda, was heavily inundated
and was filled with silt and sand. Belakuchi (Fig. 2) that was once

used as a halting place for thousands of pilgrims and local people en-route to Badrinath
shrine disappeared from the scene. A stretch of almost 13 km of the road between
Pipalkoti and Gulabkoti through Belakuchi (Fig.2) largely got washed away and the
leftover portion was severely damaged and irreparable. To restore the road link on
emergency basis, the existing footpath of 350m situated above the damaged alignment
was converted into an alternate alignment without proper survey and mapping. This

Fig.1-Patalganga Valley

activity made the entire stable stretch into highly landslide prone zone.
Fig.2- Road through the Belakuchi
3.
Methodology

The scope, scale and methodology of landslide hazard and risk assessment can
vary significantly depending upon the aims and objectives (Chowdhury, 1999). LHZ
mapping is being carried out using qualitative or quantitative approaches. The qualitative
methods essentially depend on expert opinion in dividing an area into different zones of
varying landslide susceptibility. The preliminary studies in the field of landslide zonation
in India, comprising identification of vulnerable slopes in the Himalaya on the basis of
parameters like geology, hydrology and slopes were initiated by Oldham (1880),
Middlemiss (1890). In a simple zonation, using the techniques of overlaying relief and
geological maps, Majumdar (1980) arrived at five, very general, landslide potential ratings
for the northeastern region of India. In the Nilgiris, Southern India, Sheshagiri et al.
(1982) carried out a five-category landslide zonation by assigning landslide susceptibility
values to different factors and landslide susceptibility index on the basis of percentage of
landslides in each category. Bhandari (1994) described graded landslide hazard maps by
overlying various state of nature maps. The operative parameters used in the preparation
of Landslide Hazard Zonation maps in Northwest Himalaya on 1:50,000 scale includes (i)
geology of the area, (ii) Morphometric features of slope segments and (iii) landslide
incidences (Narula et al., 1996., Anbalgan., 1992., Gupta and Sharda., 1996., Sharda.,
1994., Kishor., et al., 1996., Sharma., 1996). Gee (1992) suggested the term ‘blind-
weighting method’ for the qualitative/semi quantitative approach in which the expert gives
subjective weights for the contributing factors and then integrate these to derive a
cumulative influence factor representing the degree of hazard. Pachauri and Pant (1992)
demonstrated a weighted landslide hazard mapping procedure in the Aglar catchment of
Himalaya and used parameter-weighting method for LHZ mapping in part of Bhagirathi
valley of Garhwal Himalaya was used by Gupta et al., (1999) and Saha et al., (2002). In
recent times, the availability of a wide range of remote sensing data along with data from
other sources in digital form and their analysis using GIS, has made it possible to prepare
different thematic layers corresponding to the causative factors that are responsible for the
occurrence of landslides in a region (Kanungo et al., 2006; Gupta and Joshi, 1990; van
Westen, 1994; Nagarajan et al., 1998; Gupta, 2003). The integration of these thematic
layers with weights assigned according to their relative importance in a GIS environment
leads to the generation of an LSZ map (Kanungo et al., 2006; Gupta et al., 1999; Saha et
al., 2002; Sarkar and Kanungo, 2004; Saha et al., 2005).

Most of the landslide hazard maps in India referred above were prepared in the
scale of 1:50,000. The working scale for most of the landslide hazard mapping programe
was not determined according to the requirements of the users, instead merely as an
academic exercise to broadly exhibit the potentiality of such maps for regional planning
(Sikdar et al., 2002). The maps, created on scale like 1:50,000 and smaller do not posses
all the information required for preparing a LHZ map, that serve the purpose of guiding
development in hilly areas. However, if one uses more detailed map and a DGPS
(Differential global positioning System), accurate mapping is possible (Lee et al 2004). In
the present case a base map of the study area was prepared in 1:12,500 scale and was used
for the preparation of rest of the thematic maps and landslide susceptibility potential map.
The base map was prepared using stereo pair of IRS –satellite’s Panchromatic (PAN) and
extensive field survey by a DGPS. The DEM (Digital Elevation Model) which is an
excellent source to derive topographic attributes responsible for landslide activity was first
generated from satellite data. To get a better accuracy, 53 GCP’s (Ground Control Points)
(Fig.3) have been selected on the DEM for locating them on the ground with the help of
DGPS. After locating the majority of points on the ground those were registered with the
DEM and subsequently the contour map of the area was generated. The DEM was also
used to derive the slope and aspect data layers Identification and mapping of a suitable set
of instability factors bearing a relationship to slope failures requires a prior knowledge of
the main causes of landslides (Guzzeti et al 1999)
.
. Fig.3- GCPs around Patalganga Landslide

To evaluate the landslide susceptibility potential for the study area various causative
factors controlling the slope stability or influencing it have been identified and prioritized
such as lithology and structure, geomorphology, stream network, climate, vegetation,
landuse land cover , existing landslides, slope angle and slope aspect. These factors have
been identified based on the knowledge of the past events and present conditions
prevailing in the area in anticipation of future behavior of slope stability. For each of the
identified factors analytical maps were prepared using LISS III (Linear Imaging Self-
scanning Sensor and Panchromatic (PAN) merged data, base map, extensive field
mapping and other available information. All the maps have been generated in digital
platform in Arc-GIS 9.0 software. Each causative factor and their classes have been
assigned appropriate rating or weightage both in terms of alphabets (qualitative) and
numerals (Quantitative) as per their role in inducing slope instability (Kishor Kumar,
1996). The final grading is provided on the basis of the sum of the total estimated
significance of each of the factors. Each of the factor maps were crossed according to the
priority assigned which has resulted into a combined map i.e. Landslide Susceptibility
Potential Map (LSPM) to evaluate the actual hazard situation in the area.
The stepwise explanation of methodology followed in brief is given as under

1. Selection of the scale for mapping work: The working scale for a landslide
Susceptibility potential Mapping (LSPM) should be determined by the requirements
of the user for whom the survey is executed. In most of the cases, in India, the scale
of LSPM is decided on the basis of availability of the base maps like topographic and
geological maps. In the present case, the maps were available only in the scale of
1:50,000. Such maps do not posses all the information required for preparing a
purposeful LSP map, which can serve the purpose of guiding development in hilly
areas. Therefore, the base map was prepared in the scale of 1:12,500.
2. The topographic map available in the scale of 1:50,000 was used as an important
reference map and Stereo pair data used for preparing a DEM (Digital Elevation
Model) of the area. The DEM was then used for preparing a contour map
supplemented and corrected by the ground GCP’s (Ground Control Points) collected
with the help of a precision GPS (Global Positioning System). The contour map was
prepared in the scale of 1:12500.
3. The factors (thematic information layers) to be studied were prioritized as per their
role in inducing the slope instability in the area.
4. Individual factor (thematic) maps were prepared from remote sensing data, generated
contour map and available topographic data.
5. Factual information about each theme was collected through extensive field
observations.
6. Final theme maps were prepared after correction of each of them based on the factual
information collected.
7. Superimposition of each factor maps was done one by one through ArcGIS software.
8. Out puts on interaction of each factors (their combinations) were derived.
9. Rating system for each of the factors and their sub factors or classes were evolved.
10. Final Landslide Susceptibility Potential map, as per the susceptibility criteria’s was
prepared.

Preparation of thematic data layers

A number of thematic data layers corresponding to the identified causative factors


namely lithology and structure, geomorphology, stream network, landuse and landcover,
slope angle and aspect have been prepared. These causative factors have both individual
and collective influence on slope instability and are considered responsible for the
occurrence of landslides in the area. Data pertaining to intrinsic (lithology,
geomorphology, slope attributes, slope aspect) as well as extrinsic factors (landuse and
landcover) have been collected from the available resources as well as from the field.
Data regarding rainfall and earthquake which are triggering extrinsic factors could not be
taken into consideration because of the unavailability of past data in relation to landslide
occurrence. A thematic layer corresponding to landslide distribution has also been
prepared to establish correlation of existing landslides and the causative factors, which
will be helpful for preparation and evaluation of the landslide susceptibility potential map
of the area.

IRS –IC- LISS III (acquired in the month of December 2002) and ID-PAN (acquired in
the month of December 2002) data along with topographic map of Survey of India in 1:
50,000 scale and the geological map of the area in 1:50,000 scale are the main sources
used to generate these thematic data layers. Preliminary survey in the month of April
-May 2003 followed by extensive field survey and mapping in 2003-2005 was conducted
to collect data pertaining to existing landslide distribution, collection of GCP’s using
DGPS for DEM validation, validation of thematic maps and also reliability checking of
the landslide susceptibility potential map in the field (Fig.4)

4.1. Landform Classification Map of Patalganga basin

Fig.4- Base Map Patalganga Basin

Geomorphology is the out shaping of the landmass as an expression of the dynamic


evolutionary processes. Broadly speaking the Patalganga basin has a few prominent
landforms such as hills of high, medium and low relief, dissected by a number of streams;
a snow covered zone in the eastern side around Kunwarikhal; thick Quaternary deposits
lying all along the drainage courses also in the form of numerous conical talus structures;
alluvial terraces alongside the rivers mostly around low altitudinal areas, deep stream
channels, and significant precipitous hill slopes (Fig. 5) largely covered by dense to
moderate forests of Alpine Texas. As discussed in previous sections landform is
classified into 6 classes which are hills of high, medium and low, relief dissected by a
number of streams; a snow covered zone, thick Quaternary deposits lying all along the
drainage courses.
Fig. 5-Landform Classification Map of Patalganga basin

The area of morphological classes varies from 0.04 sq km of River terrace to maximum
22.2 sq km of highly dissected hills and the area of landslides ranges from 0.017 sq km
against low dissected hills to 1.72054 in the class of Highly dissected hill followed by
0.539035 in Intermountain valley fill, 0.200725 in River bed, 0.159209 in Moderately
dissected hill, 0.080589 in Snow fed zone and 0.026129 in River bed. Number of
landslides against the morphological classes is as HDH- 175, Intermountain valley fill-
140, LDH-8, MDH-51, River bed-44; River terrace- 7and Snow fed zone-2.

4.2. Drainage map

Since many of the mass wasting processes begin and continues with the help of either
ground water or surface water, it becomes important to generate and analyze the stream
network and its parameters so as to understand their role in landslide occurrences in the
valley. Many of the landslides in the area occur due to drainages. Therefore a drainage
map of the area was prepared by digitizing Survey of India (SOI) topographic map in a
vector layer. Another drainage map was extracted from high-resolution remote sensing
IRS-1C LISS-ΙΙΙ + PAN MERGED data. Both the maps were compared to obtain the
Fig.6- Drainage map of Patalganga Basin

current stream network details (Fig.6). First order streams, on which the whole ordering
system linked is required to be correctly defined. The ordering has been performed on
the basis of strahler’s classification scheme (Strahler, 1964). A significant increase in
number of first order streams has been detected from the recent satellite data. On the
toposheet (of 1965) only 137 number of first order streams have been shown while the
number detected from satellite data was 473. This increase could be because of two
reasons: first, not all the first order streams had been marked on the topographic map and
secondly there was some addition of the first order streams from 1965 to 2002.
Drainages up to 5th order have been observed in the study area. 35m buffer zones on
either side of the drainages for all the drainage orders were created using the buffer
wizard under the utility tools of Arc- GIS 9.0. When landslide distribution data layer was
superimposed over the drainage buffers, it was observed that majority of landslide
occurred in the 1st, 2nd and 3rd order drainages. The buffer was created to understand the
role of drainages in relation to landslide development, however, not used as a factor in
preparation of LSPM.

4.3. Lithological Map


There was no large scale geological map available for the Patalganga basin. The regional
geology map in 1:50,000 scale only displays approximate boundaries of the geological
formations with broad geo-structural setup. The micro structures as well as the geological
formations are missing out in the map and therefore may be inaccurate or misleading.
The map of geology prepared in 1:12,500 scale gives scope to work out relationship
between the geological formation and the landslide development. Geological details can
also have important influences on the type and activity levels of slides (wills et al., 2004).
Such geological map aims at meeting the requirements of exposing the geological factors
that affect the stability of slopes. The map involves type of rock formation mapped in the
field, attitudes of bedding/foliation, presence of major structures such as faults or
lineaments. The major part of the Patalganga drainage basin lying in south and
southwestern side comprises of Lesser Himalayan metasedimentary rock sequence
dominated by dolomitic limestones and slates with quartzites and metabasics. The
sequence is thrusted on the North and Northeastern side by Higher Himalayan
Crystalline/metamorphics along the Main Central Thrust (MCT) (Auden 1935, 1937;
Heim & Ganser 1939; Ganser 1964; Gaur et.al 1977; Valdiya 1980). The entire
geological sequence is exposed between the mouth of the Patalganga valley in the WSW
to Kunwarikhal in ESE, within the Drainage Basin (Table 1 and Fig. 7). Detailed
geological traversing was done around the low lying areas in the valley and in all
approachable areas in the higher reaches. For rest of the areas the geology and structure
were interpolated in the light of satellite data interpretation. Lithology defines the types
of rock, their attitudes, their total exposure and overall existing conditions such as
weathering erosion and landsliding. Seven types of rock were identified from in the area
viz Gulabkoti quartzite, Dolostone, Metabasics, Carbonaceous shales, Quartzite,
Magnesite and Central Crystallines. Their stratigraphic and tectonic setup has already
been defined under the section of geology. Each of the rock are assigned weightage
depending on their type, structure, weathering, erosion and other characteristics: The
grading has been assigned in such a way that, strong rock, less susceptible to landslide,
comes first followed by moderate, weak and weakest, Table.2 Gulabkoti quartzite was
found comparatively more fractured than other type of quartzites available in the area.
Since fracturing is more, weathering and erosion are also proportionately developed. Out
of 13.24 km2 area of Gulabkoti quartzite, 1.04 km2 is covered by 212 landslides
(including small slumps and slips).
Table.1 Tectono- stratigraphic sequence exposed in Patalganga Basin
Kunwarikhal (ESE)
Physiographic/ Tectonic zone Rock type/ Formation Approx. thickness
(m)
Higher Higher Low to high grade +1000 (top not
Himalaya Himalayan metamorphics (schists and known)
Crystallines granites gneisses)
------------------------------------------------------------
MCT----------------------------------------------------------------------------
Quartz mylonites 200
Magnesite and dolomites 300
Gulabkoti Quartzite 1500
------------------------------ Gulabkoti Thrust
Lesser
----------------------------
Lesser Himalayan
Metavolcanics 50
Himalaya metasedimentary
Dolomitic limestones and 400
sequence
slates
Carbonaceous slates 200
Quartzites +50 (base not
exposed)
Mouth Of Patalganga River (WNW)

Lithology of Patalganaga Basin

Lithology
Carbonaceous Slates
Central Crystallines
Dolostone/ Slates
Gulabkoti Quartzite
Magnesite/ Dolomite
Metavolcanics
Quartz Mylonites
1 0 1 2 Kilometers
Quartzite

Fig. 7- Lithological Map


Some landslides are large up to 0.3 km2 while a number of others are small (0.001 km2) to
very small (0.0002 km2). Next to the Gulabkoti quartzite, Dolostone formation has 4.14
km2 area out of which 0.39 km2 is covered by 94 landslides. Metabasics formation with
2.12 km2 area has 70 landslides covering 0.39 km2 area. Carbonaceous shales with 0.83
km2 area have 54 landslides covering 0.39 km2 area. Quartzite, out of total 3.14 km2 area,
0.25 km2 is of 29 landslides. If we compare the area vs. landslides the average size of the
landslides developed in central crystalline is the largest followed in decreasing order by
quartzite, carbonaceous shale, metabasics, Gulabkoti quartzite, Dolostone and Magnesite.

4.4 Structure and Tectonics

The lesser Himalayan sequence is folded into a large domal anticline known as Pipalkoti
anticline, with a most conspicuous vertical joints/ fracture cleavage (some times having
steep dips in north or south) developed on the northern limb approaching the MCT zone
or MCT sole thrust. These steep dipping joints also appear on the air photos and satellite
images as significant linear structures. Gulabkoti thrust is another major plane of
dislocation in the area which broadly separates the underlying dominantly carbonate
sequence from the overlying largely arenaceous facies. Two right lateral strike-slip faults
namely the Patalganga fault, which follows the drainage course towards the basin mouth
and Nauligwar fault, off setting obliquely to the Patalganga fault near the confluence of
Patalganga (Semkura Nadi) and Neo-Ganeshganga, has been mapped. After identifying
and mapping these linear structures in the field, the map has been digitized in a vector
layer. 250m buffer zones on either side of the fault/thrust have been created. Landslide
distribution data layer was superimposed over the buffer zones, it was observed that
majority of landslide occurred with in the distance of 250m from the fault/thrust zone
(Fig.8).

Major Geological structures in patalganga Basin


ù ( ( N

ù
ù
(

(
ù (ù


ù
ù
O

ù
(
ù
ù

ù
ù
ù
ù

ù
ù
ù

ùù ù
O

(ùù
ù
ùù
ù

(ùù
ù

ù
ùù
ù (ù
ù
ù
(ù ù
ù
ù
(ùù
ù(
(

(
ù

ù
ù

(
ù

(
O
ù


(
ù
ù
ù
O

OOOOOOO
ù
ù
ù

(

(
OO
ù

(
ù


ù

ù
(
ù
ùù(

ù
ù

ù
ù

ù

ù
ùùù
ù
ù (

(
ù
ù

(
ù
ù

ù
ù (

ù
(
ù

ù
ù

ù
ù

ù
ù (

(
(
ù

(
ù

ù
ùù

ù
ù (

ù

ù

ù
ù
ù
(

ù
ù
(
ùùù

ù
(
ù
ù

Geological Structure
ù
(
ù

O
ù

ù
ù

ù
ù

ù
ù

ùù


ùO

ù
ù

ù
ù

(ù Anticlinal axis
ù
ù

ù
ù
ù
ù
ù
O
ù
ù
ù
ù
ù
ù

O
O
ù
ù
ù
ù
ù

ùù
O
ù
ù
ù
ù
ù
ùù

Anticlinal axis (F)


ùù
ù
ù
ù
ù
ù
ù
O
ù

ù
ù
ù
ù
ù
ù
O
ù
ù

ù

ù
ù
ù
ù

ù ùù

Gulabkoti Thrust
ù
( ù
ù

ùù
(
(

Main Central Thrust


Nauligwar Fault
$

1 0 1 2 Kilometers
$
$

$
$

Fig 8 - Major Geological Structures In Patalganga


Basin
4.5. Landslide Inventory Map

This is an important factor map comprising of the information about the existing
landslides in the area. Such type of map not only provides information about the current
situation of landslide in the area but also provides opportunity to validate and correct the
landslide susceptibility potential map. The map can also be used for knowing recurrent
landslides and also old but quiescent landslides. Distribution of landslides in inhabitant
areas, agriculture land, forest land, etc can also be seen on the map to have an instant idea
about the gravity of landslide problem and its influence on socio-economic condition of
the region. Using an inventory of existing landslides, the expert can assess the hazard of
the area by identifying regions of similar geological and geomorphological conditions
(John Mathew et al., 2007).

Since the terrain is difficult, it was not possible to map each and every landslide through
manual field survey. The mapping was accomplished by the interpretation of satellite
images supported by field checks. The satellite images are used in the identification of
landslide scars, demarcation of areas affected by landslides and areas susceptible to
landslides (Nagrajan et al., 1998, Soeters et al., 1992,). Identification of landslides on
remote sensing images is based on the spectral characteristics, shape, contrast and
morphological expression (D.P.Kanungo et al. 2006). False color composites and High
resolution multispectral/PAN ortho-image have been used to interpret and map
landslides. In most of the cases landslides are recognized due to characteristic spectral
response (mostly show up as bright pixels), typical oval/elongated shape and proximity to
the river bank/ridge or initiation point of first order stream. Many of the old slide zones
have been extracted from their triangular/crescent pattern, fresh vegetation growth and
barrenness (Champati Ray 2004).

A total of 382 landslides of varying dimensions (200 m2 to 390,000 m2) have been
identified from remote sensing images and field surveys. A majority of landslides
mapped are small having an areal extent ranging from 100 m2 to –2800m2 and are

Landslide Inventory Map of Patalganga Basin


mD m
o d i gkndahua r
P
Pa agn au mC h a n c h e t k h a l
mL a n g s i d h a r
Rupu ng
Ma h a mo l t a mR u p u n g
m P a t a l g a n gma N
Ma h a mo l t a
Pat a l ga nga Mo l t a
m Mo l t a
Chhe l i
N a u l i g wa r mC h h e l i
m N a u l i g wa r

G a
m D a r mi
n a i
mG a n a i
Tena
mT e n a

Par a r i
mP a r a r i

D a r mi y a n a
m D a r mi y a n a

Ge r u g a r h
mG e r u g a r h
K u n wa r i k h a l
m K u n wa r i k h a l

m Settlements
Landslides
Drainage

1 0 1 Kilometers

Fig.9- Landslide Inventory Map


observed mostly in rocky slopes (rock slides and rockfall) while a number of other
landslides having comparatively larger extent (100,000 m2 to 390,000 m2) have been
observed in overburden slopes. Landslides thus identified have been digitized as
polygons. This layer has been converted to a rasterized landslide distribution map (Fig.8).

4.6. Landuse landcovers Map

The landuse and management of the areas is also one of the primary concerns because
this is one factor which not only reflects the current use of the land, pattern and type of its
use but also the importance of its use in relation to the living population and its
relationship with the existing landslide development. The categories selected for landuse
landcover (Fig.9 in the current case are based on the type of vegetation and agricultural
because these are only the current landuse management categories practiced in the area.
It is generally noticed that almost all the roads in the region, including National highway
58 are suffering from the recurrent landslide problems. These Landslides are reported as
soon as they occur or recur. But the landslides which occur away from the highways
(close to the habitation) are rarely reported and are also not seriously considered for
mitigation etc. For example, as a result of this negligence, landslides which developed in
Patalganga valley during the heavy rain of 1970 gradually got enlarged in their size and
frequency and caused serious geo-environmental as well as socio-economic problems. An
entire village, a few other scattered houses and huge cultivation land (irretrievable) was
found damaged. The Inhabitants still living there are seriously considering abandoning
the area. Most of the land in the valley is used for the cultivation of seasonal crops, fruit
bearing plants and forest. The categories of landuse selected for the current case are
therefore mainly based on type of vegetation and agricultural or grassland. Five classes
of landuse have been selected as agriculture/grassland, alnus and pine, Barren/rocky land,
Deodar/mixed forest and Rhododendron forests. Area of the each class is 14.46 sq km,
4.06 sq km, 4.63 sq km, and 6.37 sq km and 19.23 sq km respectively. Against each class
there are a number of landslides viz 95,23,82,15 and 59 respectively and area of
landslides under each class as 0.47 sq km, 0.28 sq km, 0.29 sq km, 0.07 sq km and 0.39
sq km respectively. However these numbers of landslides are misleading because they do
not take into account the extent of each unit’s exposure e.g. Agriculture and grassland
spread over 14.46 km2 area where as barren slope covers only 4.63 km 2. When the
landslides that occurred in the agricultural and grassland areas are divided by the area of
exposure, the number of landslides per unit area is 6.5. This number is a better indicator
of the susceptibility of a type of landuse, agricultural and grassland (7), Alnus and Pine
(6), Barren/rocky land (20), Deodar/mixed forest (2) and Rhododendron forests (3).
Fig.10- Landuse and
Landcover Map

4.7. Digital Elevation Model (DEM) and its derivatives

DEM is a digital representation of ground surface topography or terrain. It is also widely


known as a digital terrain model (DTM). DEM can be represented in raster (a grid of
squires) or as a triangular irregular network. The use of DEM to observe topographic
attributes, such as slope aspect and steepness allows researchers to comprehensively
examine the variables affecting landslide occurrence in a study area (Sawyer et al., 2004)
DEM of the area was generated using the stereo pair (a pair of stereophotos) through
photogrametry techniques. The quality of the derived maps such as slope and aspects
depends on the quality of the DEM representation of the earth surface. In many cases
errors in this representation are neither measured nor estimated (James et al., 1997). In
the present case, to get the accurate DEM, a real time differential GPS was used to
acquire ground truth data. This ground truth data is compared with DEM generated from
stereo pair. DEM was used for determining attributes of terrain, such as slope and
aspects.

Combination of the slope angles basically defines the form of the slope and its
relationship with the lithology, structure, type of soil, drainage and the landslides.
Previous landslide mapping studies have reported the dominance topographic attributes,
such as slope angle and aspects, on
areas of instability (Aniya, 1985).
Slope Category Map Slope angle is one of the key
factors in inducing slope
instability. There are many studies
where relationships between slope
angle and landslides have been
established (P.B.Saxena et al.,
1982). The slope map derived
from DEM consists of 6 classes
with 10° interval as per the slope
classification used in other studies
Slope Categories (Kishor Kumar, et al., 1996;
0-10
11-20 Anbalagan, 1992; Gupta et al.,
21-30
31-40 1999; Dhakal et al., 2000).
1 0 1 Kilometers 41-50
51-61

Aspect is defined as a
Fig.11- Slope Category Map direction along the maximum lope
inclination and has an indirect
influence on slope instability. There is a proved
relationship between the slope aspects and Slope Aspect Category Map
development of landslides in hilly areas. Certain
N
aspects of the slopes are considered vulnerable to
landslides while some others are considered as
safe depending upon the region. Slope aspect
influences the type of vegetation, microclimate
and various other properties of slope forming
material. It was established by Perring (1959)
that, vegetation varies systematically with region,
slope and aspect. Measurements taken at
Millington Pastures in the Yorkshire Wolds Baspect
shown that the volumetric soil moisture on south- ENE
ESE
facing slopes was typically 10–20% lower than on NNE
NNW
north-facing slopes during the summers of 2001– SSE
SSW
03 (J. Bennie et al., 2006). Tanaka (2005) while 1 0 1 Kilometers WNW
WSW
studying the Amehata river basin in the Minobu
mountain belt of Southern Japanese Alps, Honshu Fig.12- Slope Category Map
Island, Japan has noticed that the South-facing
slopes of this area are clearly steeper than north-facing slopes. He further found that
many small landslides occurred on the south-facing slopes, while landslides on the north-
facing slopes were less frequent, but larger. The slope aspect map of Patalganga valley
was derived as a new thematic layer. This theme was prepared to workout the
relationship between the various aspects of the slope and the landslides in the valley. The
aspect data layer derived here represents nine classes, namely, N, NE, E, SE, S, SW,W,
NW and flat as per the classification given in other studies (Sarkar and Kanungo, 2004;
Saha et al., 2005).

5. Rating of the factors/themes


As discussed in the preceding sections, each factor/theme has been studied in relation to
landslide development in the area. This relationship has formed the basis to assign
weightings/ratings to each of the factors and its classes as per their estimated contribution
in inducing slope failures. To be simple and understandable the theme classes were listed
in ascending order as per their contribution in inducing landslides and based on the
number of inferred landslides in each class (Table 2). For example, if there are seven
classes under a theme, the class with the minimum contribution comes first, on top of the
list and given 1(minimum rating) and the class with maximum contribution comes at the
bottom of the list i.e. at seventh place and 7 (maximum rating). The final rating is
provided on the basis of the sum of the total estimated significance of each factor.
The individual weighted theme class maps were laid over one another and arithmetically
added according to the following equation to generate a Landslide Susceptibility
Potential (LSP) map in GIS.

LSP = L + M + SL + AS + LU

Where L, M, SL, AS and LU are abbreviations for the weighted thematic layers for
lithology morphology, slope, aspect and land use land respectively. A total of 489
combinations of theme classes have been formed with weights ranging from 9 to 33.
Comparing the combinations and their distribution with the distribution of the existing
landslides realistic rating ranges were evolved to categorize landslide susceptibility
potential as 9-14, 15-21, 22-27, 27-32 and 32 and above. Since all the above exercises
have been done in GIS, simultaneous, updating and verification based on the actual field
inputs was also carried out. The boundaries of the theme classes were drawn at LSP
values to obtain five susceptibility zones. The LSP map thus obtained is shown in Fig.12.
The final LSP (Landslide Susceptibility Potential) grading on LSPM has the range from I
to V. Grading “I” & “V” indicate the area least and most vulnerable to landslide
respectively. LSPM was crossed with the existing landslide map of the area. The crossed
map has indicated that majority of existing landslides come under the most to moderately
susceptible categories. This was followed by the probability analysis as indicated in
following section.

5.1. Probability analysis


The LSPM was later verified by doing the probability analysis of the selected factors and
their classes. For probability analysis, area of each factor and active landslides (existing)
in each class of selected factors was calculated (Table 2). Probability value Pij = Nlsij /
Nlsij was calculated for each class (i) and of each factor (j). The total probability was
analyzed for all classes of selected factors using the equation by Gulakiyan G.A., kuntsel
V.V.,Postoev G.P. (1977)
t
Wls=1− Π  1− P k −−−−−−−−−−−− 1                        
k=1
t
∑ i , ¿j=1 P ij −−−−−−−−−−−−−−−−− 2
P k  = ¿ P ij = Nls ij / N ij −−−−−−−−−−−−−−− 3
Where
Wls - Geotechnical potential of a landslide,
Pk - Total probability of all classes of factors
Pij - Probability for classes i of factor j,
Nlsij - Area of landslide within the limit of a class
Nlsi - Area of a class
Landslide potential values have been derived using above procedure range from 0.20 to
0.90. This range was further divided into four sub ranges viz 0.20 -40, 0.41-0.60, 0.61-
0.80, 0.81 and above. Using these values, landslide potential value map has been
obtained. On overlapping Landslide distribution map over the Landslide potential value
map it was found that most of the landslides fall under the value category III followed IV,
II and I. By comparing the total probability value of each factor indicated in Table 2 with
the landslide potential value, geomorphological factors indicate maximum probability
value (90.0) and amongst its classes, river terrace, has the maximum value (0.75), which
is rightly so because most of the landslides in the area have occurred along the river
course in the terrace formations having high degree of development including road,
houses, agriculture etc.

Fig.13- Landslide Susceptibility


Potential Map
Factors Class (i) rati Probability Analysis
(j) Characteristic ngs
s Area of Probab Total
ility Proba
bility
Cla Landsl Pij Pk
ss ide (Nlsij/N
(Nls (Nlsij) lsi)
i)
Lithology (L) Central a(1) 11.3 0.08 0.007 0.529
Crystalline b(2) 2 0.02 0.005
Magnesite/Dolo c(3) 3.34 0.25 0.080
mite d(4) 3.14 0.39 0.125
Quartzite, e(5) 3.11 0.43 0.156
Carbonaceous f(6) 2.75 1.04 0.062
slates g(7) 16.6 0.39 0.094
Metavolcanics, , 8
Gulabkoti 4.14
quartzite,
Dolostone,
Landform River terrace h (1) 0.33 0.30 0.750 0.909
Classificatio Low dissected i(2) 0.68 0.02 0.029
n (M) hills j(3) 0.72 0.20 0.278
River bed k (4) 8.17 0.08 0.010
Snow fed zone l (5) 5.72 0.54 0.094
Intermountain m 11.3 0.16 0.014
valley fill (6) 3 1.07 0.048
Moderately n(7) 22.1
dissected hill 7
Highly dissected
hills
Slope Angle( 51-60 o(1) 0.31 0.04 0.127 0.452
SL) 0-10 p(2) 4.54 0.28 0.062
11-20 q(3) 5.58 0.49 0.088
41-50 r(4) 10.8 0.53 0.049
21-30 s(5) 0 0.76 0.071
31-40 t(6) 10.7 1.03 0.055
8
18.5
9
Land use Deodar/mixed u(1) 6.37 0.08 0.013 0.203
and land forest v(2) 4.06 0.29 0.071
cover (LU) alnus and pine, w(3) 19.2 0.40 0.021
Rhododendron x(4) 34.6 0.30 0.065
forests y(5) 3 0.47 0.033
Barren/rocky 14.4
land, 6
agriculture/gras
sland,
Structure & Buffer Distance Not taken for probability analysis
Tectonics from T/F z)1
(T-Thrust, F- 1.0-1.5 z1)2
fault) 0.5-1.0 z2)3
0-0.5
Slope ESE z3)1 1.19 0.04 0.034 0.384
Aspects NNE z4)2 3.91 0.13 0.033
(SA) SSE z5)3 2.53 0.30 0.119
NNW z6)4 7.23 0.22 0.030
WSW z7)5 6.12 0.31 0.051
SSW z8)6 6.07 0.28 0.046
ENE z9)7 12.1 0.37 0.030
Table. 2 Probability analyses of selected factors
and their classes

6. Conclusions

In the present study, simple conventional based method of subjective weighting has been
applied for Landslide Susceptibility Potential Mapping in Patalganga basin, part of North
Western Himalayan region of Garhwal. Probability analysis was carried out for verifying
the LSP map. Although the method applied for such mapping was simple, result
produced proved accurate as may be attributed to the following reasons.

1) Landslide susceptibility Potential Map prepared indicates that most of the slopes in
the valley fall in High to moderate susceptibility potential category.
2) A large part of the land in these categories is found under agricultural activities which
are indeed susceptible to sliding.
3) There are a few settlements including the partially damaged village “Ganai” indicated
in the LSP map under the high susceptible potential category which has been verified in
the field.
4) A total of 386 landslides have been encountered in the valley out of which a few are
of moderate size as that of Patalganga landslide where as others are in between minor
to moderate mostly indicated under High to moderate susceptibility potential
category.
5) During 1970, when a series of landslides occurred and devastation took place due to
unprecedented rainfall, the topographical, geomorphological and geological
conditions were different as the area was less exploited by human being, the
vegetation was comparatively more resulting in minimum erosion, the rocks were less
exposed and weathering was least. In the present situation where a major part of the
area falls under the high and moderate vulnerable class, the slopes are more dissected,
eroded and weathered with many existing landslides, if similar amount of
precipitation that occurred during July 1970 replicates, a bigger tragedy may strike as
the existing terrain conditions are extremely bad as compared to what had been 34
years back.

References

[1] Aleotti, P., Chowdhury, R., 1999. Landslide Hazard Assessment: Summary, Review
and New Perspectives. Bull. Eng. Geol. Environ.58, 21–44.
[2] Aniya, Masamu. 1985. Landslide- Susceptibility Mapping in Amahata River Basin,
Japan. Annals of the Association of American Geographers 75(1): 102-114.
[3] Anbalgan, R., 1992. Terrain evaluation and zonation mapping in mountainous terrain.
Engineering Geology 32, 269-277.
[4] Auden, J.B., 1935. Traverses in the Himalaya. Records of Geological survey of India.
Vol. 69 (1), pp.123-167
[5] Auden, J.B., 1937. Structure of the Himalaya in Garhwal. Records of Geological
survey of India. Vol. 71, pp. 407-433.
[6] Bhandari, R.K., Mehrotra, G.S., Nainwal, H.C. and Raiwani, K.K., 1984. Hill Roads
and Himalayan Landslides. Proceedings of Seminar on Construction of Roads in Hill
areas. Indian Roads Congress, Nainital, pp.123-141
[7] Bhandari, R.K., Weeraninghe, K., 1996. Pitfalls in Subrogating Slope Maps for
Landslide Hazard Maps. Proceeding of Asian Conference on Remote Sensing at
Colombo, Sri Lanka.
[8] Bhandari, R.K., 1994. Landslide Hazard Mapping in Sri Lanka - a Holistic Approach.
Proceeding of National Symposium on Landslides in Sri Lanka, PP: 271- 284
[9] Champati, Ray. 2004. Landslide Hazard Zonation Using Spatial Models in GIS. In:
Nagrajan, R., (Ed), Landslide Disaster: Assessment and Monitoring. Anmol.pp 81-90
[10] Chowdhury, R.N., 1996. Aspects of Risk Assessment for landslides. Proceeding
of the 7th Int. Symp on Landslides, Trondhem, Norway. Balkema, Rotterdam pp.
[11] Chowdhury, R.N., Flentje, P., Ko Ko, C., (1999). Aspects of landslide research at
University of Wollongong. Proceedings of Ninth International Conference and Field
Trip on Landslides, Bristol, U.K. pp. 17-25.
[12] Chowdhury, R.N., 1999. Landslide Hazard and Risk Assessment – An overview,
invited paper to the international Workshop cum training programme on Landslide
Hazard and Risk Assessment and Damage Control for Sustainable Development,
CSIR, New Delhi, India
[13] Dai, F. C., Lee, C. F., Ngai Y. Y., Landslide risk assessment and management an
overview. Engineering Geology, Volume 64, Issue 1, April 2002, pp 65-87.
[14] Dhakal, A.S., Amada, T., Aniya, M., 2000. Landslide hazard mapping and its
evaluation using GIS: an investigation of sampling schemes for a grid-cell based
quantitative method. Photogramm. Eng. Remote Sensing 66 (8), 981–989.
[15] Ercanoglu, M., Gokceoglu, C., 2004. Use of fuzzy relations to produce landslide
susceptibility map of a landslide prone area (West Black Sea Region, Turkey). Eng.
Geol. 75 (3&4), 229–250.
[16] Ganser, A., 1964. Geology of the Himalayas. Wiley Interscience, N.Y., 284.
[17] Gaur, G.C.S., Dave, V.K.S., and Mittal, R.S., 1977. Stratigraphy, structure and
tectonics of the carbonate suite of Chamoli, Garhwal Himalaya. Journal of Himalayan
Geology. Vol.7, 416-421
[18] Gee, M. D., Classification of landslide hazard zonation methods and a test of
predictive capability. In Proceedings of the 6th International .Symposium on
Landslides, Vol. 2, Christchurch, New Zealand, 1992, pp. 947–952.
[19] Gupta, R.P., Joshi, B.C., 1990. Landslide Hazard Zonation using the GIS
Approach—A case Study from the Ramganga Catchment, Himalayas. Eng. Geol. 28,
119–131
[20] Gupta, R.P., Saha, A.K., Arora, M.K., Kumar, A., 1999. Landslide Hazard
Zonation in a part of Bhagirathi Valley, Garhwal Himalayas, using integrated Remote
Sensing—GIS. J. Himalayan. Geol. 20 (2), 71–85.
[21] Gupta, S.K., Sharda, Y.P., 1996. Landslide zonation- a review and suggested
approaches. Proceeding of the 7th Int. Symp on Landslides, Trondhem, Norway.
Balkema, Rotterdam pp.
[22] Gupta, R.P., 2003. Remote Sensing Geology, 2nd Edition. Springer-Verlag,
Berlin Heidelberg, Germany.
[23] Guzzetti, F., Carrara, A., Cardinali, M. and Reichenbach, P., 1999. Landslide
hazard evaluation: a review of current techniques and their application in a multi
scale study, Central Italy. Geomorphology, 31: 181-216.
[24] Heim, A., Ganser, A., 1939. Central Himalaya, geological observations of the
Swiss expedition in 1936. Memoir. Soc. Helv. Sci. Nat., vol. 73, pp.1-245.
[25] J. Bennie., Mark O. Hill., Robert Baxter and Brian Huntley., 2006. Influence of
slope and aspect on long-term vegetation change in British chalk grasslands. Journal
of Ecology 94, 355–368
[26] John, Mathew, V. K., Jha, G. S., Rawat., 2007. Weights of evidence modelling
for landslide hazard zonation mapping in part of Bhagirathi valley, Uttarakhand.
Current Science, Vol. 92, N0. 5.
[27] Kanungo, D.P., M.K. Arora., S. Sarkar., R.P. Gupta, 2006. Comparative study of
conventional, ANN black box, fuzzy and combined neural and fuzzy weighting
procedures for landslide susceptibility zonation in Darjeeling Himalayas. Engineering
Geology 85 (2006) 347–366.
[28] Khire, M.V., 2004. Terrain Evaluation and Mitigation for Landslides. In:
Nagrajan, R., (Ed), Landslide Disaster: Assessment and Monitoring. Anmol.pp 37-56
[29] Kishor Kumar., Tolia D.S., Kumar Satish., 1996. Landslide Hazard Evaluation in
a part of Himalaya. Proceeding of the 7th Int. Symp on Landslides, Trondhem,
Norway. Balkema, Rotterdam pp.
[30] Kumar Kishor., Sati, D., 2005. Exploring the History of Alaknanda - Patalganga
Tragedy of 1970 & Possibility of its Recurrence and Impacts on Patalganga Basin – A
GIS and Remote Sensing Based Study. Proceedings of the 8th Annual Int. Conf., Map
India, 2005, New Delhi,
[31] Lattman, L.H. and Parizek, R.R.1964. Relationship between fracture traces and
the occurrence of groundwater in carbonate rocks. Journal of Hydrology 2: 73-91.
[32] Lee, S., Ryu, J., Won, J., Park, H., 2004. Determination and application of the
weights for landslide susceptibility mapping using an artificial neural network. Eng.
Geol. 71, 289–302.
[33] Majumdar, N., 1980. Distribution and intensity of landslide processes in North
Eastern India – A zonation approach. Proceedings of International Symposium on
Landslides, New Delhi. V.1, pp 3-8.
[34] Middlemiss, C.S., 1890. Geological sketch of Nainital with some remarks on
natural conditions governing the mountain slope. Records of Geological survey of
India. Vol.21, pp.213-234.
[35] Mihail, Popescu., 1996. From landslide causes to landslide remediation.
Proceeding of the 7th Int. Symp on Landslides, Trondhem, Norway. Balkema,
Rotterdam pp. 75-96
[36] Nagarajan, R., Mukherjee, A., Roy, A., Khire, M.V., 1998. Temporal remote
sensing data and GIS application in landslide hazard zonation of part of Western
Ghat, India. Int. J. Remote Sens. 19, 573–585.
[37] Narula, P.L., Gupta, S.K., Sharda, Y.P., Singh, J., 1996. Crustal adjustments and
related landslide hazard. Proceeding of the 7th Int. Symp on Landslides, Trondhem,
Norway. Balkema, Rotterdam pp. 995-1000.
[38] Oldham, R.D., 1880. Note on Nainital Landslides. Records of Geological survey
of India. V. 13, PP 277-281.
[39] O’Leary, D.W. Freidman, J.D., and Pohn, H.A. 1976. Lineament, linear,
lineation: Some
proposed new definitions for old terms. Geological Society of America Bulletin. 87:
1463-1469.
[40] P.B.Saxena., P.C.Bahukhandi., B.K.Paney., 1982. A study of landsliding and its
land depletion in the Alaknanda valley of Garhwal Himalaya. Him. Geol., 5, 716-723.
[41] Pachauri, A.K., Pant, M., 1992. Landslide Hazard Mapping based on Geological
Attributes. Engineering Geology, 32, pp 81-100.
[42] Perring, F., 1959. Topographical gradients of chalk grassland. Journal of Ecology,
47, 447–481.
[43] Rao P.Jagannatha., Kishor Kumar., Jai, Bhagwan. 1992. Prognostication of
Landslides in Himalaya. Proc. of Indian Geotechnical Conference, 18-20 Dec. 1992,
Calcutta. Pp 467-470
[44] Saha, A. K., Gupta, R. P. and Arora, M. K., GIS-based landslide hazard zonation
in the Bhagirathi (Ganga) valley, Himalayas. Int.J. Remote Sensing, 2002, 23, 357–
369.
[45] Saha, A.K., Gupta, R.P., Sarkar, I., Arora, M.K., Csaplovics, E., 2005. An
approach for GIS-based statistical landslide susceptibility zonation—with a case
study in the Himalayas. Landslides 2, 61–69.
[46] Sarkar, S., Kanungo, D.P., 2004. An integrated approach for landslide
susceptibility mapping using remote sensing and GIS. Photogramm. Eng. Remote
Sensing 70 (5), 617–625.
[47] Schuster, R., 1996. Socioeconomic significance of landslides. In: Turner, A.K.,
Schuster, R.L. (Eds.), Landslides: Investigation and Mitigation: Special Report, vol.
247. National Academic Press, Washington, DC, pp. 12–36.
[48] Seshagiri, D.N., Badrinarayan, S., 1982. The Nilgiri Landslides Geological
survey of India. Miscellaneous Publication. 57.
[49] Sharda, Y.P., 1994. Landslide hazards zonation in parts at Chenab basin (J & K).
Records of Geological Survey of India. Vol. 127, pp.110-114.
[50] Sharma, V.K., 1996. A Probabilistic approach for landslide zonation in parts of
Garhwal Himalaya. Proceedings of 7th International Symposium on Landslides,
Trondhem, pp. 381-388.
[51] Sikdar P.K., Kishor Kumar., Goyal Nitesh., Janarthanan, G., 2002. Critical
Review of landslide hazard zonation practices and direction for future. Proceedings of
Indo-China workshop on Natural Disaster, Central Road Research Institute, New
Delhi, pp 54-75.
[52] Soeters, R., Renger, N. and Van Westen ., C.J.,1991. Remote sensing and GIS
applied to mountain hazard analysis and environmental monitoring, in proceedings of
8th Thematic Conference on Geological remote sensing, Denver, Colorado, pp 1389-
1402.
[53] Strahler, A.N., 1964. Quantitative geomorphology of basins and channel
networks. In: Chow, V.T. (Ed.), Handbook of Applied Hydrology. Mcgraw Hill, New
York.
[54] Sawyer, Carol F., David R. Butler., 2004. Landslide Aspect: A methodological
approach to circular data for hazard analysis. Proceedings of the Applied Geography
Conference 27, St. Louis, M.O, 67 -74.
[55] Takagi, Masataka, (2006) Accuracy of Digital Terrain Model According to
Spatial Resolution. In: D.Fritsch, M. Englich and M. Sester (Eds), ‘IAPRS’, Vol 32/4,
ISPRS Commission IV Symposium on GIS – Between vision and applications,
Stuttgart, Germany.
[56] Tanaka, Y., 2005. Differences of Landslide Occurrences Behavior Due to Slope
Aspects in the Amehata River Basin, central Japan. American Geophysical Union,
Fall Meeting 2005, abstract #H51C-0378
[57] Valdiya, K.S., 1980. Geology of Kumaon Lesser Himalaya. Journal of Himalayan
Geology, Wadia Institute of Himalayan Geology, Dehradun. pp 291.
[58] Valdiya, K.S., 1992. Environmental Problem in Himalaya: Geological Aspects.
In: Himalayan environment and Development- Problem and Perspectives. Gyanodaya
Prakashan, Nainital pp- 161
[59] Van Westen, C.J., 1994. GIS in landslide hazard zonation: a review, with
examples from the Andes of Colombia. In: Price, M., Heywood, I. (Eds.), Mountain
Environments and Geographic Information System. Taylor & Francis, Basingstoke,
pp. 135–165.
[60] Varnes D. J., 1984. Landslide hazard zonation: a review of principles and
practice. Int. Assoc. of. Engg.Geol. Commission on landslides and other mass
movements on Slopes.
[61] www.em-dat.net EM-DAT: The OFDA/CRED International Disaster Database.
Universite Catholique de Louvain, Brussels, Belgium.
[62] Wills, Chris J., Manson, Michael W., Wagner, David L., 2004. Geological and
Landslide Mapping along Highway Corridors. 2004 Denver Annual Meeting, paper
No. 249-29. (site)
[63] William, M. BrownIII., 1992. Information for Disaster Reduction: The National
Landslide Information Centre, U.S Geological Survey. Proceedings of the Sixth
International Symposium on Landslides, Vol.2, Christchurch, New Zealand.
[64] Gulakiyan G.A., kuntsel V.V.,Postoev G.P. (1977). Prognozirovaniye
opolznobhikh protsessov ( Prediction of landslide processes), Text book, Ministry of
geology, Moscow, Russia, pp 135.

You might also like