Recherche Desbord

Technical papers in hydrology 15
In this series
1 Perennial Ice and Snow Masses. A Guide for Compila-
tion and Assemblage of Data for a World Inventory.
2 Seasonal Snow Cover. A Guide for Measurement,
Compilation and Assemblage of Data.
3 Variations of Existing Glaciers. A Guide to Inter-
national Practices for their Measurement.
4 Antarctic Glaciology in the International Hydrological
Decade.
5 Combined Heat, Ice and Water Balances at Selected
Glacier Basins. A Guide for Compilation and Assem-
blage of Data for Glacier Mass Balance Measurements.
6 Textbooks on Hydrology—Analyses and Synoptic
Tables of Contents of Selected Textbooks.
7 Scientific Framework of World Water Balance.
X Flood Studies—An International Guide for Collection
and Processing of Data.
9 Guide to World Inventory of Sea, Lake, and River Ice.
10 Curricula and Syllabi in Hydrology.
11 Teaching Aids in Hydrology.
12 Ecology of Water Weeds in the Neotropics.
13 The Teaching of Hydrology.
14 Legends for Geohydrochemical M a p s .
Legendes des cartes hydrogéochimiques.
Leyenda para mapas geohidroquímicos.
Jlerenna JUIA reorHApoxHMHiec¿HX xapT
15 Research on Urban Hydrology, vol. I.
A contribution to the
International Hydrological Programme
Research
on urban hydrology
Volume I
State-of-the-art reports from
Australia, Canada, U . S . S . R . ,
United Kingdom, U . S . A .
General editor : M . B . McPherson
mnesoo
The designations employed and the presentation of the
material do not imply the expression of any opinion what-
soever on the part of Unesco concerning the legal status
of any country or territory, or of its authorities, or con-
cerning the frontiers of any country or territory.
Published in 1977 by the

United Nations Educational, Scientific and Cultural
Organization
7, place de Fontenoy, 75700 Paris
Printed by Imprimerie Beugnet, Paris
I S B N 92-3-101488-9
© Unesco 1977
Printed in France
Preface
The 'Technical Papers in Hydrology' series, like the related collection of 'Studies and Reports in
Hydrology', w a s started in 1965 when the International Hydrological Decade was launched by the
General Conference of Unesco at its thirteenth session. T h e aim of this undertaking was to promote
hydrological science through the development of international co-operation and the training of specia-
lists and technicians.
Population growth and industrial and agricultural development are leading to constantly increa-
sing demands for water, hence all countries are endeavouring to improve the evaluation of their water
resources and to m a k e more rational use of them. T h e I H D was instrumental in promoting this
general effort. W h e n the Decade ended in 1974, I H D National Committees had been formed in 107
of Unesco's 135 M e m b e r States to carry out national activities and participate in regional and inter-
national activities within the I H D programme.
Unesco w a s conscious of the need to continue the efforts initiated during the International
Hydrological Decade and, following the recommendations of M e m b e r States, the Organization
decided at its seventeenth session to launch a n e w long-term intergovernmental programme, the
International Hydrological P r o g r a m m e (IHP), to follow the decade. The basic objectives of the I H P
were defined as follows: (a) to provide a scientific framework for the general development of hydro-
logical activities; (b) to improve the study of the hydrological cycle and the scientific methodology
for the assessment of water resources throughout the world, thus contributing to their rational use;
(c) to evaluate the influence of m a n ' s activities on the water cycle, considered in relation to environ-
mental conditions as a whole; (d) to promote the exchange of information on hydrological research
and on new developments in hydrology; (e) to promote education and training in hydrology; (f) to
assist M e m b e r States in the organization and development of their national hydrological activities.
The International Hydrological Programme became operational on 1 January 1976 and is to
be executed through successive phases of six years' duration. I H P activities are co-ordinated at the
international level by an intergovernmental council composed of thirty M e m b e r States. The members
are periodically elected by the General Conference and their representatives are chosen by national
committees.
The 'Technical Papers in Hydrology' series is intended to provide a means for the exchange of
information on hydrological techniques and for the co-ordination of research and data collection.
In order to co-ordinate scientific projects, however, it is essential that data acquisition, transmission
and processing be conceived in such a w a y as to permit the comparison of results. In particular, the
exchange of information on data collected throughout the world requires standard instruments, tech-
niques, units of measurement and terminology.
It is believed that the guides on data collection and compilation in various specific areas of
hydrology which have been published in the 'Technical Papers in Hydrology' series have already
helped hydrologists to standardize their records of observations and thus have facilitated the study
of hydrology on a world-wide basis.
M u c h still remains to be done in thisfield,however, even as regards the simple measurement of
basic elements such as precipitation, snow cover, soil humidity, run-off, sediment transport and
ground-water phenomena.
Unesco therefore intends to continue the publication of 'Technical Papers in Hydrology ' as an
indispensable means of bringing together and making k n o w n the experience accumulated by hydro-
logists throughout the world.
Contents
Foreword 9
Urban hydrological modelling and catchment research in U . S . A . 11

Section I Introduction 13
Section 2 Urban hydrological modelling 15
Section 3 Urban catchment research 41
Urban hydrological modelling and catchment research in Australia 65

Section I Field research 67
Section 2 Hydrological models 76
Section 3 Applications 80
Section 4 Conclusions 87
Urban hydrological modelling and catchment research in Canada 89

Section 3 S u m m a / y and conclusions 118
Appendix I Selected Canadian urban test catchments and s u m m a r y of urban
run-offfieldstudies of limited scope 125
Appendix II Instrumentation considerations 137
Methods for calculating m a x i m u m flood discharges for natural watercourses and

urban areas in the U . S . S . R . 143
Computation of peak flow 145
Application of precipitation data 145
Use of run-off coefficients 150
Flood travel time 150
Flood volumes andfloodhydrographs 153
Flood frequency 158
Urban hydrological modelling and catchment research in the United K i n g d o m 161

Section 1 General introduction 163
Foreword
This volume has been prepared under International Hydrological Programme sub-project 7.1,
'Research on Urban Hydrology'. It contains thefirstfiveof the national state-of-the-art reports
foreseen within the general framework of project 7, 'Effects of Urbanization on the Hydrological
Regime and on Quality of Water', adopted by the Intergovernmental Council of the International
Hydrological P r o g r a m m e at its 1975 session. T h e project includes also sub-project 7.2, 'Development
of Mathematical Models applied to Urban Areas considering both Water Quality and Quantitative
Aspects'.
T h e Bureau of the International Council of the International Hydrological Programme appointed
M r . M . B . McPherson ( U . S . A . ) as rapporteur of sub-project 7.1 at itsfirstsession in August 1975,
and defined the tasks to be carried out under the sub-project as follows:
1. T o prepare 'state-of-the-art' reports reviewing current research on urban hydrology based on case
studies; the reports are to include descriptions of laboratory andfieldstudies, instrumentation,
methods of processing and analysing information from experiments, development of models on
urban water systems and urban planning, water quantity and quality processes, etc.;
2. T o prepare outlines of the information manuals on urban water data collection, analysis and use.
This endeavour originated from activities and aspirations of the Unesco Subgroup on the Hydro-
logical Effects of Urbanization of the International Hydrological Decade. Part I of the Subgroup
final report1 entitled 'International S u m m a r y ' , was resolved by representatives of over thirty nations
w h o participated in an International W o r s h o p at W a r s a w , Poland, N o v e m b e r 19731 at which ten
crucial international research projects were proposed for inclusion in the Unesco component of the
International Hydrological Programme.
The American Society of Civil Engineers ( A S C E ) took early supportive action by applying for
an N S F grant to assist in two of the ten recommended projects: R.l. Catchment Studies Report,
and R . 3 . Mathematical Models Report. In April 1975, the International Council for the I H P adopted
I H P Project 7, which includes the two subjects in question and thus m a d e it possible for Unesco and
the A S C E to co-operate closely on the state-of-the-art reports.
O f particular significance was the very strong emphasis of the W a r s a w Workshop and the Sub-
group on the urgency of addressing all such reports to users of researchfindings.That is, an accentua-
tion of user participation and user orientation of I H P urban products clearly indicated that facili-
tation of the translation of research findings into implementation practice should be a central goal.
In most countries, economic growth, population growth, non-agricultural water use and pollu-
tion are intertwined. Water in its m a n y manifestations plays a vital role in the extremely complex
processes of urbanization, and thus affects a nation's health and growth. The most significant con-
clusion reached by the I H D / U n e s c o Subgroup is that most urban hydrological problems and effects
are similar in technologically and economically advanced countries. Further, m a n y problems confron-
ting the developing nations have at one time or another already been encountered by m a n y deve-
9
loped nations. This strongly suggests that great benefits would result from the exchange of infor-
mation and increased international co-operation in research and development.
T h e report for the U . S . A . served as the prototype for the series of national reports, and w a s
circulated to experts in other countries for the purpose of obtaining analogous contributions.
Unesco wishes to thank the authors for assembling the reports presented in this volume, which
should facilitate international communication on the state of the art in urban hydrology. Additional
volumes will be issued in the future, as further contributions to sub-project 7.1.
References
1. Unesco, HydrologicalEffects of Urbanization, Paris, Unesco, 1974, 280 p. (Studies and Reports in Hydrology 18).
2. A S C E , 'Report, International Workshop on the Hydrological Effects of Urbanization, Warsaw, 1973'. N e w York,
N . Y . , N S F , January 1974, 61 p.
10
Urban hydrological modelling and catchment
research in U . S . A .
M . B . McPherson
American Society of Civil Engineers
345 East 47th Street
N e w York, N . Y . 10017, U . S . A .
Section 1 Introduction
Scope of This Report
In keeping with the findings of the Subgroup and the Warsaw Workshop,
modeling and catchment research for urban drainage systems is emphasized. This is
the subject singled out as having the largest gaps in knowledge in urban hydrology.
Water quality aspects are accorded considerable attention because of the strong
interest in environmental protection in the U.S. The issue has been succinctly
stated: "When a city takes a bath, what do you do with the dirty w a t e r ? " ^ '
Precipitation removes considerable amounts of particulates from urban areas. As
an indication of the extent of this potential burden, it has been estimated that
the 330-km' of the City of Philadelphia, with a population of two million people,
typically produces, in metric tons: 2,900-tons/day of pollution emissions;
780,000-tons/year of trash; 82,000-tons/year of garbage; 560,000-tons/year of
incinerator residue; 24,000-tons/year of debris from inlets; and 83,000-tons/year
of street sweepings.'^' In addition: Philadelphia replaces 30-kra to 40-km of
asphalt paving per year; surfaces collect the wear from four million pair of
shoes; there is erosion from the 650,000 buildings in the City; the 250,000 animals
contribute their droppings; there is erosion from over two million vehicle tires;
e t c . ^ ' A large share of these numerous residuals is washed away from the land
surface and is transported to receiving bodies of water. This is over and above
the contributions from municipal and industrial wastewater treatment plant
effluents.
Figure l' 5 ' is a simplified description of the stormwater and wastewater

portion of the urban water resource system. This report emphasizes the quantity
and quality aspects of the stormwater subsystem therein, described in Figure 1,
Section 2.
References
1. Field, Richard, and John A. Lager, "Urban Runoff Pollution Control - State-
of-the-Art." J.Envir.EngrR.Div., ASCE P r o c , Vol. 101, No. EE1, pp. 107-125,
February, 1975.
2. Radziul, Joseph V., "City Pollution Loading Potential," pp. 12-13, Urban
Runoff. Quantity and Quality, ASCE, New York, N.Y., 1975.
3. Sonnen, M. B., Larry A. Roesner and Robert P. Shubinski, "Urban Water

Management Models," pp. 89-97, Urban Runoff. Quantity and Quality, ASCE,
New York, N.Y., 1975.
13
eujouj—>—sc-3 fcî
-IO
CE-
O Lü
CL
ce w
>-
Lu co
V)
L U LU
O
I- CC
I O
w
LU
CC
ce
ce LU
LU h-
<
5
ce <
o m
h-
cn ce
3
LU
ce
Lu
14
Section 2 Urban hydrological modelling
Introduction
Shown schematically in Figure l' 1 ' is a simplified description of the

major components of the urban stormwater disposal sub-system.
Natural watercourses occupy much less land area than that drained directly
by systems of underground drainage conduits. It has been estimated that approximately
one-sixth of U.S. "Urban Areas" (measure of metropolitan areas adopted for the 1970
Census) fall within natural 100-year flood plains,^ 2 ' whereas well over half of
Urban Areas are drained by system's of underground conduits. Further, national
investment for storm drainage conduit facilities appears to be more than four times
as great as that for flood plain protection works benefiting urban areas.
There is widespread interest in multi-purpose drainage facilities that

exploit opportunities for water-based recreation, provide more effective protection
of buildings from flooding, and allow for the use and re-use of storm water for
water supply. In addition, the 1972 Amendments to the Federal Water Pollution Control
Act, which established a national goal of zero-pollution, has led to considerable
interest in reducing the entry of pollutants into receiving waters from combined
sewer overflows and storm sewer discharges. All these emerging requirements call for
the use of storage facilities and special treatment works, together with some kind
of control system able to manage the sudden and brief impact of stormwater. In
sum, over the last few years urban model development has greatly intensified, with
sewered catchment models perhaps eclipsing urban receiving-water model development.
Because of this and of the tendency to use tailor-made or custom-adapted models for
urban strearaflow discharge-quality, discussion of models for simulation of under-
ground conduit system performance will predominate.
Anticipating the possible future availability of new field data, an ASCE

study team in 1968 defined considerations for modeling sewered urban catchment
rainfall-runoff-quality processes^-3' and at the same time considerations for
characterizing rainfall time and spatial distributions in future research were
explored.^'
Space does not permit presentation of a more comprehensive background,

and the reader is referred to a previous summary for a much broader coverage.^-*'
Also, problems in modeling urban catchments have been discussed elsewhere,(°) and
findings have been reported of a general study of stormwater detention usage in
the U.S.A.''' Eight lectures on urban runoff given at a training course are in-
cluded in a report'"' that is suitable for use as a text or manual in addition to
its value as a unique information source.
*: Numbered references are cited at the end of this Section. Sections 2 and 3 have
been made as self-contained as possible, with their own reference lists. As a
result, there is some duplication. The phrase "available from NTIS" in a number
of entries refers to the National Technical Information Service, U.S. Department
of Commerce, Springfield, Virginia 22151, from which reprints are available for
a cost-recovery charge.
15
K O -,
ï
r
co"
z
RIVERS
WATER
RECEIVI
LAKES,
BODIES
(STREA
OCEAN
NSEW ERED LAND
— » •
Lü
Lu
O CO
>-
CO
<TÜ CO
î 'I
CD
CO
AINAGE
ANNEL
TURAL
N-MAD
ANNEL
PROVE
1
z <
2
— » •
Q \1« m"
LU O
2 5 X
Z)
rr
ll U.
o o
cr
o < - z o 1 CO
ÜJ 1 ""7
1 LU
CE
>
X
•
>
z. o 1 LU
CO ~1 CE û.
o z * | LU
FLO^
>•. ^
1 tu til <í =î LU _J < H <
1- #» ^ _J LU <
. < i
o rr U. - S E 1 cr
RFA(
CO
(PONDS
NDW
PRE IPITA
KES
ELIBE
ORA
< CE
» N • O
LU CE 5 Q 2 < o
, Z3 y-
1 coco
<i
_i ^ et c? I O CE
Z O 1
r • 3 LU
LO-
CO
o o O
CE _l Q
1 — .J O m
"T ça cr
5 o Lü
COLLECT!«
STORAGE
SYSTEMS
GROUND
(SE\A'ERS LU <
OUTSIDE
o
FLOWS
UNDER-
UNDER-
z Z
-) z
AND
o g cr
CE o
ID
CE h-
a. CO
^
co
I z
z < <
cr
WERED
o
ENTS
CD
REET
TCH-
CO
5 CE n t-
LU
Q. cr
n LU z 3
i > < i- 2 I LU IS LU
u. O _j (O co o
LU
LU tu CE cr
LU =5
ZD O
û_ cr < O
¡2 tu z
cr Ü tr
<
CO o 5 oco
CE
16
Types of Models
Mathematical models used for the simulation of urban rainfall-runoff or

rainfall-runoff-quality can be divided into three distinct categories: planning
models, design/analysis models and operations models.
Planning Models are used in macro-scale applications, such as for

metropolitan or city-wide master plans. As an example of the scale encountered,
the City of Milwaukee has 2,200-km of separate storm drains and combined sewers
within the 246-kra2 of the City,^*' and these conduits are distributed over 465
drainage catchments having a maximum size of 740-ha and a median size of 10-ha.^^'
When dealing with so many components the model used must be as simple and as flexible
as possible. That is, data processing for planning applications becomes a much more
important practical consideration than the level of sophistication of hydrological
process modeling.
Design/Analysis Models are hydraulically more sophisticated and thus are

more detailed tools. They are used for analyzing individual catchments and sub-
catchments where the simulation of detailed performance of discrete elements within
a subcatchment must be achieved. Whereas hourly rainfall data is an appropriate
input for planning models and for simulating flows in larger urban streams, 5-minute
interval rainfall data (the shortest duration reported by the U.S. Weather Service)
is the appropriate input for simulating flows in sewers and small urban streams
for design applications. That is, the level of sophistication of hydrological
process modeling for design becomes a much more important practical consideration
than data processing, just the opposite of the emphasis imposed by planning
requirements.
Operations Models are likely to be more application-specific than planning

or design models because of wide diversities in management practices, operating
problems and individual service-system configurations. However, the most potentially
transferable technology will be for complete, "hands-off" automatic operational
control of total community runoff, a capability that probably will not be reached
for several years.v 11 ' The mathematical models required are control algorithms,
which will have to be painstakingly derived from numerous indicator applications of
both detailed design models (for generalization of the performance of individual
components by simulation) and planning models (for generalization of community-wide
system performance by simulation). Here also, design/analysis models are used as
tactical tools and planning models are used as tools of strategy.
Sewered Catchment Models
In keeping with the unanimous recommendation of the IHD/Unesco Subgroup

on the Effects of Urbanization on the Hydrological Environment that state-of-the-
art reports, including the present one, "should be addressed to users of research
findings,"' 12 ' only those models that have been tested against actual field data
will be discussed. Table 1 lists 16 models and identifies and describes the
catchments that were the source of field data which was used in their testing.
Table 1A lists the source references for Table 1 entries.
Explanation of Table 1 Entries. Instances where more than one model

was tested by a given organization against data from one or more catchments are
entered as capital letters above the diagonal indicating an entry. Singli nodel
tente by * given organization againat data from on« or more catchmenta ara enterad
as lower-cage letters below the diagonal indicating an entry, for each source
17
J¿ -H
S
.Ë o .
- i
U-l S CN
3 «M ^H
aã
U
13 > 5 (U -H
- * ~ Ed , <
•—' o t-4
(-I •s §
CJ •^
4J
B a
s . 5 <
z
<r
r*. Z
* 0\
v>
3 *-* 00 rH OO
G
* 01
C
U-l
d
O i 5
tí rH*
•H
C 13
4J 3 eu
§ u
0
T3
3
~*
kl
cd
d ~* a
(H <
s
•-H
A4
V
u
0
u
Crt
a.
Ol
i
-i I-I
0 • a i-i
4->
~ i
IS B
a *j ^H
Id * - H
* ?
£ >4 VO
03 ,-4 kl rH
tí ï so
<
H
«4 eu m
Z a^
K >>
4-1
n
CO
O tN
U. •3 O£ m « 4-Í .-4 »-S (0
< * o •4-1 00 0 sO >
(
U/J)
U rH
• r«. ^ u-i r* d ki oo w
UJ u cd a > Cd oo
_J a —*
z c < -a
m Ul •H U-l w • -H
« Ufc
h- J .s z Ù M) 1 o
00
C
-à-
CN 3 « c g> C0 0* •tf
•H O a. u >< • H k< rH Pi U Ol TJ
rH T3
-C
Ol
01
rH
pH
rH
B M
O 3
4J
rt
£ a u
m tu a.
c
o
X
U 3 C UH V3
o•
CO f
c a. CO
Ul a z = .O *H td
o• 3 -*
X Jd rH t-i 01 enI . v£J
d
a M o £ n a. o f~* cd
> u• • d* a• V g •o
* 00 h. T4
u •
o r* Id C o o e>•J -a B OdJ
£ £ .* X\ s• ce o ki
r*-
0_ o *m
J 01 U U' o1 c c" u i
£ O ki
.S Si • —
u H o 01 u 03
rH
n
• «
s- ° U H FH iJ L.
U
2 i
u U-l
O « cd
b
3 oo
n H E
ri -a
11 d
c tí U-Í
*ai< •o
M
< a »-H
o •u «
» •a
cd £ Xi
A
0 A
c
4 C U-l en Q. CE* J
c ctj CJ
n u
Z3 >>« a
4J w M a
-H
CJ
>. 1*
eu
•a -4
.c
en u ~4
o rH C -D -« -a o >. o :* (J (fl
•H
kl
cd - H
O
S 00 (H
u o-o X •• > 01
m eu n « 3 »
U w û < J
0J «
o
••4
rH-•
d
«H
>C
m -4
01 IM a
£ 13e >s
3
¿J 3 -4 s 0
rH U ce K £ •a r m u
r*
(U
rH
J) g q «c* ••
5 ta 8
#J a « .3 ë
3
r i
Vi A4
, a O
w
„ oO
cd
é rH
cd
a
cd
CO S >
0
»* *o u
Dl < a tn -1 u o o
<• £
4 r-t CO 95
H C l-l rH rH IO rJ O 1-3
o < •o
S â. x:
H V
E 3
cd VI
t^ B
3
01
CJ
tj
u •O o
u rj
0
ki
•tf —
cji J«:
Z3
C5
O
•-)
O
-«
-í
VI U-4 rH ki -4 u U-l ki i-i •o
d -. 04
J3 U a en eu « >4-1
U
4J B
O
Ul
0) H- 0
0)
UH
>>CO
4J
3 o
rH >
cd
-Q CO »H
a r^. « o u o J;
4J ^ td
M
-o B H m
2 kl CJ ki j: en u u a m o
S S i c2 g U-l S.¿? 06 ä O O S £S n w H w f c n H W
3
18
3 >' ï X >,
19
reference involved. For the latter, instances where water quality (biochemical)
parameters were also tested with the models are indicated by an underline of the
identifying letter. Two Australian catchments and one Canadian catchment are
included at the bottom of Table 1 because data from them was used in testing U.S.-
developed models. Entries are not of equal validity because a few involved
perhaps three dozen storm events whereas some involved only one or two. Entries
were restricted to cases where a report describing the tests was available to
the public. "Total Imperviousness, Per Cent," is listed to give the reader some
indication of the extent of urbanization; and the percentage may have changed
over the period of record available, but the latest estimate has been entered in
Table 1.
Comments on Table 1 Entries. Of considerable importance are the facts

that: in almost every instance field flow measurements have been made only at
one location; over half of the flow measurements have been indirect via stage
gages, only a part of which have been related to the characteristics of downstream
hydraulic controls, with the remainder depending on assumed conduit friction
coefficients; and on less than half of the catchments water quality data has been
collected, and only for a fraction of these has such data been used in model tests.
The first 14 models cited are of the analysis/design type, and all fit a distributed
classification because they take into account sub-catchment physical characteristics
in some way, and most accommodate underground conduit transport hydraulics. On
the other hand, "STORM" is a planning model and does not have any transport
reckoning capability. Unit hydrograph entries are included only where this method
was one of several tested by a given organization; and some other applications
will be discussed subsequently. Greater frequency of tests on particular catch-
ments is not accidental and can be credited to the ASCE Urban Water Resources
Research Program for rescuing and reporting field data for the Northwood,(13)
Oakdale,^ ' Grey Havení 1 ^) ancj Boneyardí-*-5^ catchments, together with details on
physical catchment characteristics suitable for modeling. Other catchment
descriptions and/or data compilations not cited in Table 1A include: a series of
reports' 1 '' for Castro Valley, Strong Ranch and Ross and Peralta Creeks in
California; papers^ 1 8 ' 1 9 ^ on water quality data for the Durham, N.C., catchment;
details on the Mt. Washington (Cincinnati) catchment;(20) data on the seven
Seattle, Washington, catchments;(^1) and data displays for the Baker Street,
San Francisco, catchment.(•*
Availability of Computer Programs
Illinois Urban Drainage Area Simulator. Entered in Table 1 is the

testing record for a U.S. adaptation of the British Road Research Laboratory
method. ILLUDAS utilizes the directly connected impervious area concept of the
RRL method but also recognizes and reproduces runoff from grassed and non-
connected impervious areas. The verification report^3^ includes a user's
manual, and the computer program is available to the public. Water quality
considerations are not accommodated, and the intended uses of the model are the
sizing of conduits and the evaluation of existing underground systems.
Environmental Protection Agency SWMM and Variants. Subroutines for

the EPA Storm Water Management Model are represented symbolically in Figure 2.(")
A user's manual'^-1' and the computer program are available to the public for the
latest version developed for EPA, by the University of Florida. A user's manual'17/
and the computer program are also available to the public for the version
*: References indicated by letters appear in Table 1A.
20
^^_
RA ALL
u.
z
5*
UJ
X
DRY WE
3
K
NOFF
FLO
INFILTRA TION
ERL INFIL TRAT ION
>
5
DEPRESS ON STORAGE 0 FLOW
•! >
¿'
*
k
'k
UTTER FLOW
i i
¡
k
a
INTERNAL
cos
l-
DECAY TRA NSP ORT
STORAGE
,k
''
21
a
¿
S
(9
PIPE ROUTIN
jk
i
• 5
k
Jk
*
ÍERNAL
COS
k-
t-
TREATM
ORAGE
Ui Z J
1
3 _J Ul
RFAC
i
'
t-
LUTA
MOVA
Í
i
-
RECEIV ING WAT

UJ
IC
§
u.
k-
CM
~k
u.
2
o
fe
_
NC INNAT DEL
IOw
cr
>-
O
evi
>
UN
cr
UJ
COMPONE NTS
C3
3
<r
CO
u.
<S>
UJ
3
m
cr
O
3
I-
z
ÜJ
CO
O
U.
LU
en
—' — s— 1
's
"' .\ .^
fWlOlï) TI37NI0KI
JO wicn
'vT
. -N
jo t n - n
\
HJVIOOMAH 00O1IO
•oiin/uis
"H
\ \
*aU O0Y3IK3
\
lElnMlON
JTTIU»!
\ S
n - \-\
3
X W K M O U jo
1 LUI IS« 1 VtWOlS
\ ,
N "
j lUlSMOO
i3<noa
\ \
VtSWtlH JO
\
1
i
All n n A i m
flMtTII
\
\ \
I3O10MDU JO liAIUMI
s \
\ \
(JSHI
JMOTOWU1
\ \ \ \ \
UVMI3MI3
jo u i s a A i m
s \
\
V K N
^ i - "s
s \ S-.'
-
wMLSKMJ I W I W i l A B
X \ " X- \ ÇN -•^ - \ \ ^ \
\-
N\ -
amiMi laoiwwvi
K3i«isn awn m i m t
\ \ \ \ \
•OiTlfNlS <IIV
^ X^ N^
jDwlvM M U I iimini
\ \ \ \ \
\ \ N\ \ \ \
N\ N \
IIIVKVAV w«r
HllWi í ? ; s a a s
m m
M • «VIM j
:
a
«mi -; Ï 3 i ü 5 5 £ Ï ä 3 á î i í -: -; Ï ï i
î 3
î 3 ï i i a
i
«niiv>n
í =¡ 3 i 5 3
Mí
S á 5 i 5 5 •i í d •i =* 5 S s 1
ï \ S ? 1 1 s 3
IT«
y S s s r;
K i ï K
g ï y t >; r: s t: s s ^ u S
! m i nj -
'cn«sfwiAn*t A i Î £ í i Z i 7 í ¡S 3 S Í
1«KU
z ^ » Z i s s = s ; s 3 S ? Í
- snviaiH
_.
:: -
i
i •Tí 3 ?. s 2
¿ s ?'n d
5 i
1 \ ÊÏ S
i ^ I i 3
ía ,- = o
i 3 a s
Ui
.e r:
Ii i
3
\\
A
••l M
t 5
ï* ,í 7Ï Î
.= .5 »! ; ï yî
M 5sä Sä r;s 5 î äï 1» f' î*
Î?: sä »3a
: i |
" aí ï 3 ;- 3
i! if !í !!
1 î
íí íí I! ¡i !lgl li ií
11¡1 1 » 53
Si S3
II îs ¡s i! il ii ÍÍ iî
22
\ \ \
\
\ \ \ \ \
s \
\* \ \
\
\ \ N
\ \ ^ \ \ \ \ \
S \ \ \ NsN \ \
N \
\
\
\ \
i
= H ; s H 5 2 E i i g Ï a B f . ï i i E E ; 8 B , , S ,
¡ 3 1l i I i l ! I i ! i ï ? s i I i i i l¡ ï 3 ? f ï 1 § 1 I
a
= 1 2 i
i I í ! \ 1 1 ! ! \ ! ! i ï i I ! 1 ! ! ! l1 !
j 1 ! ! ! ! ! 11 3 E
= Ï S
s s S ! ;
ï = = s s i i ; ! ~ 5 s " = = s = : s H I Î
"
"
í í í
i¡íí 1i!i 1
i,
É i if I i ¡i
! ;!
is
!
Ü II
i!
til
1ií!¡Iï íl¡ ¡¡S i! S ¡iS il1 i
35 íl
1
3 !
e 3
«3
_3
23
developed for the City of San Francisco by Water Resources Engineers. The
"Runoff" module of the EPA Version I I ^ 2 3 ' and the "Transport" module of the SF
version are considered to represent significant advances. The "Cost" and
"Receiving Water" modules have not yet been interfaced with the SF version; and
it has also been tested on field data from nine catchments in Hamburg, Federal
Republic of Germany.' 2 ^ Three organizations developed jointly the original EPA
SWMM, two of which developed the versions noted above. The third, Metcalf and
Eddy, Inc., developed a version used by the firm in studies in the Boston area,
and in Chicago( v ' and Cleveland.(X-J Developed expressly for quantity and quality
evaluation of sewered systems, the various versions of SWMM collectively have been
subjected to more verification than any other such models. Several catchment
tests have included receiving-water simulation, with comparisons of observed and
computed water quality parameter magnitudes in the vicinity of outfalls. It is
expected that refinement and improvement of the various versions will continue.
University of Cincinnati. Components of the University of Cincinnati

model are symbolized in Figure 3.(25) A model description and the computer
program are publicly available.(n)
Hydrocomp. The Hydrocomp Simulation Program System, outlined in

Figure 4,(26.) fundamentally simulates watershed hydrology and flow routing.(27)
The HSP is the commercial successor to the Stanford Watershed Model, first
reported in 1960.(28) Several versions of the latter have been developed by
others,(29) notably the Kentucky Watershed Model,(30,31) w hich has been tested
using data from several urban streams. The non-proprietary Stanford Watershed
Model has been tested against field data for five streams in California in
connection with simulations of the effects of changes in urbanization.
(32)
However, only the proprietary HSP accommodates water quality considerations,
and some metropolitan planning examples will be cited later. A distinctive
feature of the HSP and the Stanford Watershed Model is a continuous simulation
capability.
Massachusetts Institute of Technology. The general structure of the
MIT Catchment Model is shown in Figure 5.( n )Predominantly an analysis/design
model, the current computer program is proprietary. However, a modified version
developed for estimating sewered catchment and stream runoff for the master plan
development of the 1000-km2 County of Fairfax, Virginia,(33) may be made publicly
available.
University of Illinois Models. A user's manual with part of the

computer program listing is available to the public,(34) and other portions are
reported elsewhere.( c '
University of Nebraska. A model description and the computer program

listing are publicly available for quantity( k ' and quality^6' simulation.
Dorsch Consult. Analogous in complexity and capabilities with SWMM-type

models, the Dorsch Hydrograph-Volume-Method models-") ¿ s proprietary.
Georgia Institute of Technology. A model description and the computer

program listing are publicly available. ( O
Battelle Northwest. A general description of the model has been

published but details are in a report to a client.(37) While the model is
being utilized in the analysis of existing sewer systems in Cleveland, Ohio,
24
CONTROL
PARAMETERS
V V
Reservoirs and
Diversions FLOW ANO
INPUT DATA •> LIBRARY CHANNELS STAGE
OUTPUT
QUALITY
QUALITY OUTPUT
FIGURE 4. THE HYDROCOMP SIMULATION PROGRAM SYSTEM 1 2 6 *
RAINFALL
RAINFALL INFILTRATION AND

EXCESS EVAPORATION
CATCHMENT
STREAMFLOW GUTTER PIPE

FLOW FLOW
OUTFLOW
FIGURE 5. GENERAL STRUCTURE OF MIT CATCHMENT MODELth)
25
a major purpose of the model is for use in connection with development of
automatic operational control of stormwater runoff in that city.(38)
Chicago Flow Simulation. The published paper( w ' does not contain a
description of sewer routing capability described elsewhere.(39) jê latter
reference notes that the computer program is available to the public in the form
of a punched card deck. Water quality considerations are not accommodated.
Chicago Hydrograph Method. Based on the first sewered catchment

distributed model developed, testing(^) and appHcation(^l) have been restricted
to the Chicago area.
Corps of Engineers (SSARR). A program description with a user's manual

is available,(42) and the computer program is available to the public. The complex
Streamflow Synthesis and Reservoir Regulation (SSARR) model was designed for
operational use in hydrologie engineering studies and daily streamflow forecasting
for river basins and tributaries. Initial tests with urban catchment data( F ) have
resulted in good estimations of daily flows, indicating a potential use of the
model in urban planning applications.
Corps of Engineers (STORM). Model documentation,^' a user's manual,(43)

and the computer program are all available to the public. The Storage, Treatment,
Overflow and Runoff Model (STORM) was designed specifically for urban runoff and
quality evaluation, for master planning, and normally uses several years of continuous
hourly precipitation records. It is basically an allocation and accounting model
and does not perform routing. Erosion yields and non-urban catchments can also be
accommodated. A national assessment of pollution from urban runo
ff(m)
included the
application of STORM to fifty sample Urban Areas.
Proceedings of a national symposium^"' contains additional information
on some of the models noted above.' a,( l >t > d >33)
Comparison of Model Characteristics
An extensive analysis of various models^"' compared the characteristics

of 18 models, 10 of which have been discussed above. The findings from that
analysis for these 10 models are listed in Table 2.( B >^'
Computer requirements and programming languages for almost all of the

models discussed thus far are included in an extensive model survey.
(45)
Results of the various performance comparisons ("Multiple Model Tests,"
Table 1A) are mixed, mostly because there is no acceptable mathematical basis for
multiple-objective comparison. Peak flow is the major consideration in sizing
conduits, volume and hydrograph shape are critical for sizing and deploying storage,
and concentrations and loadings of pollutant emi-ssions are essential for evaluation
of receiving water impact and helpful in sizing treatment facilities. Each model
has its strengths, weaknesses and outright faults for a given application. Over
and above the comparisons-definition problem is the inherent difficulty with any
runoff model in the necessarily subjective separation of abstractions (infiltration,
depression storage, etc.) from total rainfall to resolve rainfall excess (amount
and pattern), which is the input from which an equal volume of direct runoff is
generated by models of one kind or another. After analyzing the performance of
a variety of models, it was concluded that the weakest link is the proper
estimation of rainfall excess.(^"'
26
aianivAV wvaooaa aainawoo • • • • •
MISCELLANEOUS
10HINOD 3HII-1V3H • •
SNOIIVMdKOD NDIS3Q • • •
IVAaaiNI 3WII 3S00H0 NVD • • 0 • • • •
N O I I V M W I S snonNiiNoo • •
Noiinnwis Ainvnb asivfi DNIAISOSH
Noiivinwis «oil aaivM ONIAISDSH • • • •
SWHOIS N33HI33 30NVlva AUIVilb •
•
WATER QUALITY
irawivaai
SNOiiovaa Ainvnb
anoos aNv NonvxNamass
ONiinoa Ãinvnb
Ainvnb asivMraois •
xinvnb H3Hiv3H-Aaa
saiiiotnsA sJNiaa
30VIS siNiaa *
SEWER HYDRAULICS
aovaois •
SNoiivis oNiawna
SNOisasAia •
«oti aanssaaa aNv DNIDHVHOHIIS
loaiNOO ñoaa wvaaismoa aNv wvsaisan
sa3«3S NI ONiinoa M O U • •
SWHOIS N33HI3B 30NVIVa aaiVM • •
HYDROLOGY
svaav snoiAH3d woaa aaowia • •

svaav snoiAasam woaa aaoNna • .
CATCHMENT
naKHONS •
SHavaÒ0I3AH 1V83A3S 30 IÍHNI • •
M O U a3Hiv3n-Aaa • •
SrtOláNI IN3HH0IV0 31dIIinW • •
ENVIRONMENTAL PROTECTION
MASSACHUSETTS INSTITUTE
BRITISH ROAD RESEARCH
CHICAGO HYDROGRAPH
ENGINEERS (STORM)
AGENCY & VARIANTS
OF TECHNOLOGY
UNIVERSITY OF
UNIVERSITY OF
CHICAGO FLOW
SIMULATION
CINCINNATI
LABORATORY
NORTHWEST
HYDROCOMP
BATTELLE
CORPS OF
ILLINOIS
METHOD
27
Parenthetically, mention should be made of a distributed-type catchment
model daveloped by the U.S. Geological S u r v e y . W i Recent modifications include
the provision for synthesizing flood runoff from small urban watersheds.^ 8 5 Also,
the Soil Conservation Service has developed urban peak flow computation procedures.(^)
Unit Hydrograph Tests
Unit hydrographs have been developed from field data for at least three-
score catchments, but almost all of these were for partially sewered catchments
where the streamflows measured included a significant contribution from non-sewered
sectors. .£*rt °f the completely sewered exceptions were catchments in Louisville,
Kentucky,i 50 ^ and Atlanta, Georgia,( 31 ) and some of these are included in Table 1.
The report on the Georgia Institute of Technology m o d e l e also contains

results of unit hydrograph tests using urban streamflow data from large catchments
nearby, as does a companion report.^ 32 '
An excellent manual on unit hydrograph analysis is available,(53) together

with a computer program user's manuali 3 ^) f o r development of unit hydrographs from
field data and for routing flows from one point to another. The computer program
has been used extensively in urban projects of the Corps of Engineers.
Unit hydrographs have been employed in recent studies investigating

effects of urbanization or improvement of predictive techniques.(55-60)
Purdue University has collected flow data of good quality from local
catchments (one is cited in Table 1) and from Indianapolis.(°1»") Water quality
sampling has been added for two of the local catchments.(63) The field data from
Indiana and elsewhere has been used in tests with various linear methods, including
the instantaneous unit hydrograph.i"1»"2)
Sewerage Operations Testing
Cited in Table 1 is a 29-ha portion of the Vicente Street catchment in

San Francisco. Colorado State University has calibrated simple distributed models
of the type tested at Purdue Universityi"1^ using stage and rainfall data for the
total 668-ha Vicente combined sewer catchment(64) as part of a program that is
assisting in the development of automatic operational control capability for the
City.(6->} Interest in automatic control has been growing.(°°^ Unit hydrographs
coupled with routing models have been used to effectuate schemes for automatic
control of combined sewer overflows in Minneapolis-St. Paul(67>68) and Seattle.(69)
Noted earlier was the use of the Battelle model in a similar scheme for Cleveland.
(38)
Metropolitan and Other Large-Scale Tests
Receiving waters are the common repository of effluents from just about
every community in a metropolis, constituting perhaps the most shared aspect of
urban water resources. Recent emphasis on regionalized wastewater treatment and
disposal has resulted in some receiving-water flow and quality simulation studies
on a grand scale, such as in San Francisco Bay,(70) metropolitan Denveo'*' and
metropolitan Seattle.O 2 ) For the first of these, descriptions are available of
some of the water quality'-'J' and
flow(74)
characteristics of the group of models
used. The Denver and Seattle studies represent planning applications of the
Hydrocomp model. Another Hydrocorap planning application^ 3 / was among the first
flood plain mapping studies undertaken with simulation techniques.
28
Water quality modeling for systems containing rivers and reservoirs
recently has been advanced through the issuance of a description of a new combina-
tion of models.(76) The Hydrologie Engineering Center is having dynamic flow routing
routines added to the model and plans to upgrade the documentation as new developments
occur.(77) Dynamic or unsteady-flow water quality modeling is particularly important
in the case of significant pulse loadings from urban runoff or when man-made controls
such as dams are involved.
A compendium,(^^) two companion reports,(78>79) a North American summary( 8 ^)

and a recent text,' 81 ) survey features of large-scale water quality models; and an
annotated bibliography of models for tidal rivers, estuaries and coastal waters is
available.(82) Tidal water models have been comprehensively classified,(83) and
capabilities for modeling estuary and streamflow water quality have been assessed.(84)
Aquatic ecosystem submodels have been delineated for process analysis.(") Effects
of risk and uncertainty in the application of operations research techniques,
including hydrologie modeling, have been critically reviewed at length.(°°)
Stochastic generation of synthetic strearaflows(87) i 8 still receiving the attention
of researchers.(88,89)
Some Other Kinds of Models
A set of conceptual models tracing pesticide movement and behavior in the

air-land, freshwater and estuarine-marine environments has been developed.(90) The
runoff and receiving-water modules of the EPA Storm Water Management Model have
been used in modeling agricultural areas under storm conditions.('!) That is,
rural pollution analysis is emerging as a new area of modeling attention.
Of course, modeling of groundwater quantity('2) has been growing all the

while, and advances are being made in the modeling of groundwater quality.(°3)
Other types of models, such as for water supply, are included in the
broad coverage of a recent review.(") Water supply modeling has ranged from
planning multiple river basin impoundment systems to drought analysis of ma instern
flows.(") Model uses by a metropolitan water supply agency have been outlined.('->)
A model for metropolitan application has been developed which considers

environmental pollution as a set of interrelated problems, using submodels for land
use, residuals and disposal-dispersion of residuals.'9"-' Interest in integrated
planning, management and control of all major pollution types is growing.
Conclusions
A questionnaire circulated a decade ago(°7) revealed that all but a small

percentage of urban storm drainage systems were being sized using a completely
empirical relation called the "rational method". Limitations of the method have
been discussed at length.('8) The reader of the preceding presentation in this
Section might have the impression that a monumental reform had since taken place
in the U.S.A. On the contrary, most analysis and design is still being performed
using such empirical methods; however, the models cited are gaining in use, and
newly accelerated metropolitan water quality improvement planning activities fore-
tell a possible explosive use of such models in the near future.
29
For models of the Storni Water Management type (Figure 2) and for STORM,
average percentage of imperviousness is the most sensitive parameter affecting the
estimation of runoff amounts and, because of current indexing of pollutants to
street loadings noted below, street gutter density is the roost sensitive parameter
affecting estimates of pollutant loadings. With the increasing interest in
metropolitan studies, it is not surprising that generalizations are being sought
between these two variables and type of land use.'"»*-'"''
We are reminded that until each internal module of an overall catchment

model can be independently verified, the model remains strictly a hypothesis with
respect to its internal locations and transformations.' 101 ' Because of the very
limited amount of field data available, just about all sewer application model
verification has been for total catchment response, at a single flow measurement
location. That is, under contemporary conditions a distributed model deteriorates
into a lumped system model, for all practical purposes. In the U.S.A., calibration
and verification are further confused by the fact that much more field data is
available for partially sewered catchments, where flow is measured in receiving
streams, than for totally sewered catchments. Adding streamflow hydraulics to
sewer hydraulics hardly simplifies the lumped system dilemma, yet much of the data
used to verify various models has been from such mixed catchments.
Relatively few runoff-quality field gagings have been made in the U.S.A.,
and these have been mostly at outfalls. Source quality has been investigated
principally as a function of street surface pollutants accumulated between rain-
falls. In order to accommodate cause-effect relationships required for modeling,
it is current practice to estimate potential street loadings, separately for
individual parameters, on the basis of the few documented solids accumulation
histories. Arbitrary allowances are then added to account for off-street
contaminant accumulations, expressed as multiples of the potential street loadings.
Thus, no direct verification of the hypothesized buildup of pollutants and their
transport to receiving waters is presently available. It is reasoned that when
"pollutographs" generated by models reasonably approximate field observations for
a catchment, that the overall accumulation and transport hypothesis is validated.
In order to resolve the lumped system hydrologie dilemma and truly validate
the pollutant accumulation and transport hypothesis there is a major research need
for simultaneous measurement of rainfall, runoff and quality at inlets and various
collection system points in field catchments. Some of the most recently instrumented
catchments may yield new process insights in the near future.
It might be concluded that model development has already greatly out-

stripped the data base for model verification, in the sense of bracketing probable
reliability. However, planning deals with alternative futures and the mechanisms
of change. As in the design and analysis of urban water resource facilities, plan-
ning by local governments involves projections of conditions expected over future
decades based on observed indications of past and current conditions. Because
anticipated future conditions cannot be simulated by manipulating actual land use,
recourse is made to simulation by some form of calculation or analogy. Even if a
given planning scenario, as defined today, was fully replicated in the future, the
hindsight precision of projected land use detail would necessarily have decreased
from one successive planning epoch to the next. While reliability of projected
conditions is a function of the precision of evaluations of current conditions, it
is not realistic to expect perfection in such appraisals at this time. If field
research and model testing continue at anywhere near the level of activity of the
past decade, substantial advances in reliability appear to be an inevitable result.
30
In closing, mention must be made of: the availability of a SWMM user-
assistance service^" 2 ); availability of the SWMM computer program^ '; a series
of papers from a national symposium'IO*) o n network and modeling aspects of the
analysis of major urban storms; a simulation study of the effects of urbaniza-
tion^0-' '; and the latest literature review of urban runoff and combined sewer
overflow pollution.(106)
Epilogue. The presentation in this Section was completed in November,

1975, and the only changes made since (October, 1976) were the updating of some
reference sources. However, some supplementary developments deserve mention.
A model developed by the University of Texas(107) has been tested on

data from Waller Creek, 23rd St., Austin, Texas (Table 1). The project report
includes the computer program.
As part of a project at Texas Tech University, the British Road Research

Laboratory Method (Table 1) was tested on data from 26th St., Lubbock, Texas,
590-ha with 547„ imperviousnesSoÔo) Flow was measured in the receiving stream
with a weir and water quality was sampled.
Versions of the ILLUDAS, EPA SWMM and MIT models (Table 1) have been
tasted on data from Crane Creek, Jackson, Mississippi (Table 1) and on Manitou Way»
Madison, Wisconsin (57-ha, 20% imperviousness), by a study group in the U.S.
Geological Survey.(1°9)
A series of papers dealt with development of unit hydrographs for gaged

catchments. (HO)
To a classification of tidal water models

(83)
has been added a review of
the predictive capability and limitations of such models.'HI)
Lastly, attention is directed to the first attempt in the U.S. to bring
together mathematical models of air pollution, water pollution, ecology, and 'other
environmental areas, in order to form a unified discipline, "environmental
modeling".(112)
31
References (Additional references are listed in Table 1A)
1. American Society of Civil Engineers, Basic Information Needs in Urban

Hydrolofty, A Study for the Geological Survey, U.S. Dept. of the Interior,
ASCE, New York, N.Y., 112 pp., April, 1969. (Available fromASCE).
2. Schneider, W. J., and J. E. Goddard, Extent and Development of Urban Flood

Plains, U.S. Geological Survey Circular 601-J, GPO, Washington, D.C., 14
pp., 1974.
3. Dawdy, D. R., R. L. Smith, N. H. Crawford, P. S. Eagleson and W. Viessman,

Jr., "Considerations for Modeling Urban Rainfall-Runoff-Quality Processes,"
Appendix A , 125 pp., in Urban Water Resources Research, ASCE, New York,
N.Y., September, 1968. (Available fromASCE).
4. Thomasell, Albert, Jr., "Considerations for Characterizing the Time and

Space Distributions of Metropolitan Storm Rainfall," Appendix B, 12 pp., in
Urban Water Resources Research, ASCE, New York, N.Y., September, 1968.
5. McPherson, M. B., "Urban Runoff," Section 1 of Part III, pp. 153-176, in

Hydrological Effects of Urbanization, Studies and Reports in Hydrology,
18, The Unesco Press, Paris, 1974.
6. McPherson, M. B., and W. J. Schneider, "Problems in Modeling Urban Watersheds,"

Water Resources Research, Vol. 10, No. 3, pp. 434-440, June, 1974.
7. Poertner, Herbert G., Practices in Detention of Urban Stormwater Runoff,

American Public Works Association Special Report No. 43, 231 pp., Chicago,
111., 1974.
8. Water Resources Engineers and The Hydrologie Engineering Center, Corps of

Engineers, "Management of Urban Storm Runoff," ASCE Urban Water Resources
Research Program Technical Memorandum No. 24, ASCE, New York, N.Y., 92 pp.,
May, 1974. (Available from NTIS as PB 234 316 or from HEC, Davis, Cal. 95616).
9. Prawdzik, Ted B., "Environmental and Technical Factors for Open Drainage
Channels in Milwaukee," ASCE Urban Water Resources Research Program, Technical
Memorandum No. 12, ASCE, New York, N.Y., 37 pp., February, 1970. (Available '
from NTIS as PB 191 710).
10. Tucker, L. S., "Sewered Drainage Catchments in Major Cities," ASCE Urban
Water Resources Research Program, Technical Memorandum No. 10, ASCE, New
York, N.Y., 71 pp., March 31, 1969. (Available from NTIS as PB 184 705).
11. McPherson, M. B., "Feasibility of the Metropolitan Water Intelligence System

Concept (Integrated Automatic Operational Control)," ASCE Urban Water
Resources Research Program, Technical Memorandum No. 15, ASCE, New York,
N.Y., 110 pp., December, 1971. (Available from NTIS as PB 207 301).
12. American Society of Civil Engineers, "Report, International Workshop on the

Hydrological Effects of Urbanization, Warsaw, 1973," to U.S. National
Science Foundation, New York, N.Y., 60 pp., January, 1974.
32
Tucker, L. S., "Northwood Gaging Installation, Baltimore - Instrumentation
and Data," ASCE Urban Water Resources Research Program Technical Memorandum
No. 1, ASCE, New York, N.Y., 36 pp., August 1, 1968. (Available from NTIS
as PB 182 786).
Tucker, L. S., "Oakdale Gaging Installation, Chicago - Instrumentation and

Data," ASCE Urban Water Resources Research Program Technical Memorandum No.
2, ASCE, New York, N.Y., 76 pp., August 15, 1968. (Available from NTIS as
PB 182 787).
Tucker, L. S., "Availability of Rainfall-Runoff Data for Sewered Drainage

Catchments," ASCE Urban Water Resources Research Program Technical Memorandum
No. 8, ASCE, New York, N.Y., 93 pp., March 3, 1969. (Available from NTIS as
PB 184 703).
Tucker, L. S., "Availability of Rainfall-Runoff Data for Partly Sewered

Urban Drainage Catchments," ASCE Urban Water Resources Research Program
Technical Memorandum No. 13, ASCE, New York, N.Y., 156 pp., March, 1970.
(Available from NTIS as PB 191 755).
The Hydrologie Engineering Center, U.S. Army Corps of Engineers, "On the
Quality of Urban Storm Runoff Entering the San Francisco Bay," Reports for
Fiscal Years 1972, 1973, 1974 and 1975, Davis, California.
Colston, Newton v., "Characterization of Urban Land Runoff," a paper

presented at the ASCE Meeting, Los Angeles, Cal., January 21-25, 1974,
Meeting Preprint 2135, 29 pp., ASCE, New York, N.Y.
Bryan, E. H., "Quality of Stormwater Drainage from Urban Land," Water

Resources Bulletin, Vol. 8, No. 3, pp. 578-588, June, 1972.
Weibel, S. R., R. J. Anderson and R. L. Woodward, "Urban Land Runoff as a

Factor in Stream Pollution," J.WPCF, Vol. 36, No. 7, pp. 914-924, July, 1964.
Municipality of Metropolitan Seattle, "Environmental Management for

Metropolitan Areas," Part II, "Urban Drainage," Appendix C, "Stormwater
Monitoring Program," approx. 100 pp., July, 1974.
Lager, John A., Robert P. Shubinski and Larry W. Russell, "Development

of Simulation Model for Stormwater Management," J.WPCF. Vol. 43, No. 12,
pp. 2424-2435, December, 1971.
University of Florida, Storm Water Management Model User's Manual, Version II.
Environmental Protection Technology Series EPA-670/2-75-017, U.S.EPA,
Cincinnati, Ohio 45268, 350 pp., March, 1975.
F. H. Kocks, KG, Ingenieure, "Berechnung von städtischen Kanalisationsnetzen

mit Hilfe von Simulationsmodellen," in Zusammenarbeit mit Water Resources
Engineers (USA), Düsseldorf, 75 pp., April, 1974.
Beers, Gary D., "Management of Stormwater Runoff in Suburban Environments,"

Engineering-Science, Inc., Cincinnati, Ohio, 110 pp., November, 1973.
Linsley, R., and N. Crawford, "Continuous Simulation Models in Urban

Hydrology," Geophysical Research Letters, Vol. 1, No. 1, pp. 59-62, May, 1974.
33
27. Metcalf and Eddy, Inc., Urban Stormwater Management and Technology; An
Assessment. Environmental Protection Technology Series EPA-670/2-74-040,
GPO, Washington, D.C., 447 pp., December, 1974.
28. Linsley, R. K., and N. H. Crawford, "Computation of Synthetic Streamflow

Record on a Digital Computer," Publication No. 51. International Association
of Scientific Hydrology, pp. 526-538, 1960.
29. Linsley, Ray K., "A Critical Revue of Currently Available Hydrologie Models
for Analysis of Urban Stormwater Runoff," Hydrocomp International, Palo
Alto, California, 83 pp., August, 1971. (Available from NTIS as PB 204 815).
30. James, L. Douglas, "An Evaluation of Relationships Between Streamflow

Patterns and Watershed Characteristics Through the Use of OPSET: A Self
Calibrating Version of the Stanford Watershed Model," Water Resources Institute
Research Report No. 36, University of Kentucky, Lexington, Kentucky, 117 pp.,
1970. (Available from NTIS as PB 198 444).
31. Liou, Earnest Y., "OPSET: Program for Computerized Selection of Watershed
Parameter Values for the Stanford Watershed Model," Water Resources Institute
Research Report No. 34, University of Kentucky, Lexington, Kentucky, 299 pp.,
32. Durbin, Timothy J., "Digital Simulation of the Effects of Urbanization on

Runoff in the Upper Santa Ana Valley, California," U.S. Geological Survey,
Water-Resources Investigations 41-73, 44 pp., February, 1974.
33. Michel, Henry L., and William P. Henry, "Flood Control and Drainage Planning
in the Urbanizing Zone: Fairfax County, Virginia," pp. 119-139 in Urban
Runoff. Quantity and Quality. ASCE, New York, N.Y., 1975.
34. Sevuk, A. S., B. C. Yen and G. E. Peterson, "Illinois Storm Sewer System
Simulation Model: User's Manual," Water Resources Center Research Report
No. 73, University of Illinois, Urbana, 111., 168 pp., October, 1973.
35. Dorsch Consult, "Urban Hydrological Processes, Computer Simulation," 81 pp.,

Munich and Toronto, 1972.
36. Brandstetter, A., R. L. Engel and D. B. Cearlock, "A Mathematical Model for
Optimum Design and Control of Metropolitan Wastewater Management Systems,"
Water Resources Bulletin. Vol. 9, No. 6, pp. 1188-1200, December, 1973.
37. Brandstetter, A., et al.. "Development of Hydraulic, Water Quality and

Optimization Models for Wastewater Management for the City of Cleveland,"
Battelle Memorial Institute, Pacific Northwest Laboratories, Richland,
Washington, April, 1972.
38. Pew, K. A., R. L. Callery, A. Brandstetter and J. J. Anderson, "Data

Acquisition and Combined Sewer Controls in Cleveland," J.WPCF. Vol. 45,
No. 11, pp. 2276-2289, November, 1973.
39. Lanyon, R. F., and J. P. Jackson, Metropolitan Sanitary District of Greater

Chicago, 111., "A Streamflow Model for Metropolitan Planning and Design,"
ASCE Urban Water Resources Research Program Technical Memorandum No. 20,
ASCE, New York, N.Y., 49 pp., January, 1974. (Available from NTIS as
PB 232 181).
34
40. Tholiii, A. L., and C. J. Keifer, "Hydrology of Urban Runoff," Transactions.
ASCE, Paper No. 3061, Vol. 125, pp. 1308-1379, 1960.
41. Department of Public Works, "Development of a Flood and Pollution Control

Plan for the Chicagoland Area, Computer Simulation Programs," Technical
Report, Part II, Chicago, Illinois, 101 pp., December, 1972. (Available
42. U.S. Army Engineer Division, North Pacific, Program Description and User
Manual for Streamflow Synthesis and Reservoir Regulation (SSARR) Model.
Program 724-K5-G0010, Portland, Oregon, 201 pp., December, 1972.
43. Hydrologie Engineering Center, Corps of Engineers, "Urban Storm Water

Runoff: STORM," Generalized Computer Program 723-S8-L2520, Davis,
California, 104 pp., August, 1975.
44. Torno, Harry C , "Storm Water Management Models," pp. 82-97 in Urban Runoff,
Quantity and Quality, ASCE, New York, N.Y., 1975.
45. Brown,- J. W., M. R. Walsh, R. W. McCarley, A. J. Green and H. W. West,

"Models and Methods Applicable to Corps of Engineers Urban Studies," U.S.
Army Engineer Waterways Experiment Station Misc. Paper H-74-8, Vicksburg,
Mississippi, 420 pp., August, 1974.
46. Hossain, A., J. W. Delleur and R. A. Rao, "Evaporation, Infiltration and

Rainfall-Runoff Processes in Urban Watersheds," Water Resources Center
Technical Report No. 41, Purdue University, West Lafayette, Indiana, 63 pp.,
January, 1974. (Available from NTIS as PB 229 642).
47. Dawdy, D. R., R. W. Lichty and J. M. Bergmann, "A Rainfall-Runoff Simulation

Model for Estimation of Flood Peaks for Small Drainage Basins," USGS
Professional Paper 506-B, GPO, Washington, D.C. 1972.
48. Boning, C. W., "User's Guide for a U.S. Geological Survey Rainfall-Runoff
Model," USGS Open File Report 74-33, Reston, Virginia, 232 pp., May, 1974.
49. Soil Conservation Service, Urban Hydrology for Small Watersheds, U.S.
Department of Agriculture, Technical Release No. 55, 91 pp., January, 1975.
50. Eagleson, Peter S., "Unit Hydrograph Characteristics for Sewered Areas,"
J.Hyd.Div., ASCE Proc, Vol. 88, No. HY2, pp. 1-25, March, 1962. Author's
closure to discussion: Vol. 89, No. HY4, pp. 193-203, July, 1963.
51. Black, Crow and Eidness, Inc., Storm and Combined S"ewer Pollution Sources and
Abatement, Atlanta, Georgia, Water Pollution Control Research Series 11024
ELB 01/71, GPO, Washington, D.C, 181 pp., January, 1971.
52. Wallace, James R., "The Effects of Land Use Change on the Hydrology of an
Urban Watershed," Georgia Institute of Technology Report ERC-0871, Atlanta,
Georgia, 66 pp., October, 1971. (Available from NTIS as PB 206 426).
53. Hydrologie Engineering Center, Corps of Engineers, "Hydrologie Engineering

Methods for Water Resources Development: Volume 4, Hydrograph Analysis,"
Publication HEC-IHD-400, Davis, California, 122 pp., October, 1973. (Available
from NTIS as AD 774 261).
35
Hydrologie Engineering Center, Corps of Engineers, "HEC-1, Flood Hydrograph
Package," Generalized Computer Program 723-010, Davis, California, 186 pp.,
January, 1973.
Dracup, John A., Thomas J. Fogarty and Sharon G. Grant, "Synthesis and
Evaluation of Urban-Regional Hydrologie Rainfall-Runoff Criteria,"
Environmental Dynamics, Inc., Los Angeles, California, 105 pp., February,
Johnson, Steven L., and Douglas M. Sayre, "Effects of Urbanization on Floods

in the Houston, Texas, Metropolitan Area," U.S. Geological Survey Mater-
Resources Investigations 3-73, Austin, Texas, 50 pp., April, 1973. (Available
Harnin, D. W., C. W. Margan and H. A. Reeder, "Statistical Analysis of

Hydrograph Characteristics for Small Urban Watersheds," Tracor, Inc., Report
No. T73-AU-9559-U, Austin, Texas, 155 pp., October, 1973. (Available from
NTIS as PB 228 131).
Chien, Jong-Song, "Urban Runoff by Linearized Subhydrograph Method,"

J.Hyd.Div., ASCE Proc, Vol. 100, No. HY8, pp. 1141-1157, August, 1974.
Schulz, E. F., and 0. G. Lopez, "Determination of Urban Watershed Response

Time," Colorado State University, Hydrology Paper 71, Fort Collins, Colorado,
41 pp., December, 1974.
Brater, Ernest F., and James D. Sherrill, Rainfall-Runoff Relations on Urban

and Rural Areas, Environmental Protection Technology Series EPA-670/2-75-046,
GPO, Washington, D.C., 98 pp., May, 1975.
Rao, A. R., J. W. Delleur and B. S. P. Sarma, "Conceptual Hydrologie Models

for Urbanizing Basins," J.Hyd.Div., ASCE Proc, Vol. 98, No. HY7, pp. 1205-
1220, July, 1972.
Sarma, P. B. S., J. W. Delleur and A. R. Rao, "Comparison of Rainfall-Runoff

Models for Urban Areas," J.Hydrology, Vol. 18, pp. 329-347, 1973.
McElroy, F. T. R., and John M. Bell, "Stormwater Runoff Quality for Urban
and Semi-Urban/Rural Watersheds," Water Resources Research Center Technical
Report No. 43, Purdue University, West Lafayette, Indiana, 156 pp., February,
Wenzel, Harry G., and Bruce H. Bradford, "Control Strategy Development Study,
San Francisco Wastewater Master Plan," Colorado State University, Fort
Collins, Colorado, 138 pp., August, 1974. (Available from NTIS as PB 239 332).
Labadie, John W., Neil S. Grigg and Bruce H. Bradford, "Automatic Control of
Large-Scale Combined Sewer Systems," J.Envir.Engrg.Div., ASCE Proc, Vol. 101,
No. EE1, pp. 27-39, February, 1975.
Field, Richard, and John A. Lager, "Urban Runoff Pollution Control — State-
of-the-Art," J.Envir.Engrg.Div., Vol. 101, No. EE1, pp. 107-125, February, 1975.
36
67. Anderson, J. J., "Real-Time Computer Control of Urban Runoff," J.Hyd.Div..
ASCE P r o c , Vol. 96, No. HY1, pp. 153-164, January, 1970.
68. Harris, Garth S., "Real Time Routing of Flood Hydrographs in Storm Sewers,"
J.Hyd.Div.. ASCE P r o c , Vol. 96, No. HY6, pp. 1247-1260, June, 1970.
69. Gibbs, C. V., S. M. Alexander and C. P. Leiser, "System for Regulation of

Combined Sewage Flows," J.San.EnRrR.Div., ASCE Proc., Vol. 98, No. SA6,
pp. 951-972, December, 1972.
70. Evenson, Donald E., and Donald J. Smith, "Mathematical Modeling of San
Francisco Bay," a paper presented at ASCE Annual and National Environmental
Engineering Convention, October 21-25, 1974, Kansas City, Missouri, Meeting
Preprint 2380, 30 pp., ASCE, New York, N.Y.
71. Soice, V. Phillip, "Application of a Water Quality Model to the Denver

Metropolitan Area," pp. 1165-1182, Volume II, Symposium on Modeling Techniques.
ASCE, New York, N.Y., 1975.
72. Lombardo, Pio S., and Ronald F. Ott, "Water Quality Simulation and Application,"
Water Resources Bulletin, Vol. 10, No. 1, pp. 1-9, February, 1974.
73. Chen, Carl W., "Concepts and Utilities of Ecologie Model," J.San.Engrg.Div.,
ASCE P r o c , Vol. 95, No. SA5, pp. 1085-1097, October, 1970.
74. Orlob, Gerald T., "Mathematical Modeling of Estuarial Systems," pp. 78-128,
Vol. 1, Proceedings of International Symposium on Modeling Techniques in
Water Resources Systems,. May 9-12, 1972, Ottawa, Ontario, Canada.
75. Hydrocomp International, Inc., "Simulation of Discharge and Stage Frequency

for Flood Plain Mapping in the North Branch of the Chicago River," Palo Alto,
California, 75 pp., February, 1971.
76. Hydrologie Engineering Center, U.S. Army Corps of Engineers, "Water Quality
for River-Reservoir Systems," Computer Programs 401-F2-L2100 and -L2100A,
Davis, California, 210 pp., July, 1974.
77. Willey, R. G., "Water Quality Evaluation of Aquatic Systems," ASCE Convention,
New Orleans, Louisiana, April 14-18, 1975, Meeting Preprint 2420, 20 pp.,
ASCE, New York, N.Y.
78. Systems Control, Inc., Palo Alto, California, "Use of Mathematical Models for
Water Quality Planning," WRIS Technical Bulletin No. 3, Department of Ecology,
State of Washington, 212 pp., June, 1974.
79. Systems Control, Inc., Evaluation of Water Quality Models; A Management

Guide for Planners. Socioeconomic Environmental Studies Series EPA-600/5-76-004,
GPO, Washington, D.C., 176 pp., July, 1976.
80. Pentland, R. L., P. J. Reynolds and Asit K. Biswas, "Water Quality Modelling:
State-of-the-Art," pp. 481-496, Vol. 3, Proceedings of International Symposium
on Modeling Techniques in Water Resources Systems. May 9-12. 1972. Ottawa.
Canada.
37
Canale, Raymond P., Ed., Modeling Biochemical Processes In Aquatic Ecosystems,-
Ann Arbor Science Publishers, Inc., Ann Arbor, Michigan, approx. 500 pp., 1975.
Gordon, R., and M. Spaulding, "A Bibliography of Numerical Models for Tidal
Rivers, Estuaries and Coastal Waters," Marine Technical Report 32, University
of Rhode Island, Narragansett, R.I. 02882, 55 pages, 1974.
Hinwood, Jon B., and Ian G. Wallis, "Classification of Models of Tidal Waters,"
J.Hyd.Div., ASCE P r o c , Vol. 101, No. HY10, pp. 1315-1331, October, 1975.
Lombardo, P. S., "Critical Review of Currently Available Water Quality Models,"

Hydrocomp, Inc., 101 pp., July, 1973. (Available from NTIS as PB 222 265).
Lassiter, Ray R., Modeling Dynamics of Biological and Chemical Components of

Aquatic Ecosystems, Ecological Research Series EPA-660/3-75-012, GPO,
Washington, D.C., 54 pp., May, 1975. (Available from NTIS as PB 241 987).
McBean, Edward A., and Daniel P. Loucks, "Planning and Analyzing of

Metropolitan Water Resource Systems," Water Resources and Marine Sciences
Center Technical Report No. 84, Cornell University, Ithaca, New York, 805
pp., June, 1974. (Available from NTIS as PB 235 257).
Hydrologie Engineering Center, Corps of Engineers, "HEC-4, Monthly Streamflow

Simulation," Generalized Computer Program 723-340, Davis, California, 113 pp.,
February, 1971.
Jackson, Barbara Bund, "The Use of Streamflow Models in Planning," Water

Resources Research, Vol. II, No. 1, pp. 54-63, February, 1975.
Mátalas, N. C., "Developments in Stochastic Hydrology," Reviews of Geophysics

and Space Physics (AGU), Vol. 13, No. 3, pp. 67-73, July, 1975.
National Environmental Research Center, Corvallis, Oregon, A Conceptual

Model for the Movement of Pesticides Through the Environment, Ecological
Research Series EPA-660/3-74-024, GPO, Washington, D.C., 79 pp., December,
1974.
Shubinski, R. P., and L. A. Roesner, "Modeling Runoff from Undeveloped Areas,"

a paper presented at the ASCE Convention, New Orleans, Louisiana, April 18,
1975, 22 pp.
Prickett, T. A., and C. G. Lonnquist, Selected Digital Computer Techniques for

Groundwater Resource Evaluation, Illinois State Water Survey Bulletin 55,
Urbana, Illinois, 62 pp., 1971; and Prickett, T. A., and C. G. Lonnquist,
Aquifer Simulator Model for Use on Disk Supported Small Computer Systems,
Illinois State Water Survey Circular 114, Urbana, Illinois, 21 pp., 1973.
Gelhar, L. W., and J. L. Wilson, "Ground-Water Quality Modeling," Ground

Water, Vol. 12, No. 6, pp. 399-408, November-December, 1974.
Water Resources Engineers, Future Direction of Urban Water Models, Environmental

Protection Technology Series EPA-600/2-76-058, GPO, Washington, D<,C0, 95 pp.,
February, 1976. (Available from NTIS as PB 249 049).
38
95. Hobbs, N., and J. D. Britton, Denver Board of Water Commissioners, "Computer
Modeling Applications in Urban Water Planning," ASCE Urban Water Resources
March, 1974. (Available from NTIS as PB 233 673).
96. Department of Economics, Georgetown University, The Integrated Multi-Media

Pollution Model, U.S.EPA Socioeconomic Studies Series EPA-600/5-74-020, GPO,
Washington, D.C., 259 pp., February, 1974.
97. American Society of Civil Engineers, Design and Construction of Sanitary and
Storm Sewers, Manuals and Reports on Engineering Practice No. 37, New York,
N.Y., 332 pp., 1969.
98. McPherson, M. B., "Some Notes on the Rational Method of Storm Drain Design,"
ASCE Urban Water Resources Research Program Technical Memorandum No. 6, ASCE,
New York, N.Y., 84 pp., January 22, 1969. (Available from NTIS as PB 184 701).
99. Graham, P. H., L. S. Costello and H. J. Mallon, "Estimation of Imperviousness

and Specific Curb Length for Forecasting Stormwater Quality and Quantity,"
J.WPCF, Vol. 46, No. 4, pp. 717-725, April, 1974.
100. Gluck, W. R., and R. H. McCuen, "Estimating Land Use Characteristics for
Hydrologie Models," Water Resources Research, Vol. 11, No. 1, pp. 177-179,
February, 1975.
101. Gburek, W. J., "Discussion of Hydrologie Consequences of Rainfall

Augmentation," J.Hyd.Div., ASCE Proc., Vol. 97, No. HY12, pp. 2114-2115,
December, 1971.
102. University City Science Center, Philadelphia, Storm Water Management Model;
Dissemination and User Assistance, Environmental Protection Technology Series
EPA-670/2-75-041, GPO, Washington, D.C., 47 pp., May, 1975.
103. Torno, Harry C , "A Model for Assessing Impact of Stormwater Runoff and
Combined Sewer Overflows and Evaluating Pollution Abatement Alternatives,"
Water Research, Vol. 9, No. 9, pp. 813-815, September, 1975.
104. American Geophysical Union, Precipitation Analysis for Hydrologie Modeling,

National Symposium of June 26-28, 1975, Davis, California, 300 pp.,
Washington, D.C., 1975.
105. Hydrocomp, Inc., "Evaluation of the Effects of Urbanization on Aquatic

Ecology and Hydrologie Regimes," Palo Alto, California, 159 pp., July, 1975.
106. Field, R., J. Curtis and R. Bowden, "¡Literature Review: Urban Runoff and
Combined Sewer Overflow," J.WPCF, Vol. 48, No. 6, pp. 1191-1206, June, 1976.
107. Skipworth, W. E„, and W. L. Moore, "Evaluation of Urban Runoff by Watershed

Simulation," Center for Research in Water Resources Technical Report HYD-14-7601,
CRWR-127, The University of Texas at Austin, 121 pp., May, 1976.
39
108. Gibson, M., R. H. Ramsey, B. J. Claborn, R. M. Sweazy and D. M. Wells,
"Variation of Urban Runoff Quality and Quantity with Duration and Intensity
of Storms," Final Report, Volume 1, "Dry Weather Flows," No. WRC-75-1, Texas
Tech.University, Lubbock, 71 pp., December, 1975; and Bailey, B. H., R. H.
Ramsey, B. J. Claborn, R. M. Sweazy and D. M. Wells, "Variation of Urban
Runoff Quality and Quantity with Duration and Intensity of Storms," Final
Report, Volume 3, "Analysis of Flow Models," No. WRC-75-3, Texas Tech
University, Lubbock, 23 pp., December, 1975.
109. Jennings, Marshall E., and Harold C. Mattraw, "Comparison of the Predictive
Accuracy of Models of Urban Flow and Water-Quality Processes," Proceedings,
National Symposium on Urban Hydrology, Hydraulics and Sediment Control,
Lexington, Kentucky, pp. 1-8, July, 1976.
110. Stubchaer, James M., "The Santa Barbara Urban Hydrograph Method," pp. 131-141;
Harm, C. T., "Comparison of Methods for Developing Urban Runoff Hydrographs,"
pp. 143-148; and-Delleur, J. W., A. R. Rao and H. Hossain, "On Modeling the
Runoff Process in Urban Areas," pp. 193-208: "Proceedings, National Symposium
on Urban Hydrology and Sediment Control, July 28-31, 1975," Office of Research
and Engineering Services report UKY BU 109, University of Kentucky, Lexington,
314 pp., November, 1975.
111. Hinwood, Jon B., and Ian G. Wallis, "Review of Models of Tidal Waters,"
J.Hyd.Div., ASCE Proc, Vol. 101, No. HY11, pp. 1405-1421, November, 1975.
112. Ott, Wayne R., (editor), Environmental Modeling and Simulation, (Conference
Proceedings, April 19-22, 1976, Cincinnati, Ohio), Office of Research and
Development and Office of Planning and Management, U.S. Environmental
Protection Agency report EPA 600/9-76-016, 847 pp., July, 1976. (Available
from NTIS).
40
Section 3 Urban catchment research
Introduction
From 1948 through 1967 The Johns Hopkins University conducted a Storm
Drainage Research Project, the most comprehensive field research project to date.
Over the 1948-1967 period a total of 52 different sewered catchments were gaged.(*)*
Data considered to be among the best accumulated during the project are available
for two catchments, Northwo od(2) (19-ha) and Gray Haven^ 3 ^ (9-ha). No water quality
sampling was performed.
Findings from the first satisfactorily instrumented sewered catchment

for ascertaining runoff water quality were reported in 1 9 6 4 ^ ) (Mt. Washington,
Cincinnati, Ohio, 12-ha).
A 1964 handbook^-*' noted a serious need for much more extensive urban
drainage field research, consistent with the huge public expenditures in these
facilities. The major finding soon afterward of the first Engineering Foundation
conference co-sponsored by what is now the ASCE Urban Water Resources Research
Council was that the acquisition of much more field data was the greatest
research need in urban hydrology.(k'
In 1969, a plan for a nationwide field data acquisition program was

published.''' Major features of the proposed plan are shown schematically in
Figure 1.(7) Principal benefits that were expected to accrue from an effective
national program of urban storm drainage research were: (1), national criteria
for more efficient and dependable planning, implementation and operation of
urban storm drainage facilities, on acceptable socio-economic bases; (2), national
quantification of pollution from urban storm drainage systems; and (3), resolution
of the only major criteria missing for effective planning, design and operation
of quantity and quality exchanges between various urban water service categories.
The "National Data Network" of Figure 1 calls for several installations repre-
senting a variety of land uses in each of several representative metropolitan
regions, with respect to data acquisition, reduction and reporting. Each
installation would have had synchronous, real-time, central recording of rainfall,
runoff and quality data. In each research catchment, flow would be measured and
water quality samples would be taken not only at the outfall but at several
junctions of the underground sewer system and at several street inlets; and
quality and flow from roofs of sample buildings was to have been determined from
time to time. Concurrent instrumentation refinement and mathematical modeling
development was strongly recommended. The minimum three-year national effort pro-
posed would have cost a total of less than one-third of one per cent of the national
investment in construction of storm drainage facilities for only one year.
The U.S. Geological Survey has long been active in urban strearaflow data
collection.
(8)
The sewered catchment "Pilot Installations" of Figure 1 have been
satisfied and progress on the "National Data Network" of Figure 1 has been made by
the USGS in cooperative projects with local governments in Philadelphia (two catch-
ments, 1974), Pompano, Florida (three catchments, 1974) and Denver (three catchments,
1975). The Philadelphia
Numbered installation
references are is described
cited at the later
end of this in this section. (Also
Section.
41
<
m
cr
Z)
U.
O
2
<
or
CD r-
nr _L
r
n o
_j <
cr
<7, Lü
T (n
O LU
cr
NAT
AGE
< 2
rr <Lt
o LJ
Lu
ZT 2
< rr
_J o
LL H
Q to
Ul
û
LU
O
O
LU
cr
LU
rr
CD
U.
-1N3W3H3 A3*
42
described are: some other advances in instrumentation and data logging; some
advances in automatic control; the development of a data bank; a new field project
in Rochester, New York; and progress in water balance inventories).
In Section 2 it was suggested that mathematical model development for

sewered system applications has seemingly already greatly outpaced the data base for
model verification. Doubt has been expressed that practicing engineers will make
any wholesale abandonments of empirical methods in favor of more scientific approaches
until more extensive verification of various models has been accomplished.("' These
regressive views should not be interpreted as indictments of failure. There have been
considerable advances in the state-of-the-art, particularly in modeling, over the
past decade. However, as subjectively represented in Figure 2,(*-"' the needs for more
and better data are growing faster than such advances are taking place. The changing
"Information Needs" of Figure 2 refer to: rising project complexity; increasing
potentials for investments required; and constantly changing "rules of the game",
such as by the environmental movement, the 1972 Amendments to the Federal Water
Pollution Control Act, the flood insurance acts, more extensive land-use planning,
temporisation on policy by enforcement agencies, etc., all of which are interrelated
and lead to greater needs for hard information. As things now stand, local govern-
ments are substantially on their own for the acquisition of field data and no
integrated or national program exists or is in sight.
Flow Measurement
Inlets. A large part of The Johns Hopkins University project^' effort

was devoted to the smaller catchments, particularly those that drained directly to
stormwater inlets.'*•*•' Instrumentation featured synchronous, continuous, automatic
recording of rainfall rate and the flow level behind a weir.^ ' Weirs were installed
in the underground catch basins of several inlets. Two of the inlet catchments are
cited in Table 1 of Section 2. While some instrumentation improvements have since been
made, the basic techniques employed still represent the state-of-the-art on inlet
catchment measurements.
In-System„ A means for measuring flow in circular storm sewers has been
M^M
developed by the U.S. Geological Survey.^ ±J -' The first field installation was in
Philadelphia, described in a later subsection. Laboratory test documentation has
not yet been published, but an accuracy of Í57» under both open-channel and full-pipe
conditions has been claimed for this device.^ •3-' As shown in Figure 3, the device
features a U-shaped insert constriction, fabricated of fiberglass in symmetrical half-
sections to facilitate transportation and installation. Subcritical and supercritical
open-channel flow and pressure flows may be gaged.
Extensive laboratory tests have been conducted on an alternative device
which also features a ramped constriction, but the constriction is restricted to
the side of the sewer and does not affect the invert.^ ' There has not yet been
an opportunity to field-test this meter.
A detailed review of various potential in-system sewer flow meters was

made in 1968.( 1 5 )
General. As noted in Section 2, almost all flow gagings have been at

sewer outfalls or downstream from outfalls. Catchments gaged at outfalls were
identified in 1969'-' and partly sewered gaged catchments were identified in
1970.( °' Many of these catchments are cited in Table 1, Section 2. Features of
various existing devices for storm and combined sewer flow measurement have
been summarized,(17) based on a new comprehensive catalogue of over 70 different
generic types of primary flow measurement devices.( iö ' Field flow measurements
43
INFORMATION NEEDS
Lü
Q
(3
<
LU
>
Lu
FIELD DATA
ACQUISITION
1965 1975
TIME
FIGURE 2 - SCHEMATIC DESCRIPTION OF QUESTION-ANSWER

RELATIONSHIPS { 1 0 )
44
c
o
u
o)
V)
m
V)
O
i_
u
o to
ce
LU
UJ
m
Lü
<
o
en
en
in
Z3
ro
Lü
ce
C3
45
for the majority of currently and formerly gaged catchments in Table 1 of
Section 2 have been indirect via stage gages, only a part of which have been re-
lated to the characteristics of downstream hydraulic controls, with the remainder
depending on assumed conduit friction coefficients. Fully satisfactory equipment
or methods for continuous measurement of flow in storm or combined sewers have
still not been developed.^'' While devices or methods for measuring storm flows
can almost always be used for wastewater flow measurement, the opposite is
unfortunately not true.
A set of primary and secondary design goals for development of devices

for continuous measurement and recording of flows in storm and combined sewers
specified by the U.S. Environmental Protection Agency (U.S. EPA) have been
listed.( '' It appears that, for the present, only two types of devices can meet
most of the U.S. EPA requirements: ultrasonic flowmeters; and the venturi meter/
flume.'13,14; ^ version of the magnetic flowmeter, when development is completed
for measurement of open-channel flow, will also meet most of the requirements.
Less costly and less complex versions of such devices are seriously needed.' ''
Indicative of the fact that stormflows are difficult to gage is the

observation of a recent analysis of devices that it is not at all uncommon to
find readings that differ from spot field checks by 50 to 200 per cent.( 1 8 '
Water Quality Sampling
Water quality data has been collected on less than half of the catch-
ments listed in Table 1 of Section 2, and for only a fraction of these has such
data been used in model tests. Features of current commercially available
automatic liquid samplers have been summarized,^ly' based on a comprehensive
study reported in 19
73(20)
which has since been considerably extended.\ ¿í >
Because of the rapid changes that occur in storm and combined sewer flows, a
minimum sample collection interval of 3-minutes or 4-minutes is desirable and
discrete samplers should be used that have a minimum of 24 containers.' •"' None
of the existing samplers had been designed specifically for stormflow applications,
and it appears that none of the commercially available equipment is eminently
satisfactory for use in storm or combined sewer applications.("' Further, there
can be marked differences in results obtained with different types of equipment.' 2 *'
However, development continues. A promising prototype automatic sewer sampling
system intended expressly for storm and combined sewer applications has been
subjected to preliminary field testing, and has been tested under controlled
laboratory conditions along with four different types of commercially available
automatic samplers.' '
In possibly the only such measurements made thus far, flow was measured
and water quality samples were collected for the lawn, driveway and roof of a
single-family residence in Cincinnati, Ohio.(23) j n addition, background rain-
fall water quality was determined.
Discussed above is sampling for determination of biological and organic
and inorganic chemical constituents, including sediments. However, sampling of
sediments from erosion provoked by construction can lead to special considerations,
as evidenced by recent field research that included a gaging and sampling program
to determine the effects of urbanization on the hydrology and water quality of
initially natural areas.( 2 ^'
Systems have been developed for automatic, on-line measurement of

total organic carbon^") and suspended solids.' °'
46
Raingage Networks
Raingage networks in the metropolitan areas of the 20 largest U.S.

cities were described in 1969.' '' Since then, two networks have been installed
that deserve particular mention.
St. Louis. From 1970 through 1975 an urban-weather research program,

the world's first major field-laboratory program aimed at assessing the effect
of a major urban-industrial complex on precipitation, was conducted in the St.
Louis metropolitan area.(^"^ Instrumentation included 250 recording raingages.
Definite inadvertent weather and precipitation modifications have been docu-
mented in initial analyses of St. Louis d a t a ^ " ' and in comparisons with data
from other metropolitan areas.S-'0-'
San Francisco. Installation of the most advanced automatic raingage

network was completed in the City of San Francisco in 1971, as part of a
hydrologie and hydraulic data acquisition system.^ 31 ' Data from 30 tipping
bucket raingages in 1/4-mm increments are recorded at a central computer as they
occur, while flow levels recorded each 15-seconds are converted into discharge
rates via rating curves. Unfortunately, an analogous capability for obtaining
water quality data was not implemented.(32.) Features of the data acquisition
system are diagrammed in Figure 4.' ' A computer program operated off-line
interpolates between raingage network readings and prints out rainfall depths for
a fine grid covering the entire city.
Other Research. Advances in the characterization of urban rainfall

variations have been summarized, with an emphasis on real-time, prediction of
rainstorms for automatic operation of runoff control facilities.(34; The
characterization bibliography contains 151 entries.
Automatic Control
Components of full automatic operational control of urban water manage-

ment systems are identified in Figure 5.(35) While perhaps several more years of
development will be required to achieve complete, "hands off" automatic control
of the facilities of any major jurisdiction, the greatest advances thus far have
been in connection with the San Francisco master plan for abatement of pollution
from combined sewer overflows.Q3O-_5Ö.) Whether only monitoring, or direct
supervisory control, or complete automatic control (Figure 5 ) , is involved,
there are opportunities for direct-reading, remote sensing of some water quality
characteristics by using or adapting instrumentation developed for monitoring
wastewaters and perennial streams.(39J
Related Findings
In reviewing the effects of urbanization on low flow, total runoff,

infiltration and groundwater recharge, an ASCE task committee noted a variety of
related research projects in its report and the accompanying citation of 89
references. (^^ Space does not permit elaboration here.
Data Bank
Clearly indicated in Section 2 is that not much field data is available

for testing models of storm and combined sewer systems, and much of the data that
is available is not of very good quality. Adding to these difficulties has been
47
Si? •m
I-
- 1 CO
r~ c >-
CO
"•5
o
tem
îon
o
i •/> Vt '
rr .9
1 C
>* 0) ço
1 co
8. I
(0 3 CO
1 X "^ 1
LU k_ i ro o
a o
o c
1 VOc i1 <
X ü o
¡ 3U !
' D hi
1 Li. * 1
Li. i <
Q
I I
<
ce
CD
<
c
IB •o
«o
O
v V I-
i* a
a> c
o
O
O
-I
O s üEo .c
m
fO
CL 0),Q) CO 8 CL
1
.S? LU
CC
»2
CD
p-
c
V) o
l
_
2 "a
s-4
nito
20
eve
_j
CD V)
*
c
48
ON-SITE FIELD
VARIARLES INSTRUMENT
DISPLAY SENSING
INSTRUMENT
SIGNALS
TELEMETRY
SIGNAL PROCESSING AT CONTROL CENTER
LOGGING OATA CHARTING WARNING

FIELD ANALYSIS FIELD AND ALARM
PARAMETERS DISPLAYS PARAMETERS DISPLAYS
CONTROL
LOGIC
QUERIES
SURVEILLANCE IY HUMAN SUPERVISOR
I
JL
MANUAL
OVER-RIDE
CONTROL
sisiygs.
I
AUTOMAT IE CONTROL ON-SITE RESPONSE
CONTROL OISPLAY OF
SIGNALS SIGNALS SIGNALS
TELEMETRY RESPONSES TELEMETRY
GENERATION
-
FIÍTÕ ¡rmn
ToTTf
CONTROL ELEMENTS
ELEMENTS RESPONSE
ACTUATION SENSING
FIGURE 5- AUTOMATIC OPERATIONAL CONTROL COMPONENTS^ 35 )
49
the necessity for each modeler to acquire raw field data almost independently.
In order to ease this last obstacle in particular, the U.S. EPA is supporting a
project by the University of Florida that is attempting to bring together data
that is currently available and formulate it in a data file such that potential
model developers and current model users may have ready access to the data.'^*'
To be included by the project will be identification of data sources, establishment
of data reliability and accuracy, data collection, arrangement of the data in
common formats, and development of a mechanism for updating and disseminating the
data. A tentative data retrieval format and an initial detailed inventory of
potential data sources were completed by the project early in 1975.'^ '
Philadelphia Project
As part of a cooperative project with the U.S. Geological Survey, the

Philadelphia Water Department in 1974 completed installation, testing and initial
operation of a unique field sampling and data recording, transmitting and
reporting System.' 13 »^ 3 ' This system essentially satisfies the criteria for a
"Pilot Installation" of Figure 1 recommended by ASCE in 1969.
(7)
Two adjoining subcatchments are involved, Figure 6 . ^ 3 ' One is a 7-ha
commercial area with an average imperviousness of 69% and the other is a 33-ha
residential area with an average imperviousness of 63%. Flow is measured by means
of USGS flow gages (see Figure 3) in a 76-cm storm sewer of the commercial area
and in a 152-cm storm sewer of the residential area just upstream of their
confluence, where the flows enter the outfall sewer of the total catchment. Three
tipping bucket raingages are located on or near the subcatchments. Water quality
sampler intakes are installed near the. flow meters. Sensor-reading and related
equipment is housed in a Transmitting Station located in a vault nearby. All of
the equipment was designed to facilitate installation in an existing system.
Shown schematically in Figure 7 ^ 3 ^ are the basic components of the

Tustin data collection system. The Receiving Station is located in the offices
of the Water Department, a distance of about 16-km from the Transmitting Station
at the catchment. Hydrographs and hyetographs are generated from data tapes by
the Department's computer.
During the initial tests, water quality samples were taken at one-rainute
intervals, but sampling frequency can be programmed at from one to fifteen
minutes and sampling is commenced when pre-set flow rate magnitudes in the sewers
are exceeded. Up to 24 two-liter samples can be collected per storm at each of
the sewers. Sample bottles are located in a refrigerated chest that is maintained
at a temperature of approximately 4° Celsius.
Pressures in the approach and in the throat of the constriction of the

flow gage (Figure 3) are sensed by transducers at the Transmitting Station, and
dry nitrogen gas is bubbled at a constant rate through connecting tubes to the
two piezometer openings to maintain a constant reference pressure. Flows in
excess of a pre-set magnitude are normally recorded at a one-minute interval.
Data, assembly and recording equipment at the Transmitting Station

comprises integrated solid-state components, including analog-to-digital signal
converters and a signal multiplexer. Analog and digital signals can be recorded
at the Transmitting Station in addition to or instead of digital recording of
telemetered data at the Receiving Station.
50
COM MERCIA L AREA
RES IDENTIA L. AREA

SEWER GAGES (TWO)
RAINGAGES (ThHREE)
X
76-cm SEWER,
SCALE, APPRO
152-cm SEWER
Af en
<r
Q "^ïSjs. <
ZU
L M II II H X
o a.
*O0 ©
UJ
-J
J
1
rA!
J^~'m m» •
•
»
K
^
^
^r~i\ Vw
\ \
1 \ \ VA
l
\i —
UJ
Q
X'
<
a.
1.' /
¡If \
1 L. Il I VÄ <"
ii V*> uj
l\[
g K
\
1 1
• ! $ u|«tni ^
T T oc
\ / \ \ *^
/ \ \ >-
f~
/
\
\—7 \A Q
Y Ã ^>
"s ,» —}\ —y' \ » »-
\ » w
\ ' W
^^^
^
\ \
<
' r
\
'.À*-
Ião:
™UI
h-
\ " • %
'i^lV \ <
t >w^ \
\ / £
\ • _ PM • « O «o*1 1 S
\
i <" •l
1
| P
«i * 1 —
(
\ CO
\ <
i\ K •fBouMOH z
t-
co
3
• A V A||o$ 1
—
ontenac St.
FIGURE 6-
astor Ave.
1
oretto
_i
\
Ji £ u \
51
LU
tr co N
AINGA GES (THREE)
*0©
ii
EWER GAGES (TWO)
H
6-cm SEWER,
H
COM MERCIAL AREA
®
152-cm SEWER
-_l
<
ÜJ
<
re
RES DENTIA
#
X
CO
:ALE, APPRO
52
-*
\K
n
CD
tr
LU
=3
co
<
CO
g
tr
LU
<
<"
CO
I-
=3
Û
>
LU
<
LU
Q.
X
-J
Q
_J
Q.
X
i
HS\
CO
•tf
UJ
i-
>-
w
o
UJ
o
o
<
<
o
<
X
CL
-I
UJ
o
<
-J
I
a.
UJ
a:
<
Q
(-
i-
UJ
a:
CD
u.
m
I I
53
Water quantity and quality monitoring of streamflows'^^ and of a
combined sewer catchment'-*) by Philadelphia, and the raingage network operated
by the City,(") were all reported in 1969. A number of significant improvements
and advances have since been made by the Water Department, particularly in
automation and data analysis.\**) For example, catchment models described in
Section 2 are being used to analyze data from the Tustin catchment, and an
increasing number of raingage and other sensor signals are being telemetered
to the Department's offices for more rapid and efficient data collection and
processing.
Rochester Project
The City of Rochester, New York, is in Monroe County. Supported by a

U.S.EPA Research and Demonstration Grant, a project for abatement and management
of combined sewer overflow pollution is being conducted under the direction of
the Monroe County Pure Waters Authority and the Rochester Pure Waters District.
O'Brien & Gere Engineers, Inc., Syracuse, New York, are consultants to the
Rochester District on the project. Only one phase of the project, monitoring
and characterization of overflows, will be discussed here.
There are thirteen combined sewer overflow outfalls involved, draining

catchments ranging in size between 65-ha and 1,300-ha, and all are instrumented
for automatic flow sensing and quality sampling.'^) Either an ultrasonic level
sensor upstream from a weir or a level sensor in conjunction with an ultrasonic
velocity measuring system has been installed in each of the thirteen oversow
conduits. Samplers are automatically activated by a relay system preset ac a
given conduit flow level, and samples are collected at a 15-rainute interval during
overflow occurrences. Rainfall is measured at ten tipping bucket raingages
distributed over the study area. Flow sensor signals are telemetered to a central
computer facility where they are processed and converted into digital tabulated
values and entered onto punched tapes. Installation of the monitoring system was
completed in 1974.
A version of the U.S.EPA Storm Water Management Model (Section 2) is

being calibrated for each catchment, using field data from the monitoring system,
for use as an analytical tool to establish guidelines for the development of
overflow pollution abatement master plans. These guidelines will be designed
such that they should be applicable anywhere in the U . S . ( ^ '
Water Balance Inventories^"^
For quite some time the water balance (or water budget) method has been
employed in traditional hydrology to account for the movement of water through
the hydrological cycle over a selected time interval for a given drainage basin,
and in national and global assessments. Water quality considerations are as
important as those of quantity in urban hydrology. The term "water balance
inventory" is used to describe the determination of the quantity and quality
aspects of the fate of water, from its appearance as precipitation through its
departure from a metropolis as runoff and évapotranspiration. It is a complete
inventory at a given time. A general accounting for the overall movement of water
and pollutants can be ascertained by comparing such complete inventories over
successive time intervals.
Satisfactory evaluation of hydrological effects of urbanization, and

related development of strategies for resource management and environmental
54
protection, have been hampered around the world because of minuscule research
investments despite the economic and environmental importance of urban water
resources. Serious obstacles have impeded advances, but progress is being made
in a few notable instances.
An ASCE task committee on the hydrological effects of urbanization^^'

highlighted in the conclusions of its final report a need for more comprehensive
and more highly systematized investigations of hydrological changes in urban areas.
The committee found that available information is of severely limited transfer
value, partially as a result of the web of complexities imposed when open land
becomes urbanized. While concluding that useful results will be obtained only
via coordinated efforts on a metropolitan scale, the unsolved central problem is
the absence of suitable means for achieving the needed coordination.
The International Hydrological Decade/Unesco Subgroup on the Effects of

Urbanization on the Hydrological Environment has arrived at very similar
conelus ions : ( ^ '
1. More metropolitan-scale water-balance inventories and their analysis

should be undertaken as a means for improving overall water resources
planning and management, and follow-on inventories should be made
periodically to document change and to provide a better understanding
of the hydrological effects of progressive urbanization.
2. The interrelation and interdependence of water and wastewater and the

competition and conflict between multiple jurisdictions have
intensified with the growth of metropolitan areas. The variety of
uses for water in metropolitan areas are continually enlarging,
particularly for recreational purposes and for esthetic enhancement.
Thus, hydrological surveys of urban areas should be updated frequently
and regularly.
Both the ASCE and IHD groups struggled with the quantification of
generic hydrological effects of urbanization on national scales. Despite the fact
that most problems and effects are very similar in technologically and economically
advanced countries, very few generalities can be drawn. To cite one of the few
successful examples, it has been demonstrated in a number of countries that
urbanization increases the local contribution of direct runoff volume and that
systems of storm drainage conduits result in greater direct runoff peaks with
shorter rise times than for pre-urban conditions. A source'of impotence in
generalization is the fact that, world-wide, the field of urban hydrology is
almost devoid of modern research investment and that there has been relatively
little study to date of the effect of human settlements upon natural hydrological
conditions.
These calls for water-balance inventories are the direct result of a

clear recognition of the interrelation, interdependence, and interconnection of
the elements of the water resources of a metropolis. That is, a total resource
systems approach is necessary if subsystem phenomena truly are to be identified,
because of the complex linkages involved. Also, metropolitan land-use constantly
changes, an occurrence which can only be accommodated for a complex system by
using a total systems approach.
The type of inventory envisaged has been partially achieved for

metropolitan San Antonio, Texas,(48,49; «here it is called a "systems description".
55
The quantitative aspects of soil water, surface water and ground water have been
inventoried and development is under way of a water quality balance, although
the latter may require several years for achievement. "Attempts to study any
inventory system apart from the total system does not enable the investigator to
see the total impact of alternatives."^*5' Another quantitative example is the
average water balance for metropolitan Chicago.(50y
Among several metropolitan environmental and resource studies undertaken

by the U.S. Geological Survey was the San Francisco Bay Region Environment and
Resources Planning Study,wlJ completed in 1975. Such comprehensive studies
greatly increase the feasibility of conducting metropolitan water resource
inventories because they facilitate the accumulation of necessary information.
(The ASCE Program has just completed a report for the USGS on its San Francisco
Study, on garth science information needs in local government water control master
planning^ 52 ^).
Conclusions
Attempts to manage water resources on a comprehensive basis at the

metropolitan level have been frustrated by the fact that the planning, implementa-
tion, and operation of facilities and- services are usually territorially or
functionally fragmented, or both, in metropolitan areas. There are in excess of
80,000 units of local government in the United States. The 50 largest metropolitan
areas have a total of close to 3,000 special districts and public authorities
involved in water-related services, greatly outnumbering the major municipal water
agencies in their midst.
Hydrological-balance surveys are grossly complicated and frequently

frustrated by a typical fractionalized-management authority. Despite the great
need for them, metropolitan water-balance inventories at a national scale will
be slow in coming. It is ironic to note that the National Water Commission has
found that the national investment thus far in urban facilities have been seven-
tenths of total investments on all water resource facilities, and the
disproportion of expenditures is expected to become even greater in the
future.C53) Yet, we will have total water balance inventories of all our river
basins well before we have them for even our larger metropolitan areas, if
present trends continue. The fact that three-fourths of our people live in
metropolitan America continually escapes the attention of framers and imple-
menters of national policy.
In the meanwhile, as this report has attempted to indicate, fractionalized,

largely independent, fretful but perhaps impressive progress is being made in
urban hydrology research, and accelerating planning activities nationwide imply
even greater attention in the immediate future. Researchers are implored to
consider the complex actualities of urban catchments when engaged in system
simulation,'-**' as has been the case in developing capabilities for automatic
control of stormwater facilities for pollution abatement and flood mitigation.^''
A recent appraisal of future modeling ne eds( 56 > has revealed that from
the standpoint of runoff quality simulation, no significant progress has been
made beyond the rather simplistic Storm Water Management Model (SWMM) that was
developed under U.S.EPA auspices (Section 2 ) . A primary reason for this lag has
been ascribed to" the lack of sufficient quality data to evaluate thoroughly
the quality portions of SWMM and its variants, let alone to evolve improvements.
"We believe that the quality of runoff from urban areas requires an immediate
56
and large amount of attention."(56)
Or, as summarized in a recent national conference on urban runoff:

"When we consider the length of time that studies have been made of treating
sanitary wastes and the principal industrial wastes, or of hydrology in general,
it is apparent that water quality analysis of urban runoff is a relative new-
comer and a neglected field."(5.7)
In the meanwhile, extensive planning for water pollution abatement is

under way, including control of urban runoff pollution. Interim guidelines for
quantification of pollutant .loadings have been offered for the use of urban
planners.('
The above emphasis on quality should not be allowed to obscure the

fact that much more also has to be learned on the quantitative aspects of urban
runoff.
Epilogue. The presentation in this Section was completed in November,

1975, and the only changes made since (October, 1976) were the updating of some
reference sources. Not to be overlooked is the successor project to the METROMEX
study in metropolitan St. Louis,(28) initiated in metropolitan Chicago in 1974.(
The Chicago project features 300 raingages distributed over an area of about
10,000-sq. km. One of the project goals is to develop a real-time prediction and
monitoring system and methodology for specifying rainfall quantity over a metro-
politan area using a weather radar.
57
Acknowledgment
Individuals who have supplied information specifically for this report

are identified below. Any errors in the report should be ascribed to misinter-
pretation by the writer and not to these generous cooperators. Citation is
alphabetical by surname.
Jesse W. Abbott (Hydrologie Engineering Center, Corps of Engineers)

C. Edward Bowers (University of Minnesota)
David J. Carleo (O'Brien & Gere Engineers, Inc.)
Stanley A. Changnon, Jr. (Illinois State Water Survey)
Harold C. Coffee (DPW, San Francisco)
Norman H. Crawford (Hydrocomp, Inc.)
Jacque W. Delleur (Purdue University)
Richard Field (EPA, Edison, N.J.)
Neil S. Grigg (Colorado State University)
Charles A. Hansen (Envirex, Inc., Environmental Sciences Division)
James P. Heaney (University of Florida)
Charles D. D. Howard (Unies, Ltd., Canada)
Stifel W. Jens (Reitz and Jens, Inc.)
John A. Lager (Metcalf and Eddy, Inc., Palo Alto)
Peter A. Mangarella (University of Massachusetts)
J. Marsalak (Canada Centre for Inland Waters)
Kenneth A. Pew (Cleveland Regional Sewer District)
Joseph V. Radziul (Philadelphia Water Department)
Robert M. Ragan (University of Maryland)
H. S. Rao (Systems Control, Inc.)
Larry A. Roesner (Water Resources Engineers, Walnut Creek)
Phillip E. Shelley (EG & G Washington Analytical Services Center, Inc.)
Robert P. Shubinski (Water Resources Engineers, Springfield, Va.)
Michael B. Sonnen (Water Resources Engineers, Walnut Creek)
John B. Stall (Illinois State Water Survey)
Harry C. Torno (EPA, Washington, D.C.)
Jekabs P. Vittands (Metcalf and Eddy, Inc., Boston)
Donald H. Waller (Nova Scotia Technical College, Canada)
Harry G. Wenzel (University of Illinois)
William Whipple, Jr. (Rutgers University)
Robert G. Willey (Hydrologie. Engineering Center, Corps of Engineers)
58
References
1. Horn, Dennis R., and Norbert Dee, "A. Description of Drainage Areas Gaged by
the Storni Drainage Research Project," Storm Drainage Research Project
Technical Report No. 5, Department of Environmental Engineering Science,
The Johns Hopkins University, Baltimore, Maryland, 71 pp., January, 1967.
2. Tucker, L. S., "Northwood Gaging Installation, Baltimore - Instrumentation

and Data," ASCE Urban Water Resources Research Program Technical Memorandum
No. 1, ASCE, New York, N.Y., 36 pp., August 1, 1968. (Available from NTIS
as PB 182 786).
3. Tucker, L. S., "Availability of Rainfall-Runoff Data for Sewered Drainage

Catchments," ASCE Urban Water Resources Research Program Technical Memorandum
No. 8, ASCE, New York, N.Y., 93 pp., March 3, 1969. (Available from NTIS as
PB 184 703).
4. Weibel, S. R., R. J. Anderson and R. L. Woodward, "Urban Land Runoff as a

Factor in Stream Pollution," J.WPCF, Vol. 36, No. 7, pp. 914-924, July, 1964,
5. Jens, St'ifel W., and M. B. McPherson, "Hydrology of Urban Arras," Section 20,
45 pp., Handbook of Applied Hydrology, V. T. Chow, ed., McGraw-Hill Book
Company, New York, N.Y., 1964.
6. "Report — Engineering Foundation Research Conference — Urban Hydrology

Research," August 9-13, 1965, Proctor Academy, Andover, N.H., Co-sponsored
by the Urban Hydrology Research Council, ASCE, 25 pp.
7. American Society of Civil Engineers, Basic Information Needs in Urban

Hydrology, A Study for the Geological Survey, U.S. Dept. of the Interior,
ASCE, New York, N.Y., 112 pp., April, 1969. (Available from ASCE).
8. Schneider, William J., "The U.S. Geological Survey Urban Water Program,"
pp. 165-168, Effects of Watershed Changes on Streamflow, edited by W. L.
Moore and C. W. Morgan, University of Texas Press, Austin, Texas, 1969.
9. VanSickle, Donald, "Storm Drainage Planning for Urban Growth," pp. 16-21,
Urban Runoff, Quantity and Quality, ASCE, New York, N.Y., 1975.
10. Work Group on User Needs for Problem Solving, "Conclusions," pp. 34-35,
11. Viessman, Warren, Jr., and John C. Geyer, "Characteristics of the Inlet
Hydrograph," J.Hyd.Div., ASCE P r o c , Vol. 88, No. HY5, pp. 245-268,
September, 1962. Authors1 closure to discussion: Vol. 89, HY4, p. 229,
July, 1963.
12. Knapp, John W., John C. Schaake, Jr., and Warren Viessman, Jr., "Measuring
Rainfall and Runoff at Storm-water Inlets," J.Hyd.Div., Vol. 89, No. HY5,
pp. 99-115, September, 1963.
13. Sraoot, George F., "A Rainfall-Runoff Quantity-Quality Data Collection

System," pp. 178-183, Urban Runoff, Quantity and Quality, ASCE, New York,
N.Y., 1975.
59
14. Wenzel, Harry G., Jr., "Meter for Sewer Flow Measurements," J.Hyd.Dlv..
ASCE Proc, Vol. 101, No. HY1, pp. 115-133, January, 1975.
15. Wenzel, Harry G., Mr., "A Critical Review of Methods of Measuring Discharge
Within a Sewer Pipe," ASCE Urban Water Resources Research Program Technical
Memorandum No. 4, ASCE, New York, N.Y., 20 pp., September, 1968. (Available
16. Tucker, L. S., "Availability of Rainfall-Runoff Data for Partly Sewered Urban
Drainage Catchments," ASCE Urban Water Resources Research Program Technical
Memorandum No. 13, ASCE, New York, N.Y., 156 pp., March, 1970. (Available
17. KirkpatricK, George A., "A Review of Flow Measuring Devices," pp. 191-198,
Urban Runoff. Quantity and Quality, ASCE, New York, N.Y., 1975.
18. EG&G Washington Analytical Services Center, Inc., Sewer Flow Measurement, A
State-of-the-Art Assessment, Environmental Protection Technology Series
EPA-600/2-75-027, GPO, Washington, D.C., 423 pp., November, 1975. (Available
19. Shelley, Philip E., "A Review of Automatic Liquid Samplers," pp. 183-191,
20. Hydrospace-Challenger, Inc., Rockville, Maryland, An Assessment of Automatic

Sewer Flow Samplers, Environmental Protection Technology Series EPA-R2-73-
261, GPO, Washington, D.C., 233 pp., June, 1973.
21. EG&G Washington Analytical Services Center, Inc., An Assessment of Automatic

Sewer Flow Samplers — 1975, Environmental Protection Technology Series
EPA-600/2-75-065, GPO, Washington, D.C., 336 pp., December, 1975. (Available
22. EG&G Washington Analytical Services Center, Inc., Design and Testing of a
Prototype Automatic Sewer Sampling System, Environmental Protection Technology
Series EPA-600/2-76-006, GPO, Washington, D.C., 96 pp., March, 1976.
23. Beers, Gary D., "Management of Stormwater Runoff in Suburban Environments,"

Engineering-Science, Inc., Cincinnati, Ohio, 110 pp., November, 1973.
24. Hittman Associates, Inc., Columbia, Maryland, Joint Construction Sediment

Control Project, Environmental Protection Technology Series EPA-660/2-73-035,
GPO, Washington, D.C., 167 pp., April, 1974.
25. Tulumello, Angelo, Automatic Organic Monitoring System for Storm and Combined
Sewers, Environmental Protection Technology Series EPA-670/2-75-067, GPO,
Washington, D.C., 70 pp., June, 1975.
26. Liskowitz, John W., and Gerald J. Franey, Suspended Solids Monitor,
Environmental Protection Technology Series EPA-670/2-75-002, GPO, Washington,
D.C.,' 40 pp., April, 1975.
60
Tucker, L. S., "Raingage Networks in the Largest Cities," ASCE Urban Water
Resources Research Program Technical Memorandum No. 9, ASCE, New York, N„Y.,
90 pp., March 17, 1969. (Available from NTIS as PB 184 704).
Illinois State Water Survey, "METROMEX: An Overview of Illinois State Water

Survey Projects," Bulletin of the American Meteorological Society. Vol. 55,
No. 2, pp. 89-100, February, 1974. (ISWS Reprint Series No. 251).
Schickedanz, P. T., "Inadvertent Rain Modification As Indicated by Surface

Raincells," Journal of Applied Meteorology, Vol. 13, No. 8, pp. 891-900,
December, 1974.
Changnon, Stanley A., Jr., "Inadvertent Weather and Precipitation Modification

by Urbanization," J.Irrig.& Drain.Div., ASCE Proc., Vol. 99, No. IR1, pp. 27-
41, March, 1973.
Anonymous, "Computerized Monitoring System Helps San Francisco Control

Overflow," Water and Sewage Works, Vol. 119, No. 8, pp. 44-45, August, 1972.
Wenzel, Harry G., Jr., "Drainage Technology — The State of the Art," pp. 72-
81, Urban Runoff, Quantity and Quality. ASCE, New York, N.Y., 1975.
Friedland, Alan 0., "San Francisco Precipitation and Flow Measuring

Network," pp. 205-212, Urban Runoff, Quantity and Quality, ASCE, New York,
N.Y., 1975.
Grigg, Neil S., John W. Labadie and Harry G. Wenzel, "Metropolitan Water
Intelligence Systems," Completion Report, Phase III, Colorado State
University, Fort Collins, Colorado, 258 pp., July, 1974. (Available from
NTIS as PB 235 532).
McPherson, M. B., "Feasibility of the Metropolitan Water Intelligence System

Concept (Integrated Automatic Operational Control)," ASCE Urban Water
Resources Research Program, Technical Memorandum No. 15, ASCE, New York,
N.Y., 110 pp., December, 1971. (Available from NTIS as PB 207 301).
Giessner, W. R., R. T. Cockburn, F. H. Moss and M. E. Noonan, "Planning and

Control of Combined Sewerage Systems," J.Environ.Engrg.Div.. ASCE Proc,
Vol. 100, No. EE4, pp. 1013-1032, August, 1974.
Friedland, Alan 0., "City-Wide Master Planning," pp. 8-12, Urban Runoff,
Quantity and Quality. ASCE, New York, N.Y., 1975.
McPherson, M. B., "Innovation: A Case Study," ASCE Urban Water Resources

February, 1974. (Available from NTIS as PB 232 166).
Metcalf and Eddy, Inc., Palo Alto, California, Urban Stormwater Management
and Technology: An Assessment, Environmental Protection Technology Series
EPA-670/2-74-040, GPO, Washington, D.C., 446 pp., December, 1974.
Committee Report, "Aspects of Hydrological Effects of Urbanization,"

J.Hyd.Div., ASCE Proc, Vol. 101, No. HY5, pp. 449-468, May, 1975.
61
Torno, Harry C , "Storm Water Management Models," pp. 82-89, Urban Runoff.
Huber, Wayne C , James P. Heaney and Amauri A. Arroyo, "Establishment of an

Urban Rainfall-Runoff Data Base," Interim Report, Phase I, EPA Project No.
68-03-0496, Department of Environmental Engineering Sciences, University
of Florida, Gainesville, Florida 32611, 65 pp., March, 1975.
Research and Development Division, "Fiscal 1974 Annual Report," Philadelphia

Water Department, Joseph V. Radziul, Chief of R. & D., 1180 Municipal Services
Building, Philadelphia, Pennsylvania 19107, 109 pp., August, 1974.
Tucker, L. S., "Sewered Drainage Catchments in Major Cities," ASCE Urban Water
Resources Research Program Technical Memorandum No. 10, ASCE, New York, N.Y.,
72 pp., March 31, 1969. (Available from NTIS as PB 184 705).
Richardson, Steven L., "A Solution to Combined Sewer Overflows," Clearwaters,

New York Water Pollution Control Association, Vol. 4, No. 4, pp. 14-15,
January, 1975.
McPherson, M. B., "Need for Metropolitan Water Balance Inventories," J.Hyd.

Div., ASCE Proc, Vol. 99, No. HY10, pp. 1837-1848, October, 1973. Author's
closure, Vol. 101, No. HY4, p. 409, April, 1975.
Subgroup of IHD, Hydrological Effects of Urbanization. Studies and Reports

in Hydrology No. 18, The Unesco Press, Paris, 280 pp., 1974.
Koch, C. T., "Systems Description for Urban Water Resources," pp. 19-41,
Proceedings, 16th Annual Conference on Water for Texas, "Urban Water Resource
Planning and Management," Texas A and M University, College Station, Texas,
Sept., 1971. (Available from NTIS as PB 210 325).
Koch, C. T., "Methodology Managing Resources," J.AWWA, Vol. 64, No. 2, pp.
122-126, Feb., 1972.
Sheaffer, John R., and Arthur T. Zeizel, "The Water Resource in Northeastern
Illinois: Planning Its Use," Technical Report No. 4, Northeastern Illinois
Metropolitan Area Planning Commission, Chicago, Illinois, 182 pp., June, 1966.
U.S. Geological Survey, "Program Design, 1971 — San Francisco Bay Region
Environment and Resources Planning Study," Washington, D.C., October, 1971.
McPherson, M. B., Regional Earth Science Information in Local Water

Management. ASCE, New York, N.Y., 155 pp., July, 1975.
Koelzer, V. A., "Urban Water Management," J.AWWA, Vol. 64, No. 9, pp. 537-544,
September, 1972.
Jens, Stifel W., and Henry M. Reitz, "Some Notes on Changing Approaches to
Urban Stormwater Management," pp. 123-130 in "Proceedings, National Sympoaiun
on Urban Hydrology and Sediment Control, July 28-31, 1975/' Office of Research
and Engineering Servicer Report UKY BU 109, University of Kentucky, Lexington,
62
55. Grigg, Neil S., and John W. Labadie, "Computing the Big Picture," Water and
Wastes Engineering, Vol. 12, No. 5, pp. 37-39 & 86, May, 1975.
56. Sonnen, Michael B., Larry A. Roesner and Robert P. Shubinski, "Urban Water
Management Models," pp. 89-97, Urban Runoff. Quantity and Quality. ASCE,
New York, N.Y., 1975.
57. Whipple, William, Jr., "Conference Summary," pp. 255-258, Urban Runoff,
58. Shubinski, Robert P., and Steven N. Nelson, "Effects of Urbanization on Water
Quality," ASCE Urban Water Resources Research Program Technical Memorandum
No. 26, ASCE, New York, N.Y., 34 pp., March, 1975. (Available from NTIS as
PB 242 297).
59. Changnon, Stanley A., Jr., and Floyd A. Huff, "Chicago Hydroraeteorological
Area Project: A Comprehensive New Study of Urban Hydrometeorology," First
Interim Report for NSF/RANN, Atmospheric Sciences Section, Illinois State
Water Survey, Urbana, Illinois, 69 pp., September, 1976.
63
research in Australia
A . P. Aitken
Senior Engineer, Hydrology
Snowy Mountains Engineering Corporation
P . O . Box 356, C o o m a North, 2360
N e w South Wales, Australia
Section 1 Field research
Introduction
Although Australia is sparsely populated, it is a highly urbanised nation.
At the present time 10.9 million people out of the total Australian population of
13.0 million live in urban areas. With two-thirds of the population in 12 urban
centres, Australian cities (with populations up to 2.9 million) present similar
urban hydrological problems to those existing in more highly populated countries
overseas. Another notable factor concerning Australian cities is that it has been
the practice, almost without exception, to separate the stormwater drainage system
from the wastewater sewer system. This situation is of course in direct contrast to
that existing in many of the older cities in the United Kingdom, Europe and the
United States of America.
Until recent years there existed very little interest in urban hydrological
studies in Australia. However, partly as a result of activities like those undertaken
in the ASCE Urban Water Resources Research Program, and partly as a result of a
greater awareness generally of environmental problems, there now exists widespread
interest and activity in this field.
Figure 1 gives a general impression of the range of urban hydrological conditions
met in Australia. In Figure 1(A) the locations of the capital cities, together with
Albury-Wodonga, a designated growth centre, are shown. Also shown is the mean
annual rainfall isohyets. It will be observed that the mean annual rainfall varies
over a wide range among the capital cities (as does also the seasonal distribution).
In Figure 1(B) the mean 5-year return period rainfall frequency curves for durations
from 0.1 hours up to 6 hours are shown. It will be observed that the intensities
vary over a wide range, the highest (in Darwin) being several times greater than
the lowest (in Hobart). As well as meteorological variations, there are very
substantial differences in the geology and topography of the capital cities.
Data .Collection Programs
Starting in the mid-1950's, six urban catchments were gauged in the cities
67
200 0 ¿00 M» MO 100 1000 k
(A) Annual isohyttal m a p
sao
I
i
i S
No. of years of record
Perth 21 y«ari
Darwin 15 years
I Adeude 67 years
2 Brisbane 58 years
>- Sydney (9 years
m Canberra 35 years
1
Melbourne 79 years
Hobort 57 years
05
02 03 04 05 06 06 10 15 20 30 10 SO 60
LEGE N O
»*rth
Darwin TIME IN HOURS
Adelaide
•rieUone
Sydney
Canberra (B) 5-year Rainfall Intensities -
Melbourne Australian Capital Cities
Hobart
RAINFALL INFORMATION
FIGURE 1
68
of Canberra and Melbourne. This marked the beginning of rainfall-runoff gauging
programs in urban areas in Australia. Expansion of this network of gauged urban
catchments was initially slow, with the inclusion of two catchments in Sydney
being the only advancement in the 1960's. A very significant expansion of the
network occurred in the present decade when most of the major Australian cities
either commenced or expanded existing programs.
Today a situation has been reached whereby data on urban catchments are
collected in seven major Australian cities with a total of 69 gauged urban catchments.
It should be noted, however, that the total number of gauged urban catchments is
dependent on the definition adopted, and for the purposes of this paper they have
been taken to be catchments in which:
. the total urbanised section of the catchment exceeds 15% of the
catchment ; and
the problem being studied is primarily the result of urbanisation.
The above definition excludes urban catchments which are gauged because they carry
floods originating in natural catchments outside the urban areas.
Table 1 gives a general summary of gauged urban catchments in Australia.
Station locations are in or adjacent to capital cities, Figure 1: Canberra, 7;
Adelaide, 8; Brisbane, 13; Darwin, 3; Melbourne, 24; Perth, 11; and Sydney, 3.
Some of these catchments have been described by Aitken (1973, 1975). With 69
catchments currently being gauged (and 5 additional installations currently proposed)
the situation is potentially a very good one, especially as the range of catchment
sizes and types is extensive. Most of the installations have been established to
develop a better understanding of the rainfall-runoff process in urban areas.
Nearly half of the total number of catchments have been established with the
intention of developing an understanding of the broader problem involving rainfall-
runoff -water quality relationships for urban catchments. Other reasons for'
installations include flood forecasting and water quality monitoring.
In general, the gauging of rainfall on urban catchments has not proven to be
difficult where sufficient pluviograph and storage gauges have been installed on
69
TABLE 1 - GAUGED URBAN CATCHMENTS IN AUSTRALIA
WATER
OPERATING NUMBER OF CATCHMENT QUALITY PURPOSE OF
AUTHORITY CATCHMENTS SIZE, ha MEASUREMENTS INSTALLATIONS**
Environmental 1 401 Yes Determination of

Protection Agency, pollutant loads
Victoria from urban
catchment
Melbourne and !3 55-39600 Yes Determination of
Metropolitan Board (for 6 R-R and R-R-WQ
of Works, Victoria catchments) relationships,
flood fore-
casting
Dandenong Valley 10 770-27000 Yes Determination of
Authority R-R and R-R-WQ
relationships
Metropolitan Water, 2 234 & 290 No Determination of
Sewerage and (4 (Yes, for 4 R-R relationships
Drainage Board, additional proposed) (Proposed catch-
New South Wales proposed) ments for R-R-WQ)
Wood and Grieve, 1 7 No Determination of
Consulting R-R relationship
Engineers,
Western Australia
Metropolitan Water 10 79-3600 Yes Determination of
Supply, Sewerage (for 1 R-R and R-R-WQ
and Drainage catchment) relationships
Board, Western
Australia
Department of 7 10-10050 Yes Determination of
Construction, R-R and R-R-WQ
Australian relationships
Capital Territory
Department of 3 22-1165 Yes Determination of
Northern R-R and R-R-WQ
Territory, N.T. relationships
Engineering and 4 1680-11840 Yes Water quality
Water Supply monitoring
Department, South
Australia
Brisbane City 13 1750-13500 No Determination of
Council, R-R relationships
Queensland
State Pollution 1 86 Yes Determination of
Control Commision, (1 R-R-WQ
New South Wales proposed) relationships
Private 4 116-641 No Determination of
consultants*, R-R relationships
South Australia
*Data not readily available. * * R - R denotes rainfall-runoff and R-R-WQ denotes

rainfall-runoff-water quality.
70
or close to the catchment. In some cases, difficulties have arisen in the
interpretation and analysis of data because of recording-time errors existing
between the pluviograph clock and the clock operating the water level recorder
which indirectly gives the runoff hydrograph. In two cases, synchronised rainfall
and runoff data are collected by recording both sets of information on one chart
and this type of instrumentation should be considered as essential on all urban
catchments.
It has been found that good quality runoff data from urban catchments can only
be obtained either by taking measurements with a current meter and developing a
rating curve or by installing a critical flow device such as a weir or flume for
which a theoretical or a laboratory-developed rating curve is available. The
former method has proven, in general, to be more satisfactory for the larger urban
catchments and the latter more successful for small catchments. Other methods,
such as the recording of water levels in pipes and channels and converting these
to discharges via Manning's equation, have been adopted for some Australian urban
catchments, but the results cannot be regarded as having satisfactory accuracies.
Table 2 summarises the types of flow gauging structures installed on the 69
existing urban catchments.
In Australia, water quality measurement programs in urban areas were not
integrated with the rainfall-runoff data collection programs until the last few
years. This was probably due to the dispersed nature of responsibilities for
measurement of rainfall, runoff and water quality. It was also in part due to a
lack of awareness of the magnitude of pollutants in stormwater runoff and their
effects on receiving waters. With the move towards the removal of nutrients from
sewage effluents in some Australian cities, more attention has been devoted to
the contribution of urban stormwater runoff to problems caused by excessive
amounts of nutrients or other pollutants reaching rivers, lakes, bays and estuaries.
Table 3 lists the most commonly measured water quality parameters in urban
catchments in Australia. Some standardisation of the types of measurements
carried out would help in the transfer of information between cities. At the
71
TABLE 2 - TYPES OF FLOW GAUGING STRUCTURES
TYPE OF BASIS OF BATING CURVE

STATION NUMBER MEASUREMENTS* THEORETICAL**
Artificial
Channel 13 9 4
Natural
Channel 30 30
Underground
Pipe 8 8
Flume 3 3
Weir 7 7
Other 8
* By current meter
** Manning's formula, weir formula, etc.
TABLE 3 - WATER QUALITY MEASUREMENTS ON URBAN CATCHMENTS
WATER QUALITY PARAMETER MEASURED NUMBER OF CATCHMENTS
Sediment Concentration Few

Turbidity Few
Suspended Solids About half
Dissolved Oxygen About half
Biochemical Oxygen Demand Most
Bacteria Most
Nutrients Most
Temperature Few
72
present time, of the 30 (approximately) catchments on which water quality data
are collected, only one catchment has an automatic water quality sampler for
collection purposes. At all other stations, manual dip samples are taken at times
of site visits during storm occurrences. Although the latter course of action is
obviously cheaper, it has the major disadvantage that the more significant runoff
events usually pass unsampled.
Telemetered rainfall and water stage information is of interest for flood
forecasting and operational purposes in a number of urban araas in Australia.
Several telemetered rainfall and water level stations have be en installed in
Melbourne and Canberra, and plans exist for expansion of the Melbourne system to
a network of 40 water level stations and 26 rainfall stations. Many of these are,
however, for catchments that would not be categorized as urban according to the
definition adopted at the beginning of this report. Telemetered installations are
also currently under consideration for installation on urban catchments in
Adelaide and Brisbane.
Good quality data on the rainfall-runoff-water quality process in urban areas
are difficult to obtain but are essential for use in modern mathematical models
of these processes. They are also essential for accurate prediction, whether for
design purposes, operational purposes or flood forecasting. Much of the data
collected in Australia has proven difficult to analyse because of inadequacies in
the quality of the data. Table 4 gives two examples of small gauged catchments
which should be satisfactory from a data quality viewpoint. Larger catchments such
as those gauged in Brisbane, where considerable attention is given to current-
meter measurements of flows, should also prove satisfactory for analysis purposes.
The regular publishing of data from urban catchments for use by research
workers and others has not been implemented in Australia. In this respect it
would be valuable to copy the practice of the United States Geological Survey which
routinely publishes data for its catchments in Texas and also that of the ASCE Urban
Water Resources Program which, in early Technical Memoranda of its regular series,
published or identified urban rainfall-runoff data available in the United States
73
TABLE 4 - GAUGED CATCHMENTS - TWO EXAMPLES
VINE STREET MAIN DRAIN, MELBOURNE (72 ha)
RAINFALL RUNOFF WATER QUALITY

MEASUREMENT BY MEASUREMENT BY MEASUREMENT BY
1 Dine1s pluviograph, Stevens A35 None

operated by float,*
1 Storage check gauge, water level measured
in a cut-throat flume
1 Barker's tipping bucket
gauge connected to runoff
gauge chart*
* Provides synchronized rainfall-runoff record
GIRALANG CATCHMENT, CANBERRA, A.C.T. (94 ha)
RAINFALL RUNOFF WATER QUALITY

MEASUREMENT BY MEASUREMENT BY MEASUREMENT BY
1 Dine 1 s pluviograph, Stevens 2A35 measures Manning water

water levels over sloping quality sampler
1 Mort pluviograph, crest weir and in a flume
when the underground
3 Stevens F type drain is surcharged
pluviographs
74
of America.
Rainfall Studies
One of the most basic items of data used by urban drainage designers is a
chart or formula giving the rainfall intensity-duration-frequency data for the
urban area under consideration. In Australia, the Bureau of Meteorology (see
Pierrehumbert (1974)) has been most active in this field and has recently published
curves for Australian capital cities based on the longest available pluviograph
records. This publication will be followed by similar publications for each State
of Australia showing rainfall intensity relationships for other centres away from
the capital cities. In a recent study for the Dandenong Valley Authority in
Melbourne, the Snowy Mountains Engineering Corporation developed a series of
rainfall-intensity curves for the Dandenong Creek catchment. These showed that,
for the same return period, rainfall intensities increase with elevation but only
for rainfall durations in excess of one day.
Rainfall temporal patterns are also of importance in many urban drainage
problems, especially those including retarding basins (Section 3 ) . The most
notable work in this field in Australia was carried out some years ago by the
University of New South Wales and published by Pilgrim et al. (1969). More recently,
the Bureau of Meteorology (Pierrehumbert (1974)) has published rainfall temporal
patterns for all Australian capital cities.
In Australia, no point rainfall versus areal rainfall relationships have been
developed. As a result, the most commonly utilised information is that published
by the U.S. Weather Bureau (1958). Aitken (1973), in a brief examination of the
problem, found that the use of rainfall depth-area curves may lead to incorrect
o
conclusions, and suggested that for catchments of up to about 250-knr in size it
may be inappropriate to apply an areal reduction factor to point rainfall frequency
values.
75
Section 2 Hydrological models
Event Models
A number of different event models, that is, models which examine isolated
rainfall-runoff events, have been tested in Australia using local data. Efforts
have been concentrated on the application of overseas models to Australian
conditions rather than the development of new models.
Several research projects have been directed towards testing the suitability
of the Road Research Laboratory Model for urban catchments in Australia. Details
of these studies are included in reports by Aitken (1968, 1975), Read (1971) and
Heeps and Mein (1974). In general, it can be said that the procedure is suitable
for cities in Australia where rainfall intensities are relatively low. As a
consequence of this situation, the procedure has been used in some drainage
studies in Hobart, Melbourne, Albury-Wodonga and Adelaide.
The Rational Formula, applied as a statistical model, is the most widely
used procedure for urban drainage design in Australia. However, only limited
testing of the method has been carried out, using data from four urban catchments
in Melbourne and two in Sydney. For the Melbourne catchments the results showed
the statistical value of the coefficient of runoff to be lower than that
generally used in design for urban catchments in that city. The results for
Sydney were inconclusive. A full description of the studies is included in a
report by Aitken (1975).
The Laurenson Runoff Routing Model was originally developed for rural
catchments. In contrast to unitgraph procedures, this technique assumes that a
non-linear relationship exists between lag time and runoff rate. The model,
with minor adaptation, has been applied by Aitken (1975) to six urban catchments
and a regression equation has been derived for the main model parameter .in
order that it can be used for the estimation of flood hydrographs on ungauged
catchments. Although much more testing is required of this model, it is
currently being used for design purposes in a number of Australian studies.
76
The Cincinnati Urban Runoff Model has been tested on two Australian urban
catchments by Heeps and Mein (1974). The results showed that the model contained
certain deficiencies which caused it to give less satisfactory results than
either the Road Research Laboratory Model or the Storm Water Management Model.
As with the Cincinnati Model, the Storm Water Management Model has been
tested on only two Australian urban catchments, the work again being carried out
by Heeps and Mein (1974). The testing did not include use of the treatment and
water quality sections of the complete model. The testing showed that the
Storm Water Management Model gave better results than either the Road Research
Laboratory Model or the Cincinnati Urban Runoff Model.
Unitgraph procedures of various types have been used on many urban catchments
in Australia, sometimes without checking whether the technique of deriving the
unitgraph was really applicable to the situation. In recent times, the simple
linear storage model developed by Rao et al. (1972) has received limited testing
and application in several urban catchment flood studies (see Morris 1975). More
recently, Bellingham (personal communication) tested Nash's unitgraph model using
the regression equation developed by Rao et al. (1972) on four catchments in
Brisbane. The results show a general under-estimation of catchment peaks
averaging about twenty per cent.
Continuous Models
Continuous rainfall-runoff models have not been used on urban catchments
in Australia. Some testing of existing models and research into new models has
recently commenced and it is expected that interest in this area of modeling
will extend to rainfall-runoff-water quality models in the near future.
Chapman (1969) presented the first results from the Australian Representative
Basin Model. Since that time very little additional research work has been
carried out except for a current project for the Australian Water Resources
Council which is investigating the modifications necessary to apply the
Australian Representative Basïn Model successfully in an urban catchment situation.
So far, the testing on one catchment in Canberra suggests that good results will
77
be forthcoming. As a part of the project for the Australian Water Resources
Council, the Hydrocomp Simulation Program will be tested on two Australian
urban catchments.
The only other current Australian work on continuous models for urban
catchments is a project at Monash University, Melbourne. The model being
developed is being tested using gauged data from urban catchments in Canberra.
Comprehensive Urban Water Models
Until recently it had not been generally appreciated that an urban catchment
has a very complicated water quantity and water pollutant balance. This is
because even in the 'separate1 water distribution, wastewater sewerage and storm
drainage systems adopted in Australian cities there exists significant
interaction between the three 'separate' systems. Complete and meaningful
balances cannot therefore be achieved without thoroughly considering the
origin and movement of all water and pollutants in an urban catchment.
The water inputs to an urban catchment are as follows:
. . rainfall falling naturally on the catchment;
domestic and industrial water transported artificially into the
catchment; and
outflows from surcharging sewers (usually only during prolonged
wet weather).
The water outputs are as follows:
stormwater runoff via the man-made drainage system;
rainfall infiltration into the wastewater sewer system via
groundwater infiltration and illegal stormwater connections;
deep percolation of rainfall into the groundwater system;
domestic and industrial wastewaters discharged through the
sewerage system; and
evaporation and transpiration to the atmosphere.
The origins of pollutant inputs include:
atmospheric pollutants which dissolve or coalesce with
78
falling rainfall;
pollutants washed from catchment surfaces;
pollutants discharged as wastewater into the sewer system; and
pollutants discharged as wastewater sewer overflows into the
stormwater drainage system.
The principal outputs for the catchment pollutants are:
. through the stormwater drainage system; and
. through the catchment wastewater sewer system.
To model the complete water and pollutant balances for an urban catchment
with the inputs and outputs described above is obviously complex. Nevertheless,
comprehensive urban water models are currently receiving consideration <n
Australia. Manash University in Melbourne is considering a comprehensive water
quantity model. As such, the model will consider domestic and industrial water
inflow and wastewater outflow relations together with rainfall-runoff relation-
ships. In a proposed study of the Parramatta River catchment in Sydney, the
data collection program will have as its objective the collection of sufficient
information to model both a catchment water quantity balance and a water
pollutant load balance.
79
Section 3 Applications
Catchments With Retarding Basins
In Australia, the term "retarding basin" has been used to describe
relatively small flood storages built by constructing an earth embankment
across the natural valley. An uncontrolled outlet in the bottom of the basin
causes inflows to pond temporarily and discharge to the downstream valley at a
reduced rate. A spillway is generally constructed to discharge safely floods
which would otherwise exceed the basin capacity.
Retarding basins have been a feature of stormwater drainage systems in
Melbourne and Perth for many years. In Melbourne they have been constructed
mainly to mitigate urban flood problems, but in some cases they also provide
areas for active and passive recreation, and in one case a permanent storage
provides a beautiful water feature in a natural bush setting. The sizes of the
storages range from 0.01x10 -m up to 2.5x10 -m with one basin presently under
consideration at a capacity of 11.5x10 - m .
In Perth the basins are called retention basins and are smaller. They are
designed to reduce runoff rates and to allow the urban runoff to recharge the
groundwater table via the permeable sand plain on which the city has been built.
The most spectacular example of a retarding basin in Australia is in Adelaide.
It is the major control for flooding on the Sturt River, which passes through the
south-western suburbs of the city. The basin has been described by Beales (1968).
The.flood control reservoir, which is a retarding basin according to the definition
in this report, is formed by a double-curvature concrete arch dam 37-m high,
capable of impounding 0.43x10 -nr of runoff from a partly urbanised catchment of
58-km 2 .
In recent years, interest in the use of retarding basin systems as a part of
stormwater drainage systems has increased. Basin schemes are currently under
80
design in a number of cities including Brisbane, Sydney, Darwin, Melbourne and
Albury-Wodonga. Because of the interest in these basin systems, improved design
procedures have been developed.
The design of retarding basin schemes is much more complicated than the
conventional urban catchment stormwater drainage system. No longer can the
critical design event be determined simply as "a time of concentration storm",
because the volumes of the rainfall events increase in importance and their peaks
become less important. Hence, a search is required to determine the duration of
the critical storm for each retarding basin system. This problem increases in
complexity wherever a catchment contains a number of basins that are either linked
into the system on parallel drainage lines or are in series on the same line. The
best solution for this type of system appears to be through mathematical models
which, in addition to estimating hydrographs at various locations within the
catchment, can also route hydrographs along creeks and channels and through
retarding basins. Two models of apparently similar capability in this respect have
been developed in Australia. Details of these models have been described by
Morris (1975) and Goyen and Aitken (1976).
Verification of models used to design complex systems including retarding
basins is difficult due to the limitations of available data. In the work by Goyen
and Aitken (1976) some verification of the model has been achieved using-data for
the May 1974 flood on the Merlynston Creek catchment in Melbourne. As shown on
Figure 2, this catchment has three retarding basins in an area of 1024-ha.
Figure 2 also shows the rainfall isohyets and hyetograph for the May 1974 storm.
A water level recorder at each basin maintains a continuous record of the variation
in storage. Figure 3 shows the following information:
. part (A) — the recorded stage-time curves and the estimated
stage-time curves from the model for each basin;
part (B) — the "recorded" outflow hydrographs and the estimated
outflow hydrographs from the model for each basin; and
81
ONCENWLE
PlUVnMVH
MERLYNSTON CREEK
CATCHMENT •6352 HKt.fl
ARMY CAMP BASIN

157 m 3 x IO3
NORMAL OUTLET - 1 2 0 0 M «• RCP
CATCHMENT t - 84« 8 Htcww

COMMONWEALTH
SERUM LABS. BASIN
509 irfx 10s
NORMAL OUTLET |- 750«« «m. RCP
CATCHMENT • 1024 3 H M W
BOX FOREST ROAD

BASIN
74 IT?X 10s
THE MERLYNSTON CREEK NORMAL OUTLET- 750««OO»» Orltte«
DRAINAGE SYSTEM
SCALE I • »0,000
DIAGRAMMATIC REPRESENTATION
OF BASIN SYSTEM
VtMrwtdoy 15 May 1974

TIME - hour«
RAINFALL HYETOGRAPH (GREENVALE)
LEGEND
ISOHYET FOR STORM PERIOD
2400 hft 14 May to 2400 hr» IS May 1974
CATCHMENT BOUNOARY
RETARD*« BASIN MERLYNSTON CREEK

PIPED DRAIN RAINFALL AND CATCHMENT
INFORMATION
OPEN UNLINED DRAW
FIGURE 2
82
. part (C) — hydrographs estimated from the model assuming that
the basins did not exist.
Considering the limitations of the basic data, the model reproductions shown in
Figures 3(A) and 3(B) are regarded as reasonable. Figure 3(C) illustrates very
clearly the mitigating effect of the basins. To reiterate, the estimated
hydrographs in Figure 3(C) were obtained using the model on the basis that no
basins-exist on the catchment.
Flood Damage Data
In general, the smaller urban stormwater drainage systems in Australia are
designed for a nominal average return period, say five to ten years, which is
arbitrarily adopted on the basis that stormwater systems are a service to the
community of urban areas and thus a 'reasonable standard1 for this service should
be adopted. On the other hand, for larger creeks and channels in urban areas, a
cost-benefit study is sometimes used to determine the design return period.
Until recently, flood damage data in urban areas has not been adequate to
carry out proper estimates of flood damage. After the 1974 flood on the Brisbane
River an extensive study of flood damage was made in order to determine a stage-
damage curve for that river. The investigation has been described by Aitken (1976)
and some of the results of the analyses could be useful for other urban flood
studies. Separate regression equations were developed to estimate flood damage
to the contents and structure of residences. The damage was related to floor area
and flood height above floor level. A separate relationship was derived for the
cost of clean-up of residences as a function of flood height over floor level.
The relationships were based on a representative sample of 400 houses. Although
a large sample of the extent of flood damage to industrial and commercial
buildings was available for analysis, the data displayed a great amount of
variability. Damage figures were therefore grouped into ranges of flood depth
over floor level and a mean damage per unit area determined for each flood depth.
83
MERLYNSTON CREEK
STAGE-TIME CURVES AND
HYDROGRAPHS
FIGURE X
ARMY CAMP
BASIN
12 18
BOX FOREST RD
BASIN
h
É2
O 6 12 18 O 6 12 16 O 6 12 18 O
T h u n * y I« M*y 1*74 FrMqr
TIME - HOURS
(A) OBSERVED AND ESTIMATED STAGE-TIME CURVES
84
8
ill
6
Mi
f «I
ARMY CAMP ARMY CAMP
BASIN BASIN
« 3
W\
01 •<-. , . . r^~
0 I 12 18 0 6 12 » 0 12 « 12 B O 6 12
j\ H
i\.M
Wi
f-i i! '!
C.S.L. C.S.L.
»I
BASIN / BASIN
i
i
,-A
—*^—-v^-Obs^rv«!
x ]]V-Eitima(«<l
0
-M6 1.2 .18 0 6 12 0 0
V
6 12 18 0 6 12 18 O
V
6 12 » O 6 12
! M
! !i
BOX FOREST R D Î BOX FOREST R D
BASIN BASIN
a *
4, V\
2
-ï— O v
0 6 . 12 18 0 6 12 18 0 6 12 18 0 6 12 m O 6 12 18 O 6 12
ThiMtoy M May «74 FrMrf Tfcmtfay M May l»TH FrMay
TIME - HOURS TIME - HOURS
IB) OBSERVED ANO ESTIMATED OUTFLOW HYDROGRAPHS (CI ESTIMATED HYDROGRAPHS WITHOUT BASINS
85
Flood Manual
In 1958 the Institution,of Engineers, Australia, published Australian
Rainfall and Runoff. The primary purpose of this report was to provide the
engineering profession in Australia with a set of guidelines for flood
estimation in both rural and urban catchments. The report has been widely used,
especially the urban drainage sections and the part containing rainfall frequency
information.
In the last few years the Institution has been working on the revision of
Australian Rainfall and Runoff and will soon replace it with a much more
comprehensive document. In the new report, the rainfall frequency information
will be comparable to that published for the United States of America, The
runoff sections will include flood frequency analysis, peak and hydrograph
estimation and rainfall-runoff models. It is expected that the revision will be
completed in 1976.
86
Section 4 Conclusions
Conclusions
After years of neglect it is concluded that urban mathematical modeling
and urban catchment research in Australia has reached a more encouraging level
of effort. Much more data are being collected now. Nevertheless, a careful
watch must be maintained to ensure that the quality of data is adequate for
future analysis purposes. Also, much more research is being conducted on
various mathematical models. Provided the present interest can be maintained,
a bright future exists for urban hydrology in Australia.
Acknowledgments
The author wishes to acknowledge with gratitude the information supplied
for this report by government agencies and other organisations. The assistance
of the Australian Water Resources Council, through its research grants in urban
hydrology, is also gratefully acknowledged.
Thanks are also due to the Snowy Mountains Engineering Corporation for
assistance in the preparation of this report and for permission to publish it.
References
AITKEN, A. P. (1968) "The application of storage routing methods to urban
hydrology". Journal of Institution of Engineers, Australia, Jan-Feb 1968.
AITKEN, A. P. (1973) "Hydrologie investigation and design in urban areas — a
review". Australian Water Resources Council, Technical Paper No. 5.
AITKEN, A. P. (1975) "Hydrologie investigation and design of urban stormwater
drainage system". Australian Water Resources Council, Technical Paper No. 10.
AITKEN, A. P. (1976) "A stage damage curve for the Brisbane River". Institution
of Engineers, Australia, Hydrology Symposium, Sydney, June 1976.
BEALES, A. G. (1968) "The south-western suburbs stormwater drainage scheme — a
technical review". Highways and Local Government Department, South Australia,
September 1968.
87
BELLINGHAM, G. (1976) "Private communication dated 23 January 1976 concerning
results of a Master of Engineering Science thesis".
CHAPMAN, T. G. (1970) "Optimisation of a rainfall-runoff model for an arid zone
catchment". International Association of Scientific Hydrology, Symposium of
Wellington, Publication 96.
GOYEN, A. G., and AITKEN, A. P. (1976) "A regional stormwater drainage model".
Institution of Engineers, Australia, Hydrology Symposium, Sydney, June 1976.
HEEPS, D. P., and MEIN, R. G. (1974) "Independent comparison of three urban
runoff models". Journal Hydraulics Division, American Society of Civil
Engineers, Volume 100(HY7), July 1974.
MORRIS, K. J. (1975) "Flood analysis — a mathematical model", Cameron, McNamara
and Partners Pty Ltd.
PIERREHUMBERT, C. L. (1974) "Point rainfall intensity-frequency-duration data -
capital cities". Bulletin No. 49, Bureau of Meteorology, Australia.
PILGRIM, D. H., CORDERY, I., and FRENCH, R. (1969) "Temporal patterns of design
rainfall for Sydney". Civil Engineering Transactions, Institution of Engineers,
Australia, CEII.
RAO, A. R., DELLEUR, J. W., and SARMA, S. P. (1972) "Conceptual hyrologic models
for urbanising basins". Journal of Hydraulics Division, American Society of
Civil Engineers, Volume 98(HY7), July 1972.
U.S. DEPARTMENT OF COMMERCE, WEATHER BUREAU (1958) "Rainfall intensity-frequency
regime, Part 2 — south-eastern United States". Technical Paper No. 29.
88
research in Canada
J.Marsalek
Hydraulics Research Division
Canada Centre for Inland Waters
867 Lakeshore Road
Burlington, Ontario, Canada
Introduction
An urgent need for urban water resources data, and particularly for
urban runoff data, has been identified under the research programme of the
Urban Drainage Subcommittee (UDS) for the Canada-Ontario Agreement on Great
Lakes Water Quality. Such data would serve for the assessment of pollution
caused by urban runoff and combined sewer overflows, for the planning, design
and operation of drainage systems; and also for the development, verification
and calibration of urban stormwater management models. It was realized at the
same time that the lack of urban water resources data would likely inhibit
achievement of advances in the management of urban water resources.
A survey of urban catchment research carried out in 1973(24> indicated

that there were only a few instrumented urban catchments in Canada, of which only
two had produced data suitable for urban runoff modeling. The UDS reacted to
this situation by establishing several new urban test catchments in the Province
of Ontario. A similar action has been taken recently by the Province of Quebec.
On a nationwide basis, however, urban catchment research still has not reached
a level consistent with the large expenditures in urban drainage facilities.
Urban catchment research in Canada seems to be plagued by a number of problems,
some of which are briefly discussed below:
a) Lack of co-ordination. There is no national network of urban test catchments

and no nationwide co-ordination of catchment research. As a result, the
current network of test catchments does not extend to all climatic regions,
and the type of information and data formats vary from case to case, thus
inhibiting comparative studies on various catchments.
b) Lack of suitable instruments. Several studies of instrumentation for

catchment research have been carried out recently.C24,26,34) ^11 of these
indicate a lack of reliable instruments for the measurement and sampling of
sewer flows. Even well-designed instrumentation systems may require up to
six months for the elimination of malfunctions before they become fully
operational.
c) High costs. Urban runoff data collection projects are rather costly and
their success is not guaranteed.
As an illustration of the cost and success question, an experience with

a typical runoff data collection project conducted by a private company
(specializing in environmental studies) for the Urban Drainage Subcommittee is
summarized below:
91
Number of Events Monitored
Activities Precipitation Precipitation & Costs
& Runoff Runoff Quantity
Quantity & Quality
1st Establishment and Instru- $10,000
Year mentation of the Catchment
2nd Data Collection & reporting 6 (>0.30") 2 $10,000
Year
3rd Data Collection, computer 2 (X).30") 5 $22,000
Year simulation with an existing
model, reporting
Totals 8 7 $42,000
The above costs do not include the cost of chemical and microbiological analyses
of stormwater samples. These analyses were performed by a government laboratory
and their commercial value is estimated at $20,000. Thus the total project cost
would be about $62,000.
At the present time, the data collection programs appear to be too

expensive for all but the largest municipalities. The government funding of these
projects also lacks continuity.
In spite of the aforementioned difficulties, the urban catchment

research has progressed significantly, especially during the last three years, and
this progress is documented in this report.
The reported research studies on urban catchments may be divided into

the following three categories: studies of the effects of urbanization on hydrolo
of undeveloped watersheds; studies of the effects of runoff from urban areas on
receiving waters; and studies of the characterization of flows in sewer networks.
Studies of the Effects of Urbanization on Hydrology of Undeveloped Watersheds
Studies of the Kanatai22) and North Pickering^3) basins dealt with the
effects of rapidly progressing urbanization on the hydrology and water quality of
runoff from undeveloped watersheds. In both cases, runoff quantity and quality
were monitored.
Kanata, one of three satellite communities that will house the expanding
population of the Ottawa area, is being developed by a single developer. Some
runoff control measures were implemented in the development. Houses were built
in a park-like setting with limitations on the extent of impervious areas. The
roof leaders are not connected to sewers. A progress report^22' summarized
preliminary results obtained in 1972 and 1973. Unfortunately, the study was
discontinued in 1973.
The North Pickering study^43^ is still active. The effects of

urbanization on the hydrology of the area undergoing development were simulated
with a Distributed Hydrologie Model. The results of these simulations will be
compared with runoff data being collected.t44'
Studies of the Effects of Runoff from Urban Areas on Receiving Waters
Two studies of this nature are presently in progress, the Pollution

from Land Use Activities Reference Group (PLUARG) study(^) and the Whitemud
Creek s t u d y . W Both studies are essentially upstream/downstream studies which
will indicate the water quality effects caused by urban land use under various
conditions including rainfall. In the PLUARG study, several urban areas in the
Grand River and Saugeen River basins (Ontario) are to be investigated. The
Whitemud Creek study (Edmonton, Alberta) is still in the planning stage.
Evidently, only a few field studies of the effects of urbanization on

the hydrology of developing watersheds and on receiving waters have been conducted
in Canada. Such studies require large resources, and typically, are carried out
over extended time periods.
Studies of the Characterization of Flows in Sewer Systems
In these studies, time varying quantity and quality of sewer flows

are studied within the sewer networks. The first studies of this nature were
limited to field observations of precipitation/runoff phenomena, very rarely
considering both runoff quantity and quality. Recent studies consider both of
these aspects of urban runoff, and frequently the runoff data collection is
pursued in support of computer modeling of sewer systems. Such a procedure enhances
the meaning and interpretation of the data collected and makes the simulation of
sewer system behaviour more reliable.
Attempts have been made to characterize sanitary flows (dry and wet
weather), combined sewer flows (wet weather), storm sewer flows, foundation drain
flows, road sewer flows, roof runoff, and winter urban runoff (including snowmelt).
Projects discussed here are divided into two categories: comprehensive

data collection projects, typically dealing with precipitation, runoff quantity
and quality; and projects of a limited scope, typically designed to produce
calibration data or first estimates of runoff quantities and pollutional loads.
Only projects in the first category will be discussed in detail.
Selected Sewer System Comprehensive Data Collection Projects
Fifteen data collection projects have been selected for inclusion in

this category. In all these cases, the projects represent well-documented
efforts providing data which either have been used for hydrological modeling and
research, or have a potential for such a use. The basic characteristics of the
test catchments are listed in Table 1 and the location of these catchments is
shown in Figure 1. Detailed descriptions of test catchments and the evaluation
and use of data collected therefrom are presented in Appendix I.
Suitability of the collected data for detailed hydrological modeling

was evaluated on the basis of two criteria: the test catchment and its physical
and operational characteristics are well defined and documented; and short
interval (say 1 to 5 minutes) precipitation data and continuous runoff measure-
ments are available. The accuracy and synchronization of both data records are
important. If water quality simulation is included, a third criterion also
applies: water quality data collected at short intervals (5 to 15 minutes) are
available. These data have to be synchronized with the precipitation and runoff
data and also have to cover the initial phase of runoff.
In many of the early studies, modeling of urban runoff was not among
the study objectives. It is therefore little surprising that the data from some
93
u "
%%
M «H N o» 00 en CM
r» •H cr> 00 wï 00 00 <T\ CM O
PB ¿4 CM •o •*
iH t-<
«» n m en •it r-l
o¿¡
C)Cj * • •
O
b 2 T3
Bu M 41
ta to m a • d (0 00 CO
|d « o 01 V 1 « ai d
<d
e-> ai e- C- ai ai o
5 a >i z X X X X •-i
X X X ¿5
«g FU
x
AVAILABILI!
aiaviivAv X X X X X X
I3A ION
DATA
aaHsnanj A X X X X
aiaviivAv X X X X X X X X X
STATUS
aaiaiawoo X f- X X
3AII0V X X X X X X X X X X X
.ï ?! • »
xirrvno ààoma K X X X X X X X X
PHENOMENA
X
MONITORED
Ja co tu
rtOTJ iiONflH X X X X X X X X si
co »v
X X
«S3
co h
X X X
TTVJNIVH K X X X X X X X X X X X X X X
XN30 y¡3i VO f- 00 r^ •o •tfm m O •tf en o» «* eo
. en co -a- CM r^ en <f m -* en 3 i-i •3-m en
ssaNsnoiAHadHi
«* «* •tf en en m rH
• * • 00 • • m •
(sajBaoaq) o vO o\ O r»
CM
CM
i-i en
i-i CO en <o r^ vO CM
CM
en
CM en CM
o en
rH
5
en
S
azis
(s^vavdas K s pu«
- U en co CO co U O CO CO co CJ co CJ CJ CO
paufcyuoo « D ) H2ISÃS TSSiaS
• • • •
• 4J ai
• • • en
ê ê S 8r
• 4J id • • • •
• co •
S u «g
CO «i 4J 4J
•E
U
1
60 U
O
Je1
O
M
tí
O
S
m
•
55
n d
d
o M
c
o
«J
00
A
rH
a]
O
«S
•» M «*
« X o o >, a o O ri
Q
¿*4
Oi
•H
>*
4J
J3
J-l
JJ
a
00
d
o
•a
s
4-1
•H
4J
r-l
•ri
4J
«
J6 -
o
41 5
I-l
O
h
«
PH
u
cl
4J
d
O
a
•g • • S
1 •3 e d
r-l 10
ol-
ri
§ 1
1
3 VI 4J
¡2 2 3 5 «
eo
* co co
s H »
4J 4J
•a 0)
d a r-l OJ ai
z U •o a) « Ol H1 »i
o
4J PL« •a
r-( d T> rH
H o u
X t
4J
oo d
bO «
•H o
4J
f-H 3 (0 O O ri
m Ai U O
5 Ï» 5 "3
s
•H 0 <4-l i-l C9 d d 01
Q U •H f-H Kl T* r-l OJ PM
d (-1
o
3 >
r4 h Y4 S CO
>
r-l o O "O
>> oo • u
a 10 « l-i OJ
CJ
«J h
« •1
•o
M w
•J 3
-ri 4J
CO o S -S
« « M U ¡s a« H H *
94
<
5
§
I
P
z
LU
I
ü
?
fe
LU
<
m
cr
<
Q
"~7! <
Z
<
ü
Û
/s • r-Yi LU
LU
CO
£
3
O
95
of these studies do not meet the above modeling criteria, but this circumstance
does not diminish the success of those studies.
Thirteen of the fifteen catchment studies identified in Table 1 were

fully or partially sponsored by the federal and provincial governments. The
remaining two studies were fully sponsored by municipalities. The actual field
data collection was done in seven cases by universities, in three cases by
private companies, in three cases by municipalities, and in two cases by the
federal and provincial governments.
Four of the catchment research projects have been discontinued, namely

the Bannatyne, Barrington, Brucewood and Windsor projects. The remaining
eleven projects are active, but for only five of these has some field data been
made available so far.
Geographical distribution
Most of the urban research catchments were established in the two most
populous provinces, Ontario and Quebec (9 and 3 respectively in Figure 1 ) . The
remaining three catchments are in Manitoba, Nova Scotia and Saskatchewan. It
would appear desirable to establish additional research catchments, particularly
in the Pacific Region, Prairies and Atlantic Region.
Size of the catchments
Size varies from about 7-hectares (Idylwyld) to 944-hectares (Toronto-

West). Four-fifths of the catchments discussed are of small size, less than
80-hectares. Smaller catchments are easier and less expensive to instrument
and document. For study of runoff routing, however, larger catchments are
preferable.
Sewer systems
Six catchments are served by combined sewers and nine by separate

storm sewers. In the latter case, storm sewers only are monitored in seven
cases, sanitary sewers only are monitored in one case and both sanitary and
storm sewers are monitored in one case.
Land use
Low density residential areas with some commercial and institutional

land use are most typical. Five of the catchments, namely Brucewood, Calvin
Park, Les Saules, Malvern, and Rigaud, are modern suburban developments. The
remaining catchments are older residential areas (30 to 70 years old) except for
the Idylwyld catchment which represents a freeway drainage area.
Effects of other than low density residential land use on runoff

quantity and quality were investigated on a very limited basis. Some information
on airport runoff, (55)
commercial plaza runoff,v33) and oil refinery runoff(55)
was collected, mostly in connection with engineering projects.
Additional studies of the effects of land use on the surface accumulation
of pollutants and quality of runoff water seem to be Justified.
Sewer System Data Acquisition Devices
The principal components of data acquisition systems are precipitation

gauges, flowmeters, wastewater samplers and recorders. Many such instruments
96
were examined in a recent report on instrumentation for field studies of urban
runoff.
(34)
An expanded summary of that report'-^' is included herein as
Appendix II.
Tipping-bucket raingauges were used in most projects. These gauges are
preferable in urban runoff studies mainly because of their superior mechanism for
actuating circuits and good accuracy of measuring rainfall of medium intensities.
The number of raingauges per catchment varies from 1 to 8. The latter number
refers to the Idylwyld catchment. Note that two gauges are recommended even for
the smallest catchments/ 2 6 ) and the largest test catchment discussed, the West-
Toronto Study Area, should be monitored with at least three raingauges.
Sewer flows are typically measured by means of constriction flow meters.
Among these, weirs are much more common than Venturi flumes. Standard rectangular
weirs are used at drainage outlets. Inside sewers, vertical slot or partial
trapezoidal weirs are used because they allow passage of solids through the
installation. Weir installations should be calibrated because the rating curves
of weirs installed in pipes, or at the pipe end, depart significantly from those
obtained for rectangular laboratory flumes.{•**' Weir head is most frequently
measured by floats or air bubblers.
Various types of flow measuring weirs were used in nine of the studies
listed in Table 1. Venturi flumes were used in several cases. Among these,
Palmer-Bowlus and Farshall flumes are common. In studies for West-Toronto and
Windsor, flow rates were estimated from the measured depth of flow and Manning's
equation. Such a procedure is very popular in engineering studies because of its
low cost, but it is not acceptable in research studies because of large uncer-
tainties in the measured flow. In all the reported studies, the sewer flows were
measured at a single point.
The above-described flow measuring techniques become inoperational for

pressure flow conditions. In sewers subject to frequent surcharge, the use of
dual free/pressure flowmeters is recommended. None of these instruments has
been installed in Canada so far, but a laboratory study of the U.S. Geological
Survey dual measuring flume was carried out recently.'™' Determination of
calibration curves for two flume sizes and flume ventilation requirements were
included in the study.
Sewer flow quality is monitored by collecting and analyzing samples.

Typically, sequential grab samples are collected either at regular time intervals
or after a constant volume of flow has passed through the installation. In the
latter case, a flowmeter and a flow integrator are used to activate an automatic
sampler.
Sequential grab sampling makes it possible to determine the variation

of constituent concentration with time and also the total constituent emission
during a runoff event. In most cases, samples are collected by automatic
samplers activated either by a raingauge or by a rise in the sewer water level.
Well-known commercial samplers are used, among these ISCO, Manning, North
Hants, Sigmamotor and Sirco. In one case, a little-known French sampler SEIN*
was used. A principal advantage of this sampler is that up to 24 sequential
2-litre samples can be collected, and 2-litre samples would be adequate in most
studies. Depending on the number of analyses to be performed, a sample size of
1-litre appears to be the minimum.
•Manufactured by Sein-Ecologie, 171 rue Veron, Alfortville, France.
97
Automatic samplers are the least reliable Instruments among those
discussed here. Where space permits, it is advisable to install and operate
two samplers independently to increase the probability of successful data
collection.(35)
Water quality and microbiological parameters investigated in the

reported studies varied from case to case. A list of parameters studied is
given in Table 2. Depending on the type and number of analyses performed,
analytical costs may represent a major component of the urban runoff study costs.
Frequently, the costs of analyses are more than 100 dollars per sample. Conse-
quently, a careful consideration of the number of samples to be analyzed is
warranted. Application of experimental design has been proposedi^) for
selection of the number of samples required to describe adequately variations in
runoff quality. Experimental design, which requires assumption of a simple
empirical model for runoff quality, has not been tested so far.
With regard to data recording, ideally a central magnetic or punched

tape recorder is used to produce computer-compatible and well-synchronized records
of precipitation, flow rate and sample collection. Such systems were used in the
Idylwyld catchment study and in the studies carried out by the INRS Institute of
the University of Quebec.
Two examples of the data acquisition systems are presented schematically

in Figures 2 and 3. The first one (Figure 2) is the Idylwyld system, a fully
automated system controlled by a minicomputer. Precipitation and runoff flow
rates are monitored. The second example, Figure 3, refers to the system designed
by the INRS and installed on the Les Saules and Rigaud catchments. Precipitation
and runoff quantity and quality are monitored.
Sewer System Data Evaluation and Use
Canadian urban runoff data of good quality are rather scarce at present.
Six out of the fifteen selected comprehensive studies are still in the initial
stage and have not yet produced any significant results.
Good precipitation and runoff quantity data suitable for hydrological

modeling are available for the Calvin Park and Idylwyld catchments. West-Toronto
data, which consist of precipitation and combined sewer overflow rates, appear to
contain significant uncertainties in the measured precipitation and flows and
this somewhat limits the use of these data.
Precipitation and runoff quantity and quality data are presently

available for six catchments: Bannatyne, Barrington, Brucewood, Fairfield,
Malvern and Windsor. The Bannatyne and Fairfield catchments are served by
combined sewers and the other four catchments are served by separate storm sewers.
Bannatyne, Barrington and Windsor data appear to have some limitations. For the
Bannatyne catchment, no precipitation data measured on the catchment are available.
The events observed on the Barrington catchment are of a minor character. In the
Windsor study, precipitation was not measured on the catchment and the runoff flow
rates were measured rather inaccurately. The remaining catchment studies,
Brucewood, Fairfield and Malvern, have produced fairly good data on runoff
quantity and quality for a number of storm events.
With regard to data use, the earlier studies, such as for the Bannatyne,
Fairfield and Windsor catchments, were instrumental in establishing the
pollutional character of combined sewage and stormwater, but particularly for the
Fairfield combined sewer catchment.
98
TABLE 2. HATER QUALITY AND MICROBIOLOGICAL PARAMETERS
INVESTIGATED IN URBAN RUNOFF STUDIES
FREQUENCY QT USX
PARAMETER
COMMON LESS COMMON
Biochemical Oxygen Demand X
Chemical Oxygen Demand X
Total Organic Carbon X
Total Solids X
Suspended Solids X
Dissolved Solids X
Volatile Suspended Solids X
Volatile Dissolved Solids X
Total Organic Kjeldahl Nitrogen X
Nitrogen - Ammonia X
Nitrate and Nitrite X
Phosphorus - Total P X
Orthophosphate Hydrolyzable X
Orthophosphate Soluble X
Oil and Grease X
Phenols X
Chloride X
Organic Chlorine X
Cadmium X
Chromium 'X
Copper X
Lead X
Mercury X
Nickel X
Zinc X
Total Coliform X
Fecal Coliform X
Fecal Streptococcus X
Fecal Sterol X
Pseudomonas Aeruginosa X
Salmonella X
99
A 39Í1V9 13A31 U31VM l
o
'Sd3i3W/\AOHd V S30nV9NIVtí L
S30VdU31NI H0V3 S13NNVH0 ZV
m
CD
CL
k-
<D
o X
X III ü <
(3 UJ z
X 0. 3
CO 0.
UJ LU
Q.
ni n
>- _l
h- OC
Lil UJ C£
UJ
»-
I
CO
CO
D
O
o
>-
>-
g
cv¡
100
i-
¡¡J?
UJ
PHONE LINE
STRIP CHART
C |
HE OFFICE)
'-
RECORDER
X
_
1 H 1
L
as
Ü
«-
—•
UJ
-i
WATEF
>cc
o
o
O
i
cc
nr
CONVERTEI ŒGRAT
SE
2 Z
rr
v^
O
2 F
z
<
MANNI IPPEI
o
z
o
<
(1 ..
<
2
EASURING
O
Z
CC
SAMPLER fEGRAT
WEIR OFF
ï
SEIN
o
ONINOIIIONOO TVN9IS
• 1
N SEWER)
-III
MAGNETIC CASSETTE
RECO RDER
1
Lu
LÜ
Cc
H
u
E
:
y
UJZ
MANHOL
TIME BASE
LOCATIC
/
101
S.C.
RECORDER
s
z
CO
z
LEVEL SE
ï
1
¿TELEPHONE
GROUND /7 LINES
Z
Sco
LEVEL SE
X
Z
CM
WATER
2
Z
CO
JL
LEVEL
i
ö!>
I 1 PVPI RF ï
2i
i
i
»
j
j
> co
INTEGRATOR TELEPHONE LINE

(TO THE OFFICE)
RAINGAUGE
S.C.
RECORDER
u_
_i cd
CO-n
SI
=j»
coS>
CO
LU C
LU ca
O
z
gure
T^
DATA ACQUISIT JE
CO
SYSTEM (Sk II inst ruments
CD
O
n%
CO
CD
CD
CO
included)
Runoff and combined sewer overflow quality data reported for various
test catchments are summarized in Table 3.
More recent studies attempted to determine the relationships between

precipitation and runoff quantity and quality. Urban runoff models are frequently
used in these studies to assist in the analysis and interpretation of the collected
data. Such an approach was taken, for example, in the Brucewood, Calvin Park and
Malvern studies, and is planned for several others. Data from well-instrumented
and documented catchments, such as Brucewood, Calvin Park, Fairfield and Malvern,
were used for the development, verification and testing of urban runoff models.
Sewer System Urban Runoff Field Studies of Limited Scope
Urban runoff studies designed to provide limited data for planning and
engineering studies fall into this category. In some cases, these studies were
in the planning or early stages and it was difficult to judge their scope and
extent. It is conceivable that some of these studies could develop in the future
into comprehensive studies providing data for urban hydrological research. Brief
descriptions of about twenty studies belonging to this category are presented in
Appendix I herein.
102
.57
Cl! m rH o CM vO rH in
O CO 00 r*. en O O O
CO r~i
r-» o • CM rH rH rH o
t-< O
m
rH * rH O • rH X O X X rH
2 O 1 T O
O
1
en
1 O 4-t CM
O Sá
•* o o • • • •
s O
• CM CM rH rH CM
a)
g. O • • en VO CM U"l
CT« O
s gj
r-l o
«a-
• rH • CM C0 O O O O
w en O rH 1 i rH rH rH
en > i-i 00 1 1
m rH o
rH X o rH
x X O X VO
l t rH 1 4-1 vO
m o O -Í O 4J en en
S CO
Z H 1 !-) r-
o
• O
O•
o
o
• o
• rH
i->
• u-1 • iH • en
•
Si co
od
u
Q
tá *
H
* | o
o S M O
"S Z
U * H
m
•4-
00
1
rH
X
rH
<§
O-i
<t
Sá S o
ON
-<
S
V©
en •*
• rH •
o co 5S H
CO
a vo
BB CJ
tO
o
rH o O
CM
co o CM en vO o CM
O
si
O
rH en
H H o <r • • • • • r~ O O
r-4 CM rH <f O en rH r- rH rH rH i-i
CO Z 5 t o%
1 en 1 1 1 1 1 i 00
& UO CMX X O X
-Í
w VO
o 3 O
u o <Ni
CM
CM CM CM
O
rH rH tfï
rH
O u en CM
s O • O• O• • O• •
rH 00« rH • r^•
o
£3 H
O
0)
01
o 5
o _
co ^
co
•H
U %
1t
l-l
g * *
Ï ilCO t tO» CM
t <r en vO
g3 H * *
o m
en rH • O
en
O
rH
00
rH o O
rH
O
rH
6 iH• • • •
z u Z s
QE
CM
CO
1
00
00 m
CM
i in|
•
rH
CM
1
•
oi
CM
O
m
rH
i
rH
CM
en
1
vO
o
vO
1 ë
rH • s «s• CM
X O
O 4J
rH • CM
X
<f
•
Cft
en CO
3
H H «Í r-l
o u"> • •
s rH o O
O•
CO Q rt 3 o O co
ou 01
§d 4J
cd
M CO
u
1 t * * 4-1
h
I-. VO
m
00
•H
o O Z
U
C3 g S
C71
00
rH X
rH H
X
ejv
«i cd
1 I».
•
u Ö co S o ICI rH
•
n Z
H
Cd
O o
»S
en
S
o S
d
r^
o
r«.
32-1
u 1 CM
U
U CO o
go en
<t
sphorus
solid
iform
iform
ml)
B s
•H
r-l ai CO en A
0 tJ
ai
O ^1
•u
O
O
rH
0
0)
Ü
£ T3 ai ai cd TJ
o
rH o
O rH O d
^»
Ú0 •H u •H a. ^•S»
co •u "25
>» u
§
u
a
«_/
o Q
u
o
r-l
TJ
cd
ai u
co
u
1-1
•H
VI
4J
•H
d
R
rH
u
cd s
a
n
Z
g
2
>-/
rH
K
u0 K
¿4
v-/
rH
Cd
O
ai
l-l
01
U-4
01
o J3 H 3
S o O •H
z Z • *
Ho CO H tu 0Í
103
National Surveys
In 1973, the Department of Environment commissioned a survey of

Canadian urban drainage practice. The surveys"J indicated that practically
all urban drainage design was based on the Rational Hethod. The Rational Method
was used even in cases which would be beyond the accepted limits of the scope of
application of the method, e.g., for the design of stormwater storage. The only
municipality using a runoff hydrograph model was the City of Toronto.
A 1976 survey reported here indicated that good progress has been made
in the field of urban hydrological modeling. A number of modeling research
projects have been executed or started, and runoff models have been used in many
engineering and planning studies. Although a trend towards using hydrological
models for drainage design has been established, most of the design work is still
being done with conventional, approximate methods. Three types of urban
hydrological modeling were reported: urban runoff; sanitary flows; and combined
sewage flows.
In this section are discussed the development of new models, modification

and interfacing of existing models, comparative studies and testing of models,
and engineering applications.
Development of New Urban Hydrological Models
Five new urban hydrological models are reported here. A brief

description of these models follows.
Queen's University Urban Runoff Model (QUURM)
The QUURM< 5 3 » 5 4 ) is a precipitation-runoff model developed for urban

catchments. At the present time it is a research-oriented model undergoing further
development and refinement. The model is non-proprietary, but a complete model
documentation (user's manual) has not yet been published. The model deals with
runoff quantity only.
The QUURM model consists of two parts, the generation of inlet

hydrographs and their routing through a sewer network. A single linear reservoir
approach is used to generate inlet hydrographs from rainfall excess for each sub-
catchment consisting of various area types. Runoff hydrographs from various
subareas form the inlet hydrographs and these are routed through the sewer network
using a modified time-offset method. The time offset is calculated for an
effective Manning n and a representative velocity corresponding to the average
discharge is calculated for the central half of the inlet hydrograph.
The QUURM was tested on data from the Calvin Park (Kingston) and Oakdale
(Chicago) catchments with very good results.(29) xhe main QUURM assets are its
simplicity and low computer requirements.
Subdivision Hydrograph Model (SHM)
The SHM has been developed by the Department of Civil Engineering^

University of Toronto, for a particular project.( lB ) It is a hybrid model
104
incorporating some features of the Storm Water Management Model (SWMM) of the
U.S. Environmental Protection Agency, The Chicago Hydrograph Method, and also
some new features such as runoff routing through various types of storage.
Rainfall excess in the SUM is calculated by subtracting the infiltration

(Horton's Formula) and surface depression capacity from the actual rainfall.
Overland flow and gutter flow are calculated by a stepwise procedure using the
Manning equation for uniform flow. For a selected time step, the runoff flow
is approximated by a successive quasi-steady state calculation in each interval.
Catchbasin outflow hydrograph and storage outflow hydrographs are combined to
form lateral sewer flow hydrographs, which are then routed through the sewer
network using a time-offset method. Flow routing through a holding reservoir is
optional. The model has built-in routing functions for various storage
alternatives, among which are roof storage and parking lot storage.
The SHM was tested on one catchment for four storms and fairly good
results were obtained. The model listing has been published but no other
documentation has been published so far. The SHM seems to be particularly useful
for analysis of on-site runoff storage in small urban subdivisions.
Distributed Hydrologie Made! (DHM)
The DHM was developed by Shully Solomon and Associates Limited for
environmental impact assessment for the North Pickering Project.(*3) It Is a
distributed, continuous simulation model requiring calibration.
Rainfall excess in the DHM is routed over the watershed, considering

various land use and soil characteristics. Numerical solutions are based on a
finite differences method. The model structure was recently expanded for runoff
quality and snowmelt. The final report is due in 1976.
An earlier version of the DHM was applied in the environmental

assessment of a new urban development at North Pickering, Ontario. No comments
on the DHM documentation or applicability can be made at this time.
Theory of Overflows from Storage and Treatment Plant Systems
This theory was developed by Unies Limited, under sponsorship of the

Urban Drainage Subcommittee.(21) Starting with the probabilistic aspects of
the climate, the statistics of untreated runoff discharge is developed using
mathematically derived probability functions. In the first step, a rainfall
excess volume is calculated by subtracting losses from the actual rainfall
volume. The excess volume then enters a storage device. The filling state of
storage is described by a probabilistic function. Between storm events, the
volume stored is depleted by diverting a constant flow to the treatment plant.
The probability of storage overflow is calculated as well as the volume of these
overflows. After introducing cost functions for storage and treatment, an
optimum combination of storage and treatment capacities can be studied.
This theoretical model was proposed for a preliminary analysis of

discharges and overflows from urban drainage networks incorporating storage and
treatment facilities. Note that by introducing a storage device, the runoff
calculation may be simplified and limited to the hydrograph volumes only, as
opposed to the traditional analysis considering entire runoff hydrographs. The
documentation of this procedure is limited.
105
Simulation of Sanitary Flows
The Department of Civil Engineering, University of Toronto, has

developed a parametric time-series model for sanitary flows.'*} The model serves
for planning and design of the deployment configuration of sanitary sewer networks.
Modification and Interfacing of Existing Urban Hydrological Models
Numerous urban hydrological models have been introduced during the last
ten years. Only limited attempts have been made to review and test existing
urban runoff models in order to assist potential model users.i-*-' »20,29,37,51)
The development of new urban runoff models is costly, and in some cases
these new models do not have any clear advantages over the existing ones, and
therefore do not advance the state of the art of modeling. Consequently, some
researchers prefer to concentrate their efforts on modification and interfacing
of the existing, verified runoff models. Such an approach was adopted, for
instance, by the Urban Drainage Subcommittee, which sponsored and directed a
study(^2) primarily dealing with the modification of the Storm Water Management
Model (SWMM) of the U.S. EPA. The selection of this model was based on the
results of a previous study dealing with several urban runoff models.(¿9)
The need to modify the SWMM for Canadian conditions was created by
several factors, namely, the Canadian climate, environmental concerns, engineering
practices and costs. Some of the principal parts of the study dealing with the
modification of the SWMM are described below. An interfacing of urban hydrological
models, as proposed in the study, is shown in Figure 4.
Data Analysis Model (DAM)
Applications of urban runoff models require large volumes of input data,

the preparation of which amounts to a significant portion of the total project
cost. It is therefore desirable to simplify and computerize this part of the
runoff modeling to the maximum possible extent.
The D A M ^ 2 ' treats only climatological data and serves as an interface

between existing data banks and both planning and design runoff models. The flow
chart of the DAM is shown in Figure 5.
The DAM was devised to provide input data for the STORM model of the
U.S. Army Corps of Engineers. For this planning-level, a continuous simulation
model using hourly precipitation and temperature data is required. Such data are
readily available for a large number of locations from the Data Banks of the
Canadian Atmospheric Service in a digital form on a magnetic tape.
Detailed runoff simulation typically requires short-interval precipitation

data (5-15 min.) which are also available from the above source, but not in a
digital form. Before using the DAM, the user would have to digitize precipitation
data in arbitrary time intervals. Once this has been done, the data processing is
the same as in the former case.
Precipitation and temperature data are read from the tape and pre-
processed by stripping and interpreting overpunches, building and sorting
precipitation and temperature files from raw data, and punching the data on
computer cards. Precipitation records of up to 6 gauges are read, combined into
a single record defined as a weighted average of all the gauges, and the data
quality is checked by plotting single and double mass curves for individual gauges.
A similar procedure is followed for the temperature data.
106
Climatological Data Bank
_ Data
Data Analysis Model
Preparation
Precipitation and Temperature Operational and Physical

Records Data for Studied Catchment
X
X
Continuous Simulation ( S T O R M )
Study of Total Volumes, Frequ-

encies, Storage and Treatment
Capacities
Identification of Critical Pjanning

Precipitation/Runoff Events Stage
Single-event Simulation (Lumped

SWMM)
I
Study of Control Alternatives
Selection of Design Alternative^)
I
Field Data Collection
Model (SWMM)Calibration-
Design/Analysis
Mainly Quality Parts
Stage
Simulations with Calibrated S W M M For Surcharged Sewers,Sim-

ulations with the W R E Version
IT]
Finalizatkxi of Storage and
Treatment Design
Study of Runoff Impact on Recei-

ving Waters (Mainly Quality)
Figure 4. INTERFACE OF COMPUTER MODELS IN URBAN RUNOFF STUDIES

(After Ref. 57)
107
STRIP CHART RECORD
MANUAL
PROCESS
•STRIP»INTERPRET OVERPUNCHES
- BUILD & SORT PRECIR & TEMP.
/AUXILLIARYV 4 _~.
m FILES FROM RAW DATA
PRECIR DATA I STORAGE P
- PUNCH DATA IN FORMAT FOR
SHORT INTERVAL LTS OR PROCESSING PROGRAM
- SCREEN DATA FILE
DATA PROCESS
READ DIGITIZED STRIP CHART RECORDS
•PRODUCE HYETOGRAPHS OF ANY MASS CURVES
TIME INTERVAL IN S W M M FORMAT PRECIPITATION PRECIPITATION PRECIPITATION
•COMPUTE WEIGHTED AVERAGE GAUGE 1 GAUGE 2 GAUGE e PLOTS SINGLE & DOUBLE
HYETOGRAPHS
-PUNCH IN PROPER FORMAT
DATA PROCESS
-READ PRECIP. RECORDS FOR ALL EVENT SUMMARIES

-ANTECEDENT CONDITIONS
GAUGES
PRECIR DATA -COMBINE RECORDS BY INPUT -TOTAL RAIN
SHORT INTERVAL WEIGHTING FACTORS -PEAK INTENSITIES ETC
- MASS & DOUBLE MASS ANALYSIS
PUNCH DATA IN LTS FORMAT
DATA PLOTS
-TEMPERATURE
BASE
^
USER DATA
I
PRECIPITATION TEMPERATURE
DATA FOR DATA FOR
LTS OR S W M M LTS
MANUAL
MODELLING
DATA FOR
EACH MODEL
Figure 5. F L O W C H A R T FOR DATA ANALYSIS M O D E L (After Bflf.42)
108
The output of the DAM consists of the punched data cards for STORM or
SWMM and of event summaries. These summaries list the tines of start and end of
each storm event, event duration, the total depth of rainfall, the peak intensity
and the antecedent dry weather period. The event summaries are useful for a
rapid review of precipitation data and, eventually, for the identification of
critical rainfall/runoff events. An example of the DAM output is shown in
Table 4.
Operation of the SWMM as a Planning Model
The feasibility was investigated of using the SWMM, in a multi-event

simulation mode, as a planning tool.'^2) The SWMM was not, however, modified to
operate in the continuous simulation mode, because this would require the
addition of a water balance accounting.
For planning purposes, the SWMM simulations could be made less expensive
by reducing the number of subcatchment and transport network elements to a
minimum and increasing the time step. Following the methodology developed at
MIT,''' even large catchments could be represented in the planning stage by a
single overland flow element. Depending on the circumstances, few or no transport
elements are used.
Parameters of the lumped overland flow element were defined as spatial

averages. The element width, directly related to the length of overland flow,
was defined as twice the total length of all main drainage pipes and gutters
serving the area. For runoff transport, the volume of pipeB in the simplified
system should be approximately equal to the volume of the real system.
The simulations made with the lumped SWMM closely approximated those
made with the discretized model. No changes were made in the Storage/Treatment
Block of the SWMM, since this block has low computer-time requirements. Simulation
time steps were sometimes extended to 60 minutes with good results.
Generalized SWMM Runoff Quality Model
In an attempt to combine the best features of the SWMM and STORM

quality subroutines, a new quality subroutine was developed.'^2) Important
features of this model are summarized below.
The model calculates runoff quality on the basis of catchment

characteristics and runoff hydrographs which have to be supplied by the user.
Runoff quality calculation may be based either on the entire runoff from the
catchment or on the runoff from impervious areas only. A single catchment can
feature up to five different land uses. The accumulation of pollutants on the
catchment surface is accounted for by specifying loading rates for each land use
and by specifying composition of "dust and dirt". Such information is provided
by the user. Catchbasin contribution to the total pollutional load is considered
in the model. Finally, the runoff quality is calculated using the SWMM algorithm.
Up to ten constituents can be simulated, namely BOD, COD, Suspended Solids,
Settleable Solids, Coliforme, N, PO4, Cl, Pb, and Oil and Grease.
Pollutogranhs, mass curves and surface load statistics for each

pollutant are produced by the model. Some verification of the model was
attempted, but appears to be inconclusive.
(42)
The generalized quality model
can be easily calibrated, which constitutes its main advantage. \It may be used
in conjunction with any method for the calculation of runoff quantity.
109
TABLE 4. DATA ANALYSIS MODEL (DAM) OUTPUT SAMPLE (After Ref. 42)
STORM EVENT SUMMARY FOR 1973,

SWMM - WEST-TORONTO STUDY AREA
(A storm has been defined as having a total rainfall greater than 0.03-in.
(0.76-mm) and having less than 3 consecutive dry hours).
Storm Number 1
Started on the 4th month, 2nd day, 2nd hour
Antecedent dry days unknown
Ended on the 4th month, 3rd day, 10th hour
Total duration in hours = 33
Total rainfall in hundredths of inches = 29 (7.3-mm)
Maximum intensity in hundredths of inches per hour = 2 (0.5-mm/hr.)
Storm Number 2
Started on the 4th month, 4th day, 16th hour
Antecedent dry days » 1,208
Ended on the 4th month, 4th day, 24th hour
Total duration in hours =* 9
Total rainfall in hundredths of inches = 14 (3.5-mra)
Storm Number 3
Started on the 4th month, 27th day, 10th hour
Antecedent dry days =» 22.375
Ended on the 4th month, 28th day, 4th hour
Total rainfall in hundredths of inches = 48 (12-ram)
Maximum intensity in hundredths of inches per hour = 8 (2.0-mn/hr.)
Storm Number 4
Started on the 5th month, 1st day, 9th hour
Antecedent dry days s 3.167
Ended on the 5th month, 1st day, 13th hour
Total rainfall in hundredths of inches = 16 (4.1-ram)
Maximum intensity in hundredths of inches per hour = 7 (1.8-ram/hr.)
Storm Number 5
Started on the 5th month, 2nd day, 21st hour
Antecedent dry days a 1.292
Ended on the 5th month, 2nd day, 21st hour
Total duration in hours « 1
Total rainfall in hundredths of inches =• 6 (1.5-mm)
110
Snowmelt Quantity and Quality
Snowmelt can be accompanied by intense rainstorms in the spring season

or occasionally in late fall, thus contributing to large runoff. At the same
time, the snowmelt water appears to be heavily polluted and this creates
environmental concerns about the impact of snowmelt runoff on receiving waters.
It was therefore decided to study the simulation of snowmelt quantity and
quality/ 4 2 ^
Following a literature survey, Anderson's snowmelt m o d e l a was

selected to be built into the SWMM as a user option. This was one of the first
attempts to simulate snowmelt in the urban environment and, mostly because of a
lack of field data, numerous approximations had to be made.
The snowpack distribution and physical parameters were assumed to be

known prior to the storm. The problem is then reduced to that of describing
the physical changes in the snow cover during the snowmelt and/or rainfall
periods and the resulting effects on the runoff. The model requires hourly air
temperature and wind speed data as climatological input data. The basic calculation
is made in hourly intervals. If required by the runoff calculation, the computed
hourly volumes are linearly interpolated for other time intervals.
Since hardly any field data were available to formulate a conceptual

model of the quality of snowmelt water, the same approach as in the SWMM quality
model was used. Pollutant accumulation on and washoff from the catchment surface
was considered. The list of SWMM water quality constituents was expanded for
chlorides and lead. The input of chlorides onto the catchment surface was derived
from typical application rates of de-icing salts.
Limited attempts have been made to verify the snowmelt quantity and
quality model on the Brucewood catchment for three events.( 2 °) Though the model
results indicated trends similar to those observed, more extensive testing will
be required to reach conclusive results.
Modification of the Storage Block
The costs of runoff control measures in the SWMM reflect U.S. economic
conditions. It was necessary to substitute the Canadian costs into the model.
(42)
These costs had to be estimated only, because very few runoff control facilities
have been built in Canada. As more experiences with building these facilities
become available, these costs will be reviewed and updated.
Behaviour of a Storm Water Retention Pond
In this study,'1-2) carried out by the Department of Civil Engineering,

University of Toronto, the regime of a storm water retention pond was simulated.
The precipitation-runoff phenomena were simulated by the STORM, which produced
simulated inflow into the siltation basin-reservoir system studied. A new
subroutine, interfaced with STORM, was written to simulate the effects of such
a system on water quality. The subroutine calculates the frequency of solids
removal efficiency of the siltation basin, monthly frequency analysis of the water
quality in the reservoir, inflow-outflow hydrographs, and pollutographs for six
constituents (BOD, DO, N, PO4, Suspended Solids, and Settleable Solids). The
model has been applied to a storm water pond in Mississauga, Ontario, and
verification of the model is planned using the same facility.
HI
Testing of Urban Runoff Models and Comparative Model Studies
Testing and comparative model studies are of utmost importance to

model users who are attempting to select a "right" model for their application.
In testing studies, the ability of a model to reproduce various runoff events
observed on test catchments is evaluated. From a scientific point of view,
neglecting the uncertainties in the input/output data and the extent of
calibration bias in such an approach is not fully satisfactory and may not fully
indicate the biased or incomplete structure of the model. This simplified
testing, however, yields a good indication of model performance and reliability,
particularly if only well-verified precipitation-runoff data are used. It has
been noted(-*6) that none of the existing test catchments in Canada allows a
fully rigorous testing of urban runoff models.
Note also that in the above procedure the overall model performance
is evaluated rather than the performance of individual model components, such
as overland flow generation, routing, etc.
The following urban runoff models have been tested to various extents
in several Canadian studies: Dorsch HVM, Queen's University QUURM, British RRL,
STORM, SWMM, University of Cincinnati UCUR, and Water Resources Engineers version
of SWMM (WRE-SWMM).
The first study(29> dealt with the HVM, QUURM, RRL, SWMM, and UCUR
models. These models were applied on several test catchments and the simulated
hydrographs were compared to the observed ones. The goodness of fit was
evaluated in two ways. Firstly, the runoff volumes, peak flows and times to
peak were considered. Secondly, the entire hydrographs were considered.
All the models performed fairly well. When comparing only the volumes,
peaks and times to peak, there was no statistically significant difference in
the performance of various models on most catchments. On average, about 70% of
the simulated runoff volumes and peak flows, and 85% of the times to peak, were
within ¿20% of the observed values. It should be stressed that only runoff
generation on small catchments (less than 37-hectares) and simple flow routing
were tested. Special features of some of the above models, such as a dynamic-
wave flow routing in the HVM model, were not fully utilized or tested.
From a user point of view, the SWMM of the U.S. EPA was rated best.
It is by far the best documented non-proprietary model which is continuously
updated and refined.
In a recent contract study(^2) sponsored and directed by the Urban

Drainage Subcommittee, the HVM, STORM, SWMM, and WRE-SWMM models were tested.
Most attention was paid to the SWMM, on which a sensitivity analysis was carried
out.
Sensitivity Analysis of the SWMM Runoff and Transport Blocks
An appreciation of the sensitivity of a runoff model to variation in

the input data is important for the preparation of input data for a detailed
simulation. The sensitivity analysis indicates what detail and accuracy of
input parameters is required. For some parameters, rough estimates may be
acceptable without decreasing appreciably the accuracy of simulations. In this
study,(42) several additional parameters were considered in comparison to the
SWMM sensitivity analysis that had been reported previously.(")
112
Repeated simulations were made for a selected storm on a 4.5-hectare
catchment. The parameters affecting the SWMM runoff quantity (Runoff Block)
can be listed in order of decreasing importance as follows: imperviousness;
catchment width; infiltration capacity; surface depression; gutter and catchment
surface roughness; and ground slope.
In the Transport Block, the conduit length, number of conduits and

conduit roughness were tested. The effect of the conduit length was rather
minor for pipes shorter than 4000-feet (1220-m). For a 10,000-ft. (3,050-m)
pipe, an attenuation of the order of 40% was obtained.
When a hydrograph was routed through a single element of a length L,

a lesser attenuation was obtained than by routing the hydrograph through several
pipes of a total cumulative length L.
Increasing roughness attenuated proportionately the peak flow.
In a similar manner, the sensitivity of the runoff quality subroutine

was studied considering the following parameters: the washoff equation
exponent b; options for the calculation of Suspended Solids (SS); number of dry
days, street cleaning, catchbasin load; pipe slopes (Transport Block); and the
specific gravity of solids (Transport Block).
The washoff exponent b affects directly the rate of the pollutant

removal, particularly in the initial period of runoff. Neither of the two
options for the calculation of SS proved to be applicable over a wide range of
antecedent conditions. The number of dry days is a very important parameter
because it affects nearly linearly the total runoff pollution load. The effect
of street cleaning is very pronounced for high cleaning efficiencies. Catchbasin
loads contribute very little to the total pollution load.
The slope of combined sewers and the specific gravity of solids affect
significantly the sediment deposition-scouring process, and consequently the
quality simulations.
Testing of the HVM„ SWMM, STORM, and WRE-SWMM Models
Three Canadian test catchments were used in these tests,\^*-' namely

Bannatyne, Brucewood, and Toronto-West. Only the SWMM was applied on all the
three catchments. An acceptable agreement between the SWMM simulated and
observed runoff quantity was found. For stormwater and combined sewage quality,
the agreement between the SWMM simulations and observations was less satisfactory
and it was evaluated as an order-of-magnitude agreement.
The STORM was applied on the Bannatyne and Toronto-West catchments.

Runoff volumes and frequency of overflows (Toronto-West) were reproduced by STORM
quite well.
The HVM and WRE-SWMM models were applied on the Bannatyne catchment for
two events. In one of these, the sewer system was surcharged, Both models
performed quite well under such conditions and produced more realistic hydrographs
than the SWMM, which produced a truncated hydrograph.
Other Studies
The SWMM and RRL models were tested and compared on the Fairfield
catchment.
(51)
The results obtained with both models appear to be, with the
exception of the calibrated runoff volumes, worse than those obtained for other
113
catchments. This may have been caused by uncertainties in the input data and
the minor character of many test events used in this study.
The SWMM was also tested on the Malvern catchment.(35) Tlie calibrated
SWMM was used to reproduce 25 precipitation-runoff events. A very good fit
between simulated and observed values was obtained: 90% of the SWMM-simulated
runoff volumes, 80% of the simulated peak flows, and 100% of the simulated times
to peak were within Ï20% of the observed values.
The runoff quality simulations were less satisfactory with only an

order-of-raagnitude agreement between the simulated and observed values.
Results of testing the SWMM on 8 urban catchments were summarized in

a recent paper.(->») The paper concluded that while the SWMM quantity simulations
were fully satisfactory for free flow in sewers, the quality simulations were in
general much less satisfactory, and additional testing and/or refinement of the
quality subroutine is required.
Engineering Applications of Urban Runoff Models
Engineering applications of urban runoff models reported in the survey

conducted for this report are briefly described below. Inevitably, the list of
applications is incomplete. The purpose of this description is only to indicate
the progress made in Canadian drainage design during the last three years and to
demonstrate the potential of urban runoff models. The projects are listed
alphabetically, according to their location.
Edmonton (Alberta)
The City of Edmonton has engaged James F„ MacLaren Limited to carry out
a master drainage study of a 4500-hectare area served by combined sewers. Two
subareas are to be studied in detail. The WRE-SWMM will be applied.( 55 ^
Halifax (Nova Scotia)
The City of Halifax engaged Canadian-British Consultants Limited, to

study the feasibility of storm water retention in the Kearney Lake Road area.(5)
The Illinois Urban Drainage Area Simulator (ILLUDAS) .was used in this study.
Hamilton (Ontario)
The Harailton-Wentworth Regional Government engaged Proctor & Redfern

Limited to study combined sewer overflows and urban runoff on two study areas.(^'
The STORM, SWMM, and WRE-SWMM were applied in this study.
Humber River Mouth Study (Toronto, Ontario)
The Metro Toronto Region Conservation Authority engaged James F.

MacLaren Limited to study combined sewer overflows, their control, and their
effects on the Humber River.(") SWMM was applied in this study.
Merivale Industrial Park Study (Merivale, Ontario)
The Region of Ottawa-Carleton, the Central Mortgage and Housing

Corporation and the Urban Drainage Subcommittee have engaged Gore & Storrie
Limited to study the changes in runoff and its pollutional loads resulting from
the development of a new industrial park and to investigate runoff control
alternatives.(3) STORM and SWMM are being used in this study.
114
Midland (Ontario)
The Town of Midland engaged Canadian-British Consultants Limited to

model urban runoff and its effects on an urban lake. The STORM is being
applied in this study.
Mississauga (Ontario)
The City of Mississauga engaged James F. MacLaren Limited to analyze

the existing storm sewer system in the Port Credit area and to develop relief
sewer alternatives.(55) The WRE-SWMM was applied in this study.
Oil Terminals Study (London and Toronto, Ontario)
The Petroleum Association for Conservation of Canadian Environment

engaged James F. MacLaren Limited to study pollutional loads in surface runoff
from oil terminal sites.(55) r|.ne modeling work was done with the SWMM and was
supported by some field measurements.
The Province of Ontario
The Urban Drainage Subcommittee engaged the American Public Works

Association to evaluate the magnitude and significance of pollution loading
from urban stormwater runoff in Ontario.(40) As a part of this study, runoff
simulations on four urban areas were performed with the STORM to estimate
pollution loadings and to evaluate various storage/treatment alternatives.
Technologically efficient combinations of storage and treatment were identified
and priced. The results were presented in a normalized form and can be used by
engineers and planners to derive preliminary estimates of costs of runoff
pollution control.
North Pickering (Ontario)
The Ministry of Housing engaged Shully Solomon & Associates Limited

to carry out an environmental assessment of the impact of urbanization on
several watersheds in the North Pickering area.(^3) A Distributed Hydrologie
Model developed by the consultant was used.
St. Catharines (Ontario)
The City of St. Catharines engaged Proctor & Redfern Limited to study
wet-weather flows in two sanitary sewerage areas.(32) ^ proprietary model SWAN
was used in this analysis.
Toronto (Ontario)
The City of Toronto was the first Canadian municipality using extensively
urban runoff hydrograph models. The City has engaged Dorsch Consult Limited to
carry out the design/analysis of a number of drainage districts.(") The
proprietary Dorsch HVM model has been used exclusively in these studies. The
Dorsch HVM pressurized flow routing capability was the main reason for selecting
this model.
Toronto Airport Study (Ontario)
A study of environmental problems at the Toronto International Airport

was commissioned by the Department of Environment to James F. MacLaren
Limited.(55). The study included some limited monitoring of runoff quantity and
115
quality, and runoff modeling with the STORM and SWMM.
Vancouver (British Columbia)
The City of Vancouver and the Greater Vancouver Sewerage and Drainage
District carried out a demonstration study with the Dorsch HVM model on a 333-
hectare area. Later, the results obtained with the HVM on one of the sub-
catchments (38-hectares. 78% impervious) were compared with those produced by
the SWMM and ILLUDAS.Í20)
Vaughan (Ontario)
The Town of Vaughan engaged James F. MacLaren Limited to develop a

master drainage concept for a proposed residential development of 1820-hectares
in the area of Thornhill-Vaughan.t55) The SWMM was applied in this study.
Winnipeg (Manitoba)
The City of Winnipeg engaged James F. MacLaren Limited to analyze

the combined sewer system in three sewerage districts and to develop relief
sewer alternatives.0°) The WRE-SWMM is being used.
The frequencies of use of various urban runoff models in the above

reported engineering applications are shown in Figure 6.
116
7"
0)
c
o
'•*-'
CO C
.0 o
"Õ.
Q.
<
o) 5-
_
-
i_
c
EPA
O
C
'o) 4- CO
LU 13
c 1
ï 2
Report
S
CO
1
rmy)
CO
2
<
s
O 2- CO'
5
3 CO
1_ ^»•r
d) E LU
n 2
ce
E
-i i
> o £
das
1 CO
z ï Ai*
ical Mo<
Hydrolo<
Distribut
Dorsc
ja
Swan
Urban Runoff Model

Figure 6. NUMBER OF REPORTED ENGINEERING APPLICATIONS
OF VARIOUS URBAN RUNOFF MODELS
117
Section 3 Summary and conclusions
The state of the art in urban hydrological modeling seems to surpass
the available calibration/verification data base. The ultimate goal of the
creation of a good urban water resources data base remains, therefore, worthwhile
and necessary. A critical survey of the Canadian urban test catchments revealed
that only five catchments have produced field data suitable, to various extents,
for the development and verification of urban runoff quantity models. These
five areas are Brucewood, Calvin Park, Fairfield, Idylwyld, and Malvern.
For runoff quality, only Brucewood, Malvern, and possibly Fairfield,

have produced field data approaching the desired standard.
Five newly established catchments hold promise for production of good

urban runoff quantity and quality data.
A variety of data acquisition systems are used in runoff studies.

Precipitation is measured commonly by means of tipping-bucket raingauges. The
number of gauges per catchment is rather limited, which may be caused by the
lack of suitable gauge sites. Sewer flows are typically measured by weirs, less
frequently by Venturi flumes. Water quality is monitored by collecting sequential
grab samples and analyzing them for a number of constituents. The most frequently
s.tudied constituents are the biochemical oxygen demand (BOD), the chemical oxygen
demand (COD), chlorides, lead, and various forms of nitrogen, phosphorus and solids.
Among microbiological parameters, total coliform and fecal coliform densities are
frequently studied.
Samples are typically collected by means of automatic samplers which

remain the least reliable component of the data acquisition systems. The recording
of all the information by a single recorder helps to maintain a good synchronization
of records. Ideally, magnetic or punched tape recorders producing computer-compatible
records are used.
Urban runoff data collected on the Canadian catchments have been used to
characterize the quality and quantity of stormwater and of combined sewage
overflows, to develop new urban runoff models, and to verify and modify some of
the existing runoff models.
Most of the data collection projects do not include measurements other

than for the drainage outfall. The impact of urban effluents on'receiving waters
is rarely studied in the field. Some studies of this nature, however, were started
recently.
The lack of urban runoff data seems to impair progress in the development,
testing, verification and calibration of runoff models. Tendencies to substitute
noncalibrated model results for actual field data, without any verification
attempts, are showing up in engineering studies. Such a trend is undesirable and
detrimental.
Several aspects of urban hydrological modeling are being developed in

Canada. Most of the interest is centered around the modeling of urban runoff,
its conveyance and control by storage and treatment, and the impact of runoff on
receiving waters. Both storm water and combined sewage are considered in this
context.
118
Though some new urban runoff models have been developed in Canada, the
major efforts seem to be directed towards the testing and modification of existing
runoff models. Two U.S. models, the STORM and SWMH, have received particularly
wide attention, and the latter model was modified, to some extent, to reflect
Canadian conditions. It is fully recognized, however, that there is a need for
an entire hierarchy of urban runoff models. Various applications require
different models having certain features and belonging to various levels in the
model hierarchy.
Although most engineering drainage design is still based on old,

approximate methods, the urban runoff models have emerged in Canadian drainage
practice as an important alternative during the last three years. The number of
engineering applications based on urban runoff models is increasing and this
trend is well documented by the examples cited in this report. The models not
only contributed to a more rational design but, in many instances, led to
significant savings in drainage costs.
Methodology
Information presented in this report has been collected and assessed by

the author, who accepts the responsibility for any misinterpretation of the basic
data supplied by the co-operators listed below. More than 300 information sources
were examined by means of a literature survey, a questionnaire and follow-up
interviews.
Acknowledgment
The co-operation of numerous persons who have responded to the project

questionnaire is gratefully acknowledged. Special thanks are due to the
individuals, listed alphabetically by surname below, who have provided information
included in the report:
B. J. Adams, University of Toronto, Toronto, Ontario

K. £. Armstrong, Alberta Environment, Edmonton, Alberta
J. C. Anderson, Gore & Storrie Limited, Ottawa, Ontario
N. Benneworth, Canadian British Consultants Ltd., Halifax, Nova Scotia
K. T. Brodersen, Environmental Protection Service, Ottawa, Ontario
R. S. Cebryk, City of Edmonton, Edmonton, Alberta
J. Dartois, Université' du Quebec, Quebec City, Quebec
H. Demard, Universitéf du Quebec, Quebec City, Quebec
P. A. Ehler, Town of Midland, Midland, Ontario
C. Eicher, Gore & Storrie Limited, Toronto, Ontario
J. Ganczarczyk, University of Toronto, Toronto, Ontario
W. P. Geiger, Dorsch Consult Limited, Toronto, Ontario
M. Gibbons, Stanley Associates Engineering Ltd., Edmonton, Alberta
J. F. Hartt, University of Windsor, Windsor, Ontario
119
J. G. Henry, University of Toronto, Toronto, Ontario
W. L. Keay, Regional Municipality of Ottawa-Carleton, Ottawa, Ontario
C. S. Kitchen, City of Toronto, Toronto, Ontario
F. I. Lorant, M. M. Dillon Limited, Toronto, Ontario
B. H. Luckman, University of Western Ontario, London, Ontario
J. E. McKenzie, City of Ottawa, Ottawa, Ontario
C. MacRae, University of Western Ontario, London, Ontario
H. Mar, City of St. Catharines, St. Catharines, Ontario
W. A. Mechler, Greater Vancouver Sewerage and Drainage District,
Vancouver, British Columbia
B. Molke, Canadian-British Engineering Consultants (1971) Limited,
Don Mills, Ontario
Z. Novak, Ministry of the Environment, Toronto, Ontario
A. R. Perks, Proctor & Redfern Limited, Toronto, Ontario
B. Ruddy, Borough of North York, Willowdale, Ontario
S. I. Solomon, Shully I. Solomon Limited, Toronto, Ontario
F. A. Tonelli, Ministry of the Environment, Rexdale, Ontario
G. L. Trider, Environmental Protection Service, Halifax, Nova Scotia
D. H. Waller, Nova Scòtia Technical College, Halifax, Nova Scotia
W. E. Watt, Queen's University, Kingston, Ontario
J. Whiting, Saskatchewan Research Council, Saskatoon, Saskatchewan
P. E. Wlsner, James F. MacLaren Limited, Willowdale, Ontario
120
REFERENCES
1. Adams, B. J., and Panagiotakopoulos, D., "Network Approach to Optimal

Wastewater Treatment System Design," Publication 75-13, Department of
Civil Engineering, University of Toronto, Toronto, Ontario, 34 pp.,
December, 1975.
2. Anderson, E. A., "National Weather Service River Forecast System— Snow

Accumulation and Ablation Model," National Oceanic and Atmospheric
Administration, NOAA Technical Memorandum NWS HYDRO-17, Silver Springs,
Maryland, 217 pp., November, 1973.
3. Anderson, J. C. (Gore & Storrie Limited, 135 York. Street, Ottawa, Ontario),
Personal Communication, 1976.
4. Armstrong, K. E. (Alberta Environment, 10040 - 104 Street, Edmonton,

Alberta), Personal Communication, 1976.
5. Benneworth, N. (Canadian-British Consultants Limited, P.O. Box 1269N,

Halifax, Nova Scotia), Personal Communication, 1976.
6. Brandstetter, A., "Evaluation of (25) Mathematical Models for the Simulation

of Time - Varying Runoff and Water Quality in Storm and Combined Sewerage
Systems," presented at the EPA Conference on Environmental Modeling and
Simulation, Cincinnati, Ohio, April 20-22, 1976 (in press).
7. Bras, R. L., and Perkins, F. E., "Effects of Urbanization on Catchment

Response," Journal of the Hydraulics Division. ASCE Proc., Vol. 101, No.
HY3, pp. 451-466, March, 1975.
8. Couillard, D., Dartois, J., Demard, H., Mascólo, D., and Potvin, L.,
"Réseaux de collecte des eaux usees. Tome 3: Caractéristiques des bassins
Les Saules, Sainte Foy et Saint-Pascal," Research on Wastewater Collection.
Vol. 3: Characteristics of Les Saules, Sainte Foy and Saint Pascal Basins
(in French). INRS - Eau Scientific Report No. 61, Quebec, Quebec, 145 pp.,
December, 1975.
9. M. M. Dillon Limited, "Storm Water Management Model Verification Study,"

Report to the Urban Drainage Subcommittee, Toronto, Ontario, 50 pp., March,
1976.
10. Droste, R. L., and Hartt, J. P., "Quality and Variation of Pollutant Loads
in Urban Stormwater Runoff," Canadian Journal of Civil Engineering. Vol. 2,
No. 4, pp. 418-429, 1975.
11. Ehler, P. A. (Town of Midland, Midland, Ontario), Personal Communication, 1976.
12. Ganczarczyk, J., "Behaviour of a Stormwater Retention Pond (A Tentative

Title)," Report to the Urban Drainage Subcommittee, Department of Civil
Engineering, University of Toronto, Toronto, Ontario, 1976.
13. Gibbons, M. (Stanley Associates Engineering Limited, 11748 Klngsway Avenue,

Edmonton, Alberta), Personal Communication, 1976.
121
14. Gore & Storrie Limited, "EPA Storm Water Management Model Verification on
Hamilton Study Area," An.Interim Report to Urban Drainage Subcommittee,
Toronto, Ontario, 46 pp., March, 1975.
15. Graham, P. H., Costello, L. S., and Mallon, H. J., "Estimation of

Imperviousness and Specific Curb Length for Forecasting Stormwater Quality
and Quantity," Journal Water Pollution Control Federation. Vol. 46, No. 4,
pp. 717-725, April, 1974.
16. Hartt, J. P., "A Study of Pollution Loadings from Urban Runoff," Proceedings
of the Canadian Symposium on Water Pollution Research, Vol. 8, pp. 16-25,
University of Waterloo, February, 1973.
17. Heeps, D. P., and Mein, R. G., "An Independent Evaluation of Three Urban
Stormwater Models," Report No. 4/1973, Civil Engineering Research Reports,
Monash University, Victoria, Australia, 92 pp., 1973.
18. Henry, J. G., and Ahern, P. A., "The Effect of Storage on Storm and Combined
Sewers," Report to the Urban Drainage Subcommittee, Department of Civil
Engineering, University of Toronto, Toronto, Ontario, 86 pp., October, 1974.
19. Hesterman, L. C , "Urban Drainage Basin Study, Idylwyld Drive Freeway

Underpass," Report E 74-9, Saskatchewan Research Council, Saskatoon,
Saskatchewan, 45 pp., July, 1974.
20. Hodgson, J., "A Comparison of Three Runoff Simulation Models," Greater
Vancouver Sewerage and Drainage District, Vancouver, B.C., 25 pp., February,
1975.
21. Howard, C. D. D., "Theory of Overflows from Storage, Treatment Plant Systems,"
Report to the Urban Drainage Subcommittee, Unies Limited, Winnipeg, Manitoba,
25 pp., 1974.
22. Jolly, J. P., "Kanata Hydrologie Research Basin," Progress Report, Department
of Civil Engineering, University of Ottawa, Ottawa, Ontario, 38 pp., August,
1973.
23. Killoran, L. B., DeBellefeuille, M., and Tennant, A. D., "Surveillance of

Storm Drainage and Effluent Discharge from Federal Properties to the Rideau
River, National Capital Area," Report EPS-5-OR-75-1, Environmental Protection
Service, Ottawa, Ontario, 62 pp., June, 1975.
24. Kirkpatrick, G. A., "A Review of Flow Measuring Devices," pp. 191-198, Urban
Runoff, Quantity and Quality, ASCE, New York, N.Y., 1975.
25. Kitchen, C. S. (City of Toronto, Department of Public Works, Toronto,

Ontario), Personal Communication, 1976.
26. Linsley, R. K., "A Manual on Collection of Hydrologie Data for Urban Drainage
Design," Hydrocomp Incorporated, Palo Alto, California, 1973.
27. Luckman, B. H., and MacRae, C. (University of Western Ontario, Department of

Geography, London, Ontario), Personal Communication, 1976.
28. James F. MacLaren Limited, "Brucewood Monitoring Programme, Vols. I-III,"

Draft reports submitted to Hydraulics Research Division, Canada Centre for
Inland Waters, Burlington, Ontario, 1976.
122
29. James F. MacLaren Limited, "Review of Canadian Design Practice and Comparison
of Urban Runoff Models," Research Report No. 26, Canada-Ontario Agreement
Research Program, 212 pp., October, 1975.
30. James F. MacLaren Limited, "Drainage Criteria Manual for The City of Winnipeg,"
(Draft) Report to the City of Winnipeg, November, 1974.
31. McPherson, M. B., "Urban Mathematical Modeling and Catchment Research in the
USA," ASCE Urban Water Resources Research Program Technical Memorandum No.
IHP-1, ASCE, New York, N.Y., 49 pp., November, 1975.
32. Mar, H. (City of St. Catharines, City Hall, St. Catharines, Ontario),
33. Marsalek, J., "Aldershot Test Catchment," Unpublished Report, Hydraulics

Research Division, Canada Centre for Inland Waters,, Burlington, Ontario,
22 pp., 1976.
34. Marsalek, J., "Instrumentation for Field Studies of Urban Runoff," Research
Report No. 42, Hydraulics Research Division, Canada Centre for Inland Waters,
Burlington, Ontario, 82 pp., 1976.
35. Marsalek, J., "Malvern Urban Test Catchment — Progress Report No. 1,"
Hydraulics Research Division, Canada Centre for Inland Waters, Burlington,
Ontario, 54 pp., June, 1976.
36. Marsalek, J., "Testing of the Storm Water Management Model of U.S. EPA,"
presented at the EPA Conference on Environmental Modeling and Simulation,
Cincinnati, Ohio, April 20-22, 1976 (in press).
37. Marsalek, J., Dick, T. M., Wisner, P. E., and Clarke, W. G., "Comparative
Evaluation of Three Urban Runoff Models," Water Resources Bulletin. Vol. 11,
No. 2, pp. 306-328, April, 1975.
38. Mechler, W. A. (Greater Vancouver Sewerage and Drainage District, 2294 West
10th Avenue, Vancouver, B.C.), Personal Communication, 1976.
39. Mills, W. G., "Water Quality of Urban Stormwater Runoff," M.Eng. Thesis
(draft), University of Toronto, Toronto, Ontario, January, 1976.
40. Novak, Z. (Ministry of the Environment, 135 St. Clair Avenue West, Toronto,
41. Perks, A. R. (Proctor & Redfern Limited, 75 Eglinton East, Toronto, Ontario),
42. Proctor & Redfern Limited, and James F. MacLaren Limited, "Storm Water
Management Model Study," A draft report submitted to Urban Drainage
Subcommittee, Toronto, Ontario, January, 1976.
43. Solomon, S. I., "A Hydrologie Model for Environmental Assessment of the Rouge,
Little Rouge, Petticoat and Duffins Watersheds," Report to Ministry of
Housing," Shully I. Solomon Limited, Toronto, Ontario, 1975.
44. Solomon, S. I. (Shully I. Solomon Limited, 50 Prince Arthur, Toronto,

123
45. Tonelli, F. A. (Ministry of the Environment, Resources Road and 401,
Rexdale, Ontario), Personal Communication, 1976.
46. Trider, G. L. (Environmental Protection Service, Halifax, Nova Scotia),

Personal Comnunication, 1976.
47. Waller, D. H., "Factors That Influence Variations in the Composition of

Urban Surface Runoff," Proceedings of the 7th Canadian Symposium on Water
Pollution Research, pp. 84-95i Hamilton, Ontario, 1972.
48. Waller, D. H. (Department of Civil Engineering, Nova Scotia Technical College,

Halifax, Nova Scotia), Personal Communication, 1976.
49. Waller, D. H., "Pollution Attributable to Surface Runoff and Overflows from
Combined Sewerage Systems," Final Report to Central Mortgage and Housing
Corporation, Ottawa, Ontario, 168 pp., April, 1971.
50. Waller, D. H., and Coulter, W. A., "Winter Runoff from an Urban Catchment,"
Report to Urban Drainage Subcommittee, Toronto, Ontario, 1974.
51. Waller, D. H., Coulter, W. A., Carson, W. M., and Bishop, D. G., "A
Comparative Evaluation of Two Urban Runoff Models," Report to Urban Drainage
Subcommittee, Toronto, Ontario, 40 pp., April, 1974.
52. Warnock, R. G., "A Study of Pollutional Loadings from Urban Storm Runoff,"
Proceedings of the Sixth Canadian Symposium on Water Pollution Research,
Toronto, Ontario, 1971.
53. Watt, W. E., "QUURM— Queen's University Urban Runoff Model," Preprints of
the Second Canadian Hydrotechnical Conference, May 14-16, 1976, Burlington,
Ontario, 1975.
54. Watt, W. E., and Kidd, C. H. R., "QUURM - A Realistic Urban Runoff Model,"
Journal of Hydrology. Vol. 27, pp. 225-235, 1975.
55. Wisner, P. E. (James F. MacLaren Limited, 435 McNicoll, Willowdale, Ontario),

Personal Communication.
56. Wisner, P. E., Jalal, K. F., Roake, A. F., and Torrens, L., "Application of
STORM and SWMM for Assessment of Urban Development Alternatives in Canada,"
presented at the EPA Conference on Environmental Modeling and Simulation,
Cincinnati, Ohio, April 20-22, 1976 (in press).
57. Wisner, P. E., Marsalek, J., Perks, A. R., and Belore, H. S., "Interfacing
Urban Runoff Models," A paper presented at the ASCE Specialty Conference on
Environmental Engineering Research, Development and Design, July 20-23, 1975,
Gainesville, Florida, 1975.
58. Zukovs, G., "An Investigation of a Sewer Flowmeter Suitable for Open Channel
and Pressure Flow Measurement," B.A. Sc. Thesis, Department of Chemical
Engineering and Applied Chemistry, University of Toronto, Toronto, Ontario,
1976.
124
Appendix I Selected Canadian urban test catchments and
summary of urban run-offfieldstudies of limited scope
SELECTED CANADIAN URBAN TES1 CATCHMENTS
1. Bannatyne Catchment. (Winnipeg. Manitoba)
In the overall assessment, the Bannatyne data are not considered suitable
for urban hydrological research, with the precipitation data being a major weakness.
However, the data were found suitable for a demonstration of an urban runoff
model. ( « )
The Bannatyne catchment'^) w a s monitored by City of Winnipeg personnel

during the summers of 1969, 1970 and 1971. The project objective was to characterize
the quantity and quality of wet-weather flows in a combined sewer. The catchment is
a well-defined drainage district of 220-hectares and is served by combined sewers.
Residential, commercial and industrial land uses are represented in the catchment.
The total catchment imperviousness is 367.»
Originally, several recording raingauges were located within the

catchment, but records from these gauges were lost. Precipitation data are available
for three gauges located 3 to 6 kilometres from the catchment. Runoff flows were
measured by means of a rectangular weir and continuous stage recording. Up to 12
sequential grab samples were manually collected at 10-minute intervals during
periods of high flows. For long duration storms, the data do not span the entire
runoff event. The samples were analyzed for two parameters.
Some of the collected data have been used in a recent study'^2) to

demonstrate the application of the Storm Water Management Model (SWMM) developed
for the U.S. Environmental Protection Agency (EPA). In total, 24 data sets for
1971 were examined for this purpose. A reasonable agreement between the three
raingauges was found only for six storms of low to medium intensity (8 to 20-mm/hr).
For these events, weighted-average hyetographs were derived and used for the
demonstration of the SWMM. In this demonstration, the observed runoff quantities
were simulated with fair agreement, and for the quality aspects the simulations
reproduced the observations within an order of magnitude. The quality data cover
only some of the periods of runoff. High pollutant concentrations were observed,
which may be attributed to the sediment deposition and scouring in low-gradiant
sewer pipes.
2. Barrington Catchment (East York. Ontario)
In an overall evaluation, the Barrington data represents a good source

of storm water quality data for very minor events. Any modeling efforts would
have to be limited to these conditions.
The Barrington catchment^") was established and monitored by the Borough

of East York, with some assistance from the Urban Drainage Subcommittee. The
project objective was to characterize the quantity and quality of storm water and,
in a later stage, to study the efficiency of various alternatives for runoff
quality control. The project has been active for two years (1974, 1975) and a
further continuation is under consideration.
The Barrington catchment is an older (50 years) residential area, in

which the sewers were separated in the 1960's. The roof drains are connected to
the former combined sewers. Only 17.4-hectares out of the total area of 22.7-
hectares contribute runoff to the modified storm sewer system.
Precipitation was measured by a single tipping-bucket raingauge located

within the catchment. Runoff flows were measured by means of a calibrated vertical
126
slot weir and continuous stage recording was accomplished by means of a Wesmar
ultrasonic sensor. Stormwater samples were collected at 5 to 15 minute intervals
and analyzed for up to 20 parameters. Some grab samples were collected manually
and others were collected by means of a Sigmamotor automatic sampler. Separate
recorders were used for precipitation, runoff flow and quality data.
A large number of events have been partially monitored. The number of

fully monitored events, for which complete precipitation, runoff flow and quality
data are available, is rather limited (about 10). Nearly all these events are
rather minor, with peak flows of about 0.056- m 3 /s (2-cfs). No checking of the
data has been reported.
(39)
The synchronization of various records may be
adversely affected by the use of separate recorders.
The quality data indicate fairly high concentrations of pollutants in
storm water. This could possibly be attributed to the minor character of events
observed and to the absence of roof runoff. During minor events, only small
runoff volumes are available to wash off surface contaminants and to dilute the
contents of catchbasins. That is, the Barrington storm sewers carry mostly the
highly polluted runoff from the streets without any dilution by the less polluted
runoff from roofs.
3. Brucewood Catchment (North York, Ontario)
The Brucewood data appear to be suitable for detailed hydrological

modeling. The scope of the data, however, is somewhat limited.
The Brucewood study'2"' was sponsored by the Urban Drainage Subcommittee

and carried out by J. F. MacLaren Limited. The study had several objectives:
firstly, to collect urban runoff data (1974-75); secondly, to monitor snowraelt on
the catchment (1975); and thirdly, to simulate the observed events with a modified
SWMM. The project has been completed.
The Brucewood catchment is a 19,44-hectare modern residential development

served by separate storm sewers. The catchment total imperviousness is 48%. Land
use is characterized by single-family detached and semi-detached residences built
in the late I960*s. The catchment is well defined and documented.
Precipitation was measured by a single raingauge (a tipping bucket)

located just outside of the catchment. Runoff flows were measured by means of a
rectangular weir located at the drainage outfall. The weir head was measured with
a Manning Dipper-Transmitter. Both precipitation and weir head data were
transmitted over leased telephone lines and remotely recorded. Runoff quality
samples were collected manually as well as automatically (by a Sirco sampler).
The sampling interval varied from 5 to 15 minutes. The samples were analyzed for
about 15 parameters. Some additional quality data were obtained, namely samples
of catchbasin contents, street dust and dirt, and sewer base-flow.
During the snowmelt period, the above instrumentation system was only
slightly modified. Temperature and wind data were obtained from a nearby
meteorological station. The snow-covered area was determined from airphotos.
Samples of snowmelt water and snow-slush were collected and analyzed.
The Brucewood instrumentation system was fairly good, but plagued by

frequent malfunctions and breakdowns. About twenty well-documented storm events
have been monitored. For half of these, precipitation (>5-mm) and runoff
hydrographs are available. For the other half, runoff pollutographs are also
available. Typically, these latter events are of a minor character.
127
Three snowmelt events have been recorded on the catchment and have
been well documented. These data have been used for the verification of a
snowmelt model.'^'
The Brucewood data were used for the verification of the SWMH.^ 28 »^ 2 )
For runoff quantities, fair agreement between the observations and simulations
was obtained. The observed pollutographs were reproduced by the SWMM only with
"an order of magnitude" agreement.
4. Calvin Park Catchment (Kingston, Ontario)
The Calvin Park Data are well suited for detailed runoff modeling. The
Calvin Park catchment^-") has been monitored by the Department of Civil
Engineering of Queen's University. The project has been partially sponsored by
the National Research Council. The project objective is to collect urban
precipitation-runoff data and to develop a computer model to simulate these
phenomena. Runoff quality has not been studied. Data collection started in 1973
and the project is still active.
The Calvin Park catchment is a 36.45-hectare modern suburban development

characterized by a low total imperviousness of 27%. The land use can be classified
as follows: low density residential, 70%; medium-to-high density residential,
157a; and institutional, 15%. The catchment is well defined and documented.
Roof leaders are not connected to the storm sewers serving the area.
Precipitation is monitored by two raingauges located within the

catchment boundaries. The output of one of these gauges (a tipping bucket) is
recorded by a Stevens Type A Water Level Recorder, which also serves to record the
stage upstream of a calibrated vertical-slot weir. This results in a good
synchronization of the precipitation and runoff data.
Ten runoff events were monitored on the catchment during the period from
1973 to 1974. All these events are well documented and verified.t29,53) <j>ne
results of more recent measurements have not been yet reported. The Calvin Park
data bave been used for the development of the Queen's University Urban Runoff
M o d e l ^ 3 ' and for a comparative study of several urban runoff models.
(29)
5. Carlina Street Catchment (London, Ontario)
This is a new project^2'' jointly sponsored by the University of Western

Ontario and the (Ontario) Ministry of the Environment. The data will be collected
by the Department of Geography, University of Western Ontario. The main project
objective is to characterize precipitation, runoff flows and their quality on a
highly impervious urban catchment.
}
The catchment area is 32.40-hectares and the total imperviousness is

75%. The catchment is served by separate storm sewers. A variety of land uses
are represented in the catchment: residential, commercial, industrial and
institutional.
One tipping-bucket and six standard raingauges were installed within the
catchment area. Runoff flows are monitored by means of a weir and recorded by a
Stevens (Type F) Water Level Recorder. Manual as well as automatic sequential
grab sampling is used. In the latter case, a North Hants sampler is used to
collect samples at 15-minute intervals. The samples are analyzed for 15 parameters.
No data are available as yet. The project holds promise to yield good
quality data on urban runoff.
128
6. East York Catchment (East York, Ontario)
The Urban Drainage Subcommittee retained M. M. Dillon Limited to establish,

instrument and monitor an urban test catchment in East York. The objectives of this
project^) are to collect urban runoff quantity and quality data on a catchment
served by combined sewers, and to use this data for the verification of the SWMM.
The project started in 1974 and is expected to continue until 1977.
The East York catchment is an older area of 155-hectares served by

combined sewers. The land use distribution is as follows: low density residential,
88%; high density residential, 17.; institutional, 67.; commercial, 17.; and parks
and open space, 47.. The catchment total imperviousness is 497..
Precipitation, dry and wet weather flows in a combined sewer, composition

of combined sewage and air temperature are monitored. A. single tipping-bucket
raingauge is used with a recorder printing the data and time, to the nearest
minute, for every 0.01-inch (0.25-mm) of precipitation. A calibrated vertical slot
weir is used to measure the flows. The weir head is measured by an air bubbler and
recorded, together with the air temperature, by a Bristol Dynamaster strip-chart
recorder. The sampler activation is also recorded on the same chart.
A Sirco automatic sampler (1000-ml) is used to collect sequential grab

samples of combined sewage at regular time-intervals of 7-minutes. The samples
will be analyzed for about 10 parameters. Because the project started only
recently, no data are available as yet. It is expected that the project will
produce good quality data suitable for hydrological modeling.
7. Fairfield Catchment (Halifax, Nova Scotia)
The Fairfield catchment was one of the first urban test catchments
established in Canada. The Fairfield data have been found suitable for urban
runoff modeling.
(51)
The Department of Civil Engineering, Nova Scotia Technical College,
carried out a number of research projects related to the Fairfield catchment.
The objectives of these projects can be summarized as follows: characterization
of quality of surface runoff and combined sewage (1969-1970) ;(^) characterization
of winter runoff (quantity and quality);(->°/ testing of urban runoff models;
(51)
deposition and scour of solids in sewer pipes (a new project starting in 1976);(^°'
and collection of precipitation-runoff data. Some of these studies have been
co-sponsored by the Central Mortgage and Housing Corporation and by the Urban
Drainage Subcommittee.
The Fairfield catchment is an older area of 67,7-hectares served by
combined sewers. The land use is residential (927.) and commercial (8%). The
imperviousness of the catchment was difficult to determine, since some roof
leaders are connected to the combined sewer whereas others are not. A linear
regression of the observed rainfall and runoff volumes indicated an imperviousness
of 347..
As many as three tipping-bucket raingauges were used in various projects
to monitor precipitation on the catchment. A critical depth meter was used to
measure combined sewage flows and the surface runoff was measured by weirs
located at the street inlets. Separate recorders were used for precipitation and
flow measurement. A large variety of water quality data have been collected,
including information on the quality of urban surface runoff, combined sewage,
roof runoff and the contents of catchbasins. Additional observations of dustfall,
dirt accumulation, and sediment accumulation in sewers were also made. Water
129
samples were collected automatically as well as manually. In the former case, a
custom-built sampler was used.'^9-' Most of the samples were analyzed for three
parameters. Bacteriological counts were also performed.
A large volume of data characterizing the quality of surface runoff

and combined sewage was collected and well documented.(^9) However, complete
records of precipitation, flow quantity, and flow quality are available only for
a limited number of events. These data were published mostly as graphs, which
makes it tedious to use them for modeling. Additional urban runoff data may
become available through the current data collection program. Winter runoff data
(snowmelt) were collected successfully for only two events.'-^)
The data collected in 1969-1970 were used to characterize the

composition of combined sewage, surface runoff and effluent from a retention tank.
Effects of dry. period length, season, land use, sediment deposition, dustfall, and
city works practice, were investigated. (")
Fifteen of the observed precipitation-runoff events were used in

comparative studies involving two urban runoff quantity models.(51) j t appears
that some of the quality data could be also used for modeling of urban runoff
quality. Other studies are presently underway.(^°'
8. Hamilton Catchment (Hamilton, Ontario)
The Hamilton test catchment was established and instrumented by Gore

& Storrie Limited in late 1975. This studyi 1 ^ is sponsored by the Urban Drainage
Subcommittee. The study objectives are to collect urban précipitâtion-runoff data
and to verify the SWMM. It is expected that the study will continue in 1976 and
possibly in 1977.
The Hamilton catchment is a 71.3-hectare development served 'by combined

sewers. The land uses represented on the catchment are single family residential
(75%), commercial and institutional (5%), and open space (20%). There are 3,050
persons living within the catchment boundaries.
Two tipping bucket raingauges are located within the catchment area.
Flow rates are measured by means of a calibrated, modified V-notch weir and a
Bristol air-bubbler. The measured weir heads are converted into flow rates.
Combined sewage samples are collected automatically by a Sirco automatic sampler.
The samples are collected sequentially at 5 to 10-minute intervals. Precipitation,
precipitation intensity, flow rate, sampler operation and temperature are recorded
by a Bristol multichannel recorder. No significant data are available as yet.
9. Idylwyld Catchment (Saskatoon, Saskatchewan)
The Idylwyld data represent a good source of data for detailed runoff
modeling.
(19) The catchment has been established, instrumented and monitored by
the Engineering Division of the Saskatchewan Research Council (Saskatoon). In
the early stages of the project, some financial support was received from the
Saskatchewan Department of Highways. The project objectives'^) are to collect
short interval rainfall data, including their variance over a small watershed,
and the resulting runoff. Such information will be used to evaluate the present
drainage design method.
The Idylwyld test catchment is a unique, well defined catchment
representing a section of a vehicular freeway and the adjacent land. The catchment
area is 6.9-hectares having a total imperviousness of 55%. A unique feature of
the catchment is that 77% of the pervious area has a slope steeper than 0.10.
130
Surface runoff is conveyed by storm sewers to a stormwater lift station which
also receives some groundwater flow.
The collection of precipitation-runoff data is fully automated and

controlled by a PDP-8L minicomputer. The minicomputer is interfaced with eight
tipping-bucket raingauges located on the catchment, three 18-inch diameter turbine
flowmeters, one 4-inch diameter turbine flowmeter and one digital shaft encoder
for water level measurements. The flowmeters are used to measure the outflow from
the lift station and the water level data obtained for the sump are used to
determine the storage volume. Inflow is determined from the station outflow and
storage data. Processing of data is fully computerized. Data stored by the
minicomputer is transmitted over telephone lines to an IBM 370 computer and
processed. The quality of storm water is not studied.
Several years of high quality precipitation-runoff data have been

produced so far and are available from the Engineering Division (Saskatoon).
These data consist of maximum rainfall intensities, the total inflow onto the
catchment, the total outflow, an average time of concentration and the runoff
coefficient.
10. Les Saules Catchment (Quebec City, Quebec)
The catchment has been established and monitored by Institut National

de la Recherche Scientifique (INRS-Eau) of Université' du Quebec, Quebec. The
project is co-sponsored by Services de Protection de l'Environnement, Quebec,
and Office de Développement de l'Est du Quebec. The objectives of this project
are to monitor time-varying sanitary flows, infiltration into sanitary sewers,
and composition of wet and dry weather sanitary flows.
The catchment represents a modern single family residential development

of 10,2-hectares served by separate sewers. The area was developed during the
period from 1955 to 1968. There are 476 people living within the catchment
boundary.
The following phenomena are measured: precipitation, sanitary flow

rates, quality of sanitary flows, sewage temperature, and water table levels.
Precipitation is measured by a single tipping-bucket raingauge. A vertical slot
weir is used as a primary flow measuring device. Weir head is measured by a
Manning Dipper and converted into flow rate by a Manning Flow Computer. For
collecting sewage samples, a Manning or SEIN sampler can be used. Water table
levels are measured in up to five boreholes. Collected sewage samples are
analyzed for a number of parameters, such as BOD, Kjeldahl nitrogen, ammonia,
suspended solids and volatile suspended solids.
Three types of recorders are used: in-situ analog recorders for visual
control; a central magnetic tape recorder; and some information is transmitted
over leased telephone lines to the INRS-Eau offices where it is recorded.
So far, only some preliminary observations have been published. The

project holds great promise to provide good data on the variation of sanitary
flows and their quality, and the effects of precipitation and high groundwater
levels on sanitary flows.
11. Malvern Catchment (Burlington, Ontario)
The Malvern data represent a good source of data for detailed modeling
of runoff.
131
The Malvern catchment has been established, instrumented and monitored
by the Hydraulics Research Division of the Canada Centre for Inland Waters.^35'
The project has been co-sponsored by the Urban Drainage Subcommittee. The
catchment was established in 1973. Hydrological studies pertinent to the catchment
are expected to continue until 1978. Main objectives of these studies are to
collect urban precipitation-runoff data, to verify the SWMM on the catchment, to
test various urban runoff models, and eventually to develop a simplified urban
runoff model.
The Malvern catchment represents a modern residential development of

23.1-hectares with a total imperviousness of 34%. The roof drains are connected
to the separate storm sewers. The catchment is well defined and documented.
(35)
The catchment instrumentation has been continuously refined and the
description below refers to the latest version. Precipitation is measured by
two tipping-bucket raingauges. Runoff flows are measured by means of a
rectangular weir which was calibrated on a scale model (for high flows) as well
as in the prototype. The weir is located at the drainage outfall. Weir head is
measured by a float placed in a stilling well. Stormwater quality is monitored
by collecting and analyzing stormwater samples. The samples are collected by two
automatic samplers which are activated by the water level rise inside the weir box.
The samples are collected sequentially at 5 to 10 minute intervals. A modified
Stevens water level recorder, Type A, is used as a central, in-situ recorder. The
output of one precipitation gauge, weir head and operation of both samplers are
recorded on the same chart.
So far, only the 1973 data have been fully analyzed and published.(•")
These are precipitation and runoff quantity data for eight medium-intensity
storms. The data collected in 1974 and 1975 are presently being analyzed and
their publication is scheduled for late 1976. A preliminary analysis indicates
that about thirty quantity-only events and twenty quantity and quality events will
be suitable for publication. The Malvern data appear to be of a good quality. All
the published data have been checked and verified.
The 1973 and 1974 Malvern data were used for the testing of the SWMM.
A close agreement between the runoff quantity simulations and observations was
reported."5,36) g u c n agreement was not found for the runoff quality.
Other studies are in progress, involving simulation of runoff quality

and development of new runoff models.
12. Rigaud Catchment (St. Foy. Quebec)

de la Recherche Scientifique (INRS-Eau) of Universite du Quebec, Quebec.
(8)
The
project is co-sponsored by Services de Protection de l'Environnement, Quebec, and
Office de Développement de l'Est du Québec. The ob-jectives of this project are
to identify and characterize various inputs to sevipr systems, such as stormwater,
sanitary sewage, and groundwater, and the total pollutional loadings carried by
sewer flows.
The Rigaud catchment is a 20-year old residential development served by
combined sewers. The catchment area is 15.7-hectares with a total imperviousness
of 44%. The total population of the area is about 800 persons.
The data acquisition system is fully integrated and consists of a
tipping-bucket raingauge, a 24-inch Palroer-Bowlus flume, air bubbler, thermometer
and automatic sampler. Dry weather flows are monitored by means of a 3-inch
132
Parshall flume. Combined sewage samples are collected by an automatic sampler
activated by the raingauge. Three types of automatic samplers have been acquired
and can be used alternately: a French sequential APAE 241F sampler (collects up
to 24, 2000-ml samples); an ISCO sampler (model 1392); and a Manning sampler
(S 4000). The samples are analyzed for five parameters.
Three types of recorders are used: analog in-situ recorders for visual
control; an in-situ magnetic recorder (made by SEIN, Alfortville, France); and
a remote recorder at the INRS-Eau offices. The data are transmitted over leased
telephone lines. The following information is recorded: precipitation, sewage
flows, sewage temperature, and sampler operation.
So far, only some preliminary data have been published. The project
holds great promise to provide high quality data for modeling of combined sewer
flows.
13. Saint Pascal Catchment (Saint-Pascal de Kamuraska, Quebec)

de la Recherche Scientifique (INRS-Eau) of Universite du Quebec, Quebec.'**) The
project is co-sponsored by Services de Protection de l'Environnement, Quebec, and
Office de Développement de l'Est du Quetec. The project objectives are to
characterize dry and wet weather flows in sanitary sewers and to characterize
quantity and quality of surface runoff in a serai-urban community of Saint-Pascal
de Kamuraska (2,500 inhabitants). The catchment serves for two types of
investigation: surface runoff is studied for the entire storm drainage area of
81-hectares; and a subárea of 39.3-hectares served by sanitary sewers is used for
the study of sanitary flows.
The 39.3-hectare subárea is a predominantly residential area (71%

residential lots, 14.5/!, streets), with some commercial establishments (7.3%) and
open space (7.2%). There are 1,100 persons living in the subárea. The data
acquisition system for the subárea is similar to that installed on the Les Saules
catchment noted earlier. Precipitation, sanitary flow rate, sanitary flow
composition and temperature, and ground water level are monitored. Sanitary flow
rates are measured by a Palmer-Bowlus flume with a built-in capacitance water
level sensor (manufactured by UES, Pleasanton, C a . ) . Composite sewage samples
are collected sequentially by a SEIN sampler.
The storm drainage area of 81-hectares consists of the sanitary subárea

described immediately above plus agricultural land and open space, having a total
imperviousness of 137.. A single tipping-bucket raingauge is common to both
districts, runoff flow rates are measured by means of a calibrated vertical slot
weir and an air bubbler. A Sigmamotor instrument is used to convert weir head
into flow rate. An ISC0 automatic sampler is used to collect sequential composite
samples of stormwater. The sampler is activated by the raingauge. Samples are
analyzed for 10 to 15 parameters. In-situ analog recorders and an in-situ
magnetic tape recorder are used.
Only some preliminary observations have been published so far. The

project holds great promise to provide useful information on sanitary flows and
good data for hydrological modeling of surface runoff in a semi-urban community.
14. West Toronto Study Area (Toronto, Ontario)
The West Toronto data appear to be suitable for the study of overflow
frequencies and demonstration of urban runoff models. The area is well
documented for such purposes. The data do not appear to be suitable for general
133
testing or evaluation of urban runoff models because of uncertainties in field
observations described below.
The catchment has been established and monitored by the City of

Toronto.'^ 2 ' Project objectives are to monitor the frequency of combined sewer
overflows and to provide data for a demonstration study of an urban runoff model.
The project has been active for the last seven years.
The West Toronto Study Area is a 944-hectare residential development

served by combined sewers. The total imperviousneso of the area is 54%.
Precipitation is measured by a single tlpping-bucket raingauge located

on the catchment and other precipitation data are available for a nearby
Atmospheric Environment Service (AES) raingauge. Combined sewage flows are
determined approximately by measuring the depth of flow in an overflow sewer and
applying the Manning equation. To obtain the total runoff from the area,
interceptor flows and diverted flows (by several diversion weirs) had to be
estimated and added to the measured flows in the overflow pipe.' ' No quality
observations have been made.
The data collected contain some appreciable uncertainties. Considering

the size of the catchment, it would be desirable to have more raingauges on the
catchment. Estimated runoff flow rates may be in error by 15 to 20%.
15. Windsor Catchment (Windsor, Ontario)
The Windsor data are useful for estimating pollutional loads in storm-
water. The data do not appear, however, to be suitable for detailed hydrological
modeling. Large uncertainties in the precipitation and runoff measurements are
the main reasons.
The Windsor catchment has been established and monitored by the

Department of Civil Engineering, University of Windsor, with some financial
support from the National Research Council. The primary study objective was
determination of the composition of stormwater.(lO) The project has been completed.
The Windsor catchment is a 12-hectare residential development served by

separate storm sewers. The housing units are mostly single family units, 25 to 30
years old, generally in good repair. The street asphalt cover is fairly worn,
which may lead to high wash-off of solids.
Precipitation data were obtained for five raingauges (weighing type)

located in the City of Windsor. None of these was located within the catchment
boundary. Runoff flow rates were measured approximately by calibrating a
measuring section and monitoring the stage at this section. No checking of the
precipitation versus runoff volumes has been reported. For some of the published
events, the runoff coefficient approached or exceeded unity.tÔ) Runoff quality
was monitored by collecting and analyzing stormwater samples. A Testing Machines
International Sampler was used to collect hourly grab samples. Nineteen analyses
were performed on the samples. In total, 25 storms were monitored on the
catchment.
134
URBAN RUNOFF FIELD STUDIES OF LIMITED SCOPE
In this section, urban runoff studies designed to provide limited data

for engineering or planning studies are briefly listed. Some of these projects
could develop in the future into comprehensive studies providing data for urban
hydrological research. The projects are listed alphabetically.
Aldershot catchment(33) is a 6.9-hectare commercial plaza that is 100%

impervious. Limited observations of precipitation, runoff flow rate and runoff
quality were made on the catchment. The feasibility of using this data for
hydrological modeling is presently being studied..
In the Dartmouth project^ ' storm water quality was monitored in

several commercial plazas for several storms.
At Edmonton^55-' a large-scale urban runoff study was initiated. Field

data are collected in support of a modeling study of the entire combined sewer
area.
At Guelph< 40 ) some observations of precipitation, runoff flow and

runoff quality were made on two test catchments. Steps were taken recently to
refine the catchment instrumentation.
In another major study,*- ' the effect of municipal pollution on the

Grand River is being studied. For this purpose, water quality upstream and
downstream of the city of Guelph (about 60,000 inhabitants) is being studied.
At Halifax^ 0 -' perfornance of a combined sewage retention tank was

studied.
McKittrick drainage area is a small drainage district on which some

precipitation and runoff measurements were made. The data were used for
calibration of an urban runoff m o d e l . ^ '
At Merivale' 3 ' a runoff retarding device ("Hydro-brake") is being

tested on a small catchment. Runoff from the area enters a small underground
storage tank and the outflow from the storage is controlled by the Hydro-brake.
Also, some water quality aspects are being studied.
At Midland the water budget of a small urban lake receiving urban

runoff is being studied.(*•'•)
At Montreal^^' combined sewage flows have been measured at a number

of locations. The best documented are the measurements carried out on the
Papineau-Curotte Catchment (1166-hectares) in 1973.
Pollutional Significance of Sanitary Bypasses^ 1 ") was a field study

of sanitary flow bypasses at four Ontario towns: Aurora, Brantford, Dundas and
Waterloo. The study was carried out by the Waste Treatment Section of the Ontario
Ministry of the Environment. Over a period of twelve months, precipitation,
bypass quantity, and bypass quality were monitored. The data will be used to
determine the pollutional loads in wet weather bypasses of waste treatment plants.
At Port Dalhousie and Port Weiler^ ' wet weather sanitary flows at a
pumping station and pollution control plant were studied.
135
In the Sawmill Creek Project,'-*) a suburban catchment partially
served by sewers will be monitored. The study objective is to determine the
feasibility of storage and treatment of stormwater runoff.
At Toronto'") t n e City has proceeded with the installation of an

automated raingauge network. Eight raingauges have been installed so far. Flow
monitors and automatic samplers will be added in the near future.
At Toronto' 55 ' quantity and quality of surface runoff from an oil

terminal site was investigated.
At Toronto International Airport^ 55 ) quantity and quality of airport

runoff was investigated. Several catchments were monitored for various time
periods.
At Vancouver'3") the City initiated the installation of an automated

raingauge network. Eighteen raingauges have been installed so far and others may
be added. Six water depth sensors will be installed in storm sewers to obtain
data for hydrological modeling. The project objective is to develop a
computer-oriented monitoring system.
In the Whitemud Creek Study'^) the impact of urban drainage and

recreational activities on a rural stream will be studied.
In the Willet Creek Project,' 55 ) dry and wet weather flows in sanitary
sewers are being studied.
At Winnipeg^ 55 ' water quality of stormwater detained in ponds is being

studied.
136
Appendix II Instrumentation considerations
A summary of ' Instrumentation for Field Studies of
Urban Runoff' by J. Marsalek, Research Report N o . 42,
Hydraulics Research Division, Canada Centre for Inland
Waters, Burlington, Ontario, 82 pp., 1976 (Reference
N o . 34).
137
INSTRUMENTATION FOR FIELD STUDIES OF URBAN RUNOFF
ABSTRACT
Instrumentation and monitoring techniques for field studies of urban

runoff were examined. In particular, the following types of instruments were
studied: recording precipitation gauges; sewer flow measurement instruments;
and automatic wastewater samplers. After reviewing the literature and surveying
equipment, some of the more promising instruments were acquired, tested in the
laboratory, and operated in the field for various time periods.
Individual instruments are discussed with regard to their technical

data, accuracy and reliability. Recommendations for the selection, interfacing
and installation of the instruments are given.
CONCLUSIONS
Assessment of the environmental impact of urban runoff on receiving

waters requires detailed data on precipitation-runoff processes. Such data is
also needed for further development and application of urban runoff models.
Precipitation, runoff flow rate and runoff quality are of major interest.
Precipitation data consists of point precipitation and of the areal

distribution of precipitation. Such information can be obtained from a network
of several recording rain gauges installed within the studied area. The tipping
bucket rain gauge of 0.01-inch (0.25-mm) per tip capacity is particularly
suitable for this purpose. A good time resolution, frequently 5-minutes or
shorter, is required. Two gauges are sufficient for catchment areas up to
10-km' (4-square miles), and for up to 50-km (20-square miles) three gauges are
recommended. Time resolutions of rainfall data recommended for urban runoff
studies are given in Table 1.
TABLE 1. RECOMMENDED TIME-RESOLUTIONS OF RAINFALL

DATA FOR URBAN RUNOFF STUDIES
WATERSHED SIZE
WATERSHED TYPE TIME RESOLUTION
Acres Hectares (Minutes)
Small experimental watersheds 10-300 4-120 1-2

(for model development or
calibration)
Large Experimental Watersheds 500-3000 200-1200 5
Data Serving for Design up to 3000 up to 1200 5-10
> 3000 > 1200 10-15
Data Serving for Planning > 3000 > 1200 60
Runoff flow rates should be recorded continuously at one or more points.

Whenever feasible, runoff flows should be measured at the outfall, outside the
sewer system. Conventional constriction flowmeters such as weirs or flumes can be
used.
138
If it is necessary to measure inside the sewer system, and the sewer
pipe is not frequently surcharged, an inexpensive vertical slot weir or a flume
(e.g., Palmer-Bowlus flume) are applicable. For frequently surcharged pipes, a
dual free-pressurized flowmeter such as the U.S. Geological Survey Sewer
Flowmeter or an acoustic flowmeter should be used.
The acceptable accuracy of runoff flow measurements is 5 to 10%.
Characteristics of selected liquid level sensors and an overview of

sewer flow measurement techniques are given in Tables 2 and 3, respectively.
TABLE 2. CHARACTERISTICS OF SELECTED LIQUID LEVEL SENSORS
INPUT POWER OPTIONS

APPLICATION TYPICAL INSTALLATION
TYPE OF (SENSOR ONLY)
LIQUID LEVEL
SENSOR FREE PRESSURE DIRECTLY IN SEWER IN A
FLOW FLOW IN SEWER BUT WITH STILLING DC AC OTHER
SOME PRO- WELL
TECTION
Capacitance Probe X X X X X
Dipper Probe X X X X
Floats X X X none
(Scow required
float)
Pneumatic Probe X X X X X X
(compressed
X
gas)
Acoustic Probe X X X
X
Runoff quality is commonly determined from the laboratory analysis of grab
samples collected in the field. Such samples are collected sequentially by
automatic samplers. A sampling interval as short as 5 to 10 minutes may be
required. The first sample should be collected as closely to the beginning of runoff
as practicable. In the currently common approach, a constant sampling interval is
selected on the basis of experience and the size of the studied area. A review of
ten urban runoff studies (i.e., storm water runoff as well as combined sewer
overflows) indicated the sampling intervals shown in Table 4.
Other factors to be considered in the selection of a sampling interval

is the precipitation time-distribution and the watershed hydrologie response.
These two factors influence the runoff flow rates to which the stormwater quality
seems to be related. Consequently, high intensity and low duration summer storms
on fast responding watersheds will call for shorter sampling intervals and vice
versa.
The first sample should be collected as closely to the beginning of

runoff as feasible. This can be achieved by activating the sampler by the first
impulse from the precipitation sensor, or better, by the rise of the water level
in the sewer by a preselected increment.
Some electronic liquid level sensors (e.g., capacitance probes, Manning

Dipper, ultrasonic probes, etc.) can be equipped with alarm relays and these are
then used to close the power supply circuit of the sampler when flow reaches the
selected level.
139
o
zo
n 01 ta en ta en ai ai ta u> U)
(U z (U OJ 0) a) 0)
u ai >•ai >4ai a) o
>" >H >< >i >4 >l >J Z
X
co o
O Z .-! X ^ i-î HÎ •J 2 S S ¿ s a
ü <
a
H o s-s fi s-s s-2 fi s-s fi i~î fi Ï~Î fi
o in m m m m in m m m
o m
ss CM
CO U
S
X X X X
W M
CO PL«
X X X X X X X
u
10
0
W •
0]
B 01
l-l
3
-H
X X X X X X X X X T3
o
Xi
S S
n
S g
4J
•a
Ci
3 3 3: a 01
CO O
i X X
«
• A
4J
01
ai
»
4J
01
Cu CM XI
o
U
Si eu
00 o
•ri u
X X X X X X X X X X X X -o
S 01
it u
3
0]
s 01
•n ai
4J
09 u
4J T5 a
4J 0 n o
O tm
U a u O G
•H 01 5U rt ta
u
60
s
r-l
S •H
c •a y, 3
t5
1 r4 £ . » M
5 3
8. m
01 00.fi O « rH 0 i co o • «)
Suu |
S U N U h • 1S •4 Ü
H
Su 35
ta
U «Irl o .c v a •
Î1 o
u
01
a
a. u
U*
•H
01 u o a.u
O S « h
eu • e • >
o hHio^
« « « »a
iJhhDS sg
4= O tl l h «
CO
U ri
a. ai
« > H >
140
TABLE 4. SAMPLING INTERVALS IN URBAN RUNOFF STUDIES
WATERSHED SIZE SAMPLING INTERVAL 24 SAMPLE CYCLE DURATION
(Acres) (Ha) (Minutes) (Hours)
10 4 5 2
50 20 5-7.5 2-3
100 40 5-10 2-4
500 202 5-15 4-6
1000 455 5-15 4-6
2000 809 15 6
3000 1214 20 8
5000 2023 25-30 10-12
The desirable size of samples is about 1000-ml. Great care has to be

devoted in order to avoid systematic errors in the sampling. The first step in
this direction is to locate the sampler intake at a cross-section where the
sampled medium is rather homogeneous. The capability of the sampling apparatus
to collect solids should be evaluated, mainly with regard to the intake
orientation and the intake nozzle and line velocities.
To reduce the loss of quality data owing to sampler malfunctions, two

samplers may have to be installed and operated in parallel.
The list of water quality parameters usually investigated in urban

runoff studies is presented herein in Section 2 (Table 2).
A good time synchronization between the recordings of precipitation,

runoff flow and sample collection can best be ensured by recording all this
information on the same chart or tape.
141
Methods for calculating m a x i m u m flood dis-
charges for natural watercourses and urban
areas in the U . S . S . R .
V. V. Kuprianov
State Hydrological Institute
23 Vtoraya Liniya, V . O .
Leningrad, L-53, U . S . S . R .
Computation of Peak Flow
In the U.S.S.R., the formula presently recommended for the computation of
peak flood flows of natural watercourses and urban areas is the so-called
"rational" formula,C 1 » 2 » 3 )
Q p - 16.7 ±T p a 6 A Eq. (1)
3
where: Q is the peak discharge in m /sec for a given probability of
exceedanee, P;
ï _ is the maximum value of mean rainfall intensity in mm/min for
the design duration ( T ) ;
T is the time of concentration (time of travel) in minutes;
Of is the runoff coefficient associated with the greatest depth
of precipitation for the design duration of precipitation ( T ) ;
6 is a coefficient that accounts for reduction of maximum
discharge attributable to the marshy or swampy character of
the drainage basin; and

2
A is the area of the basin in km .
The maximum value of mean rainfall intensity for the design duration is
obtained from precipitation reduction curves that are based on observed data
from automatic precipitation recording instruments and other types of

(4)
raingauges. Precipitation reduction curves represent regional empirical
correlations of the maximum mean intensities of precipitation for various
durations and probabilities of annual exceedance.
Application of Precipitation Data
In order to obtain precipitation reduction curves, i _ * f(t,P), the
following methodology is used for processing and drawing general inferences
from observed data:
1. For each weather station, the maximum depth of precipitation, H , is selected
from among the pluviograms for each year, for each of several durations
(t - 5, 10, 20, 60, ... 720, ... 1440 minutes), together with the
145
associated total daily precipitation, H. (Because H can occur over two
calendar days, H,,._ — H, as will be noted in Table 1 ) .

1440
2. Curves of probability of exceedance are plotted of the depth of precipitation
at various durations, H • f(P), and mean intensity of precipitation,
i t - H t /t =• f(P).
3. From these curves are obtained values of i and t for various probabilities
of exceedance (P ¡» constant), which are plotted as precipitation intensity
reduction curves, i « f(t,P).
When observed data from precipitation recorders are missing or when the
period of observation is inadequate for the plotting of exceedance curves,
recourse is made to empirical formulae or standard precipitation reduction

( 1 2 3 5)
curves. » » » ' The most widely used type of precipitation reduction curve
equation is
it p - B|l + C log 100j/tn Eq. (2)
where B, C and n are parameters of the equation determined on the basis of
mapped observed data from base stations.
Standard precipitation reduction curves are the averaged ratios of mean
maximum intensity of precipitation with a probability of exceedance, P, expressed
as fractions of the maximum daily depth of precipitation of the same probability
of exceedance, H , for a given duration, t,
*t P
- 5 * ^ - f(t,P) Eq. (3).
H
P
In order to construct standard precipitation curves, observed data from
precipitation recorders in weather stations of a homogeneous region are combined.
Using these data, a family of smoothed empirical curves is plotted showing the
frequency of the maximum depths of precipitation, H , for durations, t,
H f
t,P " l(P,t) Eq
- (4)
146
—t 00 O CM rH C» 00
CO rH O 00
•c* Ol rH CO -Ci- vO oo
in MO 00 00 CT\ en en en en
0> CO rH IA in <r
to 00 vO o . . . .
<r
CT\ t^-
00 <*•
vO •A
O o
m Cn en CTi M3 in m
o m
20.2
16.0
18.0
20.9
o
19.1
20.7
00 CM
oo o •
00 O i-» CO O 00 in co
CM rH 0~l 00 00 MO vu MO
O CM en rH m in -cl-
-cf O CO CM •cl- O en en m in
v£> MO m m in CM rH rH rH rH rH
a«
oo m co CO -Cf rH O CM rH
. . . . . .
o oo CM co co CO MO MO o vo m m m -et co
CM oo 00 m m -cr
m
<f rH .H CO M3 en m r- CM CO m 00 CM
o
XJ o o CO 00 r- MO m m -et* -et. •
co CM .
o• O •
33 *-\ CO —
r <r CO CO CO co co rH rH rH rH rH rH
a)
z 4-1
vO
o fi O
oo m •H r-^ m CM o oo O . en 00 r-o O 00
l-l
H
•H in
oo co co co CM rH « . . 00.
C ^
MO -Cj- CM en en en en
M
l-l in co CM co o
»o . en en o o -o-
in co 00 m o r-- m
a -3- C0 co rH
co
co CM
CM r»
0
r»
•
r-«
.
m
Oá
fa o o c M c o c o e o c o o o i n o o
O Z o vo m rH m CM MO
SB co m o o m 00 00 MO m 0 0 0 0 0
H MO -et co CO C0 CM CM CM CM CM r-» r^- r^- vo vo co
a rH
H CM CM vo en in
w -et -et rH m
o O en m
m CM m en va m •d- -et 0 0 0 0 0 0
•Cf co CM CM CM
CM CM
m m m m m CM
CM
en oo CM oo in
o oo —
r r- oo
CM
co \0 «H -ci- co CO CO O r- 0 0 0
co co CM CM
CM CM
CM rH -d- < r -cj- co
00 -cr r»
.
o- 0
. . r*- vo o in o en
CM 0 0 . . . 0
m CM en
ô
<r co CO CO CO CM CM O
CM 00
in -et- en -et CM
. • o . •-H 00 -d- <t m
CO a . • 0 0
CM i-l
co o en en oo rH rH O
Z „
O Bi
H W rH CM CO oo en o rH CM co
Ë" S 1 <r <t m m m m
ÖS
CO Z
147
as well as a smoothed empirical curve of frequency of precipitation H for the
same years, on which are plotted curves of H « fAt).
On the basis of the empirical frequency curves H • f (P,t) and

t,r 1
H_ » f„(P), for a series of P from 1.5% to 20%, the corresponding depths of
precipitation are determined for each duration (t =» 5, 10, 30, 60 min., e t c ) .
On the basis of the values of precipitation depths of equal frequency, H
curves are plotted showing the increase with increase in precipitation duration
of the ratio
H
-5 1 -- V i ) Eq. (5).
P
By dividing the ordinates of the curve •„(t) by the corresponding values of t,
the relative values of the mean precipitation intensity for each duration t of
the same probability of exceedance P are determined:
t_.(t) H _ ï „
Shown in Figure 1 are curves of +p(t) and ïp(t) for one of the regions of the
Soviet Union, Design values of these two parameters, which are based on the
combined data of several weather stations with a similar rainfall pattern,
constitute stable characteristics of precipitation that depend only to a small
extent on the representativeness of observation periods at individual stations.
For the territory of the Soviet Union, thirty standard reduction curves have
been constructed together with charts of depth of precipitation of a probability
of exceedance P. The precipitation depth (H ) and precipitation intensity

•-» "
(ï ) for a duration (t) and probability of exceedance (P) is determined for
t,P
any point with known precipitation (H ) according to the formulae:
Ht>p - V t ) H p Eq. (7)
and ít)P - y t ) H p Eq. (8).
148
0.050
\¡J>
0.020 fo",
^•v
a f ) _
,i,lO ^-
z o.oio T
II
* ¿ | CL
xi 0.005
•-^ 0.002
0.001
10 20 50 100 200 500 1000

t, MINUTES
FIGURE 1.-REDUCTION CURVES FOR DEPTH OF PRECIPITATION AND

INTENSITY OF PRECIPITATION
149
Equation 2 and the precipitation reduction curves make it possible to
calculate the mean maximum intensity of precipitation and the depth of
precipitation for a specific point (station).
By using Equation 1 for estimating runoff from small basins, any reduction
in precipitation intensity attributable to size of area is overlooked.
Use of Runoff Coefficients
The runoff coefficient, a, is established on the basis of synchronous
observations of runoff and precipitation at experimental water-balance stations.
Observed data are processed to give design equations and tables of runoff
coefficients for use in areas that have not been studied. For example, the
following formula is proposed for urban areas: '•*'
- 0.2 0.1
where » is the averaged value of the runoff coefficient at an exceedance

av
frequency of 1007., determined in accordance with Table 2.
TABLE 2
AVERAGED VALUE OF RUNOFF COEFFICIENT AT P = 100%
TYPE OF URBAN SURFACE a

av
Roofs and asphalt-concrete road surfaces 0.95
Crushed stone roads and macadamized road surfaces 0.60
Cobbled roads 0.45
Roads surfaced with loose stone chippings 0.40
Gravelled garden or park paths 0,30
Levelled earth surfaces 0.20
Lawns 0.10
For natural slopes and swales, it is advisable to use runoff coefficients
that are related to the nature of the surface of the basin, the depth of
precipitation and the area of the basin of the watercourse, Table 3.
Flood Travel Time
The flood concentration (travel) time, T, is determined for urban areas as
the sum of the time of travel of rainwater along the surface (T ) , gutters (T ) ,
150
% O O O O i/i ¡n o ui o o o o ui o O O
o
o vo oo o\ en <f vO I S N
cN m • * m 2 !
• • • • O O O O O O O O TI
o o o o • • • • tí
3
o o o o
u
o ID O O O o m o m Ul ui Ul ul O ul ul ul o
o *G CO Cn CT* to <ÍJ r - r-» d -tf u-| vO N N P1-Î 1-1
r-l O O O O O O O O O O O O O O O O o
CM O
m o o m m i/i o O ul vo ul O O Ul O O Ul
vo oo en en m r- oo oo •tf u i \D t*~ CN CM «tf Ul
1-1
o" o o o O O O O O O O O o o o o
< ~+ o
C O r-l
o m o m O O i/I O
<í> oo oo en
ul co O ul
ui \o r» r-
00 Ul Ul Ul o
CM m .tf ui
o r-. oo en en CM
« i-l O O O O O O O O
• • • • • o o o o
«S r-l O O O O
o o io m o ui m o ui ui ul O Ul Ul Ul O Ul
0 0 CT» CT* CTi
r» oo oo en ui vo r» oo el -tf ui ô CM
o s e o
O O O O O O O O O O O O O o o o o
V
55
O o o O O
o o O O
O H O ul O Ul O
ui O Ul CO
r-i M O
r-l CM O r-< CM O
r-l CM O O
O O
O
00 I I
O
CM
o
oo 1 1 CM O O
00 I I CM 00 I I CM
Vil rH -í A Vil Vil r-l r-l A
00 Ul Vil <-* -i A
00 ui
OO Ul
W
O
ïs m 3
e c
I
cj CU N ui
u H O
tí
U
ë
S cd o
y 00 M e ca
o r-
ol
c CU CU > cu U
CO o Xi XI TI
N ca .tí .O
tu
y tí tí CO O ca u 1 >ï
O rd .
tí
cu i >lXJ
CU tí Xl
CO
U u
y I
i-i O ca
CJ o
H
X iH H
- *
M ed
0 0 CO n)
M o « rH • H r-< y ed •ri
u 1-1 "XI U
H o
N
XI
tí Ul
«
Ul
X) XI >> o en
•i-l Ul
U-l
CO
O cd > U] O rH
Cd r-l
•ri E X)
M eu
Ul -ri
eu tí
aí
U
M
3 n)
a X(U .e
tí
n O
tí O
S Ul
N
O cd
H ia
«t
U]
H ca O M
• H en 3
m u eu eu
U
— i O r-l
Ul -ri
r-l Ul
O
i tí .OCL • H
LU
d O N cEd TD „ M O
4J 3
n. Ul
•> eu
eu
>
3
Ul XI O d cd cu cu G 4J >, cd
O i-i eö -tí ed co X) u
r-l U y
O TÍ
a XI RJ
cu V-i
&V3
ed -u
i
O
O 1
r-l 4J
U
S
Ul
00
>> >. «XI
CO >. ci) *> Ul
*o tí Xi X
Cd
Oi
X
H
I > •*
Cfl 0)
E 0 G
o i3
rJ
<rl 00 h
XI -ri Cd
xi
tí
cd
co
ca XI
eu C
xi cd
>.!-<
X I -ri
tí O
tu u u i-i y cd ui
a o co
S
151
and pipes ( T p ) : ^ ^
T =• T s + T g + Tp Eq. (10).
The time of surface concentration (T s ) is taken as 5 to 10 minutes. The time of
travel of rainwater along street gutters is determined by means of the formula:
T g = 1.25 ig/v g Eq. (11)
where ¿g is the length of the gutter in meters and v is the speed of movement
of rainwater at the end of the gutter in meters per minute. Time of travel along
pipes is taken as:
T p = Z E(l p /v p ) Eq. (12)
where ¿p is the length of the section of main drain in meters and V p is the
computed speed of movement of rainwater in meters per minute; and Z is a
'coefficient related to the climatic conditions of the locality and its relief,
varying between 1.2 and 2.8.
For natural conditions of slopes and gullies it is recommended that the
concentration time be determined by the use of the following relation:

(1)
1.1
T = 1.2TC + Tsl Eq. (13)
where T c is the time of travel along the channel of the watercourse in minutes
and T s ¿ is the time of travel over the slopes of the basin to the watercourse.
Time of travel along the channel is determined by the equation
1000L „ .,..
Tc = T—J- Eq. (14)
McT c »Q c t
where L is the length of the watercourse in kilometers, T c is the weighted mean
gradient of the watercourse in per cent, Q c is the maximum discharge in n? per
second, and M,, is a coefficient depending on the roughness of the watercourse.
152
Values of Mç are given in Table 4. The time of travel over the slopes of the
basin to the watercourse is given by:
,• (1000 J)^ „ /nr.

T
SX -^ . , / , ! . Eq. (15)
M s Ts *>[—— )
U6.7/
where I is the mean length of slope in meters, T s is the mean gradient of basin
slopes in per cent, qQ is the maximum rate of runoff per unit area from the slopes
in np/sec/km , and Mg is a coefficient depending on the roughness of the basin
slopes. Values of Mg are given in Table 5.
Flood Volumes and Flood Hydrographs
In order to simplify the computation of maximum runoff, tables have been
prepared that contain computed values for maximum runoff per unit area in relation
to the runoff coefficient and the depth of precipitation (Qp/A/0fHp) as a function
of morphological characteristics of the river channel
Dc = 1000 L Eq, (i6)

M T £A£ (£*&.„£
C C X/a
and the slope
DS =. (1000 tf ^ Eq„ (17).
Volumes of flood flow are determined as the product of the depth of flood
flow (hff) and the area of the basin (A):
Wp = 1000 hffA Eq. (18),
The design depth of flood flow is determined by one of the following two
methods:
(a), by using flood flow depth charts constructed on the basis of data from the
rivers under study (for natural watercourses); or
(b), by linking the depth of flood flow with the precipitation associated with
153
TABLE 4
CHANNEL ROUGHNESS COEFFICIENTS, M c
CATEGORY AVERAGE CHARACTERISTICS OF RIVER CHANNEL

OF RIVER CHANNELS AND FLOOD PLAIN THROUGHOUT ROUGHNESS
CHANNEL THE LENGTH OF THE RIVER FROM ITS COEFFICIENT,
SOURCE TO THE DESIGN GAUGING SECTION M
c
1 Relatively clear channels of

permanent flatland watercourses; or
Channels of periodic watercourses
(dry gulleys) in relatively
favorable conditions: 11.0
2 Channels of large and medium-sized

rivers which are partially
obstructed, tortuous and partially
overgrown, and stony with
disturbed current; or
Ephemeral watercourses, carrying a
considerable quantity of sediment
during flooding, with beds covered
with large pebbles or vegetation: 9.0
3 Pebble- and boulder-strewn channels

of the mountain type with uneven
water surface; or
Channel of ephemeral watercourses
heavily obstructed and tortuous: 7.0
154
TABLE 5
BASIN SLOPES ROUGHNESS COEFFICIENTS, M g
BASIN SLOPES ROUGHNESS COEFFICIENT, M s ,

SURFACE IN TERMS OF DEGREE OF GRASS COVER
CHARACTERISTICS
OF SLOPES N0RmL THICK
oïïSLr
Smooth surfaces
(asphalt, concrete) 0.50
Rolled and levelled soil

surface, takyr-type
plains 0.40 0.30 0.25
Well-ploughed and
harrowed surface,
unploughed without
hummocks, cobbled
roadway surfaces in
inhabited localities of
less than 20% built-up
area 0.30 0.25 0.20
Roughly ploughed surface,

taiga tremuli,
hummocky surfaces,
inhabited localities
of more than 207,
built-up area 0.20 0.15 0.10
155
the runoff in accordance with the formula
h p = QfHp Eq. (19)
where a is the coefficient of flood flow and Hp is the depth of precipitation with
a probability of exceedance P.
The design hydrographs of flood flows for natural watercourses are generally
constructed in accordance with standard hydrographs or analytical equations which
provide a relatively satisfactory description of natural hydrographs. W The
parameters used for such hydrographs are the computed values of volume of flood
flow (W ) , maximum discharge ( Q p ) , and the coefficient of hydrograph form (X),
which depends on the relationship between the volume of flow up the flood peak
( W p e a k ) and the total volume of flood flow (Wp)„ Table 6 gives values of floodwater
discharge as fractions of maximum discharge for various coefficients of flood form,
X = Qptpeafc/Wp, and for various ratios of time of flow to time of peak flow. Also
shown are ratios of volume of flow up to the flood peak to the total volume of
flood flow for the corresponding time ratios.
A start has been made over the last ten years on working out mathematical
modeling methods for calculating the runoff hydrograph. ' ' Underlying most
such methods are a series of assumptions based on the constancy of water yield
and the velocities of confluence of flows over the drainage basin area or on the
constancy of the form of a unit hydrograph.
The most flexible method for allowing for variety in runoff surfaces in
urban areas, producing variable values o£ water yield and velocities of
confluence over the area of the drainage basin, is the isochrone method. In
using this method the basin is divided into a series of component areas having a
constant time of flow to the gaging section (f^, f£, ...., f n ) and similar flow
coefficients. Discharges at the gauging section will be formed by means of
water flowing at various time intervals from the various areas.
156
0.36 0.37 0.38 0.38 0.39 0.40 0.40 0.41 0.41 0.42 0.42 0.43 0.43 0.44
0.68 0.65 0.60 0.54 0.49
2.6
0.44 0.41 0.34 0.28 0.22
0.07 0.05 0.03 0.02 0.01

0.22 0.19 0.14 0.09 0.06
0.14 0.12 0.08 0.05 0.03

0.88 0.87 0.84 0.82 0.79
0.26 0.23 0.17 0.12 0.08

1.00 1.00 1.00 1.00 1.00
0.90 0.89 0.87 0.85 0.82
0.55 0.52 0.46 0.40 0.34
0.05 0.04 0.02 0.01 0
2.4
0.03 0.02 0.01 0

2.2
0.02 0.01 0
2.0
0.01 0
1.9
0.70
0.02
0.30
0.05
1.00
0.07
0.26
0.17
0.09
0.59
0.48
0.01
0.89
0.91
1.8
0
0
0.01
0.01
0.21
0.34
0.02
0.07
0.62
0.73
0.04
0.30
0.12
0.09
0.52
0.90
0.92
1.0
1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7
H
*¿ O CMCM-tfvOrH C O V O V O O O L O L O O I L O C O C M
TIME OF FLOW TO TIMI: OF PEAK FLOW
-^ fa O r H C O v O v 0 > O C f > r v . L O C O C M r H O O O O
fa Q O O . O O . . S O . . O O 0 0 0 O
o r< O O O O O r H O O O O O O O O O O
0.04 0.02 0.01 0.01
8 cM<tcrvcT'CMOcoooOcocrvcrvcMr~.<fcM
COEFFICIENT OF FLOOD FORM,
oo r-I O r H C O v D C n O O - i r v v O - t f C M r H r H O O O
•z fa
. . . . 0 . 0 . . . . . . . . .
M fa O O O O O r H O O O O O O O O O O
H O
O -tfOOCOCMcO -tfO-tfOO-tfcOLOOvO-tf
00 OiH-tfl^CTlOCTlOOvO-tfCOCMrHrHOO
äH o
O O O O O r H O O O O O O O O O O
fa
g
0.02 0.01
vDCMOOLO-tfOLOCOOOCMtnoOOCOCTivD
oo OH
< M o
ocM-tfrâvocnoovOLncocMcMrHOo
o
. o o o o e , , , « o 0 , 0 0 0
tu fa O O O O O r H O O O O O O O O O O
So 0.01
0.04 0.02 0.01
avrv-tfcomOvouorHrv.-tf<tinooroav
HA
O OCMtOr-0>OC^00fv.L0-tfC0CMrHrHO
o
0 , 0 0 0 0 0 . 9 . . . . . . .
Ü a
oo CO
O O O O O r H O O O O O O O . OO O
H 3 cococArHvoOvorv.LncMOo>ocor--co CO CM rH
Q O rHCOLnoOtTiOCT>OOr-»vDincOCOCMrHrH O O
o o o
ARI
O O O O O r H O O O O O O O O O O O o O o
ä
H coo>-tf<tvoOr-o>oor~-vOvOfv.cTicooo o vD CO CM r-1
•«1 > rHcOvOOOCT>OCnoor-»vOLO-tfcOCMCMrH O O O
re Bi o i-H o o
o O O O O O O r H O O O O O O O O O O o O o O o
fa
8
ri w CTi
rH 00 •tf r~» CT> r
O
— f » - O r v , r H C M C M C M C M - t f v O C J v - t f LO O
C M v f v O O O C T i O C T i C J v O O r ^ v O l O - t f C O C M C M rH
vO -tf CM rH
O O
-tf
co
fa y o o o
O O o O O O O O r H O O O O O O O O O O O O o O O o o
fa 3
0.19 0.23 0.26 0.29 0.31 0.33

0.04
0.10
0.07
0.02
0.02
0.15
rH-tf-tf<T>r-»OCOCMLOrv>OOCT\rH-tfrv.rH
0.28 0.19 0.12
0.22
20
o CJ
0.8
coLor~-oocTiOcftcTioor-vOLOLo-tfcoco
00 CO o
M O O O O O r H O O O O O O O O O O
w
3 Q
0.30
0.16
0.04
0.09
CM
0.12
LO OrHO^rHOOOCO-tfOOrH-tfvOCXlCMvOO
0.22
-tfvOrvvTiOvOcr\CJvOOOOi~-vDLOLO-tf-tf
0.7
O O
o O O O O O r H O O O O O O O O O O O
0.15
0.01
0.40
0.25
0.10
0.19
0.32
0.08
CfiCTiCOnoOOcriLOrHLOCTvCOvOOLOCTN -tf rH
0.6
-tfvDCOCT*v7vOCTvCTHCAOOr-»rv.vOvOLO-tf O O
O O O O O r H O O O O O O O O O O O O
LO en <* LO CT. CO rH
3*7
co oo CftLOrv.lOO\OCTlvDCOCrv-tfCTl-tfCTl-tfCT> O
0.5
O r-i CO l f l M ) 3 0 1 C I \ O C M T i 0 v 0 0 C 0 r > r . v 0 v 0 i r i LO CO CM CM r-1 O O O

O O o O O O O O r H O O O O O O O O O O O o o O O O O O O
o\ en rH o > N H v o m O o \ i - . i n M e o i / i H h - c n o \ -tf 00 CM rv. VO Os ty\ -tf rH
VO LO -tf -tf co CM rH O O O
0.4
O CM LO v O 0 0 O \ C A C I \ O O \ O \ C T i C n C 0 0 0 C 0 | v | v v 0
O O O O O O O O r H O O O O O O O O O O O o o O O O O O O O
CM rH LOvO C 0 0 0 - t f t - * C T i O O > 0 0 r v » m c M O r v , < f r H 0 0 CO CM CO CO •tf rH CO 00
0.3
O CM -tf vO rôocncricTiOo^cncncTiCTiCftOOoooor^ r- vO vO LO LO -tf CO CM rH O
O O O O O O O O O r H O O O O O O O O O O o o o o O O O O O O
O.
[ 3.0
! 0.4
3.5
2.4
2.2
2.8
rH CM CO
2.6
O* LOvOrv.COCJvOrHCMcO-tfLOvOrv.oOlT'O O
CU o,
c O O O O O O O O r H r H t H r H r H r H r H r H r H r H C M •tf a
3
157
Flood Frequency
The value of the computed exceedance frequency, P, in computing the basic
characteristics of floodwater varies in accordance with the nature of the object
being studied, allowance being made for consequences which may arise from
rainfall exceeding computed values. For example, in the case of urban areas,
for city blocks and side streets the value of P is taken as being between ten
(3)
and two per cent, and for main streets between ten and one per cent. The
computed annual probability of maximum discharges in the planning of hydrotechnical
installations is taken as being between P = 0.01 per cent and P = 1 per cent,
depending on the category of the installation, and for temporary hydrotechnical
installations between P = 10 per cent and P = 1 per cent.
References
1. Rukovodstvo po opredeleniju rascetnyh gidrologisceskyh harakteristik (Guide
to the determination of design hydrological characteristics). Gidrometeoizdat.
Leningrad, 1973, 111 p.
2. Alekseev G. A. Shema rasceta maksimal'nyh dozdevyh rashodov vody po formule
predel'noj intensivnosti stoka s pomosc'ju krivyh redukcii osadkov i stoka
(System for the computation of maximum rainwater discharges using the formula
of the extreme intensity of runoff with the help of precipitation and runoff
with the help of precipitation and runoff reduction curves). In Tr„ GGI
(Proceedings of the State Hydrology Institute), No. 134, p. 55-71.
3. SNIP (Construction norms and regulations) 11-32-74, Kanalizacija. Naruznye
seti i sooruzenija (Drainage. External networks and installations). Moscow,
1975, p. 6-13.
4. Smirnova E. A. Rascetnye harakteristiki dozdevyh osadkov na territorii SSSR.
Mezdunarodnyj simpozium po pavodkam i ih rascetam (Design rainfall characteristics
of the USSR territory in Floods and their computations), Proceedings of the
Leningrad Symposium, IASH-UNESCO-WHO, 1969, p. 105-114.
158
Sokolovskij, D. L. Recnoi stok (Stream Flow). Gidrometeoizdat, Leningrad,
1969, p. 5-39.
Cebotarev, A. I., Serpik B. I. Vybor i obosnovanie formul dija rasceta
maksimal'nyh rashodov dozdevyh pavodkov (Selection and justification of
formulae for the computation of maximum discharges of rain floodwater). In
Sb. rabot po gidrologii No. 11 (Collected papers on hydrology No. 11).
Gidrometeoizdat, Leningrad, 1973, p. 3-47.
Zimont I. L. and Lobanskaja N. P. Izucenie processov formirovanija dozdevyh
pavodkov na zastroennyh territOjrijah (A Study of the processes of formation
of rainfall floods in built-up areas). In Tr. IV Vsesoiuznogo gidrologiceskogo
s'ezda, t. 3. Rascety stoka (Proceedings of the 4th All-Union Hydrological
Congress, Vol. 3, Runoff computation). Gidrometeoizdat, 1975, p. 445-452.
Kajurov A. V., Siklomanov I. A. Rascety maksimal'nogo dozdevogo stoka na
rekah Primorskogo kraya s ispol'zovaniem nelinejnyh elektronnyh modelirujuscih
ustrojstv (Computations of the maximum rainwater flow on the rivers of the
Maritime Territory). Primorskij Kraj (Using non-linear electronic modeling
devices). In Tr. GGI (Proceedings of the State Hydrology Institute), No. 211,
1973, p. 138-168.
Kucraent L. S. Matematiceskoe modelirovanie recnogo stoka (Mathematical
modeling of streamflow) Leningrad, 1972, p. 191.
159
research in the United Kingdom
M . J. Lowing
Institute of Hydrology
MacLean Building
Crowmarsh Gifibrd
Wallingford, Oxon
O X 1 0 8BB, United Kingdom
Section 1 General introduction
In the United Kingdom as elsewhere, urban hydrology is only now

beginning to receive the attention it deserves. Although the Road Research
Laboratory (RRL) hydrograph method of sewer design is known internationally, it
was until recently the only significant United Kingdom contribution to 'modern*
urban hydrology with regard both to data collection and analysis (Section 2 of
this report) and to the design method itself (Section 3 ) .
Although the total research effort is increasing, it is still small by
comparison, for example, with the U.S.A. However, it is possible that work is
more closely co-ordinated in the United Kingdom and that benefit derives from the
frequent informal meetings organised by the active researchers in the field. The
Department of the Environment/National Water Council Working Party on the Hydraulic
Design of Storm Sewers was formed in 1974 and now acts as a focal point for
research co-ordination and information exchange.^' Previously a national
colloquium^) a t Bristol University had helped to crystallise the growing
dissatisfaction with existing design methods and with the inadequate body of
knowledge relating to urban runoff, quantity and quality. Current research at
central research stations and universities is concerned with the development of
new methods but most results are, as yet, only tentative.
MDst of the research is into rainfall-runoff relations and is described
in this paper. Economic aspects of sewer design and construction are also under
study'-*' but are not described here. There is also a growing body of opinion
which is interested in total planning rather than sub-division planning. Why not
prevent the increased flows rather than design against them? Should we
concentrate on standards of maintenance rather than design? There is clearly a
significant interdependence between hydraulic design, pollution prevention and
maintenance work; and recent re-organisation in the United Kingdom Water Industry
has allowed the broader view to be seen at an early stage in planning. But such
a view is not the aim of this paper, which is limited to the relatively narrow
field of urban hydrology.
Urban hydrology covers a wide range of catchments: the small, totally
impervious parking lot draining to inlets; medium sized areas where the dominant
process is pipe flow; and larger, mainly natural catchments, with a significant
163
proportion of urban development. The latter case, of a catchment being urbanised
but still drained by a natural watercourse, has been studied for several years,
but again the pace of research is accelerating as more data are gathered and more
workers turn to the topic. The Construction Industry Research and Information
Association (CIRIA) has recently initiated a project to provide engineering guidance
on the subject, and it is expected that design criteria derived from the recent
Flood Studies Report^) will provide useful input to the project. This five-volume
Report represents four years work at the Institute of Hydrology, Meteorological
Office, and Hydraulics Research Station. It describes the outcome of an
exhaustive study of floods, their magnitude, timing and frequency. Although
concentrating on natural catchments, the Report provides a platform of experience
and methodology from which to launch further projects in applied hydrology.
As requested by UNESCO, the format of the present paper follows that of
the U.S.A. contribution^) to the UNESCO reports. This paper is much shorter,
not only because of the much lower level of activity, but also because some
material has been omitted where it is adequately described in the first Technical
Memorandum of this special IHF series.
(5)
The preparation of this report at the Institute of Hydrology was
supported by the United Kingdom Department of the Environment under contract
number DGR/480/38.
References. Section 1
1. National Water Council (1976). Working Party on the Hydraulic Design of
Storm Sewers, review of progress to June, 1975. Department of the
Environment/National Water Council Standing Technical Committees Report
No. 1. (From NWC, Queen Annes Gate, London, U.K.)
2. Construction Industry Research and Information Association (1974). 'Rainfall-
runoff and surface water drainage of urban catchments'. Proceedings of
Bristol colloquium, April 1973. (From CIRIA, Storey's Gate, London, U.K.)
3. Local Government Operational Research Unit (1975). 'Economics of sewerage

design'. Report No. C218. (From LGORU, Norman House, Kings Road, Reading,
U.K.)
4. Natural Environment Research Council (1975). 'Flood Studies Report,1 5

volumes. (From Institute of Hydrology, Wallingford, Oxon, U.K.)
5. McPherson, M. B. (1975). 'Urban hydrological modeling and catchment research

in the U.S.A.»reprinted in this volume.
164
Introduction
This section of the report is intended to describe catchment-based
research in urban areas in the United Kingdom. Four broad categories are
distinguished by the primary objective of the research:
(a)» To provide data on rainfall and runoff from urban areas with flow monitored
at some convenient point in or at the outfall from a sewer system,
(b). As in (a), but flow monitored at the entry to the sewer system with the
specific aim of calibrating an above-ground model of runoff.
(c). Water quality studies.
(d). Studies of the effect on flow in natural watercourses as a catchment is
urbanised.
The Map which follows, expanded in detail from an earlier survey, (1)
shows the location of all known catchment studies. The accompanying Table gives
details of the sewered and urbanising catchments (pages 6 and 7).
Before listing the research in the four areas defined above, the next
two subsections deal with methods of flow measurement and rainfall aspects
respectively. Described in the final subsection are methods of data collection
and processing.
Methods of Flow Measurement
Inlets
,3
There have been several studies' ' where a small (< 1-ha) paved area
has been gauged, but in each case a significant length of pipe has taken flow
from the inlet to the gauging point (V-notch weir or flume). However, an instrument
has recently been developed at the Institute of Hydrology(^) which is designed to
fit into a road gully and measure inflow to the pipe system. The meter is basically
a hinged plate in a vertical, square-section tube through which all flow is forced
to pass. Plate rotation is linked to a potentiometer and may be calibrated
against discharge.(5) ^ filtering system and careful design detail minimise the
effects of sediment and trash but a high level of maintenance is necessary.
In-Pipe. by Water Level Measurement
All in-pipe monitoring of water level has been at manhole sections with
conversion to flow based on assumed friction coefficients. Coefficient values
have sometimes been adjusted after calibration, at low flows only, by dilution
gauging. The most usual method of water level measurement has been with an air
reaction system (Arkon) but a sonar system (Echolot) is being tested in Edinburgh
and a water surface following device (Manning Dipper) is under investigation
165
Instrumented urban catchments in the
United Kingdom
GAUGED RIVER CAT
> 2S¥urban development
HANDLE AT COWÖLLYS KILL

EEVfcHLEY BROOK AT bL^lEDCK
nxsaix AT kEicntï;
DOLLIS ¡HCOK AT IÍJÍJON
WS3.T AT M 3 Ö 5 PARK
BHJÍT AT H W X L L
RAVDISBOUKNL AT EIITIÜY
RAVENSBOUTCiE AT M&SaCUBE TT)
(JUACCY AT H'JOR líXüc
H O C AT RECTO» KD
CHAFFINCH BR30K AT ¡EODSttK
SILK STREAM AT COLLI.'.T*EP LA'CE
UPPER LEE AT LUTON ICG
HXUKGTCK STREAH AT ¡OLLD^TC
»ERMITAGE STREAM AT I1AVAHT
TAME AT KATER ORION
TAME AT LEAHARSTON
LEEK AT NorraaiA*
MCKER WO0K AT a/LADUE
MERSEY AT liXAH ttIR
K-KÍIY AT ASHIDN KEÍR
IRK AT SCOTLAND KEIR
ROTTEN CALOER AT K-D I.EES
NORTH CALDER AT CALLEK PAEK
M U T E C A , " AT IIAKHÏL'J)
KATER OF LEITH AT HKKAYFIELD
CRAY AT CRAYFORD
CORBY CENTRAL
CORBY s a m i
ÎUJLOCX AT AROHiaC
STEVENAGE BROOK 6 BRAGEURY END
STALL RIVE?. LEE t ENFIELD
LODGE
SALMON BROOK t ENFIELD
FYMCS BROOK Í IPPER EIMKICK
BEAM RIVER i ELM PAP¡Í
RIVER INCREBOIK.T; 8 HAVERING
CATCHMENTS
RRL SPONSORED 00€ SPONSORED MISCELLANEOUS
CATCHMENTS CATCHMENTS CATCHMENTS
CATCHMENTS UNDERGO»« 1 BLACKPOOL 9 BRAaSELL 50 BIRMlNai'«
URBANISATION 2 niaiousE 10 DERBY 51 NORTWI.T
3 DONCASTER 11 STEVENAGE 52 SOUIUÏÎTCI
1S1 WLTOK KEYNES 4 raeiNcroN 12 BWLIHCTON
nDBROOKE 13 HDEERSFIELD 53 BRISTOL
lil HARLOW »
é LEICESTER 14 LEICESTER 54 CRAMLING1CN
153 amjst 7 OKHBY IS MANCHESTER 55 NOTTINUIVI
1*4 EXETER
UI SKMUCUSE • STEVENAGE 16 SiEFHELD
1S6 H3CASTLE 17 anrnm
11 SMDERLñND
19 nució*
zo COVENTRY
166
in Stevenage. Resistance-type gauges (Eurogauge) and pressure bulb types have
also been used.
The Construction Industry Research and Information Association is
currently sponsoring an evaluation by the Hydraulics Research Station of methods
of flow measurement in rivers, open channels and sewers.
Special Structures, by Water Level Measurement
The more reliable data on sewer flows derive from purpose-built flumes
either within a specially enlarged sewer section or at the outfall from the
system. A good example of the former is at Lodge Farm, Milton Keynes, where a
1.65-m square culvert was heightened to 3.225-m and a trapezoidal flume included
during sewer construction. Water level is normally monitored by float in a
conventionally separate stilling well, and theoretical stage/discharge relationships
are applied.
Dilution Gauging
Most of the U.K. expertise in dilution gauging in sewers was developed
in the early I9601s at the Water Pollution Research Laboratory (now Water Research
Centre, Stevenage Laboratory) and jointly by the City and University of Newcastle.
Lithium chloride was and still is the main tracer employed in sewers. A British
Standard describing the constant-rate-of-injection method has been published.'''
Further development of techniques has taken place at the Water Research
Association^8) (now the Water Research Centre, Medmenham Laboratory) using sodium
dichromate and, more recently, at the Institute of Hydrology using sodium
iodide. In both cases, the emphasis has been on gauging in natural streams.
Further developments in sewer gauging were reported by Blakey.t*)
The practical application of radioisotope dilution gauging techniques

to systems where conduits are flowing full was pioneered by the U.K. Atomic
Energy Authority¿( 38 )
Most gauging has been for calibration of pipe stage/discharge
relationships and has involved manual techniques which have worked well in
wastewater sewers. But in storm sewers it is seldom possible to be on the site
and prepared for gauging during a high discharge, and only rarely can similar
flows be introduced artificially by pumping. An automatic technique, triggered
by a predetermined discharge, is therefore essential to achieve calibration over
a wide range of flowB. Alternatively, automatic gauging can be thought of as
providing direct flow measurement during a storm event. In this context,
TuckerdO) nas described an apparatus using gulp injection. This method requires
a longer sampling time (three minutes in his case) than the constant rate
injection and, because the derived flow figure is an average value over the
167
B co 73
cd j¿ u
u d O
c
cu o o o
eu 73 tj cu i-i eu
E CU 4J X\ 4-1 u
JS d CO U
cd •Ü
CJ i-l CU 7 3 u a eu Ü
73 4-> CU u e co
u O 0) •i •ri 4J eu >, d 3 d x> o >-i eu U
cd C C o ca o 4J 4J CU
o cd x: eu
o o cd •W E CO 1-1 a. a Xt
73 a) 3
4-1 4J co 73 o
ôT "S CU r-l D . ¿£
i-i
cd
a,
o
d cd
eu
cu 00 O cd a cd
cd ca
U 3 r-l
i-l i-i
o xi — B 4J
73
U U 4-1
i-i
4-1 cd U CU eu co
Q. •H 3 x> cr eu D,-ri O 4J Q. C0
E
o
.c .-
u co
cd "
o
i cd
73
ri CO
«73
eu
>
r-l
oo cd
O ca cU o 4J 3
Q. O
73 4J O . S>1 4J 1-1
CJ co ! - H o 4-1 co • *• CM tí 73 CU d o
U 00 CO 4J XI 6 i -l
• H i-l O O 73 oo u
c tí 3 - O CO o d I-l cd 3 eu xi
l4 eu 73 T-t U
•o cd T) CU
7 3 i-l
O JJ tí 00 •« 73 4-1 -ri tí eu 73 4-> 3
•H 13 cd eu 4-1 73
1-1 00
m ^>i 4J
Xi
O o
73 E c o cd
cd
a . 4J xi -
.173
cd
d
cd
•ri 7 3
3
73 tí cd
• " tí
•» c 4J o cd js i-l eu co0 4J
cu
c 3 0 0 tí
« CU
o o co cd 0 0 0 i-l
3 cd o o
u
«S 00 73 cd
u
Ol M 4J
4J
i eu tí co tí i-i § cd
OO S CO CO
CU
•H E
cd
4-1 4-1
u
U-l ü
cd
i-l cd
1-1 4-1 eu oo co
•H
73
a , i-i o
§"5 U
cu o
tí
00 cu
4-1 1-4
f l CU
äu
73 cd co 00 d 4j • H «d CO 7 3 cd
-3
i-I O co C CU o E eu cd i-l o ri J3
x: i-i E i-l
CS U 4J 4-1 4J tí U
3
CO S
3 JS
73
¿tí CU
àea 3 oo cd 3
j! i» «
C4-I
73 4J
O 3
tí
O ri
3 CU
eu
O
eu x>
cr . « O O 4-1 4-1
gê
- a CU O cw
iH CO
r-l 1-4 IH 73 1-1
11 1-1 Cd 1-1 CM
0 0 i-I O 1-1 Xi O 3 r-l
n) d E r-4 K eu 3 73 4J
• a • " ai - H 3 W o cd
o cd 3 S X u
eu eu eu eu
O 73
OÙ
3 8 § 6
E
3
r-l
•H d
CU
E
3
i-H 73
3 3 3 3 14-1
* I-l I-l
u-i # *
•
r-l i-H
eu
IM
CU U-l
cu cu 14-1 «4-1 eu o Ü eu o o
M > M
eu eu
¿t
3
00
cd
+
Ioo
00
d
•H
S*
cd
00
1-1
eu
3
3 3
00 00
cd
3
00
tí +
cd
s
3
cd
+
MW
+
cd
c e i-l 0 -
c d i-l 00
J2 d
X\ 1-1 73
U -H 4J ri
•ri .ri 73 x\ .d x: c x¡ x\
4J 4J 73 d 4J 73 O -H 73 73
tí 4J 4-1 u 4J 4J
a a. a cd C 3 tí 4)
CU Cd 3 &. CL a •H CU O.
cu cd 4-1 3 4J 4J
cd CU eu CU
I 4J CU 73
73 cu Cl) > co
1 » CO C0 4-1 73 73 73
tí
73 cu
CO
73 > - 73
CO o 00 Cn 00 00 en en en en ô eu eu CU
m m mcd m m
in m m m m 4J 4J
4-1 ri ri ri ri
en vD
en en en en en en mm m cd cd cd cd 4-> en en en en
en cd
en en en Q O Ö
a o
ri ri \0 r-i oo «a- e n r l H
m <t i-l
m en CO o- en co
m m io m m m m m m ri ri ri riri ri ri r—
IO en
en en en cri en en en en en en en IT» en en
en en en
en
CM CM o m co vo o o co r-i o m -* m o m vi- m
CN n en vo en vo CM -í co CM <f C-. o« m CO m •-i
co v£>
CM n en CO
co <* •* CO 00 O 1-1 CO CO V0
• •
• O
• ri
• • • CM CM
• •
-4* CO
o r-l < f CO r-l O r-l O m o CO
VO •H O CO iH i-l CM r-l
CM I-l r-l CO
m
cd j3 cd xi
d tí cu
eu u o u cu 0 •H 1-1
o co 4J cd x¡ oo > cu 1-1
cu cu
o 00
o co x¡ CU o x> > I-l 73 ü
«d 1-1 73 ü J3 > cd u CU u 73 1-1
u tí i-l a) u CU 4J cd 3 CU
«s I
CU 4J
CO
a co o a rJ
en
co <t" m *40 ri co co -*
168
¡
atchment
nt data
t over
oped
eir
tí
o 1 >, CJ
•w CJ Xl o rH c 4-1 rH XI 3
4J M E tí 3 tí CJ O CJ CJ
3 cd to X) O Ë rH > o. U-l
i-l
•H
3 ts
4-1 ?•, co
CJ
4-1 XI 10 CO
X C.
XI
CJ
rH
o
00
O
00 cd r-l V—' •H CJ 4J 4J
o
4-1
E CJ tí 00
•u tí rH 4J tí tí tí td
o
CJ ft •H tí
rH •W
o cd tí co xi
B
•H tí CJ CJ u • •
>
0 3 •H
3 4J o 4J CI .H cd E > >. 4-1 -o tí 4-1
CO
E_t
U-l
0]
rH
1-1
c 2 .Ci tí tí
• H 1-1 • ft
rH
C
x;
u
CJ
o
tí rH
rH
3
O #1
•H
4-1
rH
•H
2Î ca ta m •> o O XI XJ 4-1 3 cd XI •-XI tí CO
3
§
CJ
u rH
ca
3
•a
o cd
4J n. E
o
O
tí
tí
O
cd
u
CJ
IH
4J
c
3
4-1 XI
cd
CS
O
CJ
g O
CJ rH 0 a. u GO O •H •H cj « I-l tí o 5
4-J 4-1
S 3 U cd •i-I
o £2 4J XI 10 cd E U •H 3 CU 00
o
CJ ca o XI r-l - •H ••% "i cd cd 3 o. n) > O tí 3 CJ
tí o « 3 CJ CO u •—. O I-I o O Ë O CJ
u X. •H CJ 3
3 Ou ca •a > tí O co (J XI u 3 O rH ï* 4J co XI tí XI
cu -d CJ i-I 4J 4-1 •H 00 u CI tí M 3 rH
•M M 4J CJ • n •H tí rH U-l •H U-l > 3 CJ •H XI cd
• o co • ft
o o
-Ö 60 •H C TI i-i tí ÇU a> cd o 4J 0 CU E •H 1-4 3 OJ 4J
CJ C U) O CJ 4J tí >> CJ tí CO •a uj D. 3 o X 4-1 cd
tí - H
• W 00
Ti
OJ tí
tí
•H
o cd
>-,Xi
g lCJ CU
rH Cd XI
XI
4J
O
o
TI
I-I t-l 4J
O
rH
rH
CJ
XI
rH 01
CJ
o
X)
XI 3 CJ cd u3 U 4J tí 4J r—• 00 O o x 0J M o co ••o u
O
E cd
00
>-i
J3
ja
cd
"iO 4J
o -
3
O
cd
ü
cd
U
cd
XI
CJ
•H
tí
CJ
o
3
CJ
CJ
" - ) CJ
c
>
OJ Ä
cd 0 3
O
o
1-1
o
o
CJ 4-1 CJ c U-l U-l rH 4-1 u
% X) s x: o. a,
CJ O
rH
E
3 CJ
E
3 Cd
3 rH 00 Ë rH I-)
O H U-l tí 3 U-l 4-J
•J ñ # * * *
Pu W • • • • •H rH tí
U O CJ O CI CJ 00 U-l CJ O I-J
s O M •H
9 h â
o ai u-i
•
m
•
U-l m
• u
•H
CJ
E
3
rH
>
cd
3
cd
00
rH
t«
>
cd 1-1
CJ
CJ
p cd
• •
ed
u
•H cd
• •
cd «4-1
3 T>
3 x;
4J
3 u
•H
a co 3 3
O -a¡ CJ 00 tí •H 00 •H CJ CJ
3 + + r* •H tí o 0 tí 3 x; OJ 3
+ +
o
o
gl .tí
ca
cu X. a.
Ë
x:
4J
CJ x:
4-1
TJ
4J
O
M
3
4-1
•H
XI
tí
•H
4-1
3
N
CJ
Cu
•H
TI
tí
CU ü x;
4-1 E
O
u
4-J
O
>1
4J
a,
Ë
§ u a3
u B
3 O. tí tí cd rH cd cd 3
E O 3 tí id 3
a, o. a.
o
"0
i-I
U-l
CJ
73 Ü
B CJ
XI
rH
U-l
CJ
XI >
1
>
CJ 4-1
CO
•H
XI
r-l
4-1
4J
co
rH
•4-J
tí rH
U-l >
1 rH
U-l
M
O
w
•H m LO m U"> m LO CJ <f CJ CJ <]• CJ CJ CO CJ CJ OJ OJ
r- f» t'- r~- r-. f- 4-1 fi- 4-1 4J r» 4J 4J r-» 4-1 4-1 4J 4-1
CO o
H o Cf\ en en cn cr\ cd en cd cd <j\ cd cd CT> cd cd cd id
rH rH r-l rH rH rH a rH Q Q rH Q Q rH Q Q Q
§ a
o S rH CI -tf U-l rH CM m CM <f vD <t rH O -* CM CO u-| CM
u q r- f» n r- —
r l'- l'- —
r —
r u-i r» r- u-i u^ u-l vO r» r~-
ai ai CJ> O* Oí en CTi en en CTi en en o\ CT« Oí CTl o\ o\ (T> CT\
Pu r-l •H r-l rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH
CO
H O
m co o CM O O O CM rH u-\ í-l u-l m O CM rH
ig
MP ER
PERC
n m CM <r €>• U-l O

<r O» f> CO CM rH rH CM
rH
H
CM
^J A in CO
B e
H W cd O r«.•
rH
o
00
o
CM
o
rH
m
-d-
O O O o
rH
CM
vO
CM
O
O
O
O
O
O O
CM
o O
CM
O S »tí O m CM rH u-l
o o
CM CO rH rH CM CM CO
H 5 rH CM •0- CM
V)
CI
tí »~\ .—N
55 U XI E o C!l XI OJ
H a)
4J
TJ tí cd
4-1
4-1 el Cl v-' \s LO
3
CJ
rH
r-l cd 3 X! X.
S H 0) 0) tí rH O fr 00 rH i- rH 00 u. >. o 4-1
Í2 S 0) 1-1 ..
o 1-1 rH 4J tí O cd O tí tí 3 O U Xl ca
Si r*
"Õ
u-i
U-l
•o
tí
CJ
TI
CO
E
tí
O
•H
Ë
XI
4J
X!
4J
4-1
10
•H
4J
o
+J
O
rH
rH
3
a
4-1
CJ
tí
cd
o
3 cq
CJ
x:
•H
3
tí
3
rH
•A
>
o •H
U
O
3
O
•H
t-l
4J
O
rH U
cd
cd
l-i
o
X
o
4-J
3
OJ
* w co CO 3 o pq
z C/3 M Z 'a PC u w CO Z
• § «i
H w a m «43 r- 00 en o O rH CM co >* rH CM CO <r in \D
t-t r-l r-l rH rH CM U1 u-l m u-i m Ul U1 u-> u-i u-i u-i
8Eë
rH rH rH rH rH rH
äSaas
169
sampling period, it is likely that constant rate injection and 'grab' sampling
is more appropriate for calibration requirements. Development work is continuing
at the Water Research Centre (Stevenage Laboratory) and the Institute of
Hydrology, with the aim of producing a simple, reliable, and portable package for
automatic chemical injection and downstream sampling.
Other Methods of Flow Measurement
Both ultrasonic and electromagnetic methods of flow measurement have
been applied to natural watercourses (11) and to pipes that are completely filled.
Commercial instruments of either type are available. The difficulty of using
either method to gauge flow in conduits flowing partly full has so far prevented
their exploitation in sewers, although they could be used where there is an
inverted siphon.
Rainfall Aspects
The requirements of urban hydrology are such that recording raingauges
are needed for determination of rainfall in time periods as short as one minute,
A recently published catalogue^ iZ ' of such gauges shows that there are about 1200
of them in the U.K. and that they are predominantly in urban areas. Nevertheless,
it is unusual for an existing gauge to be in a suitable location when an urban
catchment study is begun, and new gauges are usually deployed. The Dines tilting
siphon gauge is the commonest instrument but increasing use is being made of
tipping bucket gauges connected directly to magnetic tape data loggers.
In some studies, the recording raingauge is the sole source of rain data
but in others it is used only to distribute in time an average catchment rainfall
derived from a larger number of daily gauges. In some cases, gauges are installed
with the orifice at ground level in a surrounding anti-splash configuration.
Despite the greater accuracy of data from a ground level installation, it is
debateable whether they should be used to produce rainfall-runoff relationships
for subsequent design applications using rainfall statistics based on the regular
network.
Studies of storm movement require dense networks of gauges and/or a

radar. The South London/Surrey area is well endowed with gauges and a number of
these have had their entire period of record digitised to form an archive of data
from which to extract the statistics of storm velocity and direction. This work
forms part of current research at the Meteorological 0ffice(13) where methods of
analysis are based on the work of Marshall^1*) at the University of Bristol. The
use of radar.to study storm movement in urban areas is being actively pursued at
Birmingham University.( 15 ^
The possibility of urban development inducing convective rainfall is
under study at Queen Mary College, London.(1&)
170
Gauged Urban Catchments
The most significant contribution to urban hydrology in the U.K. has
been the programme of data collection initiated by the Road Research Laboratory
in the early 1950's.
(2) Twelve catchments (see Map above) of various sizes
and types were gauged for up to five years to provide the data upon which the
Laboratory's hydrograph method of sewer design was based. (The RSL Model is
described in Section 3 ) . Some of the data have now been digitised at the
Institute of Hydrology and are available for testing with alternative models.
A more recent exercise was commissioned by the Department of the
Environment. A number of local authorities collected rainfall and runoff data
for a few years starting in 1972. Unfortunately, when the scheme was later
scrutinised by hydrologists, most of the catchments were found to suffer in one
way or another from problems of calibration, siting, or supervision. Subsequently,
the emphasis was changed to study a few catchments in more detail and with
improved instrumentation and procedures. The Hydraulics Research Station and the
Institute of Hydrology are now responsible for data processing and general
supervision on sewer catchments in three towns (Derby, Bracknell, Stevenage);
and thirteen catchments in nine towns were dropped from the scheme (Nos. 12-20
inclusive on the Map, page 4 ) . Other data have been collected by researchers in
Southampton,^) Birmingham, (17) and Nottingham. (10)
All these catchments are
gauged in or at the outfall from a pipe sewer system.
Other catchments, equally 'urban' in character, are gauged in an open
watercourse receiving discharge from several sewer systems. Most of those which
are more than 60% developed are in the Greater London area. All gauged catchments
with more than 257. development are shown on the Map.above.
Small Sub-Catchments Gauged at Inlet Point to Pipe System
It is generally recognised that a further advance in knowledge of urban
runoff processes requires that the above-ground and below-ground phases be
considered separately.(1°) As stated previously, some small areas have been
studied but the gauging has always been made at the downstream end of a short
length of pipe. In order to gauge exactly at the phase boundary, the Institute
of Hydrology is installing gully meters (see "Inlets," above) at eleven sites
within three catchments that are already gauged downstream (Bracknell, Stevenage,
Southampton). It is intended to collect data throughout the summers of 1976 and
1977. Additional sites, with different slopes and surfaces, might be gauged in
later years. In the meantime, it is planned to commission a complementary study
on the laboratory catchment at Imperial College, London,i*-?' where the effects of
catchment and storm characteristics can be studied under controlled conditions.
From 1960 to 1965 the Road Research Laboratory turned its attention to
171
motorway drainage and instrumented eight sites on three motorways. Runoff was
monitored by flumes in the outfall pipe to each length of road. Imperial College
0
undertook analysis of the d a t a . ^ '
Water Quality Studies
Some catchment studies are primarily directed to problems of water quality
but they invariably require the same effort applied to the measurement of quantity
as do rainfall-runoff studies»
The first published account of a storm runoff quality investigation
based on systematic analysis of recorded data seems to be the work of Wilkinson^ -^
who took advantage of the gauged catchment at Oxhey (No. 7 in the Map) set up by
the Road Research Laboratory in 1953. He found that first flushes were not much
more polluting than subsequent flows except after long dry periods.
Hedley and King,(22) investigating a combined drainage system in the
Haunch Valley, Birmingham, emphasised the importance of the highly polluted initial
runoff and discussed means of providing temporary storage. They also drew
attention to the need to identify the separate sources of pollution and their
relative significance. Their techniques of data collection were described
separately „("-J More recently, \¿^J runoff from an urban motorway has been gauged
and analysed. Motorway runoff quality was studied at Lancaster University for
several years prior to December 1974. Three sites on the M6 motorway were
monitored with particular emphasis on salt dispersal, and Patrie
k (36)
reported on
the nature of sediment collected in settling tanks elsewhere on the same motorway.
A further extensive study of motorway runoff quality is under way at Imperial College
using new sites on the ML motorway. This study is partly sponsored by the Transport
and Road Research Laboratory, which has a particular interest in the deposition of
oil on the road surface and its ultimate fate.
Tucker'2-*' used dilution gauging for flow measurement at Nottingham and
analysed the samples to give not only the discharge hydrograph but also the
concentration-time curve of polluting constituents. He also emphasised the
importance of 'first flush' pollution.
The changes in sediment production resulting from building activity
have been studied on a 0.26-km2 catchment near Exeter.'2°,27) This i s clearly a
highly significant source of sediment; and some of the large sewers in Birmingham,
for example, have up to a 0.5-m depth of building debris in the invert.
The Water Research Centre is working on the Shephall catchment in
Stevenage (also being 6tudied by Institute of Hydrology and Hydraulics Research
Station teams) and are collecting data with the aim of devising and calibrating a
mathematical model of the storm-water pollution process. They are looking at the
ways in which pollutants accumulate on and are freed from urban surfaces. Also,
172
they are studying the effects of intermittent discharges of polluted surface
water on water quality and the contamination of sediments and biota in receiving
streams.
A team at Middlesex Polytechnic have begun a programme^2"' of data
collection, chemical analysis and urban land use surveying related to storm-water
pollution in the catchments of the River Brent and Silk Stream in North London.
Here they will be installing some new instrumentation to complement river gauging
stations operated by the Greater London Council, (Nos. 105, 106, 112 in the Map,
above).
Catchments Undergoing Urbanisation
This subsection describes active or recent work involving the collection
and/or analysis of data with the aim of learning more about the hydrological
consequences of urbanisation.
The Gloucester Joint Surface Water Study^ 2 "' was established in 1969 with
the aim of assessing the magnitude-frequency relationships of flows in North
Gloucestershire rivers and how these relationships would be affected by the
anticipated development of the region. A network of gauging stations and
autographic rainfall recorders was established.
Gregory^JU-' has analysed over 600 storm events occurring in four years,
during which the subject small catchment near Exeter was progressively built over.
The study continues.w-W
At Milton Keynes, the Institute of Hydrology is studying two catchments:
one is to remain rural for several years and the other is to be completely developed.
Rainfall and runoff gaugings started in 1972 and, on the catchment being urbanised,
flow is presently being gauged both in the original natural stream and in the
parallel main storm sewer (see "Special Structures," above),
n
The Scottish Development Department is gauging the 26-knr Calder Water

(No. 155 in the Map) with the aim of studying the effects of a projected new town
on runoff distribution and water quality. The University of Newcastle and
Northumbrian Water Authority are studying runoff from a developing industrial
estate at Cramlington.
The above studies have each been designed specifically for the purpose
in hand, but other workers have used data from established gauging stations and
drawn conclusions about the effects of urbanisation on high flows, low flows, and
hydrograph shape.
Hollis has made an extensive study of the flow regime of the 21-km2 Canons
Brook catchment which changed from a rural state in 1951 to a fully developed state
in 1970. He has s h o w n ^ 2 ) that while the frequency of winter flood peaks has
remained unchanged, the frequency of summer floods has increased markedly.
173
Urbanisation of this clay catchment does not seen to have altered the magnitude-
frequency relationship at return periods greater than 20 years. Hollis has
subsequently(33) reported on the effects on the low flow regime.
Hall'3^' studied the change in shape of the derived unit hydrographs
for both urban and rural catchment areas in the headwaters of the River Mole
near Crawley (No. 153 in the Map). Whilst confirming the expected trend for lag
times to decrease with urbanisation, he concluded that a simple measure of
•percentage impervious1 was inadequate and that changes in the channel system
and the distribution of impervious surface within the catchment should be
considered. Packman^37' followed Hall's techniques in analysing unit hydrographs
in two urbanising catchments in North London. He found that non-linear effects
tended to obscure the changes in lag time due to urbanisation. While the
expected reduction in lag time could be observed in the initial stages of
urbanisation, including the establishment of the basic sewer system, later
infilling had little extra effect.
Data Collection and Processing Systems
r
Compared with the U.S.A., the expenditure on gauging of urban runoff in

the U.K. has been small. There have been so few data that sophisticated methods
of handling and processing have not been required. Records have been obtained on
charts and extracted manually. Only recently has the use of analogue-to-digital
conversion equipment been justified, and magnetic tape logging systems are being
used in urban hydrology for the first time this year (1976)0
As part of the data gathering exercise commissioned by the Department
of the Environment (see "Gauged Urban Catchments," above), Coventry City developed
a system^-3-''' of processing and archiving data using a digitiser. Rainfall
increments, water levels, or flows from any number of recorders, can be extracted
for any specified time interval and stored on magnetic tape for retrieval in
various formats for different purposes. One of the main users of the system has
been the Gloucester Joint Surface Water Study (see the preceding subsection). A
similar system is in use at the Institute of Hydrology and has recently been
extended to cope with data received from magnetic tape data loggers.(4) xh e
Microdata logging system is used, which can record analogue voltage inputs on
up to 12 channels in digital form on compact cassettes. Input may be from a
raingauge, a gully meter (see "Inlets," above), or a water level recorder, either
float type or Manning dipper (see "In-Pipe," above).
References. Section 2*
1. INSTITUTE OF HYDROLOGY (1975). Survey of Instrumented Urban Catchments.
Unpublished-report - also document 75/25 of DOE/NWC Working Party on
Hydraulic Design of Storm Sewers.
*N0TE: Addresses for unpublished report acquisition are given in the following
subsection.
174
WATKINS, L. Ho (1962). The design of urban sewer systems. Road Research
Laboratory Technical Paper 55, HMSO.
KLDD, C. H. R. (1976). A non-linear urban runoff model. Institute of

Hydrology Report, No. 31.
BLYTH, K. and KIDD, C. H. R. (1976). The development of a meter for the

measurement of discharge through a road gully. Institute of Hydrology.
WOOLDRIDGE, G.(1975). Calibration of a flow metering device for road

drainage gullies. Hydraulics Research Station Report, INT 144.
DEPT. OF CIVIL ENGINEERING, UNIVERSITY OF NEWCASTLE (1964).. Dilution

Techniques for flow measurement. Bulletin No. 31, February.
BRITISH STANDARDS INSTITUTION (1964). Methods of measurement of liquid

flow in open channels. Part 2, Dilution methods. 2A, constant rate-injection.
H.M. Stationery Office.
WATER RESEARCH ASSOCIATION (1970). River flow measurement by dilution

gauging. Technical Paper 74.
BLAKEY, A. W. (1969). Flow measurement in sewers using a constant rate of

injection dilution technique. Journal of Institution of Municipal Engineers,
96, 44-52.
TUCKER, C. G. J. (1974). Stormwater pollution - sampling and measurement.

Journal Institution of Municipal Engineers, 101, 269-273.
WATER RESEARCH CENTRE (1975). Proceedings of Reading Symposium on River

gauging by ultrasonic and electronic methods. December 1974.
INSTITUTE OF HYDROLOGY (1976). Recording raingauge catalogue of the U.K.
SHEARMAN, R. J. (1976). Second progress report on storm movement project.

Informal and unpublished paper presented to DOE/NWC Working Party on
Hydraulic Design of Storm Sewers, Document 76/6.
MARSHALL, R. J. (1975). A stochastic model to simulate moving storms.

Symposium on precipitation analysis for hydrologie modeling at Davis,
California. June. (In Press).
KELWAY, P. S., WARNER, G. and TAYLOR, B. C. (1974). Rainfall analysis and

radar prediction. Proceedings Instn. of Civil Engineers, Informal Discussion,
Part 2, June,
ATKINSON, B. W, (1971). The effect of an urban area on the precipitation from

a moving thunderstorm. Journal of Applied Meteorology. 10.
HINDI, W. N. A. (1976). A computational method for the optimal design and

analysis of urban sewer systems. Unpublished Ph.D. thesis, University of
Birmingham.
NEWTON, S. G. and PAINTER, R. B. (1974). A mathematical examination of urban

runoff prediction. Proceedings Instn. of Civil Engineers, Part 2, 57, 143-157.
HALL, M. J, (1970). Laboratory catchment studies of Imperial College.

'Spanner', magazine of the City and Guilds Union, Imperial College, London,
73-77.
175
20. SWINNERTON, C. J., HALL, M. J. and O'DONNELL, T. (1973). Conceptual model
design for motorway stonnwater drainage. Civil Engineering and Public Works
Review, 68, 799, 123-129, 132.
21. WILKINSON, R. (1956). Quality of rainfall runoff water from a housing estate.
Inst. Pub. Health Eng.. Vol. 55, Part 1, 70-84.
22. HEDLEY, G. and KING, M. V. (1971). Suggested correlation between storm

sewage characteristics and storm overflow performance. Proceedings Instn.
of Civil Engineers, 48, 399.
23. GALLIERS, R. and KING, M. V. (1970). Developments in sewer monitoring

equipments and techniques. Institution of Municipal Engineers, January.
24. HEDLEY, G., and LOCKLEY, J. C. (1975). Quality of Water discharged from an
urban motorway. Journal of Water Pollution Control, 74, 6, 659.
25. TUCKER, C. G. J. (1975). Polluting loads in runoff from urban areas.

Unpublished Ph.D. thesis, Trent Polytechnic Department of Building and Civil
Engineering.
26. WALLING, D. E. and GREGORY, K. J. (1970). Measurement of the effects of

building construction on drainage basin dynamics. Journ. of Hydrology, Vol.
II, 129-144.
27. WALLING, D. E. (1974). Suspended sediment production and building activity

in a small British basin. IAHS Publication 133, 137-144.
28. ELLIS, J. B. (1975). Urban stonnwater pollution - Research Report 1, Middlesex

Polytechnic.
29. SHAW, T. L. and WALLER, R. S. (1973). The Gloucester Joint Surface Water
Study, Journ. Inst. Municipal Engineers, 100, 120-123.
30. GREGORY, K. J. (1974). Streamflow and building activity. Inst. British

Geographers. Special Pub. No. 6, 107-122.
31. WALLING, D. E. (1976). Modification of hydrological processes consequent

upon suburbanisation. To be presented to IGU/IHP Symposium on man made
transformations of the water balance. Leningrad, July 1976.
32. HOLLIS, G. E. (1974). The effect of urbanisation on floods in the Canons

Brook, Harlow, Essex. Inst. British Geographers. Special Publication No. 6,
123-138.
33. HOLLIS, G. E. (1976). Water yield changes after the urbanisation of the Canons
Brook catchment, Harlow, England. To be published in IAHS Bulletin.
34. HALL, M. J. (1973). Synthetic unit hydrograph technique for the design of
flood alleviation works in urban areas. International symposium on design
of water resources projects with inadequate data. Madrid. Proc. Vol. 1,
145-161.
35. CITY OF COVENTRY (1974). Stonnwater research - computer system manual. City
Treasurers Department.
36. PATRICK, C. (1976). Nature and significance of sediment derived from motorways.
Proceedings 9th International Conference of Sedimentology, Nice, July 1975.
Int. Assoc. Sedimentologists, in press.
176
37. PACKMAN, J. C. (1974). The application of unit hydrograph theory in
catchments subject to urbanisation. Unpublished M.Sc thesis, Imperial
College, University of London.
38. CLAYTON, C. G. and EVANS, G. V. (1968). The constant rate injection and
velocity methods of flow measurement for testing hydraulic machines.
United Kingdom Atomic Energy Authority Research Group (AERE - R 5872).
Addresses for Section 2 References
Institute of Hydrology (1, 3, 4, 12)

Crowmarsh Gifford
Wallingford
Oxon, U.K.
Transport and Road Research Laboratory (2, 38)

Crowthorne
Berks, U.K.
Hydraulics Research Station (5)
Howbery Park
Wallingford
Oxon, U.K.
University of Newcastle (6)

Claremont Road
Newcastle upon Tyne NE1 7RU
U.K.
Her Majestys Stationery Office (7)
York House
Kingsway
London WC2, U.K.
P. J. Colyer, Esq. (13)
Secretary, DOE/NWC Working Party
on Hydraulic Design of Storm Sewers
at Hydraulic Research Station
Howbery Park
Wallingford
Oxon, U.K.
Water Research Centre (8, 11)
Medmenham
Bucks, U.K.
University of Birmingham (17)
P.O. Box 363
Birmingham B15 2TT, U.K.
Imperial College (19, 37)
South Kensington
London, U.K.
Trent Polytechnic (25)

Burton Street
Nottingham NG1 4BU
U.K.
177
Middlesex Polytechnic (28)
lhe Burroughs
Hendon
London, U.K.
Atomic Energy Research Establishment (38)

Harwell
Didcot
Oxon, U.K.
178
Introduction
This section contains a brief account of urban hydrolgical modeling in

the U.K. Answers to a recent questionnaire^•'••' revealed that the overwhelming
majority of sewer designers used either the rational or TRRL (Transport and
Road Research Laboratory) methods or both. The TRRL method (formerly called the
RRL method) was developed by the Road Research Laboratory (now TRRL) and is the
best known British contribution to post-war urban hydrology. A brief history of
the evolution of the TRRL method is given in the next subsection. Despite its
U.K. origins, all but one of the several independent and published studies of
comparative evaluation have been performed in other countries. Studies in the
U.S.A. have been listed,^) and these and others are also described in a recent
literature review^) produced at the Hydraulics Research Station, which also
reproduces published data on the performance of the various methods.
Despite the lack of formal evaluation in the U.K., there has been a
fair amount of published criticism founded on theoretical objections to the basis
of the model. This has prompted the development of alternative design models, but
the RRL method's simplicity and orientation to design have been recognised as
important attributes and any new developments will inevitably be measured against
it. The newer models are described in a later subsection.
As noted in Section 2, water quality modeling is being actively pursued
in the U.K. but has not yet been included in a hydrological model.
Also noted in Section 2, the hydrological effects of urbanisation have
been studied in the U.K. Most of the work has concerned single catchments and,
although unit hydrographs and regression equations have been used to illustrate
trends, the studies have not been concerned with model calibration. In one case,
however, positive recommendations are made for application to an urbanised and
ungauged catchment and this is given a brief mention in a later subsection.
The TRRL Method
The TRRL method of sewer design can be categorized as a design/analysis
model in the sense used in the first of the special IHP Technical Memoranda
series.' ' It was developed at the Road Research Laboratory using data collected
from 1952 to 1960 (see Section 2) and was described in detail in 1962,(4) W ith a
scientific paper^) and engineers' guidei") in 1963.
These early descriptions of the model are recognised by hydrologists as
being similar to the Clark-Johnstone method of unit hydrograph synthesis, except
that the area-time curve is routed through a non-linear storage rather than a
179
linear storage. Later, the concept of routing through a single storage was
(-] a)
replaced^ » ' by a method whereby a separate routing was applied at every pipe
length. By thus emphasising that the storage allowance was linked to storage in
the pipes, the model's designers left themselves open to criticism regarding the
need for storage routing at all. It is argued^ 9 » 10 » 11 ) that the pipe storage is
implicitly allowed for in the area-time diagram construction and need not be
considered again. The reply to this is that, whatever the cause, there is
clearly an effect of reservoir-type storage and the TRRL model matches this
effect in practice. But several workers have queried the philosophy of 'two
wrongs make a right', not only in the matter of storage but also with regard to
percentage runoff assumptions (i.e., 100% from paved surfaces, 07. from the rest),
the assumed relationship between rainfall and runoff frequencies, and the neglect
of surcharging.
These criticisms raise the general point of whether or not it is
necessary to understand the hydrological processes at work in sewered catchments.
It is likely that the majority of engineers are satisfied with an empirical
method, such as the TRRL method, despite the general awareness that, when used
with the recommended design rainstorm, it over-estimates the discharge for the
specified return period.( 12 ) In recognising that it is the specification of a
design rainfall input which is at least as important as the rainfall-runoff model
itself, the Transport and Road Research Laboratory is in the process of issuing a
revised guide(13) on the use of the method. This incorporates some of the new
information on rainfall statistics and design profiles which has become available
following publication of the Flood Studies Report.( 1¿ ^
Using the same assumptions as for the TRRL method, King^"' substituted
simple functional relationships for the design storm profile, area-time diagram,
and flow-retention relationship. He was thus able to give a functional form for
the design hydrograph and to produce a set of design curves for quick evaluation
of peak discharge.
Other MethodB Developed in the U.K.
Sarginson^^) has focussed attention on the problem of identifying the
source of the storage which the TRRL method models so convincingly, and he proposed
that the above-ground and below-ground storages be treated as two separate linear
reservoirs in series (compared with the TRRL method, which represents them
together by a linear channel and non-linear reservoir in series). He has
subsequently suggested'*"' that the assumption of 100% runoff from impervious
surfaces with no contribution from pervious areas is an over-simplification, and
that depression storage and infiltration should be more realistically modeled.
Further work(l') has led to a prediction equation for percentage runoff in terms
180
of catchment slope; and the two conceptual reservoirs, originally unequal, are now
taken to be equal.
Workers at Birmingham University have applied automatic optimisation
techniques^18) i n refining the TRRL method by consideration of above-ground storage
and by an iterative approach to find a 'true' mean pipe velocity rather than the
use of a pipe-full velocity.'19'
Kidd(20) has further emphasised the importance of separating the above-
ground and below-ground phases of runoff. In his model, depression storage is
deducted and a loss rate is applied to the total rainfall to give the rainfall
excess. This is then routed to the sewer inlet through a single reservoir. Linear
and non-linear reservoirs are compared and the superiority of the latter is
clearly demonstrated.
A single linear reservoir is used by the Transport and Road Research
Laboratory^21' to modify the rainfall excess hyetograph generated on pervious
surfaces only, when applying their method in areas of tropical rainfall.
Models of motorway drainage may be considered as special cases of the
above-ground runoff model. With their regular geometry and single surface type,
motorways are more susceptible to deterministic modeling of physical processes
than to the lumped parameter approach used in the models described above. It is
interesting therefore that, even in this situation, workers at Imperial College
preferred to base their proposed method firmly on empiricism. They developed a
22
dimensionless hydrograph design method,V. ) for single peaked hydrographs only,
with scaling factors determined from prediction equations whose coefficients had
been determined by multiple linear regression on catchment and storm characteristics.
They also describ
e(23)
a conceptual model using a single linear reservoir for the
impervious areas only. The storage constant takes one value if it is raining and
another if it is not raining; furthermore, the values depend on storm rainfall
and site characteristics so the model is more accurately described as quasi-linear.
Current work at the Hydraulics Research Station and Institute of
Hydrology is designed to produce a recommended set of design methods for use in
different situations. The rational method and TRRL method will probably continue
to be the most suitable in preliminary design of small and large schemes,
respectively, but for high accuracy and large schemes in particular, it is likely
that a new design method incorporating economic design criteria will be proposed.
The Hydraulics Research Station is studying various methods of pipe flow routing
and ways of allowing for surcharging. The Institute of Hydrology is concentrating
on the above-ground phase, extending the work of Kidd,^ 20 ' seeking prediction
equations for runoff volume taking account of depression storage and unpaved
areas.
(24)
181
Progress in hydrological modeling inevitably appears to involve more
complicated procedures for the designer to implement and more information to be
gathered. It is vital for the researcher to be aware of this and to ensure that
recommended improvements are truly beneficial. For example, the present use of
TREL is probabilistically unsound and too simple in terms of scientific hydrology.
But unless a new method can be shown to give more accurately sized pipes and less
costly protection against surface flooding, no amount of technical elegance will
persuade the engineering profession to adopt it. It is this reluctance to accept
anything which appears more complicated than is considered necessary that is
sometimes responsible for recommendations that we return to simpler techniques.
The most notable of recent calls to that effect was by Hepworth,'25) w ho pursued
the claim that the storage routing correction in the TRRL method was spurious- and
needed only to counteract overestimation of flow rates during the first half of
the storm. He suggested that a simple time-area calculation should be applied
just to the part of the rainfall profile occurring after the peak intensity.
The Urbanised Catchment
It is sometimes difficult to make a neat separation between the hydrological
modeling of urban catchments (by which we tend to mean sewered catchments) and that
of natural catchments either undergoing progressive urbanisation or already
substantially urbanised. In practice, however, the distinction is whether or not
any attempt is made to simulate flow through pipe systems. If not, and if the
extent of sewered areas is represented simply by one or two catchment characteristics,
then we have an 'urbanised catchment model'. The characteristics can be included
as independent variables in regression equations to predict model parameters. This
approach has been used frequently in the U.S.... with the simple unit hydrograph
modela) but only one such study has reached a similar stage in the U.K. Halloo)
studied unit hydrographs for several urban and rural catchments near Crawley, and
also the changing shape of the unit hydrograph during progressive urbanisation of
three of the catchments. He developed a dimensionless unit hydrograph scaled by
only one parameter and related this parameter to basin ratio (L//5-; where L -
main channel length, S - main channel slope) and degree of urbanisation.
Packman,(27) applying the same techniques to catchments in North London, obtained
reasonably consistent values of lag time, and his dimensionless unit hydrograph is
similar to that subsequently published in the Flood Studies Report.d^)
In response to queries arising since publication of the Flood Studies

Report,' ^' the Institute of Hydrology re-examined basic data and produced
recommendations^28' for adjusting estimates of the mean annual flood, the T-year
flood and hydrograph shape for urbanising catchments.
182
Conclusions
Urban hydrological modeling in the U.K. continues to be geared primarily
to the improvement of sewer design methods. The common aim is to seek a compromise
between the mainly old, established, easily applied but theoretically unattractive
methods, and the highly complex analytical models based on physical laws.
References, Section 3*
1. NATIONAL WATER COUNCIL (1976). Working Party on the hydraulic design of
storm sewers. Review of progress to June 1975. Department of the
Environment/National Water Council Standing Technical Committee Report No. 1.
2. McPHERSON, M. B. (1975). Urban hydrological modeling and catchment research

in the U.S.A.,reprinted in this volume.
3. COLYER, P. J. and PETHICK, R. W. (1976). Storm drainage design methods - a

literature review. Hydraulics Research Station Report INT 154.
4. WATKINS, L. H. (1962). The design of urban sewer systems. Road Research

Laboratory. Technical Paper 55, H.M.S.O.
5. WATKINS, L. H. (1963). Research on surface water drainage. Proceedings

Instn. Civil Engineers, 24, 305-330.
6. ROAD RESEARCH LABORATORY (1963). A guide for engineers to the design of

storm sewer systems. Road Note 35, DSIR.
7. WATKINS, L. H. and YOUNG, C. P. (1965). Developments in urban hydrology in

Great Britain. Unpublished report (LN 885) of Road Research Laboratory.
8. YOUNG, C. P. (1974). Urban drainage in the United Kingdom. CIRIA/Bristol

University Colloquium on rainfall, runoff, and surface water drainage of
urban catchments. April 1973.
9. ESCRITT, L. and YOUNG, A. J. M. (1963). Economic surface-water sewerage:

a suggested standard of practice. Journal Inst. Pub. Health Engnrs.. 62, 4.
10. SARGINSON, E. J. and BOURNE, D. E. (1969). The analysis of urban rainfall,

runoff and discharge. Journal Institution of Municipal Engineers. 96, 81.
11. NEWTON, S. G. and PAINTER, R. B. (1974). A mathematical examination of

urban runoff prediction. Proceedings Instn. Civil Engineers. Part 2, 57,
143-157.
12. STEEL, P. H. (1974). Summary of the present position and areas in which
further research is needed. CIRIA/Bristol University Colloquium on rainfall,
runoff, and surface water drainage of urban catchments. April 1973.
13. TRANSPORT AND ROAD RESEARCH LABORATORY (1976). Road Note No. 35, 2nd edition.
14. NATURAL ENVIRONMENT RESEARCH COUNCIL (1975). Flood Studies Report, 5 Vols.
*N0TE: Addresses for unpublished report acquisition are given in the following
subsection.
183
15. KING, M. V. (1967). Storm runoff from urban areas. Proceedings Instn. of
Civil Engineers, 37, 34.
16. SARGINSON, E. J. (1973). A statistical treatment of urban drainage design.
Journal Institution of Municipal Engineers. 100, 152.
17. SARGINSON, E. J. and NUSSEY, B. B. (1975). A mathematical model for urban

runoff. Symposium on the application of mathematical models in hydrology
and water resource systems. Bratislava IAHS Pub. No. 115, 209.
18. WOOD, S. R. (1975). A catchment simulation model developed for urban and
urbanising catchments with particular reference^ to the use of automatic
optimisation techniques. International Federation of Automation and
Control, Symposium on Modeling and simulation of Water Resources Systems.
Ed. Van Steenkiste, North Holland Publishing Company.
19. HINDI, W. N. A. (1976). A computational method for the optimal design and
analysis of urban sewer systems. Unpublished Ph.D thesis, Department of
Civil Engineering, University of Birmingham.
20. KIDD, C. H. R. (1975). A nonlinear model of urban catchment response.

Unpublished Ph.D thesis, Department of Civil Engineering, University of
Southampton.
21. WATKINS, L. H. (1976). The RRL hydrograph method of urban sewer design
adapted for tropical conditions. To be published in part 2 of Instn. Civil
Engineers proceedings.
22. SWINNERTON, C. J., HALL, M. J. and O'DONNELL, T. (1972). A dimensionless

hydrograph design method for motorway stormwater drainage systems. Journal
Institution Highway Engineers, 19, 11, 2-10.
23. SWINNERTON, C. J., HALL, M. J. and O'DONNELL, T. (1973). Conceptual model

design for motorway stormwater drainage. Civil Engineering and Public Works
Review, 68, 799, 123-129, 132.
24. HELLIWELL, P. R., KIDD, C. H. R. and LOWING, M. J. (1976). Estimation of

the above-ground runoff hydrograph for storm sewer design purposes. Pending
publication, available from Institute of Hydrology.
25. HEPW0RTH, R. (1974). A direct approach to storm sewer design. New Civil
Engineer, 14 November, pp. 44-46.
26. HALL, M. J. (1973). Synthetic unit hydrograph technique for design of

flood alleviation works in urban areas. International symposium on design
of water resources projects with inadequate data. Madrid, P r o c , Vol. 1,
145-161.
27. PACKMAN, J. C. (1974). The application of unit hydrograph theory in catchments

subject to urbanisation. Unpublished M.Sc. thesis, Imperial College,
University of London.
28. INSTITUTE OF HYDROLOGY (1976). The effect of urbanisation on flood estimates.

Unpublished report - also document 76/17 of DOE/NWC Working Party on Hydraulic
Design of Storm Sewers.
184
Addresses' for Section 3 References
P. J. Colyer, Esq. (1)

Hydraulics Research Station
Howbery Park
Wallingford
Oxon, U.K.
Hydraulics Research Station (3)

Howbery Park
Wallingford
Oxon, U.K.
Transport and Road Research Laboratory (4, 6, 7, 13)

Crowthorne
Berks, U.K.
Construction Industry Research and (8, 12)

Information Association
6, Storey's Gate
London, U.K.
Institute of Hydrology (14, 24, 28)

Maclean Building
Crowmarsh Gifford
Wallingford
Oxon, 0X10 8BB, U.K.
University of Birmingham (19)

P.O. Box 363
Birmingham B15 2TT, U.K.
University of Southampton (20)

Highfield
Southampton, S09 5NH
U.K.
Imperial College (27)

South Kensington
London, U.K.
185
[A.35JSC.76/XXI.I5/A
I S B N 92-3-101488-9
ERRATUM
W e apologize for the repetition of Figure 6 which
has been inadvertently printed on pages 51 and 5 2 .
In this series
1 Perennial Ice and S n o w Masses. A Guide for Compila-
tion and Assemblage of Data for a World Inventory.
2 Seasonal Snow Cover. A Guide for Measurement,
Compilation and Assemblage of Data.
3 Variations of Existing Glaciers. A Guide to Inter-
national Practices for their Measurement.
4 Antarctic Glaciology in the International Hydrological
Decade.
5 Combined Heal, Ice and Water Balances at Selected
Glacier Basins. A Guide for Compilation and Assem-
blage of Data for Glacier Mass Balance Measurements.
6 Textbooks on Hydrology—Analyses and Synoptic
Tables of Contents of Selected Textbooks.
7 Scientific Framework of World Water Balance.
8 Flood Studies—An International Guide for Collection
and Processing of Data.
9 Guide to World Inventory of Sea, Lake, and River Ice.
10 Curricula and Syllabi in Hydrology.
11 Teaching Aids in Hydrology.
12 Ecology of Water Weeds in the Neotropics.
13 The Teaching of Hydrology.
14 Legends for Geohydrochemical M a p s .
Légendes des cartes hydrogéochimiques.
Leyenda para mapas geohidroquímicos.
JlereHßa w w reoniApoxHMHHecwux KapT
15 Research on Urban Hydrology, vol. 1.
16 Research on Urban Hydrology, vol. 2.
International Hydrological Programme
Research
on urban hydrology
Volume 2
State-of-the-art reports from
France, Federal Republic
of Germany, India,
Netherlands, Norway,
Poland, Sweden
(uresco
The designations employed and the presentation of the
material do not imply the expression of any opinion what-
soever on the part of Unesco concerning the legal status
of any country or territory, or of its authorities, or con-
cerning the frontiers of any country or territory.

Organization
Printed by Imprimerie J. Floch, 53100 Mayenne
I S B N 92-3-101555-9
© Unesco 1978
Printed in France
Preface
T h e 'Technical Papers in Hydrology' series, like the related collection of 'Studies and Reports in
Hydrology', w a s started in 1965 when the International Hydrological Decade w a s launched by the
General Conference of Unesco at its thirteenth session. T h e aim of this undertaking was to promote
hydrological science through the development of international co-operation and the training of specia-
lists and technicians.
Population growth and industrial and agricultural development are leading to constantly increa-
sing demands for water, hence all countries are endeavouring to improve the evaluation of their water
resources and to m a k e more rational use of them. T h e 1 H D was instrumental in promoting this
general effort. W h e n the Decade ended in 1974, I H D National Committees had been formed in 107
of Unesco's 135 M e m b e r States to carry out national activities and participate in regional and inter-
national activities within the I H D programme.
Unesco was conscious of the need to continue the efforts initiated during the International
Hydrological Decade and, following the recommendations of M e m b e r States, the Organization
decided at its seventeenth session to launch a n e w long-term intergovernmental programme, the
International Hydrological Programme (IHP), to follow the decade. The basic objectives of the I H P
were defined as follows: (a) to provide a scientific framework for the general development of hydro-
logical activities; (b) to improve the study of the hydrological cycle and the scientific methodology
for the assessment of water resources throughout the world, thus contributing to their rational use;
(c) to evaluate the influence of m a n ' s activities on the water cycle, considered in relation to environ-
mental conditions as a whole; (d) to promote the exchange of information on hydrological research
and on new developments in hydrology; (e) to promote education and training in hydrology; (f) to
assist M e m b e r States in the organization and development of their national hydrological activities.
T h e International Hydrological P r o g r a m m e became operational on 1 January 1976 and is to
be executed through successive phases of six years' duration. I H P activities are co-ordinated at the
international level by an intergovernmental council composed of thirty M e m b e r States. The m e m b e r s
are periodically elected by the General Conference and their representatives are chosen by national
committees.
The 'Technical Papers in Hydrology' series is intended to provide a means for the exchange of
information on hydrological techniques and for the co-ordination of research and data collection.
In order to co-ordinate scientific projects, however, it,is essential that data acquisition, transmission
and processing be conceived in such a way as to permit the comparison of results. In particular, the
exchange of information on data collected throughout the world requires standard instruments, tech-
niques, units of measurement and terminology.
It is believed that the guides on data collection and compilation in various specific areas of
hydrology which have been published in the 'Technical Papers in Hydrology' series have already
helped hydrologists to standardize their records of observations and thus have facilitated the study
of hydrology on a world-wide basis.
M u c h still remains to be done in thisfield,however, even as regards the simple measurement of
basic elements such as precipitation, snow cover, soil humidity, run-off, sediment transport and
ground-water phenomena.
Unesco therefore intends to continue the publication of 'Technical Papers in Hydrology" as an
indispensable means of bringing together and making k n o w n the experience accumulated by hydro-
logists throughout the world.
Contents
Foreword
Urban hydrological modelling and catchment research in France 11

Section 1 Urban catchment research in France 13
Section 2 Urban hydrological modelling in France 26
Appendix A A general study of urban runoff drainage 31
Appendix B Mathematical model of flood and pollutant propagation in a
meshed drainage network 45
Appendix C Extension of Caquot's model to non-homogeneous catchments 61
Urban hydrology studies and mathematical modelling in the Federal Republic

of Germany 73
Section 1 General information 75
Section 2 Hydrological studies in urban catchments 82
Section 3 Models 106
Urban hydrological modelling and catchment research in India 129

Section 1 Introduction 131
Section 2 Urban drainage design in India 135
Section 3 Field research 141
Section 4 Hydrologie modelling 147
Section 5 Conclusions 150
Urban hydrological modelling and catchment research in the Netherlands 153

Section 2 Urban catchment research in the Netherlands 156
Section 3 Urban hydrology modelling in the Netherlands 182
Urban hydrological modelling and catchment research in Norway 199

Section 2 Urban catchment research in Norway 203
Section 3 Urban hydrological modelling in Norway 213
Urban runoff research in Poland 219
Introduction 221
Research on the relations between rainfall and the runoff in urban drainage
networks 222
Research on the influence of the storage capacity in drainage networks and
flow-balancing tanks on the diminution of pollution loads transported by runoff
to receiving waters 224
Research on the influence of short undetermined rainfalls on the quality of
receiving waters 225
Urban hydrological modelling and catchment research in Sweden 229

Section 2 Urban catchment research in Sweden 238
Section 3 Urban hydrological modelling in Sweden 255
Foreword
This is the second volume prepared under International Hydrological Programme sub-project 7.1
entitled 'Research o n Urban Hydrology'. It contains seven national state-of-the-art reports fore-
seen within the general framework of project 7, 'Effects of Urbanization on the Hydrological
Regime and on Quality of Water', adopted by the Intergovernmental Council of the International
Hydrological Programme at its 1975 session. T h e project includes also sub-project 7.2, 'Develop-
ment of Mathematical Models applied to Urban Areas considering both Water Quality and
Quantitative Aspects'.
The Bureau of the International Council of the International Hydrological Programme appointed
M r . M . B . McPherson ( U . S . A . ) as rapporteur of sub-project 7.1 at itsfirstsession in August 1975,
and defined the tasks to be carried out under the sub-project as follows:
1. T o prepare 'state-of-the-art' reports reviewing current research on urban hydrology based on case
studies; the reports are to include descriptions of laboratory andfieldstudies, instrumentation,
methods of processing and analysing information from experiments, development of models on
urban water systems and urban planning, water quantity and quality processes, etc.;
2. T o prepare outlines of the information manuals on urban water data collection, analysis and use.
This endeavour originated from activities and aspirations of the Unesco Subgroup on the Hydro-
logical Effects of Urbanization of the International Hydrological Decade. Part I of the Subgroup
final report1 entitled 'International S u m m a r y ' , was resolved by representatives of over thirty nations
w h o participated in an International Worshop at W a r s a w , Poland, N o v e m b e r 19731 at which ten
crucial international research projects were proposed for inclusion in the Unesco component of the
International Hydrological Programme.
The American Society of Civil Engineers ( A S C E ) took early supportive action by applying for
an N S F grant to assist in two of the ten recommended projects: R . l . Catchment Studies Report,
and R . 3 . Mathematical Models Report. In April 1975, the International Council for the I H P adopted
I H P Project 7, which includes the two subjects in question and thus m a d e it possible for Unesco and
the A S C E to co-operate closely on the state-of-the-art reports.
O f particular significance was the very strong emphasis of the W a r s a w W o r k s h o p and the Sub-
group on the urgency of addressing all such reports to users of researchfindings.That is, an accentua-
tion of user participation and user orientation of I H P urban products clearly indicated that facili-
tation of the translation of research findings into implementation practice should be a central goal.
In most countries, economic growth, population growth, non-agricultural water use and pollu-
tion are intertwined. Water in its m a n y manifestations plays a vital role in the extremely complex
processes of urbanization, and thus affects a nation's health and growth. The most significant con-
clusion reached by the I H D / U n e s c o Subgroup is that most urban hydrological problems and effects
are similar in technologically and economically advanced countries. Further, m a n y problems confron-
ting the developing nations have at one time or another already been encountered by m a n y deve-
9
loped nations. This strongly suggests that great benefits would result from the exchange of infor-
mation and increased international co-operation in research and development.
The report for the U . S . A . served as the prototype for the series of national reports, and was
circulated to experts in other countries for the purpose of obtaining analogous contributions.
Unesco wishes to thank the authors and the National Committees for t h e T H P of the seven
countries concerned for the reports presented in this volume, which should facilitate international
communication o n the state of the art in urban hydrology. Additional volumes will be issued in
the future, as further contributions to sub-project 7.1.
10
research in France
M . Desbordes
Laboratoire d'Hydrologie Mathématique
Université des Sciences et Techniques du Languedoc
Place Eugène Bataillon
34060 Montpellier Cedex
France
and
D . Normand
Service Scientifique d'Application
S O G R E A H Ingénieurs Conseils
B P 172 - Centre de Tri
38042 Grenoble Cedex
France
Section 1 Urban catchment research in France
Introduction
French research on the effects of urbanization on the water cycle has undergone
a new development since 1969, prompted by the Ministry of Equipment.' ' This has
resulted from the ever-growing complexity of urban sewerage problems: new techniques
must be perfected and better adapted to the nuisances of all kinds caused by
stormwater disposal . Experimentation with these new techniques is necessary when
existing sewer networks are extended or when new builtup areas are created.
(2) It
is also necessary to find new methods for sewer calculations, consistent with these
new techniques and the economic problems related to the struggle against nuisances.
From 1970 to 1975 there were two research goals: to bring up to date the
official regulations on urban sewerage (these rules are set forth in Technical
Instruction C G . 1333, February, 1949^'); and to improve knowledge on rainfall-
runoff transformations for urban watersheds'^»5' by making runoff mathematical models
adapted to the design of complex sewer networks. As new concepts on drainage systems
evolved,'""'-' researchers and technicians who had been previously preoccupied with
runoff physical characteristics became gradually aware of the importance of water
quality in such systems.'^' Therefore, attention to research on water quality

aspects of drainage is very recent in France.
(2,10,11)
At present, water planning specialists have taken a greater interest in urban
hydrology. They have agreed to study the relation between urbanization problems
and the water cycle in a comprehensive way, and are trying to perfect water planning
models and techniques without separating any longer water quantity from water quality
considerations . I " J
13
The French Ministry of Equipment Profiram
The starting point of French research has been the experimental program, begun
as early as 1969, by the Ministry of Equipment.'*' It has served as the basis for
most theoretical research on the subject and has been followed by a few other
efforts.
By the end of 1968, six small urban watersheds, situated in the vicinity of
Grenoble, had been equipped with rainfall and runoff gaging stations. In the same
way three other watersheds were equipped in Montpellier by the end of 1969. The
rainfall and runoff gaging period came to an end in 1970 at Grenoble and continued
at Montpellier until 1973. Table 1 gives the basic characteristics of these
watersheds. Because of their small size, each watershed was equipped with a single
water-level recorder and a single tipping-bucket raingage.
Water levels in the sewer at the outlet of each watershed were indicated by
float-actuated recording gauges, placed either in a stilling well in close relation
to the bottom of the sewer (watersheds 5 through 9) or in the sewer itself
(watersheds 1 through 4). For the latter, the float was connected with a rotating
axis by a rigid bar where the rotation angle was related to the water level. These
indirect discharge measurements required calibration of the sewer to establish a
rating curve. Calibration was achieved by the use of fire-hydrants, but it cannot
be said that the reliability of the rating curves for watersheds 7, 8 and 9 was
very high because large pipes were involved.
Rainfall was recorded with a classical tipping-bucket raingage producing an
electrical signal.
Double-tracked recorders in parallel operated by a single clock were used.
14
e
W 0) oo 10 r-. -tf O
Oc U •
o o
o
• • o
•
CM CS• •
O O -tf o m
HJ U
c/> eu
a.
H
o
vO 00 O o CO o o o o
00 iH vO CTV co u-i 00 -tf vO
cg u-l vO CT» -tf •tf CM vO r»
ESS
'-N
Ï5 4-1
00 tí
3 (U
O Ü O 00 CM CT\ CM O CM CM r-»
H ï-» r-» ¡n co CM O co -tf CM
> M
cu
Ö a.
ë
^5 «~\ m vo o IO CO vo a\ -tf
W cd • « •
• • CO • •
55 i-l -tf CO CM
i-l
i-H
r-f
CO
h h (k
3 3 3
cg NJ Ni
CM CO -tf U"> vO f^
15
Recording was accomplished by the simultaneous punching of the two tapes with each
tick of the clock (a tick every two minutes and a half or every five minutes). The
punching of the tape corresponded to a classical Telex Code. The recording could
simply be reproduced or divided into the desired time interval.
If the data from watersheds 1 through 6(13) could be considered of fairly good
quality, that is not the case for watersheds 7 through 9 which were more difficult
to calibrate. Moreover, the sort of recorders that were chosen were not durable
and required considerable maintenance. Lastly, the absence of an on-site graphic
recorder as a means for checking observations made the data unreliable when
equipment malfunctions might have occurred.
First Program Data Treatment
Data from the experimental watersheds led to two principal types of studies:
theoretical studies on urban runoff modeling on the one hand, and studies made from
the viewpoint of testing and adapting the French method of drainage design^-*) on
the other hand.
Theoretical studies were initially conducted at the request of the French
Ministry of Equipment by the engineering firm SOGREAH, Grenoble, and by the
Laboratoire National d'Hydraulique (L.N.H.) (Electricité de France) associated with
the Laboratoire d'Hydrologie Mathématique (L.H.M.) (Montpellier university). They
succeeded in perfecting or testing various mathematical runoff models.

(4,5,14-16,25)
At the same time, the L.H.M. studied different one-dimensional and two-dimensional
mathematical models of infiltration in the zone of aeration so as to make a
connection between surface water and groundwater models.'41-44)
The original contributions of the French theoretical studies lie in the
statistical analysis of thunderstorms. These studies were aimed at the perfection
of design storm models that would be used as the input for hydro logical models.
16
Without desiga storm models there would be limited interest in the study of
drainage projects for ungaged watersheds. These rainfall models do not yet take
into account the spatial distribution of rain, owing to the nature of the presently
available data.(I?,18) Studies of the sensitivity of runoff models to the definitive
characteristics of thunderstorms'•'-"•' have enabled, more particularly, the elaboration
of surrogate stochastic models for the simulation of design storms.(20) Spatial
analysis of thunderstorms has been the subject of a few studies despite the lack
of available data.(21)
Practical applications have been made by the Loriferne Commission, which is
responsible for revision of Technical Instruction C.G. 1333. This commission is
composed of specialists from the main ministries responsible for problems of
water planning, University researchers and engineers from public and private research
firms. The Commission has just completed revision of the Instruction, Caquot's
method for French sewer design (inspired by the so-called Rational Method used in
the U.S.A.) has been experimentally checked and adapted, and its applicability in
its renovated form has been better defined.(4,22-24)
Studies of Instruments for Hydrological Measurement
Very quickly it became evident that the main problem in the advance of urban
hydrology was the absence of good data. Theoretical research on modeling of
hydrological phenomena has very quickly exceeded the data usually available. The
tempo of these research efforts quickened as they benefitted, as early as 1970, from
papers published in foreign countries, especially ones from America. This fact is
substantiated by hydrologists all over the world. Unfortunately, there are as yet
no rainfall and runoff gauges designed specifically for urban hydrology, nor is
there any outstanding device to sample the water quality in a storm or combined sewer.
The rapid changes in these variables and their random nature, and the difficult
17
measuring conditions encountered in sewers or in urban space, have prevented
definition of an ideal measurement assemblage, which in any case would probably be
quite expensive. So, different public and private firms have devoted their efforts,
since 1970, to research on data measuring and recording instruments and on their
adaptation to experimental conditions encountered in urban hydrology.
The most widely used instrument in France is the tipping-bucket raingage. This
instrument tends to underestimate the highest instantaneous rainfall intensities
(more than 60- to 80-mm/h) because of water losses, response lags, etc. Research
on how to correct the information given by this instrumentado) as well as tests of
different graphic recorders^2''' has been undertaken (cylinders, unrolling tables,
automatic piling of papers, etc.). Tests of punched-tape recorders'*-^' have not
led to very interesting conclusive results. Those of mini-tape recorders' ' have
proved quite promising. However, there has not been any important testing of
electronic synchronization instruments. It would be advisable in this field to
make a great effort using ingenuity. The assemblages on sale are too costly and
are fairly difficult to use.
A bibliographical study of sewer measurement processes' '' has shown that
there is no instrument perfectly adapted to that kind of problem. The greater part
of flow measurement is carried out by means of a water level measurement associated
with a stage-discharge curve. Most of the time, this rating curve is established
ignoring upstream or downstream conditions, for it is not very common to find
hydraulic control sections at places suitable for gaging. The water-level measuring
instruments in most wide use are float-actuated recording gauges or pressure-
actuated recording gauges (or bubble gauges).(",30) Stage-discharge curves are
generally obtained with the help of radioactive tracers^30™33' or by the use of
induced injections of known discharges in the case of small pipes.V1^) Tests on
18
electromagnetic Instruments measuring continuous velocity"4-) have been carried out.
These instruments estimate the velocity within a few centimenters of the pipe wall.
It is not always a simple matter to compare this velocity to the mean velocity.
Some prototypes of ultrasonic flowmeters'35' and others using radioactive tracers^ '
have also been studied. These instruments are quite costly and the specificity of
the latter precludes its continuous use. Recording and synchronization problems
are very similar to those encountered in rainfall measurements.
Pollution gaging is a very recent field that has been little explored and
samplings are generally not continuous.t10»30-) There is no specific sampler
adapted to that problem. Besides, it may not be possible to improve this type of
sampler until we have more precise and clearer notions on water quality, pollution
variables which ought be measured and sampling time intervals (which are often less
than five minutes for rainfall intensity and discharge measurements).
Thus, despite the numerous efforts of researchers, technological studies seem
to be needed that would lead to the design of new instruments and recorders fully
adapted to problems encountered in urban hydrology. Only an international symposium
comparing various experiences would lead to a quick resolution of this problem.
Future Research Aims
Experience gained during the last five years has oriented research towards the
following main themes:
Rainfall studies. Analysis and modeling of spatial distributions of rainstorms,
analysis of influences due to altitude variations, study of regional rainfall data,
analysis of the effects of rainstorm travel, etc.
Runoff studies. Analysis and modeling of runoff from urban pervious areas.w7)
Water quality studies. Model and attempt to define the quality of runoff
pollution.
These studies cannot be undertaken without important experimental help. That
19
is why the Ministry of the Interior and the Ministry of Equipment are financing for
1976 an equipment scheme and a preliminary measurement campaign of ten watersheds.
First, these watersheds, specially chosen for research purposes, will be provided
with two discharge stations and a few recording raingages. At the end of the first
period, five or six watersheds that are among the most representative will be
chosen to be better equipped in order to assure a high quality of measurements.
This first experimental year should expose management and instrument maintenance
problems, enable testing the quality of the instruments selected and the reliability
of planned technical solutions (recording, synchronization, and so on). Furthermore,
on account of the fairly important land-use characteristics of these v/atersheds,
there should be opportunity to verify the validity of the models that have been
studied so far. The investigation should extend ever about ten years.
Conclusion
As a new activity, research on urban hydrology In France has led to Interesting
results, proportionally speaking, that can be compared with those acquired arouud
the world. The Importance of that particular component of the water cycle has
been emphasized. Furthermore, the necessity to acquire reliable data very quickly
has been shown and thence the specification of necessary data-collecting instruments.
The research scheme under way and international comparison of results should
encourage the use of urban water planning and management models^»"««» which at
present are seldom used.
References
1. AFFHOLDER M. "Contribution d'une etude expérimentale a l'amélioration

des méthodes d'évaluation des débits de ruissellement en
milieu urbain" ("Contribution of an experimental study
to better urban runoff evaluation").
Association Française pour l'Etude des Irrigations et du
Drainage. Journées d'Etudes de COLAR, 13 p. (1970)
20
LELONG F. "Les problèmes de l'eau de la ville nouvelle du VAUDRÊDTL"
("Water problems of che new town VAUDREUTL").
E.P.A. du VAUDREUIL, BP 5 27690 LERY, 178 p. (Hal 1975)
MINISTERE DE IA RECONSTRUCTION ET DE L'URBANISME

"Instruction Technique relative a l'assainissement des
agglomérations" no. C.G. 1333 ("Technical Instruction No.
C G . 1333").
Imprimerie Nationale PARIS, 50 p. (Février 1949)
4. NORMAND D. "Etude générale du ruissellement urbain - Verne partie"

("A study of urban runoff, Part V").
Rapport SOGREAH R. 10943, Ministère de l'Equipement et
du Logement, 4 rue Leon Bollee PARIS, approx. 30 p.
(1971)
5. DESBORDES M. "Réflexions sur les méthodes de calcul des réseaux
urbains d'assainissement pluvial" ("A study of methods
used in sewer design").
These de Docteur Ingénieur - Université des Sciences et
Techniques du Languedoc, MONTPELLIER, 224 p. (Février 1974)
6. CORMARY Y. DESBORDES M. "Etude des écoulements urbains, conceptions des reseaux

d'assainissement" ("Study of urban runoff, sewer design").
Xllemes journées de l'Hydraulique - Question IV Rapport 1,
Société Hydrotechnique de France, 199 rue de Grenelle
75007 PARIS, 9 p. (Septembre 1974)
7. TRICHARD M. "Role des bassins de retenue en assainissement urbain"

("The use of storm basins in urban sewerage"). Ministère
de l'Equipement, Direction de la Construction, 2 av. du
Parc de Passy, PARIS, 208 p. (Juin 1972)
8. JACOBSEN E. "La régularisation des écoulements de pluie: Etude

fréquentielle des pluies" ("Regulation of runoff:
rainfall frequency analysis"). Le Génie Rural no. 1,
p. 27-31. (Janvier 1973)
DESBORDES M. "Quelques méthodes de calcul des bassins de retenue des

eaux pluviales" ("Some calculating methods for storm
basins"). Centre Belge d'Etude et de Documentation des
Eaux. Tribune du CEBEDEAU no. 377, p. 168-174.
(Avril 1975)
10. DEUTSH J.C. "Etudes générales de pollution des eaux dans le

Département de Seine St Denis" ("Study of water pollution
in the Seine-St. Denis Region"). Direction Départementale
de l'Equipement, 99 av. du Général de Gaulle, 92 R0SNY SOUS
BOIS, 6 p. (Janvier 1975)
21
COYNE & BELLIER, Ing. Cons.
"Etude d'ouvrages de separation et de collecte des
premieres eaux pluviales" ("Study of hydraulic structures
for separating and collecting first runoff waters").
COYNE & BELLIER, 19 rue Alphonse de Neuville 75017,
PARIS, 50 p. (1974)
DESBORDES M. "Hydrologie urbaine; programme de recherche complémentaire

("Urban hydrology: new research program"). Sous groupe
1 Commission LORIFERNE. Ministère de l'Intérieur D.G.CL.,
Services Techniques, 1 bis place des Saussaies, 75 PARIS,
15 p. (Février 1975)
NORMAND D., RENE R. "Etude generale du ruissellement urbain: annexes Al et

A2, hyétogrammes et hydrogrammes pris en compte dans
l'étude" ("Study of urban runoff: Appendices Al and A2").
Ministère de l'Equipement, Rapport SOGREAH no. R 10737
(1971)
N0RMA.ND D., RECOURA J., RENE R.

"Etude générale du ruissellement urbain - IVeme partie"
("A general study of urban runoff: Part IV"). Ministère
de l'Equipement Rapport SOGREAH no. R 10737 59 p. (1971)
DESBORDES M., BARA B., LAVABRE J., MASSON J.M., NGUYEN VANKY S«, RA0US P.
"Réflexions sur les méthodes de calcul des réseaux
urbains d'assainissement pluvial," 11 parties. ("Studies
of sewerage calculation and design methods") - in 11
parts. Ministère de l'Equipement. S.H.C., 4 rue Léon
Bollée, 75 PARIS, Approx. 660 p. (Décembre 1972 a
Décembre 1975)
DESBORDES M. "Un essai de modélisation des phénomènes de ruissellement

pluvial urbain" ("Urban runoff modeling"). Techniques
et Sciences Municipales, 70seme année, no. 3, p. 121-126.
(Mars 1975)
NORMAND D., RECOURA J., RENE R.

"Etude statistique des hyétogrammes types" ("A
statistical study of design storms"). Ministère de
l'Equipement, Rapport SOGREAH no. R 10653, 20 p. (1970)
DESBORDES M., LAVABRE J. "Pluie de projet: résultat des recherches" ("Design

storm: research results"). Ministère de 1'Equipcnent
S.C.H., 59 p. (Septembre 1973)
DESBORDES M., RAOUS P. "Réflexions sur les méthodes de calcul des réseaux
urbains d'assainissement pluvial 9%eme partie Etude de
sensibilité du modele de ruissellement" ("Sensitivity
study of an urban runoff model"). Ministère de
l'Equipement S.C.H., 68 p. (Décembre 1974)
22
DESBORDES M., MAS SON J.M., RUIZ L.
"Reflexions sur les méthodes de calcul des réseaux
urbains d'assainissement pluvial 10 "eme partie
Amélioration de la définition de pluie de projet"
("A better definition of design storm"). Ministère de
l'Equipement S.C.H., 90 p. (Décembre 1975)
MASSON J.M., RAMPEREZ A., SAURÍN J.

"Etude des pluies orageuses se produisant sur la région
parisienne a partir des relevés pluviométriques" ("Study
of rainstorms in the area of Paris"). Note L.H.M. no.
3/75, Agence Financière de Bassin Seine-Normandie, 10 rue
du Capitaine Menard 75015, PARIS, 64 p. plus annexes.
(Juin 1975)
NORMAND D . "Modele de CAQUOT. Confrontation entre la formule mise

au point par le L.H.M. et les mesures effectuées par
SOGREAH" ("CAQUOT model: analysis of the L.H.M. formula
and of the SOGREAH measures"). Ministère de l'Intérieur,
D.G.C.L., Services Techniques, 30 p. (Mars 1974)
DESBORDES M . "Verification expérimentale du modele de CAQUOT" ("CAQUOT

model experimental testing"). Ministère de l'Intérieur,
D.G.C.L., Services Techniques, 31 p. (Avril 1974)
DESBORDES M . "Extension du modele de CAQUOT aux bassins versants

hétérogènes" ("CAQUOT model extrapolation for
non-homogeneous catchments"). Ministère de l'Intérieur,
D.G.C.L., Services Techniques, PARIS, 48 p. (Septembre
1975)
C.T.G.R.E.F. AIX-EN-PROVENCE
"Etudes hydrologique et hydraulique de l'assainissement
pluvial des terres basses d'Arles" ("Hydrological and
hydraulic studies of the drainage of low lands in ARLES").
Ministère de l'Agriculture, C.T.G.R.E.F., Le Tholonet
13603, B.P. 92, AIX EN PROVENCE, 150 p. approx. (Juin 1975)
C.T.G.R.E.F. PARIS "Etalonnage de pluviographs *a augets basculeurs modele

précis mécanique MN R5 302 a tambour" ("Tipping-bucket
recording raingage MN R5 302 testing"). Ministère de
l'Agriculture, C.T.G.R.E.F., Parc de Tourvoie, 75,
ANTONY, approx. 70 p. (Février 1974)
O.M.M. "Comparaison d'instruments hydrométriques participation

aux tests du Laboratoire Régional de l'Ouest Parisien
(MM CHERON - BELIN) du Centre Technique du Génie .Rural
des Eaux et des Forêts (M. OBERLIN) et du Laboratoire de
Mécanique des Fluides de TOULOUSE (M. ALDEGHRIRI)"
("Comparison of hydrometrical Instruments. Tests made
by three agencies"). Approx. 25 p. O.M.M., B.P. 1,
GENEVE 20.
23
ELECTRICITE DE FRANCE "Enregistreur magnétique séquentiel code sur minicassette
a grande autonomie pour collecte de données sur le
terrain" ("A mini-cassette tape recorder for long time
outdoor data collecting"). E.D.F., Direction Technique
Générale, Fiche no. 50, LA HOUILLE BLANCHE no. 4,
p. 231-232 (1975)
DESBORDES M. "Etude bibliographique de quelques étapes d'un calcul

de réseau" ("Bibliographie study of some aspects of
sewer network calculations"). Rapport L.H.M. no. 1/72,
Ministère de l'Equipement S.H.C., p. 69-81 (Septembre
1972)
CHERRON J., RANCHET J. (L.R.O.P.)

"Debits et pollution des deux principaux reseaux de
l'agglomération de MANTES" ("Discharges and pollution
in the two main sewer networks in MANTES"). Bulletin
de liaison des Laboratoires des Fonts & Chaussées,
no. 78, p. 33-44 (Juillet 1975)
BELIN J.F., DKLION J. "Compte rendu de mesure de vitesses et de débit par

traceurs radioactifs dans un réseau d'assainissement"
("Velocity and discharge measurements with radioactive
tracers in a sewer"). Rapport des Laboratoires des
Fonts et Chaussées, Ministère de l'Equipement, 15 p.
(Octobre 1972)
BELIN J.F., DELION J., ROUSSELIN R.S., GAUTHIER C. (L.R.O.P.)

"Mesures de vitesses et de debits d'écoulement en
canal avec 1'Indium 113 M" ("Velocity and discharge
measurements with Indium 113 M"). Bulletin de liaison
des Laboratoires des Ponts et Chaussées no. 74, p. 97-109
(Décembre 1974)
LABORATOIRE DE L'OUEST PARISIEN

"Compte-rendus de mesures de débits par traceurs
radioactifs" ("Velocities with the aid of radioactive
tracers"). 12 rapports internes L.R.O.P. 78190, TRAPPES
(1972-1975)
LACROIX J.L. "Contrôle des écoulements dans les les égoûts et

installations réalisées dans les egouts de Marseille"
("Flow control and gaging stations in MARSEILLE sewers").
Bulletin d'information de l'Agence Financière de Bassin
Rhône-Méditerranée-Corse no. 2, p. 45-52 (Novembre 1971)
LENORMAND J. "Debimetre a ultra-sons M0L2. Compte-rendu d'essais"

("MOL 2 ultrasonic flowmeter. Testing report").
Ministère de l'Equipement S.V.N.N.P.C., 92 av. Pasteur,
LILLE, 24 p. (Mars 1974)
24
JANIN P. "Mesure des vitesses de fluide par la méthode des
bouffées radioactives" ("Fluid velocity measurement
by the use of radioactive tracers"). E.D.F.
Laboratoire National d'Hydraulique, Département Essais,
6 quai Watier, 78400, ChATOU, Note Jll/D 05, 5 p.
(Mars 1973)
DESBORDES M. "Estimation des coefficients de ruissellement urbain"

("Estimating urban runoff coefficients"). La Tribune
du CEBEDEAU no. 376, p. 106-110 (Mars 1975)
C.E.T.E. de BORDEAUX "Programme CERA: calcul électronique des réseaux

d'assainissement" ("CERA Program: sewerage electronic
computing"). C.E.T.E. 472, av. du Maréchal Delattre de
Tassigny, BORDEAUX-CAUDERAN, 40 p. (1973)
C.E.T.E. d'AIX EN PROVENCE

"Programme RERAM" ("RERAM Program"). Ministère de
l'Equipement, Les MILLES, 13, AIX EN PROVENCE, 70 p.
(1971)
ROUSSET M. LORGERE M. "Modele mathématique de simulation des écoulements dans
un reseau d'assainissement" ("A mathematical model for
flow simulation in a sewer network"). LA. HOUILLE
BLANCHE no. 415, p. 363-368 (1974)
SAUCEROTTE H. "Un modele bidimensionnel d'infiltration dans la zone
d'aération. Utilisation de la méthode des éléments
finis" ("Two-dimensional model for infiltration in the
zone of aeration. Use of the finite element method").
Rapport L.H.M. no. 30/70 (Mars 1970)
CORMARY Y., THORE Ph. "Traitement de mesure de succion et d'humidité au moyen

d'un mod'èle bidimensionnel d'infiltration aux éléments
finis" ("Treatment of suction and soil moisture
measurements with the help of a finite element two-
dimensional model"). Secrétariat Permanent pour l'Etude
des Problèmes de l'Eau, 14 bd. du Général Leclerc,
92521, NEUILLY-SUR-SEINE, Congres de READING (G.B.)
(Juin 1973)
CORMARY Y., THORE Ph. "Un modele monodimensionnel d'infiltration dans la zone
d'aération" ("One-dimensional model for infiltration
in the zone of aeration"). Laboratoire d'Hydrologie
Mathématique, Report L.H.M. 17/73, 30 p. (Avril 1973)
THORE, Ph. "Simulation numérique des phénomènes d'hystérésis"

("Numerical simulation of hysteresis phenomena").
L.H.M. Rapport 26/73 (Septembre 1973)
25
Section 2 Urban hydrological modelling in France
Caquot Formula for Calculating Maximum Discharges
In France, since 1949 the discharges taken into account in the analysis of
storm drainage networks have been calculated by means of a statistical model
developed by Caquot.^ ^' The Caquot model is derived from a formula of the
following type:
Qp = K I u Cv A w
where Qp is the peak discharge in m /sec, I is the average slope in m/m,
A is the catchment area in hectares and C is a coefficient reflecting the degree
of impermeability of the catchment. The parameters K, u, v and w are adjusted as
a function of regional rainfall characteristics and the rainfall frequency
considered. The method of analyzing the phenomenon is similar to the "rational
method" but is complemented and refined with regard to the following three
factors i
(a) consideration of the temporary storage capacity of the catchment area and
drainage network;
(b) consideration of an areal reduction in rainfall in relation to the size of
the surface area of the catchment; and
(c) consideration of the peak discharge in the time of concentration expression.
A simple formula of this type can be used to calibrate correctly small drainage
networks under the condition that the constants related to physical factors have
been experimentally adjusted and that the parameters related to local rainfall
conditions can be adjusted on the basis of direct observations. As noted in
Section 1, the Loriferne Commission is responsible for revision of Technical
26
Instruction C G . 1333.(1' The adjustment values that have been taken into account
by the Loriferne Commission are valid for:
A < 200 ha (area);

0.002 < S < 0.05 (slope); and
C > 0.2 (impermeability coefficient).
Simulation Models
For studies concerning complex catchment areas, basin storm damping phenomena
and circulatory and storage possibilities of meshed networks, simulation models
have to be used to describe the physical processes involved.
Rainwater runoff simulation over a complex urban area comprises three stages:
(a) the rainfall itself, which constitutes the "input" of the model;
(b) the "rainfall-runoff" transformation over the elementary basins (subcatchments);
(c) the composition and propagation of the flood waves from each of the elementary
basins into the drainage network.
Concept of "Pesian Rainfall"
In France, preliminary studies of this concept were carried out by SOGREAHC 5 )
and general descriptions of the results have been published by the writer.(2,6)
This concept was also discussed by Desbordes

(3)
(see Appendix A ) .
i
The construction of typical storms, with their inherent temporal and spatial
problems, is a fundamental factor because storms constitute the input to any
model. Unfortunately, we have found that there has been very little research
carried out on the "storm shower" phenomenon anywhere in the world.
Hydrological Models
In order to convert rainfall intensity curves into discharge hydrographs we
use the single-reservoir linear model in France, for which the characteristic time
27
constant, K, can be calculated from the experimental adjustment made in France by
Desbordes
(3)
(see Appendix A ) .
We also use a "Muskingum-Caquot" model recommended by SOGREAH for small
elementary basins, with a damping factor in the Muskingum equation of X = 0o2 and
a response time, K, adjusted in such a manner that, for a given frequency, the
peak discharge of the hydrograph is equal to the discharge calculated from the
Caquot formula. Through this adjustment of the simulation model, on the basis of
a statistical formula, it is possible to achieve design homogeneity in the
drainage network between the Caquot formula, which serves for sizing the drains
within the elementary basins, and the simulation study, which serves for sizing
the network draining the inflows from these various elementary basins.
Propagation and General Models
Discharge propagation in main drains can be simulated perfectly by
hydrodynamic models based on the Barre de St-Venant equations. We feel that such
models are a vital part of our applied research facilities.
In this respect, SOGREAH constructed the CAREDAS model which is valid for
both branched and meshed networks and which can also take into account pressure
flow conditions('' (see Appendix B). The CAREDAS model is made up of five
interconnected programs:
(a) the model program, for preparing and checking the data defining the network;
(b) the Hydurb program, for the rainfall-discharge transformation of elementary
basins, which can be based on a simple method if the elementary basins are
small with respect to the overall catchment area;
(c) the Puma model, which calculates the composition and propagation of the flow
of the discharge hydrographs from the elementary basins;
28
(d) the Convec program, which follows the change In pollution level In the network
as a function of pollution intensity curves from the elementary basins; and
(e) the Exprès program for processing the results and graphically printing them
with automatic plotters.
Model operation is made particularly flexible by the use of intermediary files
which can store information.
When applied to the 100- km2 Seine-Saint-Denis catchment area, northeast of
Paris, which was divided into 180 elementary basins, this model included 400
sections and 2000 computation points. The CAREDAS model is used for structural
design studies of complex highly meshed networks and will be used at a later date
to simulate networks in real-time, on the basis of rainfall and discharge information
collected by a remote-controlled measurement network.
Optimization Model Development
For the case of a branched network draining a basin subjected to progressive
urbanization, S0GREAH has developed the Abac model which associates a development
design optimization program with the flood origin and propagation program'0-' so
that runoff can be controlled via storm reservoirs and pipes.^'^
The Abac program, which depends on a dynamic programming algorithm, has
already been used in the study of a 200-knr suburban basin, which was divided into
33 elementary basins. The network comprised 26 sections to be calibrated and 21
possible sites for storm reservoirs to be sized. As a result of using this program,
it was possible to define the optimum time schedule for the construction work
required for drainage purposes, with due consideration given to future town-planning
projects in the catchment area up to the year 2000, and with a minimum overall cost
target.
29
References
1. MINISTERE DE LA RECONSTRUCTION ET DE L'URBANISME

"Instruction Technique relative a l'assainissement des
agglomerations" no. C.G. 1333 ("Technical Instruction No.
C G . 1333").
Imprimerie Nationale PARIS, 50 p. (Février 1949)
NORMAND D. "Ajustement experimental de la formule de Caquot en

hydrologie urbaine" ("Experimental adjustment of
the Caquot formula in urban hydrology"). LA HOUILLE
BLANCHE no. 4/5, p. 357-361 (1974)
DESBORDES M. "Une étude générale de l'assainissement pluvial urbain"

("A general study of urban runoff drainage"). LA
HOUILLE BLANCHE no. 1, p. 37-43 (1975)
4. DESBORDES M. "Extension du modele de Caquot aux bassins versants

hétérogènes" ("Extension of Caquot1s model to non-
homogeneous catchments"). TECHNIQUES ET SCIENCES
MUNICIPALES L'EAU, p. 223-229 (Mai 1976)
5. NORMAND D. RECOURA J., RENE R.

"Etude statistique des hyétogrammes types" ("A ,,
statistical study of design storms"). Ministère de
l'Equipement, Rapport SOGREAH no. R 10653, 20 p. (1970)
6. NORMAND D. "Modeles pour l'étude du ruissellement urbain"
("Models for the investigation of runoff in towns").
LA HOUILLE BLANCHE no. 3, p. 231-238 (1971)
7. CHEVEREAU G. "Un modele mathématique de propagation des crues et

des polluants dans un système maillé de réseau
d'assainissement" ("Mathematical model of flood and
pollutant propagation in a meshed drainage network").
SOGREAH
8. NORMAND D. "Modele mathématique permettant de simuler la relation

averse ruillellement sur un bassin versant soumis a
une urbanisation progressive". XlIIe Journées de
l'Hydraulique - SHF (1974)
9. DUJARDIN J.M., DARNIGE J. "Planification des aménagements de protection contre

les crues dans un bassin versant soumis a une
urbanisation progressive". XlIIe Journées de
l'Hydraulique - SHF (1974)
30
Appendix A A general study of urban runoff drainage
M . Desbordes
This is an English version of the paper which wasfirstpublished
in La Houille Blanche, N o . 1, 1975.
Permission has been kindly granted by the publisher.
(This paper is Reference 3 of Section 2.)
31
A GENERAL STUDY OF URBAN RUNOFF DRAINAGE
Urban growth makes sewage problems more and more difficult. Since 1972,
the Mathematical Hydrology Laboratory of Montpellier University^1) has been
studying sewer networks better adapted to these problems and new methods of
calculation consistent with their complexity and economic importance.
1. Urbanization and sewage
Rapid urbanization modifies the physical and economic data of the sewage
problem requiring new network approaches.
1.1 PHYSICAL DATA MODIFICATION
Waterproofing of surfaces, the first consequence of urbanization, leads

to rapid increases in peak discharges and volumes of runoff. As a result, the peak
runoff per unit area coming from a fully urbanized zone may be ten or twenty times
greater than those from a rural area of similar size.(2'
Population, convenience and industrial growth lead to an increase in

wastewater discharge rates and volumes and, in consequence, to an increase in sewage
works sizes and costs. These sewage works are more and more necessary as nuisances
of all kinds are generated by this evolution.
Runoff pollution adds to wastewater nuisances, coming as it does from the

washing of pavements, streets and roofs after a time without rain. This pollution
is concentrated in runoff waters at the beginning of storms and may be very high in
large industrial towns.(-*)
1.2 MODIFICATION OF ECONOMIC DATA
Changes in investment costs are mainly the consequence of physical

modifications.
The drainage of growing quantities of all kinds of waters leads to an

important increase in the investments for new sewer networks. Moreover, many old
networks must be reinforced under difficult and costly conditions. So, in France,
about 100 to 200 million Francs per year are used for the extension and creation of
sewer networks. These investments have an annual growth of about 15%.
(4)
The present pollution level of rivers calls for the construction of larger
and more efficient sewage treatment plants which are more expensive. Waste discharge
control is needed, especially at the storm overflows in combined sewerage. Discharge
control may lead to the construction of costly special devices limiting the water
runoff pollution effects or the frequency of storm overflow incidents.
Lastly, a new view of convenience favors a diminution of sewer failure
33
risks and thus an increase of investments independently of urbanization consequences.
These consequences act in the same way that industrial and convenience growth
increase the values of economic parameters of land occupation.
1.3 RESEARCH ON SOLUTIONS
Growth of nuisances and difficulties encountered in the adaptation of

classical sewage systems to this problem might, in our opinion, turn the choice of
urban planning specialists towards particular new separate sewerage systems:
separate systems with storm runoff basins.(-*'
The wastewater system would be calculated from an estimate of probable

water consumption. It would be provided with sewage treatment plants progressively
built. These plants would have specific parts according to the kind of wastewater
(industrial or domestic) handled.
The runoff drainage system would be calculated from an estimate of the

extent of ground waterproofing. It would be provided with storm runoff basins.
They allow a better and more rational working of the system, because they increase
the discharge duration at full sewer capacity (present systems are theoretically
calculated in France, for a discharge at full capacity one time every ten years).
In this way some problems can be solved in reinforcing old systems, problems which
are generally quite costly. Storage basins contribute to the struggle against
runoff nuisances: they limit the effects of pollutant concentrations at the
beginning of storms and allow treatment of accumulated waters if their natural
outlet needs particular protection. Lastly, these basins may be useful in urban
arrangements: they are natural or artificial ponds and may enhance green spaces
in towns. In this case, runoff pollution must be controlled.
This kind of network also has economic advantages. If sewage plant costs
are not included, about 807. of present network costs is used for temporary rainfall
drainage. A separate sewerage system allows better calculation of the wastewater
system and sewage works and their better performance. Storm runoff storage basins
lead to important savings for the same protection level, because sewer dimensions
are reduced below the basins. Some authors' ' think that separate systems with
storage basins may be more economical than combined sewerage independently of their
other advantages, the economic estimate of which is not very easy. Further, the
same storm would cause smaller damages with that kind of system than with a classical
one.
However, the calculation of sewer systems with storage basins could not be
accurately done using the present method included in C.G. 1333 National Technical
Instruction (1949). This method is an adjustment of the so-called rational method
and has been developed by M. Caquot. It permits determination of only a given
frequency peak runoff at the outlet of an urban catchment, whereas the calculation
of a system with storage basins requires knowledge of the runoff hydrograph at each
point of the network. So we have been studying, since 1972, new calculation methods
founded on a better understanding of urban hydrological phenomena. The complexity
of hydrological processes has turned our research towards conceptual mathematical
modeling. The chief difficulty is that networks should be economically calculated
34
in terras of risk of failure because rainfall is random. However, the first research
results make possible new methods better adapted to urban evolution conditions.
These results are summarized hereinafter.
2. A study of the validity of M. Caquot's model
For the first time, we have studied the validity of Caquot's model and
have made a new numerical values adjustment of its parameters. We have experimentally
tested the quality of this adjustment.
2.1 FORMULATION OF THE MODEL
Conservation of runoff volumes during the period between the beginning of

rainfall and the time to peak runoff at the outlet is given by Equation 1 in which:
V;*HAC = [8(tl + t2) + mQP (1)
H = rainfall (mm) during times 0 and 8, at the storm center

A =» catchment area (ha)
C =» runoff coefficient (HC = net rainfall)
or =* coefficient for spatial rainfall distribution = A - 8
t^ = water traveling time in network (minutes)
to — water traveling time on the ground (minutes)
t
l+t2 = wa
tershed time of concentration = t c
Q = peak runoff at the outlet (nrfys)
C(t]^+t2)Qp — water volume stored on the ground and in the drainage system
during times 0 and 0.
P0Q = water volume that has flowed at the outlet during times 0 and
0.
For t^+t2 = 8 = t c , Equation 1 becomes Equation 2. H/t c is the mean

rainfall intensity over t c . For a given return period T, the mean maximal intensity
for each time t may be given by Equation 3; so Equation 2 becomes Equations 4 and 5.
1
aCA T- =QP (2)
6(ß + 5) " t,
/(/, D = a 00/"Cl (3)
aC/1
QP(T)= ¿"(ßVs) ''""•* ( 'c , r ) (4)
35
ß» (r) = 6(ß + S)C " {T) 'cHT) A1~C <5)
tt = \iI'A*QJ=tl + ta (6)
M. Caquot has given the theoretical relation, Equation 6, for the time of
concentration in which \i is an adjustment parameter and I the mean watershed slope
in m/m. Equations 5 and 6 lead to Equation 7. So when the numerical values of the
nine parameters (a, b, c, d, f, u-, ß, Ç, e) are known, Caquot's model may be
summarized by Equation 8. This relation may be shown by way of an easily used
linear diagram when the physical watershed parameters are known without ambiguity.
1/(1-6/)
yn m
u - F -* 1 X IMa
av
( M ~"
n
1 £, X
Cini W
( 1 i > / )~ m
K
» ' ~ |_6(ß + 5)_| x^ + - ' - '
op (7') = /C (T) X '" i r ) X C W X ^,<,(T) (8)
For simple networks, only requiring a knowledge of peak runoff, Caquot's

model seems useful if its hypothesis is objective and its parameters are well
estimated.
2.2 MODEL HYPOTHESIS ANALYSIS
Equation 4 shows that this kind of model is an adaptation of the rational

method (1851) used in the United States and given by Equation 9.
Q,iT) = CXHte,T)XA (9)
This method is based on the time of concentration concept. This concept

assumes that peak runoff is observed at the outlet only when the whole watershed has
given its contribution to this peak; that is to say, under a storm the duration of
which is at least equal to the time of concentration.
This method assumes linearity of the rainfall-runoff transformation. It

involves the supposition that the catchment response, under a storm having any
shape and lasting less than tc, must be the same as the response under a rectangular
storm of the same duration and the same volume. In the Caquot model this linearity
hypothesis appears with the use of constant values of parameters ß and Ç.
However, M. Caquot has made three modifications to Equation 9. First, he

assumes a damping of the intensity i (t c ) with the temporary storage of water on
the ground and in the sewer system. This damping effect is expressed by the
parameters ß and g. Then he has taken into account spatial rainfall distribution
effects by means of a simple expression (parameter c). Lastly, he has included an
36
interesting non-linear element, assuming that the time of concentration is n o t a
constant but is a function of peak runoff. T h u s , the model is pseudo-linear.
T h e model also has a probabilistic hypothesis because it is assumed that

mean rainfall intensity and peak runoff have the same frequency.
The deterministic hypotheses of the Caquot model probably are near

reality if the watershed's physical characteristics are homogeneous (slopes and
runoff coefficients). They may lead to good peak, runoff estimates if the numerical
values of the nine model parameters are well estimated. However, as the rainfall-
runoff transformation is non-linear, the numerical values of these parameters are
dependent, in o u r view, on watershed dimensions. S o , it may b e useful to give the
applicable limits of each parameter's numerical adjustment. Particularly, the
results of our research seem to indicate that this kind of model could not b e used
in the case of important watersheds (heterogeneity of physical characteristics).
The probabilistic hypothesis implies that the runoff coefficient C is

not a random variable. Our work has shown that, in urban space, the coefficient C
may be estimated by the percentage of imperviousness when it is more than 2 0 % .
So, in urban hydrology, the runoff coefficient seems to b e a deterministic variable
consistent with the Caquot model probabilistic hypothesis.
2.3 NUMERICAL ADJUSTMENT AND VALIDITY DOMAIN
[X = 0,93; C = —0,363; ¿ = 0,366;

/ = — 0,2; ß + 5 = 1,5; E = 0,1.
We have given new estimates for the model parameters. Using the works
of Izzard.(7) Hicks(8) and Barnes
(9) on overland flow and flow in pipes and sewers,
and the results of numerous experiences, w e have determined a n e w time of
concentration expression (Equation 1 0 ) .
tc (ran) = 0,65 X I~0M (m/m) X A 0 * 1 (ha)
Xß p -°' 29 (m 3 /s) (10)
The relation given by Equation 10 increases the t c v a l u e given by M.

Caquot by more than 20 to 8 0 % , a value otherwise giving too high a v e l o c i t y . ^ 9 '
Analyzing simple observed hydrographs from small SOGREAH urban

catchments,i 1 0 ^ w e have established that ß and £ parameters are greatly changing
even for the same watershed (non-linearity). T h e mean value of their sum is given
by Equation 1 1 . W e think that the ß+£ value increases with the dimensions of a
watershed.
ß + 8=l,l (11)
37
The analysis of recent experiences^^ has shown that the value for e
of 0.1 was too high. This is so true that we think that the storm traveling over
a watershed may sometimes have the same effect as an intensity increase in the
relation given by Equation 4. Furthermore, the rainfall information used in the
estimates of the a and b coefficients is now so dispersed that it could not allow a good
calculation of the maximum rainfall intensity at the storm center. This information
holds implicity the spatial rainfall distribution effects. For e we give the value
shown in Equation 12.
E = 0,05 (12)
We have made an experimental verification of our adjustment using SOGREAH

data. This verification was made by way of Equation 2, with the numerical values
of parameters given by Equations 10, 11 and 12 and assuming that the runoff
coefficient C is equal to the imperviousness percentage. It has been shown that
our adjustment was much better than those of the C.G. 1333 Technical Instruction,
but rather underestimated peak runoff. This improvement was obtained using a new
time of concentration given by Equation 12. D. Normand has studied this improvement'•",)
and has concluded that it was good for watersheds up to 200 hectares in size. So the
validity domain of our adjustment for homogeneous watersheds would be given by
Equation 13.
'„• = 0,8 tc (12)
A < 200 ha ; 0,2 % < / < 5 % ; 0,2 < C < 1 (13)
For heterogeneous watersheds (mixtures of rural and urban parts with steep
and low slopes, etc.) the model would give too approximate results.
3. Modeling of urban hydrological phenomena
The model last described allows only for the estimation of peak runoff
for small homogeneous watersheds. We are studying new methods based on general
modeling of urban hydrological phenomena encountered in the different stages of a
network calculation. The models under development are:
- A stochastic model of rainfall, the input of the watershed-network system,

allowing for the performance of drainage calculations in terms of risks of failure
for economic purposes.
- A model for rainfall losses. These losses may change in time and space.
- An elementary hydrological model giving the runoff hydrograph at the outlet

of a small watershed equipped with an elementary drainage system. This model would
38
be applied to small homogeneous catchments of an important hydrological entity for
which one wants to design the drainage system.
- A hydraulic model for routing elementary hydrographs through the main sewers
of the entity.
- An economic model for balancing between investments and damages following a

drainage system failure for a, given rainfall event.
Only the hydrological models are discussed hereinafter. Hydraulic models

are numerous and one of the more simple ones could be used when that would be
consistent with the precision level needed for a drainage project. Economic
modeling is theoretically without difficulty, but estimation of damage costs is
rather awkward because, at present, information is poor in this domain.
3.1 STUDY OF RUNOFF PROCESSES
Analysis of experimental data leads to an awareness of the preponderance

of impervious areas involved in urban runoff, so much so that the contribution, of
pervious areas can be neglected when pervious areas are less than 20% of the whole
watershed area.
A necessity for modeling runoff losses has not arisen in present experiences.
However, we think that, under heavy storms and for slightly pervious soils with steep
slopes, the pervious areas' contribution to runoff must no longer be neglected. An
experimental study is still necessary to solve this problem. It may be locally
important when urban space is, for example, below an important rural watershed that
is partly drained by an urban sewer system.
Our research has given a physical meaning to the runoff coefficient, C,

of the Caquot model. By relating C to imperviousness, it would be possible to take
into account for planning purposes the effects linked to different urbanization
hypotheses. For that, it would be sufficient to study the correlations between
imperviousness of soils and types of urbanization. The French Ministry of Equipment
is looking into this question.
3.2 ELEMENTARY HYDROLOGICAL MODEL
Recognition of the relative complexity of physical processes governing

the rainfall-runoff transformation, and the difficulty encountered in the estimation
of hydraulic parameters involved in that transformation, has turned our research
towards a model derived from systems analysis. The watershed is then taken as a
transformation system in which rainfall is the input and the discharge at the
outlet is the output.
We have studied different conceptual linear models commonly used in urban

hydrology. (1) The general shape of experimental hydrographs we have studied have
led us to retain a model with a storage effect only. Such a model results from the
39
association of the equation of conservation of volumes between the input and the
output and of a storage equation whose general form is given by Equation 14. In
s ( 0 = ; £ ^ ( / , ß , o ^ + 2ß,<;,ß,o§ (i4)
this equation S(t) is the storage at time t in the system, i is rainfall and Q the
discharge at the outlet. As we desired that the model may be used for design
purposes on ungaged catchments, we retained a very simple storage equation with the
least number of parameter.:, possible,, This is Equation 15 which, when associated
with continuity, Equation 16, leads to the differential equation of the transformation
process whose general solution is given by Equation 17. The model pulse response,
sometimes called the Instantaneous Unit Hydrograph, is given by Equation 18.
S(t) = KQ(t) (15)
dS{i)/dt = i{t) — Q{t) (16)
Q (/) = Q0 e-V->°»K + i/K /*'(„) e-(t-H)iKdu (17)

J to
/i{f) = (l/K)e-«« (18)
The K model parameter, homogeneous over time, is the lag time between the
mass centers of net rainfall and discharge.
The input i(t) is generally not known in an analytic form, so the numerical
integration of Equation 17, for a separate empty system at the beginning time, may
be done by means of recurrent use of Equation 19, for each time interval dt.
Q (« dt) = e - M Q [(n — 1) dt] + (1 — e"1'*) i (n dt) (19)
In Equation 19, K must be expressed in time interval units.
Calculating the observed K values for the physical time lag definition of
K, and using Equation 19, we have tried to reproduce 55 hydrographs observed on
French experimental catchments.( 10 ^ We have seen that, for each catchment, the
observed value of K greatly changes from storm to storm and must be reduced by 5 to
20% in order to have good hydrograph regeneration (less than 5% error in peak
discharges).
The changes in K, from storm to storm, have led us to study a relationship

between this parameter and the physical characteristics of storms. We have thus
established by means of multiple-regression for 13 experimental catchmentsi 1 0 » 1 ^ the
relation given by Equation 20.
K = 5,07 X ¿°'18 X P~0,36 X [1 + (/MP/100)]-1.»
X TEW X L°-i5 X HPE-ow (20)
40
In this relation, K is in minutes, A is the catchment area in hectares, p is the mean
slope as a percentage, IMP is the imperviousness percentage, TE is the storm duration
in minutes, L is the catchment slope and HPE is the rainfall depth in millimeters.
The variation range for these parameters is given by Equation 21.
0,4 ha ^ A ^ 5 000 ha
2% ^ IMP^ 100%
HOm^ L ^17800m (21)
0,4 % sS p ^5 %
5mn^ TE ^ 180 mn
5 mm ^ HPE ^ 240 mm
Because K changes with the H P E and T E storm v a l u e s , the model is pseudo-linear.
W e h a v e used Equation 20 in reproducing observed hydrographs taken from among

those which have n o t been used for the establishment of this equation. S o , about 8 0 %
of the reproduced hydrographs had a maximum error o f peak discharge less than 2 0 % .
T h e most important errors were associated with very small storms that would b e o f
little interest in sewer network calculations.
The model w e have studied offers advantages because it m a y b e used for

ungaged catchments. Moreover, it allows the simulation o f consequences related to
different urbanization hypotheses. L a s t l y , it allows us to obtain a whole runoff
hydrograph resulting from a given storm, and so to calculate precisely complex
networks with storage basins, pumping stations, e t c . It may b e used for the planning
and management o f important networks: so some cities in the United States h a v e a n
automatic management system, using a computer for their sewage networks.(•"-->) Rainfall
and flow data a r e transmitted to a computer, which foresees the evolution o f flows
in the system during a rain using a runoff and sewer flow modelization, then reroutes
flows in the sewers so as to u s e the full network capacity.
3.3 TOWARDS STOCHASTIC MODELING OF RAINFALL
W e h a v e studied the sensitivity o f the rainfall-runoff transformation o f

the model and rainfall parameters.' '
First, the K. value is only a statistical parameter. I f the sampling error

o f K would lead to an important change in time to peak o r in peak runoff for a given
storm, the model would b e without interest for ungaged catchments. S o , w e h a v e
calculated that for the extreme combinations o f the sample domain (Equation 2 1 ) in
Equation 2 0 , the 8 5 % confidence limit allowed resolution o f K. with a precision less
than 407.. W e have studied cases with K changing from 5 to 150 integration time u n i t s ,
and the effects of incremental change» in K o n the model response to storms very
different in shape and with very different durations. W e have established that a 4 0 %
error o f K leads to changes in peak runoff less than 2 0 % ; that is to say, an error in
pipe liametars less than 8%. These change« a r e smaller for time to peak and o n l y reach
a few per cent. So the model is not very sensitive to the K v a l u e and has a n adequate
precision level for the design of network projects. F o r existing sewers, if o n e could
obtain discharge measurements, the precision level would b e quite a b i t higher.
41
In the second place, our attempts to determine a design storm'l) have not
been very conclusive because rainfall is essentially random. So we have also studied
the model transformation sensitivity to rainfall parameters, in order to estimate their
relative importance in the transformation and to retain the most important ones for a
design storm definition. We have built rainfall with simple geometric shapes
(triangular, rectangular, exponential, . . . ) . Each rainfall event was also known by
its duration, volume and position in time of its maximum instantaneous intensity.
Volume and duration were related by the so-called "intensity-duration-frequency"
curves.
For a large range of K values, we have obtained the following results:
- The model shows that there exists a critical duration B m always smaller than
the classical time of concentration. The model is not very sensitive to B m , which is
the rainfall duration leading to the maximum peak discharge for a given K and a
given storm shape. Beyond B = 3 or 4 hours, the changes in peak discharge or time
to peak may be neglected. The "design storm" could have such a maximum duration.
- Triangular and exponential storm shapes give approximately the same model
responses even with differences in instantaneous peak intensities of more than 50Z.
- Maximum instantaneous intensity position is an important parameter for time

to peak discharge.
- The rainfall time discretization does not have an important effect when the
rainfall duration exceeds 6 to 8 time steps.
- When we superimposed, on the simple geometrical storm shape, log-normally

distributed variances, with very important coefficients of variation (reaching 300%
beyond 30 minutes from each side of the maximum intensity), the maximum changes in
peak discharge were about 5 to 10% for 75% of the 600 simulations made.
Then it would seem that a design storm would be accurately defined by

means of three random variables: the mean maximum intensity during 3 or 4 hours
(hydrograph volume), the mean maximum intensity during a 15-minutes to 1-hour
intense rainfall period (peak runoff) whose duration depends on watershed dimensions
and the time position of this intense rainfall period over that 4-hour duration.
Study of the statistical relationships between these three parameters would allow us
to assemble a design storm stochastic model. It would allow us to make a simulation
of rainfall events over a given period and to analyze statistically the resulting
hydrograph variables. One could give the frequency of a given discharge or volume
independently of the whole rainfall frequency, which cannot be defined.
4. Conclusions
Studies made by the Montpellier University Laboratory of Hydrology in the

urban hydrology domain show that mathematical modeling is an interesting way to
approach research on new calculating methods for networks. Experiences are still
necessary and should give validity limits of models for lightly urbanized catchments,
the part played by pervious areas in runoff and the importance of rainfall time and
space distributions.
42
Bibliography
[1] D E S B O R D E S ( M . ) . — « Réflexions sur les méthodes de calcul

des réseaux urbains d'assainissement pluvial ». Thèse d'Ingé-
nieur-Docteur, 224 p., Montpellier (février 1974).
[2] M O O R E (W. L.), MORGAN (C. W . ) . — «Effects of Watershed
changes on streamflow ». Water Resources Symposium n° 2,
290 p., University of Texas, Austin (1969).
[3] E N V I R O N M E N T A L P R O T E C T I O N A G E N C Y . — Projects of the M u n i -
cipal Technology. Branch through June 1972. E.P.A.R.,
n" 2-72-080.
[4] A F F H O L D E R (M.). — Contribution d'une étude expérimentale à
l'amélioration des méthodes d'évaluation des débits de ruis-
sellement en milieu urbain. Association Française pour l'Etude
des Irrigations et du Drainage. Journée d'études de Colmar,
13 p. (1970).
[5] D E S B O R D E S ( M . ) . — Les réservoirs d'accumulation des eaux
pluviales. Construction (Dunod éd.), n" 1, pp. 19-25 (janvier
1975).
[6] JACOBSEN (E.). — La régularisation des écoulements de pluie.
Etude fréquentielle des pluies. Le Génie Rural, n° 1, pp.
27-31 (janvier 1973).
[7] IZZARD (C. G . ) . — Hydraulic of runoff from developed surfaces.
Proceedings of the 26th annual meeting. Highway Research
Board, vol. 26, pp. 129-146 (1946).
[8] H I C K S (W. I.). — A method of computing urban runoff.
Transactions of the A.S.C.E., vol. 109, pp. 1227-1253 (1944).
[9] B A R N E S (A.). — Flood routing through storm drains. Colorado
State University, Hydrology papers n" 43-44-45-46, Fort
Collins (1970).
[10] N O R M A N D (D.), R E N É (R.). — Etude générale du ruissellement
urbain; Annexes A l et A 2 . Ministère de l'Equipement at
du Logement, D . A . F . U . , Rapport R . 10737 (1971).
[11] O S B O R N (H. B.). — Thunderstorm runoff in south-eastern Ari-
zona. Journal of the Hydraulics division. Vol. 99, n° H Y 7,
pp. 1129-1145 (1973).
[12] D E S B O R D E S (M.). — «Vérification expérimentale du modèle de
Caquot ». Direction Générale des Collectivités locales. Note
L H M 9/74, 31 p., Ministère de l'Intérieur (1974).
[13] N O R M A N D (D). — Ajustement expérimental de la formule de
Caquot en hydrologie urbaine. La Houille Blanche, n° 4 / 5 ,
pp. 357-361 (1974).
[14] S A R M A (P. S.), D E L L E U R (J. W . ) , R A O . — A programm in urban
hydrology, Part II. Technical report n" 9, Purdue University
Indiana (1969).
[15] GiBBS ( V . C ) , A L E X A N D E R ( M . S . ) , LEISER (C.P.). — System for
regulation of combined sewage flows. Journal of the Sanitary
Engineering Division, pp. 951-972 (décembre 1972).
[16] D E S B O R D E S (M.), RAOUS (P.). — «Etude de sensibilité du
modèle de ruissellement ». Laboratoire National d'Hydrauli-
que. Direction des Etudes et Recherches d'E.D.F., Note
C O 45/7/74, 68 p. (décembre 1974).
43
Appendix B • Mathematical model offloodand pollutant
propagation in a meshed drainage network
G . Chevereau
This paper is a special report of S O G R E A H ,
Consulting Engineers, Grenoble, France.
English translation was by M r . D . Normand.
Permission for reproduction has been kindly granted by
S O G R E A H . (This paper is Reference 7 of Section 2.)
45
MATHEMATICAL MODEL OF FLOOD
AND POLLUTANT PROPAGATION IN A MESHED
DRAINAGE NETWORK
Introduction
Due to the fact that expanding urban concentrations are creating larger
and larger impermeable areas, urban drainage engineers are having to face increasingly
complex and urgent problems.
; Numerous Municipalities are having to cope with the now familiar problems
of linking up new estates to an old network, generally of insufficient capacity, or
designing separate networks.
Until the introduction of mathematical models, urban drainage authorities

were badly equipped to tackle these problems and often had to resort to using only
their common sense and experience for deciding what work should be undertaken for
improving stormwater flow in their network. No equipment existed for objectively
verifying the validity of their schemes.
Laying large diameter pipes in the increasingly congested sub-soil under

built-up areas is expensive, and it would be advantageous to know beforehand whether
the proposed work would satisfactorily solve the technical problem or whether less
expensive schemes would have sufficed.
Traditional methods, such as that of Caquot which is used extensively in

France, or the "rational" method, are used for calculating acceptable dimensions of
the smallest pipes of a branched network, but prove to be much too rudimentary for
solving the problems described above.
Progress was made with adoption of Muskingum-type flow calculation methods

which take into account the damping of flood waves due to friction and due to the
staggering of partial waves coming from different pipes. Although such methods give
improved dimensioning of branched networks, their field of application is seriously
limited due to the difficulty in taking into consideration the many particular points
and backwater phenomena and due to the incapacity of the methods for processing
meshed networks and pressure flow. As such, new calculation methods had to be
developed based on the integration of complete equations.
By applying an implicit method for integrating the complete Barré de Saint

Venant equations, SOGREAH developed the PUMA program for calculating free surface or
pressure flow in complex meshed or branched drainage networks. In conjunction with
this program, other programs for calculating the rainfall-to-discharge transformation
47
on subcatchments and the transport of pollution by convection, and additional service
programs, constitute the overall CAREDAS program which is a useful tool capable of
solving all urban drainage problems. (CAREDAS: CAlcul des REseaux D'ASsainissement •
Calculation of Sewerage Networks).
The CAREDAS Program
CAREDAS is a set of five interconnected programs, the general structure of

which is represented in Figure 1. The great flexibility of the program and the
ability to store information is due to the fact that intermediate files are used.
The main program (PUMA) calculates the rainfall-induced flow in the network
from the data contained in the data sets defining the network and the input
hydrographs. As the input hydrographs are generally calculated from a representative
hyetograph of rainfall, this calculation is performed by the HYDURB program, which is
upstream from the PUMA, program in the flow chart.
Downstream from the PUMA, program in the flow chart is the CONVEC program,
used for following pollution movement in a network. CONVEC has an obvious advantage
for the study of networks linked to a treatment plant.
To these three functional programs should be added two service programs

(MODELE and EXPRES) whose essential role is to make it easier for the other programs
to be put into operation. In this way, CAREDAS can be used by unspecialized
operators.
48
/
Network fCatchment basin
description 1 description
data
1 data
CARECASrfnodel
7
CAREÜAS/hydurb
Control
Control
listing
listing
Computation
r
data
Appurte-
CAREDAS/Puma nances
processing
Printed
results
Automatic A
I 'Pollutographs' plotter J
/
CAREOAS/ConwK CAREOAS/Expres
Printed
results
*A Printed
results
FIGURE 1- PROGRAM CARE DAS - FLOW CHART
49
PUMA ProRram
PUMA. is the basic program which enabled CAREDAS to be developed. By taking

into account the complete Barré de Saint Venant equations and using an implicit
integration method, the program is based on rational factors thus ensuring valid flow
simulation. The Barré de Saint Venant equations are as follows:
1. Continuity equation: 1 d(,v)

+ —= o
g 3x
Ot
2. Dynamic equation: 3y 1 3v 1 3w .
— + — v — + + K v |v = 0
3K g 3x g 3t
FIGURE 2-GEOMETRIC TERMS
50
Where (see Figure 2 ) ,
b = width at water surface

s =s discharge section
-4
v = average velocity along the x-axis
y — water elevation
K = friction coefficient which depends on the shape, dimensions and
roughness of the pipes
x = abscissa along the pipe
t = time
These equations do not have analytical solutions but can be solved numerically. In
order to do this, the PUMA, program uses an implicit method of finite differences.
Only an implicit integration method can calculate both free surface and pressure flow:
the calculation stability condition inherent in explicit methods prohibits their use
in simulating infinite wave propagation velocity when the pipes are pressurized.
There are essentially two kinds of data necessary for putting the program
into operation: data describing the pipe network compiled in coded form in the
MODELE file (because of the complexity of meshed networks, this file is produced by
the MODELE program designed for this purpose); and data describing the input
hydrographs in the network for each catchment basin (these hydrographs are generally
calculated by the HYDURB program).
The program can process all appurtenances usually encountered in a drainage

network such as weirs, flap valves, siphons, storage reservoirs, etc. Due to the
program's modular structure, it is easy to add a particular new type of device. The
program can also simulate automatic equipment such as a gate, the opening and closing
of which is controlled by a given water level.
All the calculation results, i.e., water level, discharge and velocity at
each calculation point of the model and for each time step, are stored on magnetic
tape ready for future use. Partial results are printed during the calculation.
HYDURB Program
The HYDURB program is used for rainfall-to-discharge transformation

calculations but is interchangeable and any of the usual urban hydrology methods can
be used. The choice of method depends on factors such as the dimensions of the
51
catchment basins. (The smaller the basins, the simpler the method may be). In most
cases, the "Caquot-Muskingum" method, developed by SOGREAH's urban hydrology department
for the French Ministry of Equipment, can be applied. Using design storm hyetographs
and the definition of each of the catchment basins, the program produces hydrographs
corresponding to each basin by applying the "Muskingum" method such that the may-frami
discharge value corresponds to that given by the Caquot formula. A homogeneous
catchment basin is defined by its area, average slope, length, runoff coefficient
and the name of the basin outlet.
CONVEC Program
Using the velocities calculated by the PUMA, program at each point of the
model and for each time step, this program can simulate the convection transport of
a conservative substance (or assumed to be so as a first approximation) across the
network. At each point of concentration, it is possible to introduce "pollutographs"
or concentration curves as a function of time and to obtain a resultant pollutograph
for a given point in the network, especially at storm sideflow weirs in combined sewer
networks or at storm network outfalls. These resultant pollutographs can be used later
on for a pollution study of the receiving waters (river, canal, sea).
MODELE Program
The implementation of the implicit flow calculation method by the PUMA

program implies that the various sections of the network be arranged in a well
defined order. The automatic determination of this order of calculation is the
essential role of the MODELE program. In general, this program is Intended to
assist as much as possible in the preparation of data defining the network srch
that the program can be made available to users not having a good knowledge of the
calculation method. The network is described section by section and the data
required for each section is as follows; length in meters; names of the points
upstream and downstream of the section (the idea of upstream and downstream Is
conventional in this case: a discharge traveling from a upstream point to a
downstream point is assumed positive); type of cross-section and its characteristics;
the roughness coefficient of the section; invert elevations of the upstream and
downstream points; ground elevations at the upstream and downstream points (optional);
and number of intermediate points in the section. For each intermediate point the
data required includes: its distance from the upstream point and possibly its name;
the type of point if it is concerned with a hydraulic device (siphon, weir, flap
valve, etc.) and, In that event, the characteristics of the device; and the invert
elevations upstream and downstream from the intermediate point if there is an invert
discontinuity. The various sections are given In any order, one after the other.
In this way, the network model can be assembled or modified without difficulty.
In addition, the program assesses the data supplied and an error message
is printed if the data is not consistent. The model is then put Into a coded data
set which improves the performance of the PUMA program.
52
CTPttES Program
This is a service program for presenting the results, stored on a file, in

the form of graphs obtained automatically on a plotter.
Application Example
The CAREDAS program was applied to the combined sever network of the Seine
Saint-Denis department, northeast of Paris, for the "Direction Départementale de
l'Equipement de la Seine Saint-Denis". The model of this complex network, represented
in Figure 3, includes not less than 400 sections and 2,000 calculation points. The
drained area, approximately 100-km , was divided up into 180 unit basins.
One part of the departmental network, in the Montreuil district, is a small

independent network which was handled separately (Figure 4). The main stages of a
CAREDAS standard calculation are shown in chis simple example —
(a). Division of the drainage area into 14 homogeneous catchment areas whose
characteristics were defined (Figure 5). A sketch was then made of the pipe
network to be considered (Figure 6 ) .
(b). Punching of data cards defining the network, intended for processing by the
MDDELE program. This program prints out a copy of information on the cards
introduced.
(c). Calculation of discharge hydrographs by the "Caquot-Muskingum" method from

the standard ten-year frequency rainfall (Figure 7 ) .
(d). Flow calculation using the PUMA, program. The calculation results are stored
on magnetic tape and partial results are printed (Figure 8). An example of the
change in water level with time calculated at a point on the network is given
in Figure 9.
(e). Calculation of pollution propagation from a simple fictitious pollutograph

introduced at each of the basin outlets (Figure 10).
(f). Tracing of resulting curves by automatic plotter.
53
Fig. 3 DEPARTMENT OF SEINE-SAINT-DENIS, SEWERAGE NETWORK
54
WHJ
Me. /
"^>
• î"^*-
• Oil
/
S à #
Y )
J 1
20(yi05
1. fr^\ 1
/Á
t
r 2ûC/!05 _
,^J f
/
MONTREUILI
I
rso
? o 9 ^
I ^ , i rfô
F ^tó^" '- ' ¿? « •
"L-*
N \1
FIGURE 4 - DEPARTMENT OF SEI NE-SAINT-DEN IS, MONTREUIL NETWORK
55
FIGURE 5-MONTREUIL NETWORK-CATCHMENT BASINS
Conclusions
The above application example indicates the capacity and flexibility of

the CAREDAS program.
The preliminary stage of a drainage network study generally consists of

simulating as accurately as possible the operation of the existing network by
adjusting coefficients such as the runoff coefficients of the catchment areas or the
roughness coefficients of the pipes. As soon as the model has been satisfactorily
adjusted, thereby guaranteeing its reliability, it can be used for various purposes:
Comparative study of various possible solutions for improving water

flow for typical rainfalls.
(continued)
56
57
, Rainfall Discharge
(mm) (m3/s)
Hyetograph
|—IX
f \ Resull ant Hydrograph
~1 ^ ^
0h30 lh lh30 Time
FIGURE 7 - RAINFALL-TO-RUNOFF COMPUTATION

Example: Catchment basin No. 1
DOE SEINE SAINT DENIS T 1 HE o H 30 m a s . 1800. S.I. TIME STEP ' 30 S . , CYCLE 60
PIPE U P S T R E A M P O I N T D O U N S T R E A H P O I N T
NO
* NAME Y (HI Q I M3/SI V («/SI SUKCH. FLOOD • NAME Y (MI Q IM3/S) V (M/SI SURCH. FLOOD
•
1 • QB 4 85.444 2.705 3.206 • P B 85.041 2.677 5.560
2 P 8 85.0+1 0.53« 1.316 « OB 8 72.355 6.517 8.821
3
• P 8 85.0«! 2.043 4.242 • QB 3 72.332 1.680 2.140
4
* QB 3 72.88? 1.722 2.193 • QB 8 72.355 1.243 0.815
5
* QB 8 72.355 3.530 2.542 • • P 5 67.637 4.779 4.223
6
* QB 8 72.355 6.376 4.591 » • P 5 67.637 2.933 2.592
7
• QB 3 72.882 2.972 4.726 • P 9 68.537 2.786 3.442
8
• P 9 68.537 2.786 3.578 • OB 9 66.020 2.598 3.796
9
• QB 9 66.020 0.153 -0.172 • AVL5 58.400 6.384 9.002
10
* 08 9 66.020 3.409 2.398 • P 2 63.189 3.111 3.4 76
11
* P 2 63. 189 2.088 2.333 • P 6 62.574 3.079 1.622
12
* P 4 68.837 2.774 2.0S1 • P 10 62.919 2.401 2.680
13
• P 2 63.189 1.02 J 1.781 • P 11 61.182 0.565 0.816
14
* P I 62.935 1.270 1.741 * P 11 61.182 3.651 6.572
15
* P 5 67. 637 2.217 1.959 « P 6 62.574 2.175 1.867
16
• P 5 67.63 7 4.424 3.650 • P 7 66.178 3.664 1.613
17
• P 5 67.637 0.571 0.505 • P 7 66.178 -0.199 -0.143
18
* P 7 66. 178 3.470 2.498 * • 0813 64.610 4.660 3.3 55
IS
• 0913 64.610 8.005 3.533 • AVL1 10.700 2.456 3.8 81
20
• P 6 62.574 2.661 2.282 • P 12 61.S25 0.713 0.939
21
• P 12 61.825 0.713 0.784 • AVL2 60.803 2.207 2.992
22
*0 P 12 61.825 O.COO 3.000 • AVL3 56.&C0 0.050 0.072
• P 13 59.547 2.87« 1.713
23 P 11 61.182 3.085 4.161
2*
* P II 61.182 1.132 2.325 • P 13 59.547 I.015 5.130
25
*m QS 2 62. 1(4 1.7J1 2.476 • P 1J 59.547 1.452 7. 3 M
• AVL4 58.000 6.753 9.022
26 P 13 59.547 5.346 2.556
*
FIGURE 8 - P U M A PROGRAM-EXAMPLE OF PRINTED RESULTS
58
Piezometric
head
(m)
68
67
ground elevation
////**/rj?f
66
65
crown
64
63
invert elevation
62 ' »»
0h3O In Time
FIGURE 9 - FLOW COMPUTATION

Example- piezometric head computed at
point 0813
, i Concen-
tration
10
. Resultant pollutograph
\ ^ - a t point A V L 1
Input pollutograph
at every catchment
basin
0
10 20 30 40 5 0 Time (min)
FIGURE 10- POLLUTANT TRANSPORT
59
. Network extension study to serve new bousing estates.
. Study of pollution movement in a network and at storm sideflow weirs.
. Dimensioning of sewers.
. Dimensioning of flood retarding basins.
. Dimensioning of polluted water storage reservoirs upstream from treatment

plants.
. Network automatic control study. Definition of operating instructions for

manual or automatic control equipment. Development of network operation
programs.
. Operating a proposed network control. Cost comparisons of various alternatives.
60
Appendix C Extension of Caquot's model to
non-homogeneous catchments
M . Desbordes
This is an English version of the paper which wasfirstpublished
in Techniques et Sciences Municipales, L'Eau, M a y 1976.
Permission has been kindly granted by the publisher. (This
paper is Reference 4 of Section 2.)
61
EXTENSION OF CAQUOT'S MODEL TO NON-HOMOGENEOUS CATCHMENTS
1. Introduction
Studies made for the French Ministry of Equipment,(*»^) and more

specifically for the French Ministry of Interior by the Loriferne Commission revising
French Technical Instruction C.G. 1333,(3>^' have allowed for specification of the
validity of Caquot's runoff model. They have led to a good numerical adjustment of
that model for small homogeneous urban watersheds.
The validity domain of the numerical adjustraent,^»^^ for catchments having

a length coefficient equivalent to 2, is presently limited to: 1-ha ^ A ^ 200-ha;
0.002-m/m ^ I ^ 0.05-m/m; and 0.2 ^ C < 1. A is the catchment area, I is the mean
slope and C is the runoff coefficient.
Calculation of an important sewer network by means of Caquot's model must

be done as follows: the watershed is cut into smaller homogeneous catchments; then,
from upstream to downstream, one calculates the discharge at the outlet of catchment
sets whose A, C and I characteristics may be very different. One can be suspicious
of the model validity in that case, because it is a very global model expressing the
particularity of each catchment by means of only these three parameters. So, the
peak discharge Cv,, for a return period T, at the outlet of the j t n catchment of a
set is given by Equation 1. The K, u, v and w coefficients change with the return
period T and the regional rainfall characteristics. When one groups together several
non-homogeneous catchments, what parameters must be retained for the estimate of peak
runoff at the outlet of the set? That is to say, what equivalent mean slope I e q ,
equivalent runoff coefficient C equivalent area A must be retained?
Qp(T).Kiju Cj" Aj" (equ. 1)
We hereafter discuss an attempt at the definition of such equivalent

parameters, a study made for the Loriferne Commission.
2. Study of physical parameters of a non-homogeneous set of catchments
2.1 Relative importance of parameters
Equation 1 is a useful digested form of the Caquot model whose general

expression is given by Equation 2.
be I bd » I - £ , -, ,
"TbT TTbT i.bf ICqU.2l
For a catchment having a length coefficient equivalent to 2 (see hereafter),

we made the following numerical adjustment:
(2)
63
u- a 0.5; c = 0.41; d - 0.51; f = -0.29 (time of concentration); P +• Ç » 1.1 (storage
effect); and « » 0.05 (rainfall spatial distribution effect). The a and b
coefficients are those of the so-called "intensity-duration-frequency" curves. They
are dependent on the return period T.
The works of the Loriferne Commission have shown that, for a ten year
return period, which is commonly used in France for sewer network calculations, the
b parameter had a range given by Equation 3 for France.
_ 0,59 < b s .0,44. (equ.3)
So it is evident that the importance of physical watershed characteristics should

be rated by means of Equation 2, using the numerical adjustment given above.
The most important is C, then A and lastly I. For two watersheds with A, C and I
parameter ratios equal to 5, the influence of a change in C on discharges is twice
as great as for A and four times as great as for I.
2.2 Equivalent area
For catchments in series or in parallel, it seems logical to retain the

equivalent parameter given by Equation 4.
Aeq = V A ] avec Aec^ < 200 h (eqU.4)
It results from volumetric balance of the Caquot model. However, one may note
that the compounded effects of rainfall spatial distribution (« coefficient) and of
the decrease in critical intensities with a lengthening of the tine of concentration,
would lead to an equivalent surface smaller than those of the whole catchment set.
Such cases cannot occur with Equation 1, which is constantly increasing as A
increases, and, in our opinion, are unreal for areas of less than 200 hectares.
Studies of spatial rainfall distribution and of model application for important
watersheds would provide an answer to that question.
2.3 Equivalent runoff coefficient
Previous studies^, *>) have shown that the runoff coefficient C could be
estimated by the ratio of impervious area to the whole watershed area (Equation 5 ) .
r - A IMP iequ.5)
A
Experience is still necessary to determine the precise validity of Equation 5 for

catchments that are not very urbanized (C less than 2071) that have rather steep
slopes when they are subjected to heavy storms. One can refine the C definition
using in Equation 5 impervious areas directly connected to sewers.I ' As the
equivalent area is given by Equation 4, the equivalent runoff coefficient is defined
by Equation 6.
64
, ¿LA IMPJ ZAÍ Ci ,
¿ J
|- —-i- (equ.b)
Z. Aj Aeq
2.4 Equivalent mean slope
Mean slope is used i n the time o f concentration estimate, so one must

distinguish between watersheds i n series and those i n parallel.
2.4.1 Watersheds i n series
Mean slope has a n influence o n the mean maximum intensity leading to t h e

peak runoff. Its estimation is rather awkward as the Caquot model is non-linear,
contrary to the character of the greater part o f hydrological models used for the
same purpose. T h u s , the time o f concentration t c is given by Equation 7 in w h i c h Q p
is the desired unknown variable.
tc = u Ie A d Q p (equ 7)
If one retains the physical meaning o f the time o f concentration, one can give a n
approximate hydraulic definition for the equivalent mean slope.
Let us assume that the longer flow traveling time is composed o f sections
having a length L^ and a slope I. almost constant for each section; let u s
also assume that, a t equilxbrium^ there ale uniform flows over the different sections;
then the total traveling time t c will b e given by Equation 8 .
•Ï-H (equ. 8)
V j is the nean velocity o f the uniform flow for the j t a section; and it may b e
regarded as being roughly proportional to the square root o f I f and then as being
given by Equation 9 .
tc Xh-*±\^.-L^ (equ. 9)
Tit supposes that hydraulic roughness and hydraulic radius slowly change from o n e section
to another. The first hypothesis is quite real. The second is less real b u t may n o t
result i n important errors for small catchments because the mean flow velocity i s
proportional to the square root o f the hydraulic radius. S o , for watersheds i n
series, the œ a n equivalent slope would b e given by Equation 1 0 .
i
\J± (equ. 10)
¿/ÏÏ.
65
A better formula would give a velocity proportional to a slope to a power less
than 0.5. Equation 7 shows that this would lead to an implicit equation,
complicating model use. Moreover, this equation gives a power equal to -0.41 near
the -0.5 power of Equation 9. Lastly, the discharge influence in Equation 7.
remains moderate.
2.A.2 Watersheds in parallel
The equivalent mean slope definition is now quite awkward because watersheds
are hydraulically independent units as far as their common outlet.
In a search for a solution, we will first look at the different A, C and I

parameter combinations, over their variation domain as defined above. In order to
simplify, we will consider two watersheds having respective parameters A-i, Cj, Ii and
A
2> c2» I 2 * Qpl» ccl a n d Qp2» fcc2 a r e their peak runoffs and timas of concentration
when they are considered separately. Two cases are possible and are given by
Equations 9-1 and 9-2.
Cas n°l ; tel ^ U 2 e\ Qpl s Q p 2 (eqU. 9.1)
Cas n'2 : U 1 » tc2 el Q p 1 < Q p 2 (equ.9.2)
Let us look at the first case: t c l ^ t c 2 and Q p l ^ Q p 2 . The time of

concentration inequality in which Q p has been replaced by its expression given by
Equation 2, leads to Equation 10.
ii - Rl : A , / A , c R A . C , / C 2 = RC
'» ' (equ. 10)
Log R A > 1,75 log RI + 1,22 lo5 R C
In the coordinate system (log RA; log RI), Equation 10 represents a straight
line set with RC as parameter. For the present validity domain of the model
numerical adjustment, we have Equation 11.
0,005 < R A < 200 ;0,2 < RC < 1. 0,04 < RI £ 25 (CC|U. 1 1 )
The tci ^ tc2 inequality is satisfied, for a given RC, in the domain part
above the straight line of Equation 10 corresponding to that RC value.
The second constraint, Q ^ > Qp2» leads to Equation 12.
Log R A > ! (bclog.RUlogRC) (eqU. 12)

I _ £ + bd
66
It depends on the b parameter value, and for a ten-year return period and
the maximum range of variation for b (Equation 3) we have Equations 12-1 and 12-2.
l o S R A > . 0 , 3 7 3 logRI_l,54 loir RC (cqU. 12.1)
log RA > .0.248 log ni. l,3B log RC (eqU 12 2)
The inequality of peak discharges is satisfied on the domain part above

the straight lines of Equations 12-1 and 12-2. Equations 10 and 12 limit the
existing zones for Case Number 1. These zones are given in Figure 1»
For Case Number 2, Equation 10 is almost true, and the inequality sense
in Equation 12 is inverted and leads to Equations 12-3 and 12-4.
The domains in Figure 1 show that the RC ratio has a greater influence
on peak discharge than on time of concentration. The points on the straight
vertical and horizontal lines limiting the domains, show the equalities Qpi =» Q p 2
or t c l =« t c2 ; and the intersections of equation lines 10 and 12 with the horizontal
lines passing through RA - 200 and 0.005 and the vertical lines passing through
RI = 25 and 0.04 correspond to the extreme combinations of A, C and I parameters,
as defined above.
Thus the research of an expression for the mean equivalent slope I e q

requires us to find a function given by Equation 13
le., =r(RA,RC,RI,RQ p ) (eqU. 13)
67
which would satisfy the constraints imposed by the Caquot model's formulation and
hypothesis. These constraints are given by Equations 14 to 17.
Q n ! = Kxlet| , '«Cecj , xAe^«a (eqU. 14)
Ma,(Qpi,Qp2)< Q T I J (equ. 15)
Q T 1 , < Qpi-Qp2 (equ. 16)
U T ( j î s Mín(ici,te?) (equ. 17)
In Equation 14, for watersheds in parallel, Q e q and A e q are given

by Equations 6 and 5; and or is a correction factor taking into account the
watershed set's length. Equations 15 and 17 correspond to physical aspects of runoff.
Q T 1 2 ^ s t*ie P e a k discharge of the set, and t c T 1 2 ^ t s t^-me °^ concentration. A
more*precise method in the research of Equation 13 would come down to the following
elements: verification of constraints (15) to (17); calculation of I e q by Equation
14; and research, by multivariate techniques, of an expression such as 13 in which
RA, RC, RI and R Q p are the extreme ratio of the catchment set's A, C, I and Q p
parameters.
The main difficulty remains in the first element. Its verification

supposes existing design storm models. They are not very numerous and have not been
experimentally tested. One can search for a simpler relation if one observes that
I _ must be homogeneous to a slope, and that the main element in the QT value is
the relative importance of peak discharges of each catchment. Thus, in a first
approximation, we propose Equation 18.
leg• EljQpi (equ. 18)
This equation must of course verify the relations 14 to 17. Equation 14

becomes 14.1.
QT, a = K , i»^» / C . A . - C 2 A * )VX(A,. A,)"» g iequ. 14.1)
Equation 18 leads to Equations 18.1 and 18.2.
(C(
Uc, - l,x(l*RI , R Q ) / ( U R Q ) 1U- 1 8 - "
= i, x ( i • (I/RLRQ))/(UI/RQ) (equ. 18.2)
The constraint 16 combined with 18.2 leads to Equation 19.
s
° V. RIXRQ/ \ u / V R Q / u Ö V. RAx R C / " M RA' u s
OL '
68
The verification of constraint 15 requires us to study two possible cases;
Qpl ^ Qp2 a n d Qp2 ^ Qpl» T n e first case leads to Equation 20.
lo
«( u TiïW) >Ua ( U W^ , ^( U SS™) + J ! S S , 0 * ( U S ï ) ^ , o a ^ û-2°)
One may observe that Equations 19 and 20 are consistent. Using Equation
18.1 in constraint 16, we have Equation 19.1. The second case gives Equation 20.1.
Equations 19.1 and 20.1 are also consistent.
logiURl X R Q ) * (l*i) log ( U R Q ) _ £ I O S (URAx BC) . 1£L log (l * RA) + i log ELL (eqU. 19.1)
lo S (uRlxRQ)»lo S (uRQ).^lo S (t*RAxRC)*2I¿!¡iIo S (uRA) + l.lo 3 ^L (eqU. 20.1)
For the last constraint, 17, as t c l is always greater than tc2

(Equations 9.1 and 9.2), we must verify Equation 17.1 using Equations 7.1 and 7.2.
UT>ic, (equ. 17.1)
K = tVe<T*(Al'A2)d*QTf (equ. 7.1)
te,. p 2 l2c d
A2 Q p ; (equ. 7.2)
The combination of Equations 17.1, 7.1 and 7.2, supposing that M-eq is
not very different from M<2 (catchments lengthening function), gives Equation 21.
ios hi < ¿.er lo* (i • R A ) + _ L ios ( i + R Ã » Re) + r(i-bf) t CL_ , qu 2 n

•* 1 1 | c | ^v ' |c| -^ ' |c| Ä
a2 vï /
One could show that Equation 19.1 can also be written as Equation 19.2.
U g i^..;llElog(uRQ)-Llo s (l>RA»RcK £ - b d log (uRA) . i ^ l log g l (¿qU. 19.2)
If Equations 21 and 19.2 are of the same type then one must verify the
most obvious one. Now Equation 19.2 is lower in rank than Equation 21 and so we
have Equation 22.
loS(l,RA«Rc);> log(wRQ)+ lo5(l*RA)*(t.br)logfli (equ. 22)
This last equation may be approximated by Equation 23 because c is small

and the <* ratio in 22 not very important.
I03 (URAxRC)> log ( U R Q ) (23)
69
Thus Equation 23 becomes Equation 24, which will be verified if Equation
10 is verified.
Inj Al Id | 11 , f , CI l*br | . ga,

Ä Ä Ä
A* d-f.-L. 12 d.tv.í. C 2 £.b(d.r) ct2 feau ->4)
V C U
ibi ibi H - -tí
In short, formula 18 for the mean equivalent slope will be adequate for
most cases if Equations 10 and 12 and constraint 15 are simultaneously satisfied
over the variation domain of the A, C and I parameters. We have studied the
validity of Equation 18 at different points in the domain given by Figure 1 and
particularly at the extreme point A to L. For them, constraints 15 and 17 are always
satisfied» Constraint 16 is not satisfied for point J to L but the differences are
rather small«, The peak discharge Q<£ i 2 ^ c r these points is slightly greater than
the peak discharge sum of the two catchments studied separately,, The difference does
not exceed 107.« Mareover these points correspond to extreme and not very frequent
situations. The approximative formula 18 would so be used for non-homogeneous
catchments in parallel, in the present limits of Caquot's model validity.
2.5 Length coefficient
The numerical adjustment proposed in paragraph 2.1 above, supposes

catchments with length coefficients equal to 2. For other values, one must increase
or decrease peak discharges by the use of a correction factor. The length coefficient
E is given by Equation 25 in which L is the flow length corresponding to the longest
E.L/»/Ä" (equ.25»
water traveling time over a watershed. We have s h o w n ^ ' that the time of concentration
may be given by Equation 26.
'-^""'""(^•f'^-) .cq...26l
In order to have a monomial form for t c , we look for a relation such as

Equation 27.
..ap.A!Jl.i„E).(Av7Qp,-"),<í,.ríE) .equ.27|
Thus the p- parameter of the time of concentration is given by Equation 28.
P(C|~0,243, EKIJÍE) (equ.2&l
For E equal to 2, we have shown< 2 ) that T(E) was equal to 1.34 and 8(E)
to 0.096. However, if we retain for c, d and f, the numerical values estimated
when E is equal to 2, Equation 27 must be consistent with Equation 29.
Êí^.^.^^íEjÂ-'/op'")— leqii. 211
70
For 1 to 200 hectare areas, 0.35 to 60-m3/s discharges, and 0.5 to 6 length
coefficients, we may approximate 11(E) by Equation 30. Then we have Equation 31.
W E ) - 1.49 E-*•' lequ 301
p(E),0,3€l E*'* (cqu.31l
As V> is proportional to the 0„84 power of E, and peak discharge is

proportional to the (b/1 - bf) power of M<, the discharge Q p given by Equation 1 must be
multiplied by a correction factor given by Equation 32.
'nr lequ.321
•<*)"
m*
However, if it is not difficult to estimate the length coefficient for
watersheds in series, L being well defined, it is not the same thing for watersheds
in parallel. We once more consider that one must retain the relative importance of
peak discharges when catchments of a set are considered separately. The length
coefficient of the set will be so calculated, taking as equivalent length that of
the catchment having the greater individual peak discharge. Equation 33 gives the
length coefficient for watersheds in parallel.
L(Qp. MAX)
(cqu.33l
3. Conclusions
Use of Caquot's model may be made, in the present validity limits of its
numerical adjustment and for non-homogeneous catchments, using the formulae given in
the recapitulative table at the end of the text.
In most cases, these formulae will give results consistent with the
precision level required for sewer planning projects.
Basins in series Basins in parallel
A
•q
Z'i r*j
c•q Iej*j / Z A
j Z'/j'E A
J
ï
•q «z-'^yîK ZVV'ZV
E
E"-J/V¿*J MQ„K*X) / y Z*j
•q
71
Bibliography
[l] N O R M A N D D . — « Etude générale du ruissellement
urbain ». Ministère de l'Equipement. Rapport S.O.G.R.E.A.H.
R.10943, 25 p. (1971).
[2| D E S B O R D E S M . — « Réflexions sur les méthodes de
calcul des réseaux urbains d'assainissement pluvial ». Thèse
d'Ingénieur Docteur — Université des Sciences Montpellier,
224 p. (février 1974)
[3\ D E S B O R D E S M . — « Vérification expérimentale du
modèle de Caquot ». Ministère de l'Intérieur D.G.C.L. 30 p.
(mars 1974).
[4] N O R M A N D D . — « Confrontation entre la formule mise
au point par le L . H . M . et les mesures effectuées par
S . O . G . R . E . A . H . ». Ministère de l'Intérieur D.G.C.L. 30 p. (mars
1974)
[5| D E S B O R D E S M . — ci Extension du modèle de Caquot
aux bassins versants heterogenes ». Ministère de l'Intérieur
D.OfC.L. Rapport 9/1975, 48 p. (septembre 1975).
[6| D E S B O R D E S M . — « Estimation des coefficients de
ruissellement urbain ». Centre Belge d'Etude et de Documenta-
tion des Eaux. Tribune du CEBEDEAU n° 376, p. 106-110
(mars 1975).
72
Urban hydrology studies and
mathematical modelling
in the Federal Republic of Germany
H . Massing
Landesanstalt für Wasser und Abfall,
Northrhein-Westfalen
Postfach 5227
D-4000 Düsseldorf 1
Federal Republic of Germany
Section 1 General information
Purpose
The National Committee for the IHP of the Federal Republic of Germany
decided at its meeting in November, 1975, to establish a national working group for IHP
Project No o 7, and requested the author to act as chairman and undertake the
preparatory work.
As expressly requested by the IHD Subgroup and emphasized by the Warsaw

Workshop, this paper is intended for practioners in urban hydrology, to give them
a concentrated survey of available research results. It is hoped that this will
facilitate translation of scientific results into actual practice and communicate
unsolved practical questions and problems to researchers.
Procedure
Due to requisite organizational preliminaries and the very close timing,

only about five months were available for assembling this Technical Memorandum. In
view of the scope and diversity of the themes involved, this time was too short for
even an approximately faultless and complete survey. It is hoped that in the course
of the further work under IHP Project No. 7 other experts from the Federal Republic
of Germany will have an opportunity to suggest corrections and to offer additional
information.
The U.S.A. Technical Memorandum in its final draft of November, 1975, has
been available since January, 1976 (McPherson, 1975). Between December, 1975, and
February, 1976, the writer requested about 40 experts and institutions to submit
their contributions to the present report. In most cases the response was very
positive. A rich fund of material in the form of personal notes, citations of
literature, documentary information, replies to questionnaires, etc., was made
available. The first draft was submitted by the writer in July, 1976, to more
than 40 experts and institutions with a request for comments and suggestions for
amplification and revision. After an intensive revision of the manuscript on the
basis of the numerous recommendations received, the present version of the Technical
Memorandum was resolved early in September, 1976. Unfortunately, there are still
information gaps in the scope of knowledge circumscribed by the original outline,
because sufficiently comprehensive material is not yet available and will have to
be procured as part of later work for the IHP.
Background
After World War II, "urbanization in the Federal Republic of Germany has
progressed more intensively than in other industrial countries" (Hillebrecht, 1976).
In 1910j already 26.64 million inhabitants, i.e., 45.57. of the residential

population, lived in 24 urban agglomeration areas. 55.4% of all places of
employment were located in these areas. The agglomeration areas covered
7.3% of the territory of the Federal Republic of Germany.
75
Ever since man ceased being a nomad, he has interfered with the cycle of
water, energy and materials between the atmosphere and the earth (Plate, 1976).
None of his measures (e.g., clearing for cultivation) had such aggravating local
and regional consequences as urbanization (Zayc, 1973; Kalweit, 1976). The
complexities of the water cycle of an urban area are suggested in Figure 1.
Present-day man witnesses an exorbitant escalation of interferences which

threaten to lead to permanent adverse effects on the global water budget. At the
same time, and as never before, he has the means to document interferences and
consequences and to assess their development, for example: by investigations and
research, with the aid of new methods and modern equipment, into the actual water
resources situation (e.g., National Committee for Geodesy and Geophysics of the
Federal Republic of Germany, 1975), discussed in Section 2 herein; and by models
simulating processes taking place in the environment for the purpose of, for
example, determining causalities and estimating future events (e.g., Hahn, 1972;
Lecher, 1968), discussed in Section 3 herein.
In recent years, extensive urban hydrology research has been undertaken

in the Federal Republic of Germany. An indication of these activities is
illustrated by the conduct of two important national meetings on subjects related
to the "Effects of Urbanization on the Water Cycle": German Hydrological Conference,
September, 1972, at Duisburg (Ruhr industrial region) (cf. Deutsche Gewässerkundliche
Mitteilungen, special edition, Koblenz, 1973); and Convention of the German Water
Management Association (Deutscher Verband für Wasserwirtschaft e.V.), September,
1975, at Wiesbaden (agglomeration area at the confluence of the rivers Rhine and
Main) (cf. Wasserwirtschaft, Vol. 66, No. 1/2, January/February, 1976).
To compensate for the consequences of residential agglomerations in the

water management field, a number of far-reaching and farsighted measures were taken
early. Typical examples from the catchment area of the Rhine River are: measures
concerning the public water supply and the recreation potential at the Ruhr river,
and drainage and receiving stream capacity in the Emscher region, as taken by the
Ruhrverband and the Emschergenossenschaft, which are the most important water
management associations in the Ruhr region (Kneese, 1972; Ko enig, Imhoff, 1973;
Emschergenossenschaft and Lippeverband, 1975); and integration measures in the
Rhine/Main agglomeration area (Hammel, 1976).
In water management, the questions of water quantity and water quality

are interrelated. Both aspects are therefore addressed in this Technical
Memorandum.
76
77
Notes
German-language papers and studies produced in the Federal Republic of

Germany were used almost exclusively as a base for the present report. From the
large number of hydrological reports on investigations, research and practical
measures, only papers dealing with the influence, effects, or consequences of
urbanization, or being at least directly related to these topics, were taken into
account in the assembly of this Technical Memorandum.
To some extent, recourse was made to computer-stored literature (Schmidt,

K., 1973).
Acknowledgment
The following water management experts assisted with advice and information,
and as authors, in the compilation of this Technical Memorandum:
Wolfgang F. Geiger of the firm Dorsch Consult, Munich

Johannes Mühlbauer of the firm F. H. Kocks KG, Düsseldorf
Friedrich Wilhelm Renz, of the Ruhrtalsperrenverein, Essen
Erich Treunert, of the Landesanstalt fUr Wasser und Abfall
Nordrhein-Westfalen, Düsseldorf
The following experts edited parts of the Memorandum:
Heinrich Kalthoff, of the Landesanstalt für Wasser und Abfall

Christian Vorreyer, of the Landesanstalt für Wasser und Abfall
The following persons and institutions provided a great variety of

information for the Memorandum:
1. Abwassertechnische Vereinigung e.V., Bonn

2. Bernhard Anderl, Emschergenossenschaft, Essen
3. Bertold Baitsch, Institut für Städtebau, Bodenordnung
und Kulturtechnik, Universität, Bonn
4. Heinz-Christian Baumgart, Lippeverband, Essen
5o Wolfgang Bischofsberger, Institut für Wasserwirtschaft und
Gesundheitsingenieurwesen, Technische Universität München
6. Herbert Billib, o. Professor em., Hannover
7. Botho Böhnke, Institut für Siedlungswasserwirtschaft,
Technische Hochschule, Aachen
8. Bundesministerium des Innern, Bonn
9. Franz Fischer, Ingenieurbüro F. Fischer, Erftstadt-Liblar
10. Hermann Flohn, Meteorologisches Institut, Universität, Bonn
11. Klaus Haberer, Stadtwerke Wiesbaden AG, Wiesbaden
12. Hermann H. Hahn, Institut für Siedlungswasserwirtschaft,
Universität Karlsruhe
13. Gerhard Holfelder, Dr. G. Holfelder Beratende Ingenieure GmbH,
Freiburg
14. Klaus R. Imhoff, Ruhrverband, Essen
78
15. Hans-Hinnerk Johannsen, Deutscher Wetterdienst, Offenbach
16. Karl Hofius, Secretariat for the International Hydrological
Programme, Koblenz
17. Hans-JUrgen Karpe, Institut für Umweltschutz und Umweltplanung,
Universität, Dortmund
18. Reiner Keller, Geographisches Institut I, Universität, Freiburg i.Br.
19 o Jürgen Keser, Institut für Wasserwirtschaft, Hydrologie und
landwirtschaftlichen Wasserbau, Technische Universität, Hannover
20. Länderarbeitsgemeinschaft Wasser (LAWA), Mainz
21. Kurt Lecher, Institut für Wasserwirtschaft, Hydrologie und
22. Hans-JUrgen Liebscher, Bundesanstalt für Gewässerkunde, Koblenz
23. Dieter Londong, Einschergenossenschaft, Essen
24. Ulrich Maniak, Leichtweiß-Institut für Wasserbau, Technische
Universität, Braunschweig
25. Rolf Mull, Institut für Wasserwirtschaft, Hydrologie und
26. Wolfgang Neumann, Institut für Wasserwirtschaft und Gesundheits-
ingenieurwesen, Technische Universität, München
27. Heinz-Dieter Olbrisch, Ingenieurbüro Abwassertechnik GmbH, Essen
28» Rolf Pecher, Ingenieurbüro Lautrich und Pecher, Düsseldorf
29. Erich Plate, Institut Wasserbau III, Universität, Karlsruhe
30. Günther Rincke, Institut für Wasserversorgung, Abwasserbeseitigung
und Stadtbauwesen, Technische Hochschule, Darmstadt
31. Franz Roloff, Wetteramt Essen
32. Gregor Scheich, Landasamt fUr Immissions- und Bodennutzungsschütz
Nordrhein-Westfalen, Essen
33. Heinz Schiller, Bayerisches Landesamt fUr Wasserwirtschaft, München
34. Karlheinz Schmidt, Institut für Wasserforschung der Dortmunder
Stadtwerke AG, Dortmund
35. Klaus-Wilhelm Schulze, Firma F.H. Kocks KG, Hamburg
36. Friedhelm Sieker, Institut für Wasserwirtschaft, Hydrologie und
37. Volker Stalmann, Lippeverband, Essen
38. Klaus-Jürgen Ueker, Fachverband Steinzeugindustrie e.V., Köln
39. Peter Unger, Ingenieurbüro R. Sonnenburg, Hungen
40. Gerhard Werner, Institut für Umweltschutz und Umweltplanung,
Universität, Dortmund
79
References
Emschergenossenschaft und Lippeverband, 1975. Die Wasserwirtschaft im

Emscher- und Lippegebiet (Water Management in the Emscher and Lippe Area).
Selbstverlag Essen, 31 p.
Hahn, H.H. 1972. Operations Research und seine Anwendung in der Siedlungs-
wasserwirtschaft (Operations Research and its Application to Urban Water
Management). Series "Wasser und Abwasser in Forschung und Praxis", Vol. 5,
Erich Schmidt Verlag, Bielefeld, 331 p.
Hammel, H. 1976. Wasserwirtschaftliche Probleme im Ballungsraum Rhein-Main
(Water Management Problems in the Rhine-Main-Area). In: Wasserwirtschaft,
Vol. 66, No. 1/2, pp. 30-36.
Hillebrccht, R. 1976. Entwicklung der Urbanisierung in Deutschland (Urban
Development in Germany). In: Wasserwirtschaft, Vol. 66, No. 1/2, pp. 1-7.
Kaiweit, H. 1976. Auswirkungen der Urbanisierung auf die Wasserwirtschaft
eines großen Flußgebietes - Modell Rhein (Effects of Urbanization on
the Water Resources of a Large River Basin - Rhine Model). In: Wasserwirt-
schaft, Vol. 66, No. 1/2, pp. 14-24.
Kneese, A.V. 1972. Die Wassergütewirtschaft - Wirtschaftstheoretische Grund-
lagen, Technologien, Institutionen (Water Quality Management - Theoretical
Economic Basis, Technology, Institutions). Verlag R. Oldenbourg, München
und Wien, 328 p.
Koenig, H-W., Imhoff, K.R. 1973. Wassergüte- und Wassermengenwirtschaft
an der Ruhr (Water Quality and Quantity Management in the Ruhr Area). In:
Gas- und Wasserfach, Vol. 114, No. 9, pp. 406-412.
Lecher, K. 1968. Die Simulation als Teil der Unternehmensforschung
(Operations Research) zur optimalen Lösung komplexer Wasserwirtschafts-
systeme (Simulation as a Part of Operations Research for the Optimal
Solution of Complex Water Management Systems). Mitteilungen aus dem
Institut für Wasserwirtschaft, Hydrologie und landwirtschaftlichen Wasserbau,
Technische Universität Hannover, Vol. 13, 150 p.
McPherson, M.B. 1974. Hydrological Effects of Urbanization. Series Studies
and Reports in Hydrology, No. 18. The UNESCO Press, Paris, 280 p.
McPherson, M.B. 1975. Urban Hydrological Modelling and Catchment Research
in the USA. ASCE Urban Water Resources Research Program, Technical
Memorandum No. IHP-1, November 1975. American Society of Civil Engineers,
New York.
National Committee for Geodesy and Geophysics of the Federal Republic
of Germany. 1975. National Report on Hydrological Research 1967-1974.
Koblenz, 68 p.
Plate, E.J. 1976. Auswirkung der Urbanisierung auf den Wasserhaushalt
(Effects of Urbanization on the Hydrological Cycle). In: Wasserwirtschaft,
Vol. 66, No. 1/2, pp. 7-14.
80
Schmidt, K. 1973. EDV-Literaturspelcherung - Ein Hilfsmittel für die
Forschung (Computerized Storage of References for Research Purposes).
Veröffentlichungen des Instituts für Wasserforschung GmbH und der Hydro-
logischen Abteilung der Dortmunder Stadtwerke AG, Vol. 18, Dortmund, 35 p.
Zayc, R. 1973. Die Einwirkung der Verstädterung und Industrialisierung auf
den Wasserkreislauf (The Influence of Urbanization and Industrialization
on the Hydrological Cycle). In: Deutsche Gewässerkundliche Mitteilungen,
special edition, pp. 7-12.
81
Section 2 Hydrological studies in urban catchments
Introduction
Characteristics of Urbanization and Urban Hydrology.
The social process of urbanization, as far as it affects hydrology,

manifests itself mainly through the following symptoms: high local population
density; intensified industrial and trade activity; changes in ground surface;
heavily increasing water demand; increased energy consumption; physical and
chemical changes in the quality of surface and subsurface waters; air pollution;
great need for protection from natural phenomena (e.g., floods); need for the
disposal of increasing quantities of waste of all kinds; and recreational
requirements to be met by surface waters.
Urbanization affects all phases of the water cycle in settlement areas,

with far-reaching changes taking place in precipitation, evaporation, évapotrans-
piration, infiltration and runoff. Urban hydrology is that part of the
comprehensive field of hydrology which deals with effects and phenomena in human
settlements.
Extent of the Studies, and Relating Remarks.
This section is intended to provide an insight into the hydrology of

urban catchments, and covers measuring and research instruments, investigation
methods, assessment techniques and research results. The extent to which models
are used for the description of hydrological processes in urban areas is discussed
in Section 3. However, the boundary line between the uses of models and other
methods may come close to vanishing as, for example, in the case of precipitation/
runoff computations. Further, the treatment of the various partial aspects of the
urban water cycle covers both water quantity and quality. Possible large-area
influences of urbanization on supraregional water balances, e.g. of the entire
Federal Republic of Germany (Keller, 1971), are not discussed here. Also excluded
are the public water supply and sewage disposal sectors of water management, even
though they represent central socio-economic aspects of urban water problems.
Precipitation
The fact that local climatic conditions are undergoing profound changes
in all urban centers has been corroborated by the results of modern climatic
research (Eriksen, 1972). The local modification of atmospheric processes
affects all climatic elements (Eriksen, 1971).
Precipitation measurement technology is still in need of further

intensive development (Dien, 1967; Otto, 1971). Densely populated areas offer
fewer possibilities than do the open country for the installation of measuring
instruments in a way that will yield results which are representative of the
82
environment. Almost without exception, the measured precipitation is lower (57» to
207.) than the precipitation which actually reaches the ground. Pluviographs even
record 57. to 107. less precipitation than from daily-read rain gauges. Conventional
raingage networks usually are not dense enough to indicate the areal extent of
rainstorms (Aniol, 1972); but results of related studies, undertaken within the
framework of special research sector 150 of Braunschweig Technical University, are
available (Grobe, 1974). This gap will probably be closed by radar measurements
(Attmannspacher, Hartmannsgruber, Riedl, 1974).
In a number of large cities in the Federal Republic of Germany (Kiel,

Bremen, Hamburg, West Berlin, Munich, etc.), comparative precipitation measurements
relative to the rural surroundings have been made. In Kiel, a mean increase of 10%,
and in Bremen of 16%, in the amount of precipitation was indicated (Eriksen, 1972).
In Hamburg, a mean increase of 5% in total precipitation (Flohn, 1973a), and of 10%
in rainfall intensity (Freie und Hansestadt Hamburg, 1976) has been measured, and
the assumed cause is frequent rain showers induced by local heat sources (Flohn,
1973a). For Hamburg, it was found that the probability of heavy rainfall (for 100
cases where 20-mm. or more were measured at least at one station with 24 hours)
rises from 25% to 30% in the environs to 50% to 55% in the center of the city
(Flohn, 1973b). The high pluviograph network density (one pluviograph for about
every 11-sq.km.) of the heavily urbanized territory of the city state of Hamburg
afforded an excellent data source for study (Reidat, 1971a; Reidat, 1971b). Very
comprehensive studies of precipitation conditions were also made in West Berlin
with the aid of 75 gages (Schlaak, 1972), where the "tip of the precipitation
pyramid" was rarely above the center of the city and was usually displaced toward
the leeward side of the prevailing wind direction.
Observations over recent decades have shown that similar phenomena, that
is an increase in amounts and a spatial displacement, are also apparent in snowfalls
(Kienle, 1952).
In the catchments covered by Emschergenossenschaft and Lippeverband,

regional water management associations in the center of the Ruhr region, which is
one of the most densely populated and industrialized areas of the world, extensive
precipitation measurements have been made for decades. A network comprising 71
pluviographs now covers an area of 2850-sq.km. By way of comprehensive computerized
evaluation of pluviograph values obtained over an observation period of 18-years
(about 54300 entries), the parameters required for design and dimensioning for
receiving streams and municipal drainage systems were determined: basic areal
rainfall isolines (for example, 15 minute rain of 1-year frequency) for summer
and winter; areal rainfall; rainfall durations; a frequency diagram and a time
coefficient table (Stalmann, Schoss, Chilla, 1974). In addition, the aforementioned
occurrence of maximum areal rainfall values on the leeward side of areas with highly
concentrated city development was confirmed (Stalmann, 1976). Further evaluation of
this excellent basic material holds promise for development of important new
information on the urbanization/water management complex.
Generally, it may be said that there is a growing tendency in urban areas

to record precipitation events accurately and Intensively over adequate periods
(more than ten years). In conjunction with modern, computerized calculation
methods, a higher information level allows better development and operation of
83
drainage systems (Landeshauptstadt Düsseldorf, 1975). And the development of new
techniques will permit far-reaching optimization of drainage systems and their
operation in the foreseeable future.
Surface Runoff
General.
Apart from the influence on depth and distribution of precipitation

already mentioned, that portion of precipitation which evaporates is higher in
urban areas than in the natural landscape before urbanization. This is due to the
rise of the average urban temperature relative to that of the hinterland (Monthly
Weather Reports for North-Rhine Westphalia, 1970-1976).
Sealing of the ground surface by buildings, streets, roads and parking

lots results in a reduced infiltration rate, and consequently, a reduction in
groundwater regeneration. Inflow of groundwater into wastewater sewer systems,
artificial lowering of groundwater levels, e.g. for subway construction (Krems,
1975), and drawdown resulting from underground mining, further reduce the
groundwater supply. To compensate for this reduction, and to meet the extremely
high and continuously rising water demand in urban areas (Federal Ministry or the
Interior, 1972; Asemann, Wirth, 1974; International Union for Geodesy and Geophysics,
1975), sometimes considerable water quantities are drawn from adjacent catchments
(Ruhrverband, Ruhrtalsperrenverein, 1973; Imhoff, Walter, 1975). Thus, surface
runoff in urban areas comprises three components: surface runoff from rainfall
excess (Mendel, Ubell, 1973); groundwater extracted in the area; and water from
adjacent catchments.
Surface Runoff from Rainfall Excess.
For deploying and dimensioning stormwater sewerage networks in urban

areas, of great importance is a knowledge of the amount and distribution over time
of surface runoff from precipitation, and of the changes it undergoes in consequence
of progressing urbanization. For this reason investigations have been made of the
influence of ground slope, evaporation and storage on rainwater runoff (Pécher,
1971b), and the variation in time of the runoff coefficient with rainfall duration
and intensity (Pecher, 1970a; Lautrich, 1967). Subsequent studies have dealt with
the question of whether or not the stormwater runoff quantities traditionally used
as input values for the design calculation of sewerage networks must or can be
determined more accurately (Annen, 1969; Pecher, 1973b). Besides time-coefficient
and summation methods, mathematical simulation models were suggested for sewerage
design calculation (KBniger, 1972), and results from their use were compared with
conventional results (Haendel, 1974; Maas, 1975). In addition, dimensioning rules
derived from runoff values measured in densely populated areas were compared with
results obtained from the use of unit hydrograph methods (Annen, Schoss, 1972).
An attempt was also made to develop, for that portion of precipitation which runs
off from permeable ground surfaces, a theoretically exact infiltration model
which, while basically simple, would take into account the original moisture and
structure of the soil and the intensity variation of a rainfall event (Neumann,
1975).
84
Changes in Surface Runoff Due to Increased Urbanization.
Certain aspects of this topic have been under investigation for years at
various university institutes, in particular at Braunschweig, Darmstadt, Hanover,
Karlsruhe and Munich. In this context, special publications on the influence of
expanding city development on flood runoff should be mentioned (Billib, Draschoff,
Keser, Koehler, Sieker, 1972; Wittenberg, 1974; Technische Universität Braunschweig,
1974).
Meteorological, topographical and other influences on stormwater runoff

in sewer systems are treated in another publication (Pfeiff, 1971). An opinion
expressed in that investigation was that "it would seem more adequate to proceed in
future from measured statistical runoff values in defined networks instead of from
measured statistical rainfall intensities of specific frequency which, in part,
involve uncertainties resulting from actual measuring conditions, and are influenced
in a very complex way by meteorological, topographical and civil engineering
factors". Similar considerations underlie the arguments in various other publica-
tions concerning the determination of conclusive runoff data for small receiving
streams with the aid of short-terra measurements (Sieker, 1969; Koehler, 1971).
These results of hydrological investigations are also the basis of the "elements"
method for the computation of rainwater runoff in channels, where analysis of
measured results with the aid of the unit hydrograph method serves as a concept
for the computation of surface runoff (Keser, 1973).
New Calculation Methods for Sewers, Rain Outlets and Stormwater Basins.
The stormwater sewer systems of most German cities are of the combined
type. They were built essentially within a period of 100 years. During that
period, land use and population densities have changed continuously, as have
wastewater volumes per inhabitant and per unit produced in industry and trade.
As a result, the original design values and the underlying planning concepts
often are no longer adequate. And in many cases, new sections have been added
to existing old networks without any practical possibilities for taking the
frequently insufficient capacities of the latter into account as a limiting
factor. Therefore, rain outlets (outfall sewers) and stormwater basins must now
be installed in city drainage systems to prevent detrimental effects. The
determination of their design dimensions requires comprehensive measurement
programs and investigations (Erode v.d., Hoffman, 1969). In particular, the
question of overflow frequency (Brunner, 1971a) must be studied and necessary water
protection measures must be considered (Schloz, 1971; Ruhrverband, Ruhrtalsperren-
verein, 1973). Water quality considerations are amplified in a subsequent sub-
section.
New calculation methods for sewer systems (Billmeier, 1970; Pécher,

1971a) and rain outlets (Brunner, 1970) have been suggested and thoroughly discussed
by experts. In the course of these discussions, the need for revising the dimen-
sioning rules applied so far became obvious. The Abwassertechnische Vereinigung
(Waste Water Engineering Society), taking the discussed considerations into
account, drafted in 1976 new directives for the calculation of wastewater, rainwater
and mixed water sewers, and in 1975 issued directives for the calculation and design
of stormwater relief arrangements (including the requisite stormwater basins and
85
sewer storage spaces) in combined sewer systems. These directives are now under
profession-wide discussion. The results of research and investigations have also
been incorporated in handbooks for actual practice (Imhoff, Imhoff, 1976).
Groundwater Regeneration and Recharge
Infiltration.
In heavily urbanized areas the demands on groundwater rise continuously,

requiring intensified resources management to ensure an adequate public water supply
for domestic and industrial use. Important factors in the groundwater budget of a
city area are groundwater regeneration, groundwater flow, continuous and temporary
withdrawals and the interaction between groundwater and surface waters. Areal
groundwater regeneration (via infiltration) depends, among other things, on the
type of soil and the character of building development. Groundwater volume and
local distribution can be determined more accurately with the aid of a suitable
model (Battermann, 1973; Battermann, Hill, 1975).
A study in the Hanover urban area proved that at the present stage of
building and construction development, possible groundwater regeneration has
already been reduced by the amount withheld from the groundwater budget by
continuous artificial extraction (Battermann, 1975). The study was motivated by
the construction of the municipal subway which, through the lowering of the
groundwater level during construction and through impoundment and drawdown induced
by the structures, affects the groundwater budget (Mull, 1975).
Artificial Recharge.
Artificial recharge is an excellent way to compensate for groundwater

losses, extractions and deficits, and to make sure that the required water quantities
can be supplied economically, hygienically, and in a technically feasible way.
However, a suitable terrain and the availability of sufficient quantities of water
of specific quality are prerequisite conditions.
In the Federal Republic of Germany, research in the field of artificial

groundwater recharge is carried out by, among others, the Wiesbaden Waterworks AG
(Haberer, Normann, 1974), the Water Management Institute of Hanover Technical
University (Bettaque, 1958; Mohle, 1967), and the Institute for Water Research of
the Dortmund Waterworks AG (Frank, Schmidt, 1965).
Examples of the practice of groundwater recharge with conditioned river

water are Berlin (Tessendorff, 1973), Frankfurt (Mevius, Wirth, 1973), Krefeld
(Czerwenka, Seidel, 1965), and Wiesbaden (Haberer, 1975).
Two-fifths of the drinking water of the North-Rhine Westphalian industrial

agglomeration area is derived from the Ruhr River. Industries and about 5-million
inhabitants are supplied from a river/public supply network/drainage network
system with a maximum circulation of 22% (Koenig, Rincke, Imhoff, 1970). The
largest part of the drinking water is withdrawn in the form of Ruhr river valley
86
groundwater artificially recharged from infiltration basins (Ruhrverband,
Ruhrtalsperrenverein, 1973), Figure 2.
FIGURE 2-SCHEMATIC VIEW OF WATER PROCUREMENT IN THE RUHR RIVER VALLEY.

(Ruhrverband, Ruhrtalsperrenverein, 1973)
The waterworks at the Rhine River procure the water mainly from groundwater
recharged with bank-filtrated river water. The wells usually supply a mixture of
"genuine" groundwater and bank filtrate (Kludig, 1970). In consequence of the
concentration of large industrial plants and cities on the banks of the Rhine,
rising water demand is accompanied by intensified river water pollution. The latter
is responsible not only for the pollution load of the procured raw water (Holluta,
Bauer, Ko'lle, 1968) but also for noticeable reductions in the hydraulic conductivity
of the bank underground, in particular on account of the infiltration resistance of
the river bottom (Kludig, 1968).
Water storage in the subsoil is another new development. It can be used

for water supply purposes, for raising the low water level, for flood control,
waste water disposal, etc. (Marotz, 1968).
Influence of Urbanization on the Water Quality of Receiving Streams
The numerous influences of densely populated areas on receiving surface

streams include: (a), inflow from sewer systems and sewage treatment plants;
(b), inflow from stormwater drainage systems; (c), direct industrial wastewater
inflow from individual factories; and (d), diffuse sources (inflow of pollute^'
groundwater and of uncaptured surface water, unauthorized or criminal actions,
transportation accidents, unforeseeable cases of damage, etc.). The range of
national and international literature on points (a) and (c) is comprehensive, so
it is neither possible nor necessary to go into details here. The influences due
to stormwater drainage systems, point (b), are defined below, and in this
connection point (d) is discussed briefly.
The diversion of stormwater from built-up areas considerably affects the

quality of receiving-stream water (Muller, 1971; Krauth, 1973). In view of the high
standard of domestic and industrial wastewater treatment in the Federal Republic of
Germany (by 1985 there will be complete biological or equivalent treatment of a.11
87
wastewaters), this pollution factor is of special importance. For the assessment
of this influence, a comprehensive study was undertaken at the request of the
Federal Ministry of the Interior (Lautrich, Pécher, 1974). With a mean population
density of about 80 to 100 inhabitants per hectare and a mean daily domestic waste
water volume of 150 litres per inhabitant, the domestic waste water quantity
diverted from built-up areas is estimated at 5000-m/ha. Assuming a mean annual
precipitation depth of 803-ram (corresponding to 8030 nrVha/year) over 680 hours per
year and approximately 280 rainfall events, a sealed (impervious) surface total of
507. of the built-up areas and an annual mean runoff of about 70% of precipitation
on sealed ground, the annual surface runoff is 281-mm, or approximately 2800 nP/ha/
year (Pécher, 1974). This means that the annual amount of runoff from rainfall is
only about 56% of the domestic waste water volume. The concentrations of the
pollution constituents in stormwater discharges from sewer systems fluctuate
extensively (sometimes at a rate of 1:100). The annual pollution load in terms of
BOD5 and chemical oxygen demand is within the order of magnitude of that for
completely treated wastewater.
In recent years, stormwater quality has been the subject of comprehensive

studies in the Federal Republic of Germany. These studies were undertaken, for
example: at the separate sewerage system of Pullach, a community on the southern
outskirts of Munich, by the Institute of Water Management and Sanitary Engineering
of Munich Technical University (Director: Prof. Dr. Bischofsberger) (Brunner, 1975);
and at the combined sewer system of Stuttgart, by the Institute for Hydraulic
Engineering and Water Quality Management of Stuttgart University (Director: Prof.
Dr. Hunken) (Krauth, 1971). As mentioned earlier, drainage in the Federal Republic
of Germany is effected predominantly by means of combined sewer systems.
For example, since 1975 precipitation and runoff measurements have been
made at a stormwater outlet of the combined sewer system of a suburb of Munich. The
gaging area covers 4.35-sq.km. with a population density of 50 to 100 inhabitants per
hectare. Three chemical and six physical wastewater parameters are measured both
during dry weather and during precipitation-runoff periods. Measurement frequency
follows theoretical information requirements (Marr, Gottle, Pieper, 1975).
Admittance into a treatment plant of the entire stormwater quantity to be

diverted is neither economical nor practical, and it is impossible with combined
sewer systems. During rainy weather periods, the settling basins designed for dry
weather inflow and a retention time of two hours can temporarily cope with five
times the normal flow, but in the normally dimensioned biological section the
design capacity may be exceeded only by up to 507. if a significant loss in
purification efficiency is to be avoided (Imhoff, Imhoff, 1976). The treatment
of stormwater from the sewer system (Lautrich, 1971) therefore requires provision
of additional and expensive installations (Kiefer, 1976), such as stormwater
overflow basins, stormwater settling basins and stormwater retention tanks (Krauth,
1976; Schloz, 1976), in some cases accompanied by oil retention arrangements (Krauel,
1967).
As mentioned earlier, for economic-technical reasons inherent in sewer

and treatment plant dimensioning, rain outlets, usually with relief discharge into
the nearest receiving stream, are incorporated in combined sewer systems (Knollmann,
1969; Krauth, 1971b). A dimensioning rule, which formerly was followed in most
88
cases, provided for quantities exceeding the fivefold dry-weather runoff to be
diverted to the receiving stream. After preliminary treatment of the fivefold dry
weather runoff, another overflow structure reduces the runoff to twice the dry
weather runoff. This reduced runoff is then treated in the biological section of
the treatment plant.
In recent years, the method of dimensioning stormwater outlets on the

basis of a mixing ratio has been subjected to growing criticism, because the
procedure underrates the local precipitation situation and overemphasizes actual
wastewater runoff. Proposals have been made to replace it by design dimensioning
based on critical areal rainfall, Figure 3, (Abwassertechnische Vereinigung, 1975),
and the method is gaining acceptance (Abwassertechnische Vereinigung, 1976). In
this context, critical areal rainfall is a function of receiving stream discharge,
wastewater runoff from the sewer system and, possibly, the long-term depth of
local precipitation in summer (Brunner, 1973). Also discussed in this connection
was the development of new calculation methods for stormwater overflow basins
(Pécher, 1970b; Pécher, 1973a) and for their design and dimensioning (Günzel, 1971),
with a view towards obtaining at the lowest possible cost an optimum in the way of
operational reliability and efficiency.
FIGURE 3-DETERMINATION OF THE CRITICAL AREAL RAINFALL

(Àbw*«a*rt«chnÍ8che Vereinigung, 1975)
The provision of stormwater settling basins was also suggested (Brunner,

1971), and a retention time of at least 10 to 20 minutes was considered necessary
to accommodate the peak of the first inrushing wave and to eliminate sediments.
The elimination rate for dissolved substances is very low, and for bacterial
pollution it is zero. Complete emptying of the basin and sludge removal after
rainfalls are necessary.
89
Facilities and operation of sewage treatment plants must also be
adapted to the increased load resulting from storm water (Kneipp, 1976). To
prevent additional coarse material, which can be disposed of in an economically
more favorable way, from being introduced into the sewer system, the use of
kitchen garbage grinders has been prohibited in the Federal Republic of Germany
(Licht, 1974).
The inflow into surface waters of rain water from stormwater outlets
of large traffic arteries involves a special pollution threat. Accidents, especially
of tank trucks carrying potential pollutants, in particular petroleum products, have
caused severe damage in recent years. In the valley of the Ruhr River on which the
drinking water supply for more than 5 million inhabitants is based, oil separation
basins are provided at transit highways. They are dimensioned for an areal rainfall
of 100 1/s/ha, runoff coefficient 0.9, 12-cu.m. inflow per sq.ra. per hour, and a
retention time of 10 minutes (Imhoff, 1967). In some cases, they are designed as
retention basins and combined with an oil collection tank (Annen, 1967).
A variety of water quality monitoring equipment has been developed

(Eckoldt,' 1967), including stationary and mobile measuring stations (Massing, 1971)
on water and on land, operated manually or automatically, continuously or on call
(Landesanstalt für Wasser und Abfall Nordrhein-Westfalen, 1972; Malz, 1971), Based
on such stations and the water parameters they measure, alarm and emergency plans
have been developed (Ottmann, 1973). For example, in the rivers Main and Regnitz
in Bavaria, which are subject to very intensive pollution from domestic and
industrial sources. When a tendency towards critical water quality situations
(oxygen concentration) becomes apparent the alert phase is initiated, and when the
critical situation is at hand a state of emergency begins, i.e., measures are taken
(sewage treatment plant operation, heat emission, wastewater drains, surface water
aeration, etc.).
Related Investigations
Diffuse Sources of Urban Groundwater Pollution.
The sources of groundwater pollution in built-up areas are manifold and

intensive, and correspond to the diversity of urban sectors, such as housing, road
traffic, trade, industry, etc.
In 1971 and 1972 the influence of urbanization on the subsoil in the

Cologne city area was investigated (Balke, 1974). In the uppermost soil layers
and in the zone of groundwater close to the ground surface, a positive thermal
anomaly with a maximum temperature rise of 4°C was observed. In some places, the
thermal influence reached down to depths as great as 30-m. Residential quarters,
commerical and industrial buildings, the sewer system, cooling water inflow,
infiltration of surface water, and the city climate were identified as heat sources.
In another investigation, positive thermal anomalies in the subsoil under an
abandoned factory were still encountered many years after the plant was closed down
(Balke, Kussmaul, Siebert, 1973). In this case, chemical reactions in the
chemically polluted groundwater zone are assumed to be the heat sources.
90
Roads are an important source of groundwater pollution. Over a period of
more than two years (June, 1970, through July, 1972), extensive investigations were
carried out in the area of the Frankfurt airport (Golwer, Schneider, 1973).
Hydrocarbons, heavy metals and sodium chloride were found to be road-specific
indicators of soil and subsurface, water pollution. The extent of pollution
depends primarily on traffic density, solubility of the substances involved,
purification effect of top layers, and dilution in the subsoil (Golwer, 1973).
Comprehensive results of an investigation at the Hanover-Kassel motorway for
mineral oils, common salt and lead compounds, showing high substance concentrations,
are available (Tiemann, 1972).
During the last decades, the steady increase in the consumption of

petroleum products such as fuel oil, Diesel fuel and gasoline for heating and
motor vehicle operation has produced a growing threat to groundwater quality,
especially in built-up areas. This threat arises from (Bartz, 1966):
transportation accidents; damage in transfer; damage resulting from inadequate
storage tank facilities above and below ground (e.g., leaks); and damage by major
forces. Numerous studies of petroleum product behavior in the ground are
available (Becksmann, Billib, v. Engelhardt, Zimmermann, 1965; Schwüle, 1971).
A fundamental study covering the assessment and treatment of oil spill accidents
was carried on by a large group of experts (Arbeitskreis "Wasser und Mineralöl",
1970). A guide describing immediate measures to be taken for the limitation and
elimination of damage from oil spill accidents has been published (Arbeitskreis
"Lagerung und Transport wassergefährdender Stoffe", 1973).
Another diffuse source of groundwater pollution is organic environmental

substances (Quentin, Weil, Udluft, 1973). Not only are they used on a large scale
as pesticides in agriculture (Frank, Schmidt, Bauer, 1970), but in the course of
the last decades they have also found their way, as a broad palette of chemical
products, into private households (Chemische Industrie, 1975; Althaus, Jung, 1972).
Contamination of groundwater is possible in a great variety of ways (Weil,

Quentin, Rimicke, 1973). A computerized "Data Bank for Pollutants" is being set up
at present (Institut für Wasserforschung, 1975).
Influence of Solid Wastes on Water Quality.
Increased accumulation of domestic garbage and industrial waste is a

characteristic of urban areas (Siedlungsverband Ruhrkohlenbezirk, 1971), and a
further increase must be expected (Bundesminister des Innern, 1971). Domestic
garbage is disposed of by deposition (787.), composting (2%) and combustion (20%)
(Bundesminister des Innern, 1971).
A severe impairment of water quality must be expected if seepage water

from solid waste deposits infiltrates groundwater or surface waters (Bucksteeg,
1969). With domestic garbage, the seepage rate is 0.05 to 0.1-1/s/ha (Schenkel,
1974). Quantity and quality of the seepage water depend on such factors as
preliminary garbage treatment (Collins, Spillmann, 1974), deposition technique
(Stegmann, Knoch, 1975), properties of the deposited garbage (Helmer, 1974),
admixture of sludge (Reuss, 1971), and the oxygen balance in the deposits (Pierau,
1968). Extensive investigations have shown that the seepage water from domestic
garbage deposits carries high concentrations of organic and inorganic substances,
91
and that these concentrations may be up to 100 times the concentrations encountered
in domestic waste water (Stegmann, 1974). Tests have proved that seepage water can
be treated together with community sewage, or separately on the mechanical/biological
pattern (Knoch, Stegmann, 1971; Knoch, 1972; Wilderer, Hartmann, 1972). Usually,
however, high residual concentrations of organic substances remain that are very
difficult to degrade biochemically; but recourse can be made to chemico-physical
methods (Stegmann, Knoch, 1974).
The protection of groundwaters and surface waters against pollution from

waste deposits requires the observation of certain rules for dumping area layouts
(Hantge, 1975), as well as precautionary measures in deposit operations (Schoder,
1975), and constant control (Schrammeck, 1973). Numerous publications on the
question of self-purification processes in the ground relative to the influences
emanating from waste deposits are available (Knoll, 1969; Golwer, Matthess,
Schneider, 1969). While the extent of groundwater pollution from waste deposits
should not be overrated in large-area assessments, especially if compared with other
pollution sources such as agricultural fertilizers (Klotter, Hantge, 1969), very
significant local pollution may result, particularly from cumulative effects
(Thews, 1971).
Many investigations have been carried out to study groundwater pollution

at existing waste deposits. During the period from 1964 to 1973, the effects of
several large deposits near Frankfurt on the chemical, physical, micro-biological
and hygienic quality of the groundwater were examined (Golwer, Knoll, Matthess.
Schneider, Wallhäuser, 1976). The Frankfurt case represents deposits in areas with
low groundwater flow velocity, a nearby receiving stream, and top layers of high
air permeability, where groundwater pollution was found to extend only over several
hundred meters from the deposit. By comparison, at a waste deposit in South
Bavaria where groundwater flow velocity is considerably higher and aquifers are
comparatively thin, investigations revealed that groundwater pollution there
spreads over several kilometers (Exler, 1972).
The deposition of clinker, ash and compost from municipal garbage

disposal facilities may also affect groundwater and surface water significantly
(Wolfskehl, Boye, 1966; Btfke, Sturm, Zubiller, 1973).
A large number of investigations have been concerned with the deposition

of industrial wastes, such as salt clinker (Grammel, 1972), blast furnace gas
sludge (Schewe, 1969), cyanic salts (Birk, Geiersbach, Muller, 1973) and mine
dumps (Heitfeld, Schüttler, 1973).
References
Abwassertechnische Vereinigung. 1975. Richtlinien für die Bemessung und

Gestaltung von Regenentlastungen in Mischwasserkanälen (Guidelines
for the Calculation of Stormwater Outlets in Combined Sewerage Systems).
Arbeitsblatt Abwassertechnische Vereinigung e.V., Bonn, 35 p.
Abwassertechnische Vereinigung. ,1976. Richtlinien für die Berechnung von
Schmutz-, Regen- und Hischwasserkanälen (Guidelines for the Calculation
of Sewage, Stormwater and Combined Sewerage Systems). Arbeitsblatt A 118,
Entwurf. Abwassertechnische Vereinigung e.V., Bonn, 22 p.
92
Althaus, H., Jung, K.D. 1972. Wirkungskonzentration gesundheitsschädigender
bzw. toxischer Stoffe in Wasser für niedere Wasserorganismen sowie kalt-
und warmblütige Wirbeltiere einschließlich des Menschen bei oraler Auf-
nahme des Wassers oder Kontakt mit dem Wasser (Effective Concentration
of Substances in Water which are Injurious to Health or Toxic to Fresh-
water Evertebrates as well as Poikilothermie and Homiothermic Vertebrates
Including Man, by Oral Uptake or Contact with Water). Hygiene-Institut
des Ruhrgebietes, Gelsenkirchen, 72 p.
Aniol, R. 1972. Beitrag zur zeitliehen und räumlichen Struktur des Nieder-
schlages (On the Temporary and Local Structure of Precipitation). In:
Meteorologische Rundschau, Vol. 25, No. 6, pp. 182-185.
Annen, C.W. 1967. Zur Gestaltung von Rückhaltebecken bei der Entwässerung
von Fernstraßen (On the Construction of Retention Basins for Highway
Drainage). In: Cas- und Wasserfach, Vol. 108, No. 2, pp. 46-48.
Annen, G.W. 1969. Das Zeitbeiwertverfahren zur Hochwasserberechnung (The
Rational Method for the Computation of Flood Frequency). In: Gas- und
Wasserfach, Vol. 110, No. 12, pp. 317-320.
Annen, G.W., Schoss, H.D. 1972. Das Einheitsganglinienverfahren bei stark
bebauten Gebieten (The Unit Hydrograph Method as Applied to Densely
Built-up Areas). In: Gas- und Wasserfach, Vol. 113, No. 2, pp. 78-81.
Arbeitskreis "Wasser und Mineralöl". 1970. Beurteilung und Behandlung
von Mineralölunfällen auf dem Lande im Hinblick auf den Gewässerschutz
(Assessment and Treatment of Oil Spill Accidents on Land with a View
to the Protection of Waters). Bundesministeriura des Innern, Bonn,
2nd Edition, 138 p.
Asemann, K., Wirth, H. 1974. Der Wasserverbrauch Frankfurter Privathaus-
halte in Gegenwart und Zukunft (Present and Future Domestic Water Demand
in Frankfurt). Stadtwerke und Statistisches Amt der Stadt Frankfurt,
Frankfurt, 2nd Edition, 119 p.
Attmannspacher, W., Hartmannsgruber, R., Riedl, J. 1974. Eine Methode
zur halbautomatischen Flächenniederschlagsmessung mit X-Band Radar
(A Method for Semi-automatical Rainfall Gauging of Areas by X-Ray
Radar). In: Archiv für Meteorologie, Geophysik und Bioklimatologie,
Series B, Vol. 22, No. 1/2, pp. 27-38.
Balke, K.-D. 1974. Der thermische Einfluß besiedelter Gebiete auf das Grund-
wasser, dargestellt am Beispiel der Stadt Köln (The Thermic Influence
of Urban Areas on Groundwater, as Demonstrated by the Example of the
City of Cologne). In: Gas- und Wasserfach, Vol. 115, No. 3, pp. 117-124.
Balke, K.-D., Kussmaul, H., Siebert, G. 1973. Chemische und thermische
Kontamination des Grundwassers durch Industriewässer (ChemienL and
Thermal Pollution of Groundwater by Industrial Waste Water). In:
Zeitschrift der Deutschen Geologischen Gesellschaft, Vol. 124, Part II,
pp. 447-460.
Bartz, J. 1966. Gefährdung des Grundwassers durch Mineralölprodukte
(Petroleum Products as a Threat to Groundwater Quality). In: Gesundheits-
Ingenieur, Vol. 87, No. 9, pp. 254-257.
93
Battermann, G. 1973. Grundwasserhaushalt eines Stadtgebietes untersucht
mit einem Widerstands-Kondensator-Netzwerk (Groundwater Budget of an
Urban Area as Analysed by a Network of Rheostats and Condensers).
In: Mitteilungen aus dem Institut für Wasserwirtschaft, Hydrologie und
landwirtschaftlichen Wasserbau, Technische Universität Hannover, Vol. 28,
pp. 1-206.
Battermann, G. 1975. Grundwasserhaushalt eines Stadtgebietes (Groundwater
Budget of an Urban Area). In: Zeitschrift der Deutschen Geologischen Ge-
sellschaft, Vol. 126, Part II, pp. 253-260.
Battermann, G., Mull, R. 1975. Groundwater Budget of an Urban Area.
Proceedings 2nd World Congress on Water Resources "Water for Human Need",
New Delhi, India, Vol. II, pp. 497-502.
Becksmann, E., Billib, H., von Engelhardt, W., Zimmermann, W. 1965.
Gutachten - Verhalten von Erdölprodukten im Boden (The Behaviour of
Petroleum Products in the Soil). Bundesminister für Gesundheitswesen,
Bonn, 130 p.
Bettaque, R. 1958. Studien zur künstlichen Grundwasseranreicherung (Research
on Artificial Groundwater Recharge). In: Veröffentlichungen des Institutes
für Siedlungswasserwirtschaft, Technische Universität Hannover, Vol. 2,
105 p.
Billib, H., Mull, R. 1971. Beeinflussung der Grundwasserverhältnisse durch
bauliche Maßnahmen, untersucht mit einem Widerstands-Netzwerk (Effects
of Construction Measures on the Groundwater Situation, Investigated with
the Aid of a Rheostatic Network). In: Bautechnik, No. 2, pp. 53-56.
Billib, H., Draschoff, R., Keser, J., Koehler, G., Sieker, F. 1972: Der Ein-
fluß zukünftiger Bebauung auf das AbfluSverhalten eines natürlichen Ein-
zugsgebietes (The Effect of Development on the Drainage Behaviour of a
Natural Catchment). In: Mitteilungen aus dem Institut für Wasserwirtschaft,
Hydrologie und landwirtschaftlichen Wasserbau, Technische Universität
Hannover, Vol. 24, pp. 279-385.
Billmeier, E. 1970. Bemessungsverfahren zur Planung von Kanalisationsnetzen
(Methods for the Design of Drainage Systems). 1. Abwassertechnisches
Seminar: Berechnungsmethoden für Abwasserkanäle, Regenüberläufe und
Regenbecken. In: Berichte aus dem Institut für Wasserwirtschaft und
Gesundheitsingenieurwesen, Technische Hochschule München, Vol. 4,
pp. 3-53.
Birk, F., Geiersbach, R., Müller, W. 1973. Die Auswirkungen der Verkippung
und Lagerung von cyanidhaltigen Härtesalzen in Bochum-Gerthe auf das Grund-
und Oberflächenwasser (Effects of Dumping and Storage of Cyanic Salts in
Bochum-Gerthe on Groundwater and Surface Water). In: Zeitschrift der
Deutschen Geologischen Gesellschaft, Vol. 124, Part II, pp. 461-473.
Böke, E., Sturm, G., Zubiller, C.-O. 1973. Die Ablagerung von Schlacke und
Asche aus kommunalen Müllverbrennungsanlagen im Hinblick auf den Gewässer-
schutz (Deposition of Clinker and Ash from Municipal Refuse Incinerators
with a View to the Protection of Water Resources). In: Deutsche Gewässer-
kundliche Mittellungen, Sonderheft, Koblenz, pp. 55-60.
94
Brunner, P.G. 1970. Neue Bemessungsgrundlagen zur Berechnung von Regenüber-
läufen (New Design Criteria for Stonnwater Overflows). 1. Abwassertech-
nisches Seminar: Berechnungsmethoden für Abwasserkanäle, Regenüberlaufe
und Regenbecken. In: Berichte aus dem Institut für Wasserwirtschaft und
Gesundheitsingenieurwesen, Technische Hochschule München, Vol. 4, pp.
57-122.
Brunner, P.G. 1971a. Uberlaufhäufigkeit von Regenüberläufen und ihre Er-
mittlung mittels Häufigkeitslinien und Punkte-Regendiagrammen (Overflow
Frequency of Rain Outlets and its Determination from Frequency Curves
and Point-Rainfall Diagrams). In: Wasser und Boden, Vol. 23, No. 5,
pp. 115-117.
Brunner, P.G. 1971b. Belastung der Gewässer durch künstliche Ableitung von
Niederschlägen - Belastung durch Abwässer der Trennkanalisation (Pollution
of Receiving Waters by Artificial Drainage of Stormwater - Pollution
by Sewage Water from Seperate Sewerage Systems). In: Berichte der Ab-
wassertechnischen Vereinigung e.V., Vol. 25, Gesellschaft zur Förderung
der Abwassertechnik e.V., Bonn, pp. 75-96.
Brunner, P.G. 1973. Der Einfluß der örtlichen Niederschlagshöhe auf die
Belastung der Gewässer mit Mischwasser (The Influence of Local Precipitation
Depth on the Pollution of Receiving Waters by Water from Combined Sewerage
Systems). In: Berichte der Abwassertechnischen Vereinigung e.V., Vol. 26,
Gesellschaft zur Förderung der Abwassertechnik e.V., Bonn, pp. 163-172.
Brunner, P.G. 1975. Die Verschmutzung des Regenabflusses im Trennverfahren
(The Pollution of Stormwater Runoff in Seperate Sewerage Systems). In:
Berichte aus Wasserwirtschaft und Gesundheitsingenieurwesen, Technische
Universität München, No. 9, 243 p.
Buckstecg, W. 1969. Abfalldeponien und ihre Auswirkungen auf Grund- und
Oberflächenwasser (Waste Deposits and their Effect on Groundwater and
Surface Water). In: Gas- und Wasserfach, Vol. 110, No. 20, pp. 529-537-
Bundesminister des Innern. 1971. Materialien zum Umweltprogramm der
Bundesregierung (Information on the Environmental Program of the .Federal
Government). Deutscher Bundestag, Drucksache VI/2710, Bonn, 661 p.
Bundesministerium des Innern. 1972. Wasserbedarfsentwicklung in Industrie,
Haushalten, Gewerbe, öffentlichen Einrichtungen und Landwirtschaft -
Prognose des Wasserbedarfs in der Bundesrepublik Deutschland (Development
of the Water Demand in Industry, Private Households, Trade, Public
Institutions and Agriculture - Prognosis of the Water Demand in the
Federal Republic of Germany). Bericht des Battelle-Institutes e.V.
Frankfurt, Bonn, 211 p.
Chemische Industrie. 1975. Umwelt und Chemie von A bis Z (Environment and
Chemistry from A to Z). Verlag Herder, Freiburg i.Br., 148 p.
Collins, H.-.J., Spillmann, P. 1974. Lagerungsdichte, Abbautemperaturen,
Sickerwasseranfall und -belastung in einer Rottedeponie in Abhängigkeit
von der Vorbehandlung des Mülls (Compactness of Deposition, Degradation
Temperatures, Quantity and Quality of Seepage Water in a Rotting Deposit
as Functions of Preliminary Waste Treatment). In: Müll und Abfall, Vol.6,
No. 3, pp. 71-81.
95
Czerwenka, W., Seidel, K. 1965. Neue Wege einer Grundwasseranreicherung
in Krefeld (New Methods of Groundwater Recharge in Krefeld). In: Gas-
und Wasserfach, Vol. 106, No. 30, pp. 328-833.
Deutscher Wetterdienst. 1970-1976. Monatlicher Witterungsbericht für
Nordrhein-Westfalen (Monthly Weather Report for North-Rhine Westphalia).
Wetteramt Essen.
Dien, M. 1967. Die Genauigkeit von Regenmessungen (The Accuracy of Rain
Gauging). In: Zur Struktur der Niederschläge, Teil 1. Archiv für
Meteorologie, Geophysik und Bioklimatologie, Series B, Vol. 15, pp.
39-51.
Eckoldt, M. 1967. Meßstationen zur Registrierung der Wasserbeschaffenheit
von Fließgewässern (Monitoring Stations for the Recording of River Water
Quality). In: Deutsche Gewässerkundliehe Mitteilungen, Vol. 11, No. 1,
pp. 3-12.
Emde, W. v.d-, Hoffmann, S. 1969. Untersuchungen an Regenüberlaufen eines
großstädtischen Kanalnetzes (Investigations at the Stormwater Overflows
of a City Drainage System). In: Gas- und Wasserfach, Vol. 110, No. 12,
pp. 321-325.
Eriksen, W. 1971. Die stadtklimatischen Konsequenzen städtebaulicher
Entwicklung (The Urban Climatic Effects of City Development). In:
Städtehygiene, Vol. 22, No. 11, pp. 259-262.
Eriksen, W. 1972. Zur Niederschlagsmodifikation im Bereich von Großstädten
(Changes in the Precipitation Regime of Urban Areas). In: Städtehygiene,
Vol. 23, No. 7, pp. 164-166.
Exter, H.J. 1972. Ausbreitung und Reichweite von Grundwasserverunreinigungen
im Unterstrom einer Mülldeponie (Spread and Extent of Groundwater
Pollution Downstream of a Waste Deposit). In: Gas- und Wasserfach,
Vol. 113, No. 3, pp. 101-148.
Flohn, H. 1973a. Natürliche und anthropogene Klimamodifikationen (Natural
and Anthropogenous Modifications of Climate). In: Annalen der Meteorologie,
New Series, Vol. 6, pp. 59-66.
Flohn, H. 1973b. Der Wasserhaushalt der Erde - Schwankungen und Eingriffe
(The Water Recourses of the World - Variations and Interventions). In:
Naturwissenschaften, Vol. 60, pp. 340-348.
Frank, W.H., Schmidt, K.H. 1965. Biologische Studien zur Weiterentwicklung
der künstlichen Grundwasseranreicherung (Biological Research for the
Further Development of Artificial Groundwater Recharge). In: Gas- und
Frank, W.H., Schmidt, K.H., Bauer, U. 1970. Pestizide und Wasserversorgung
- eine Literaturauswertung (Pesticides and Water Supply - An Evaluation
of Relating Literature). Institut für Wasserforschung GmbH, Dortmund,
81 p.
Freie und Hansestadt Hamburg. 1976. Tabellen über die Prognostizierbarkeit
von relevanten Einflußfaktoren im Aufgabenbereich der Stadtentwässerung
(Tables of the Prognostic Possibilities of Relevant Factors in the Field
of Urban Drainage). Baubehörde Hamburg, Amt für Ingenieurwesen III, 2 p.
96
Colwer-, A., Matthess, G., Schneider, W. 1969. Selbstreinigungsvorgänge im
aeroben und anaeroben Grundwasserbereich (Self-Purification Processes
in Aerobic and Anaerobic Groundwater Zones). In: Vom Wasser, Vol. 36,
pp. 64-92.
Golwer, A., Knoll, K.-H., Matthess, G., Schneider, W., Wallhäuser, K.H.
1967. Belastung und Verunreinigung des Grundwassers durch feste Abfall-
stoffe (Groundwater Pollution by Solid Wastes). In: Abhandlungen des
Hessischen Landesamtes für Bodenforschung, Wiesbaden, Vol. 73, 131 p.
Golwer, A. 1973. Beeinflussung des Grundwassers durch Straßen (Effects
of Roads on Groundwater). In: Zeitschrift der Deutschen Geologischen
Gesellschaft, Vol. 124, Part II, pp. 435-446.
Golwer, A., Schneider, W. 1973. Belastung des Bodens und des unterirdischen
Wassers durch Straßenverkehr (Pollution of Soil and Subsurface Water by
Road Traffic). In: Gas- und Wasserfach, Vol. 114, No. 4, pp. 154-165.
Grammel, U. 1972. über die Auswirkungen einer Salzschlackeablagerung auf
das Grundwasser (On the Effects of a Salt Clinker Deposit on the Ground-
water); In: Gas- und Wasserfach, Vol. 113, No. 3, pp. 113-116.
Grobe, B. 1974. Niederschlagsgebiete in Abhängigkeit von Niederschlagshöhe
und Eintrittswahrscheinlichkeit (Drainage Basins as Functions of
Precipitation Depth and Occurrence Probability). In: Wasserhaushalt
und Bodennutzung, Schriftenreihe des Sonderforschungsbereiches 150,
Technische Universität Braunschweig, Vol. 3, 179 p.
Günzel, G. 1971. Die Bemessung und Gestaltung von Rückhaltebecken in Misch-
wasserkanälen der Ortsentwässerung (Planning and Design of Retention
Basins in Combined Sewerage Systems). In: Wasser und Boden, Vol. 23,
No. 8, pp. 230-234.
Haberer, K. 1972. Trinkwasser im Umweltschutz (Drinking Water in the
Protection of the Environment). In: Gas- und Wasserfach, Vol. 113,
No. 10/12, pp. 486-493/556-561.
Haberer, K., Normann, S. 1974. Qualitätsbetrachtungen zur künstlichen Grund-
wasseranreicherung in Wiesbaden-Schierstein (Quality Considerations about
Artificial Groundwater Recharge at Wiesbaden-Schierstein). In: Künstliche
Grundwasseranreicherung am Rhein, Wissenschaftliche Berichte der Stadt-
werke Wiesbaden AG, Vol. 2, pp. 131-163.
Haberer, K. 1975. Probleme der Wassergewinnung durch Grundwasseranreicherung
aus der Sicht der Stadtwerke Wiesbaden AG (Problems of Water Procurement
by Groundwater Recharge from the Point of View of Stadtwerke Wiesbaden AG).
In: Gas-Wasser-Abwasser, Vol. 55, No. 12, pp. 828-836.
Haendel, B. 1974. Die Möglichkeiten zur Ermittlung maximaler Regenabfluß-
mengen in der Kanalisationsnetzberechnung und deren Zuverlässigkeit
(Possibilities of Determining Maximum Stormwater Flow for the Design
of Drainage Systems and Calculation Reliability). In: Gas- und" Wasserfach,
Vol. 115, No. I, pp. 28-30.
Hantge, E. 1975. Mülldeponie und Schutz des Grund- und Oberflächenwassers
(Waste Deposits and the Protection of Groundwater and Surface Water).
In: Müll und Abfall, Vol. 7, No. 1, pp. 1-4.
97
Heitfeld, K.-H-, Schüttler, U. 1973. Versickert Wohin? - Kontamination des
Wassers im Bereich von Abfallhalden durch Spurenmetalle (Pollution of
Water within the Reach of Waste Dumps by Trace Elements), In: Umwelt,
Vol. 1, pp. 57-58.
Helmcr, R. 1974. Menge und Zusammensetzung von Sickerwasser aus Deponien
verschiedenartiger Abfallstoffe (Quantity and Quality of Seepage Water
from Deposits of Diversified Wastes). In: Müll und Abfall, Vol. 6, No. 3,
pp. 61-71.
Holluta, J., Bauer, L., Kölle, W. 1968. Über die Einwirkung steigender
Flußverschmutzung auf die Wasserqualität und die Kapazität der Ufer-
filtrate (Effects of Increasing River Pollution on Water Quality and
Bank Filtration Capacity). In: Gas- und Wasserfach, Vol. 109, No. 30,
pp. 1406-1409.
Imhoff, K.R. 1967. Ölabscheidebecken für Fernstraßen im Ruhrtal (Oil
Seperation Tanks at the Transit Highways in the Ruhr Valley). In:
Imhoff, K.R. 1974. Gedanken zur Steigerung der Reinhaiteleistung, ausgeführt
am Beispiel eines Flußsystems (Reflections on the Increase in
Purification Capacity of a River System). In: Gas-Wasser-Abwasser,
Vol. 54, No. 3, pp.-61-74.
Imhoff, K.R., Walter, E. 1975. Ruhrwasserwirtschaft 1974, Bericht über die
Wasserwirtschaft im Abflußjahr 1974 (Report on Water Management in the
Ruhr Region During the Flow Year 1974). Eigenverlag Ruhrtalsperrenverein,
Essen, 41 p.
Imhoff, K-, Imhoff, K.R. 1976. Taschenbuch der Stadtentwässerung (Handbook
of Town Drainage Systems). Verlag R. Oldenbourg, München, 24th Edition,
392 p.
Institut für Wasserforschung GmbH. 1975. Untersuchung zur Entwicklung einer
Datenbank für wassergefährdende Stoffe - DABAWAS - (On the Development
of a Data Bank for Pollutants). Eigenverlag, Dortmund, 60 p.
Internationale Union für Geodäsie und Geophysik, Internationale Assoziation
für hydrologische Wissenschaften. 1975. Hydrologische Bibliographie für
das Jahr 1969 (mit Nachträgen, Bundesrepublik Deutschland)(Hydrological
Bibliography 1969 for the Federal Republic of Germany,with Supplement?).
Bundesanstalt für Gewässerkunde, Koblenz, 167 p.
Keller, R. 1971. Wasserbilanz der Bundesrepublik Deutschland (Water Balance
of the Federal Republic of Germany). In: Umschau in Wissenschaft und
Technik, No. 3, pp. 73-78.
Keser, J. 1973. Folgerungen aus dem Abflußverhalten befestigter Flächen
für die Bemessung von Regenwasserkanälen - Elementenverfahren (Conclusions
from the Behaviour of Stormwater Flow from Impervious Areas for the
Calculation of Drainage Systems - Element Method). Mitteilungen aus dem
Institut für Wasserwirtschaft, Hydrologie und landwirtschaftlichen Wasser-
bau, Technische Universität Hannover, Vol. 27, 185 p.
98
Kiefer, W. 1976. Problematik und Zielsetzung der Regenwasserbehandlung
(Problems and Aims of Stormwater Treatment). In: Berichte der Abwasser-
technischen Vereinigung e.V., Vol. 28, Gesellschaft zur Förderung der
Abwassertechnik e.V., Bonn, pp. 851-856.
Kienle, J. v. 1952. Ein stadtgebundener Schneefall in Mannheim (An Urban-
specific Snowfall at Mannheim). In: Meteorologische Rundschau, pp. 132-133.
Klotter, H.-E., Hantge, E. 1969. Abfallbeseitigung und Grundwasserschutz
(Waste Disposal and Groundwater Protection). In: Müll und Abfall, Vol. 1,
No. I, pp. 1-8.
Kludig, K.H. 1968. Die Gewinnung von uferfiltriertem Grundwasser und der
Einfluß der Rheinverschmutzung (Winning of Bank Filtrated Groundwater
and the Effects of Rhine River Pollution). In: Gas- und Wasserfach,
Vol. 109, No. 50, pp. 1401-1405.
Kludig, K.H. 1970. Ermittlung des Anteils von Uferfiltrat bei Mischgrund-
wasser (Determination of the Portion of Bank Filtrated River Water in
Mixed Groundwater). In: Gas- und Wasserfach, Vol. Ill, No. 2, pp. 77-79.
Kneip, H.J. 1976. Regenwasserbehandlung im Klärwerk (Stormwater Purification
in Sewage Treatment Plants). In: Berichte der Abwassertechnischen Ver-
einigung e.V., Vol. 28, Gesellschaft zur Förderung der Abwassertechnik
e.V., Bonn, pp. 895-912.
Knoch, J., Stegmann, R. 1971. Versuche zur Reinigung von Müllsickerwasser
(Experiments for the Purification of Seepage Water from Waste Deposits).
In:" Müll und Abfall, Vol. 3, No. 6, pp. 160-166.
Knoch, J. 1972. Reinigung von Müllsickerwasser mit belüfteten Teichen
(Purification of Waste Deposit Seepage Water in Aerated Ponds). In:
Müll und Abfall, Vol. 4, No. 4, pp. 123-133.
Knoll, K.H. 1969. Hygienische Bedeutung natürlicher Selbstreinigungsvorgänge
für die Grundwasserbeschaffenheit im Bereich von Abfalldeponien (Hygienic
Importance of Natural Self-Purification Processes for Groundwater Quality
within the Reach of Waste Deposits). In: Müll und Abfall, Vol. 1, No. 2,
pp. 35-41.
Knollmann, R. 1969. Mischwasserkanalisation und Regenüberläufe (Combined
Drainage Systems and Stormwater Overflows). In: Veröffentlichungen des
Institutes für Siedlungswasserwirtschaft, Technische Universität Hannover,
Vol. 31, pp. 259-285.
Koehler, G. 1971. Ermittlung maßgebender Abflußdaten für kleinere Vorfluter
mit Hilfe kurzzeitiger Naturmessungen (Determination of the Distinct
Flow Data of Small Receiving Streams on the Basis of Short Time
Measurements). In: Mitteilungen aus dem Institut für Wasserwirtschaft,
Hydrologie und landwirtschaftlichen Wasserbau, Technische Universität
Koenig, H.W., Rincke, G., Imhoff, K.R. 1970. Der Sonderfall des Wasserkreis-
laufes an der Ruhr im Trockenjahr 1959 (The Particular Case of the Water
Cycle in the Ruhr Area in the Drought Year 1959). In: Gas- und Wasserfach,
Vol. 111, No. 7, pp. 369-373.
99
Königer, W. 1972. Anwendung eines komplexen Abflußroodells in der Stadtent-
wässerung (Application of a Complex Stormwater Flow Model to a Town
Drainage System). In: Lehrgangs-Symposium Elektronische Berechnungen von
Rohr- und Gerinneströmungen, Nürnberg, October 1972. Eigenverlag Lehrstuhl
für Wasserbau und Wasserwirtschaft, Technische Universität München, 25 p.
Krauel, A. 1967. ölrückhalt in Regenwasserpumpwerken und Regenbecken (Oil
Retention in Stormwater Pumping Stations and Basins). In: Gas- und Wasser-
fach, Vol. 108, No. 2, pp. 48-51.
Krauth, Kh. 1971a. Der Abfluß und die Verschmutzung des Abflusses in Misch-
kanalisationen bei Regen (Stormwater Flow and Pollution in Combined Sewers
during Rainfall). Stuttgarter Berichte zur Siedlungswasserwirtschaft,
Vol. 45, Verlag R. Oldenbourg, München, 251 p.
Krauth, Kh. 1971b. Belastung der Gewässer durch künstliche Ableitung von
Niederschlägen - Belastung durch Abwässer der Mischkanalisation (Pollution
of Receiving Waters by Artificial Drainage of Stormwater - Pollution by
Sewage from Combined Sewerage Systems). In: Bericht der Abwassertechnischen
Vereinigung e.V., Vol. 25, Gesellschaft zur Förderung der Abwassertechnik
e.V., Bonn, pp. 97-115.
Krauth, Kh. 1973. Die Auswirkungen der Einleitung von Regenwasser auf die
Vorfluter (Effects of Stormwater Inflow on Receiving Streams). In:
Deutsche Gewässerkundliche Mitteilungen, Sonderheft, Koblenz, pp. 50-54.
Krauth, Kh. 1976. Die Auswirkungen von verschiedenen Bauarten von Regenüber-
laufbecken auf die Schmutzbelastung des Vorfluters und die Wirtschaft-
lichkeit des Kanalnetzes (Effects of Various Types of Stormwater Settling
Tanks on the. Pollution of Receiving Streams and on the Operational Economy
of Sewerage Systems). In: Berichte der Abwassertechnischen Vereinigung e.V.,
Vol. 28, Gesellschaft zur Förderung der Abwassertechnik e.V., Bonn,
pp. 583-591.
Krems, G. 1975. Die Grundwassersituation in Berlin (West) und die Auswir-
kungen von Grundwasserabsenkungen im Rahmen von Baumaßnahmen auf den
Grundwasserhaushalt (The Groundwater Situation in Berlin (West), and the
Effects of Artificial Lowering of the Water Table for Construction
Purposes on the Groundwater Budget). In: Zeitschrift der Deutschen
Geologischen Gesellschaft, Vol. 126, Part II, pp. 215-222.
Landesanstalt für Wasser und Abfall Nordrhein-Westfalen. 1972. Untersuchungs-
einrichtungen der Landesanstalt für Gewässerkunde und Gewässerschutz
Nordrhein-Westfalen (Water Monitoring Facilities of the State Institute
for Water Management and Water Pollution Control North-Rhine Westphalia).
Eigenverlag, Düsseldorf, Vol. 31, 43 p.
Lautrich, R. 1967. Ist der 15-Minuten-Berechnungsregen heute noch zweckmäßig?
(Is the 15-Minute Design Storm Still Appropriate?). In: Wasser und Boden,
Vol. 18, No. 12, pp. 367-368.
Lautrich, R. 1971. Die Schmutzwasserbelastung des Vorfluters durch Regenüber-
läufe mit und ohne Zwischenschaltung von Regenüberlaufbecken (Receiving
Stream Pollution from Stormwater Overflows with and without Intermediate
Stormwater Retention Tanks). In: Wasser und Boden, Vol. 23, No. 8, pp.
234-235.
100
Lautrich, R., Pecher, R. 1974. Gutachten über den Regenwasserabfluß von be-
bauten Gebieten und sein Einfluß auf die Gewässerverschmutzung (Report on
Stormwater Runoff from Built-up Areas and its Influence on the Pollution
of Receiving Streams). Bundesminister des Innern, Bonn, 295 p.
Licht, H. 1974. Abschwemmung in die Kanalisation - Küchenabfallzerkleinerer
(Domestic Waste Washed into the Sewerage System - Kitchen Garbage Grinders).
In: Kumpf, W., Maas, K., Straub, H.: Müll- und Abfallbeseitigung, Erich
Schmidt Verlag, Berlin, 32. Lieferung, VI/74, No. 2330, pp. 1-7.
Maas, S. 1975. Eine instationäre Kanalsimulation im Vergleich mit dem Zeit-
beiwertverfahren (Dynamic Simulation of Drainage Systems in Comparison
to the Rational Method). In: Korrespondenz Abwasser, Vol. 22, No. 7,
pp. 203-209, No. 11, pp. 351.
Malz, F. 1971. Moderne Methoden der Gewässerüberwachung und der Abwasser-
einleitungen (New Quality Control Methods for Water Bodies and Sewage
Outlets). In: Berichte der Abwassertechnischen Vereinigung e.V., Vol. 25,
Gesellschaft zur Förderung der Abwassertechnik e.V., Bonn, pp. 61-74.
Malpricht, E. 1976. Entwurf und Bemessung von Rückhaltenetzen (Design and
Calculation of Sewerage Retention Systems). In: Korrespondenz Abwasser,
Vol. 23, No. 7, pp. 195-205.
Marr, G., Göttle, A., Pieper, L. 1975. Analyse und Simulation von Nieder-
schlagsabfluß und Beschaffenheit in städtischen Gebieten (Analysis
and Simulation of Rainfall Run-off and Quality in Urban Areas). In:
Drucksache Sonderforschungsbereich 81, Vortragsveranstaltung 13. Juni
1975, Technische Universität München, pp. 115-147.
Marotz, G. 1968. Technische Grundlagen einer Wasserspeicherung im natür-
lichen Untergrund (Technical Fundamentals of Water Storage in Natural
Subsoil). In: Schriftenreihe des Kuratoriums für Kulturbauwesen,
Verlag Wasser und Boden, Hamburg, Vol. 18, 228 p.
Massing, H. 1971. Meßstationen zur Untersuchung der Wasserbeschaffenheit
(Water Quality Monitoring Stations). In: Wasser und Boden, Vol. 23,
No. I, pp. 14-15.
Mendel, H.G., Ubell, K. 1973. Der Abflußvorgang (The Runoff Process).
In: Deutsche Gewässerkundliche Mitteilungen, Vol. 17, No. 2, pp. 33-39.
Mevius, W., Wirth, H. 1973. Grundwasseranreicherung im Frankfurter Stadt-
wald (Groundwater Recharge in the Municipal Forest of Frankfurt). In:
Wasser und Boden, Vol. 25, No. 3, pp. 56-58.
Möhle, K.A. 1967. Die Versickerung von Oberflächenwasser mittels offener
Anreicherungsbecken (Infiltration of Surface Water into the Underground
Through Open Recharge Basins). In: Veröffentlichungen des Institutes
für Siedlungswasserwirtschaft, Technische Universität Hannover, Vol. 26,
119 p.
Mull, R. 1975. Simulation der durch den U-Bahn-Bau in Hannover beeinflußten
Grundwasserströmung (Simulation of Groundwater Flow as Affected by Subway
Construction in Hanover). In: Zeitschrift der Deutschen Geologischen
Gesellschaft, Vol. 126, Part II, pp. 261-271.
101
Müller, W.J. 1971. Der Beitrag der Regenabflüsse zur Verunreinigung der Ge-
wässer (Contribution of Stormwater Runoff to the Pollution of Receiving
Streams). In: Gas- und Wasserfach, Vol. 112, No. 1, pp. 15-17.
Neumann, W. 1975. Der Oberflächenabfluß in städtischen Einzugsgebieten
- Ein Beitrag zur Kanalnetzberechnung (Surface Runoff in Urban Catchments
- A Contribution to the Calculation of Drainage Systems). Dissertation
im Fachbereich Bauingenieur- und Vermessungswesen der Technischen
Universität München, 141 p.
Ottmann, E. 1973. Alarm- und Einsatzpläne bei Unglücksfällen und kritischen
Gewässergüteverhältnissen an der Regnitz (Alarm and Emergency Plans for
Meeting Accidents and Critical Water Quality Conditions at the Regnitz
River). In: Deutsche Gewässerkundliehe Mitteilungen, Sonderheft, Koblenz,
pp. 98-102.
Otto, G. 1971. Zur Ungenauigkeit von Niederschlagswerten und ihre Auswir-
kungen (On the Inaccuracy of Rain Gauging and Its Consequences). In:
Wissenschaftliche Zeitschrift der Universität Greifswald, Mathematisch-
naturwissenschaftliche Reihe, Vol. 20, No. 1/2, pp. 55-61.
Pecher, R. 1970a. Die zeitliche Abhängigkeit des Abflußbeiwertes von der
Regendauer und der Regenintensität (The Time Dependence of the Runoff
Coefficient on Duration and Intensity of Rainfall). In: Gas- und Wasser-
fach, Vol. Ill, No. 8, pp. 444-450.
Pécher, R. 1970b. Neue Berechnungsinethoden für Regenbecken (New Methods
for the Calculation of Stormwater Tanks). In: Berichte aus dem Institut
für Wasserwirtschaft und Gesundheitsingenieurwesen, Technische Hochschule
München, Vol. 4, pp. 125-185.
Pecher, R. 1971a. Das Zeitabflußfaktorenverfahren zur Bemessung von Kanal-
netzen (The Time-Runoff-Coefficient Method for the Design of Drainage
Systems). In: Gas- und Wasserfach, Vol. 112, No. 6, pp. 312-319.
Pecher, R. 1971b. Einfluß von Geländeneigung, Verdunstung und Speicherung
auf den Regenwasserabfluß in Kanalnetzen (Influence of Slope, Evaporation
and Retention on Stormwater Runoff in Sewerage Systems). In: Gas- und
Pécher, R. 1973a. Einfluß der Niederschlagsereignisse und des Einzugsge-
bietes auf die Bemessung von Regenrückhaltebecken, Regenklärbecken und
Keflenüberläufen (Influence of Rainfall and Catchment on the Design of
Retention Basins, Stormwater Tanks and Stormwater Overflows). In:
Die Wasserwirtschaft, Vol. 63, No. 2, pp. 47-52.
Pécher, R. 1973b. Die praktische Bemessung von Kanalnetzen und Regenrück-
haltebecken mit dem zeitlich veränderlichen Abflußbeiwert (Practical
Design of Sewerage Systems and Stormwater Retention Basins with the Aid
of Time-Variable Runoff Coefficients). In: Korrespondenz Abwasser, Vol.
20, No. 2, pp. 29-33.
Pécher, R. 1974. Der jährliche RegenwasserabfluB von bebauten Gebieten und
seine Verschmutzung (The Annual Stormwater Runoff from Built-up Areas and
its Pollution). In: Korrespondenz Abwasser, Vol. 21, No. 5, pp. 113-120.
102
Pfeiff, S. 1971. Meteorologische, topographische und bautechnische Einflüsse
auf den Regenabfluß in Kanalisationsnetzen (Meteorological, Topographical
and Civil Engineering Influences on Stormwater Runoff in Sewerage Systems).
Wasser und Abwasser in Forschung und Praxis, Bielefeld, E. Schmidt Verlag,
Vol. 3, 340 p.
Pierau, H. 1968. Ergebnisse der Untersuchungen an Versuchsdeponien und be-
stehenden Ablagerungsplätzen (Results of Research at Experimental and
Operational Deposits). In: Stuttgarter Berichte zur Siedlungswasserwirt-
schaft, Verlag R. Oldenbourg, München, Vol. 41, pp. 27-51.
Quentin, K.-E., Weil, L., Udluft, P. 1973. Grundwasserverunreinigungen durch
organische Umweltchemikalien (Groundwater Pollution by Organic Environment
Substances). In: Zeitschrift der Deutschen Geologischen Gesellschaft,
Vol. 124, Part II, pp. 417-424.
Reidat, R. 1971a. Das Klima (The Climate). In: Akademie für Raumforschung
und Landesplanung: Deutscher Planungsatlas, Band 8 (Hamburg), Lieferung 7,
Gebrüder Jaenecke Verlag, Hannover, 14 p.
Reidat, R. 1971b. Über den Einfluß der Stadt auf die Niederschlagsverteilung
bei starken Regenfällen in Hamburg (Urban Influence on the Local Distri-
bution of Heavy Rainfall in Hamburg). In: Wetter und Leben, Vol. 23,
No. 1/2, pp. 1-6.
Reuß, K. 1971. Untersuchungen zur Herabsetzung der Sickerwassermengen bei
der gemeinsamen Ablagerung von Abwasserschlamm und Hausmüll (Reduction
of Seepage Water Quantity from Combined Deposition of Sewage Sludge and
Domestic Garbage). In: Mitteilungen des Leichtweiß-Institutes;für Wasser-
bau und Grundbau, Technische Universität Braunschweig, Vol. 30, pp. 193-
204.
Rother, K.H. 1974. Der Einfluß von Veränderungen im Abflußsystem auf den
Hochwasserablauf in kleineren Einzugsgebieten (Effects of Changes in the
Runoff System on the Flood Runoff of Smaller Catchments). Wasserhaushalt
und Bodennutzung, Schriftenreihe des Sonderforschungsbereiches 150,
Technische Universität Braunschweig, Vol. 2, 222 p.
Ruhrverband, Ruhrtalsperrenverein. 1973. Die zunehmende Gefährdung der
Trinkwasserversorgung aus der Ruhr (Growing danger to the Public Drinking
Water Supply from the Ruhr River). Eigenverlag, Essen, 35 p.
Schewe, L.-D. 1969. Cyanhaltiger Gichtgasschlamm als Ursache einer Grund-
wasserverunreinigung (Cyanic Sludge from Blast Furnace Gas as a Cause of
Groundwater Pollution). In: Gas- und Wasserfach, Vol. ,110, No. 26, pp.
702-706.
Schenkel, W. 1974. Die geordnete Deponie von festen Abfallstoffen
(Controlled Deposition of Solid Wastes). In: Müll und Abfall, Vol. 6,
Beiheft 9, 80 p.
Schlaak, F. 1972. Mittlere und extreme Niederschlagsverhältnisse in Berlin
(Average and Extreme Precipitation Conditions in Berlin). In: Beilage
zur Berliner Wetterkarte, January 13th 1972, 24 p.
Schloz, G. 1971. Regenüberläufe in Mischwasserkanalisation
(Stormwater Outlets in Combined Sewers). In: Wasser und Boden, Vol. 23,
No. 5, pp. 118-120.
103
Schloz, G. 1976. Bemessungsgrundlagen für die Regenwasserbehandlung im Misch-
verfahren (Basis of Calculation for the Treatment of Stormwater in Combined
Sewerage Systems). In: Berichte der Abwassertechnischen Vereinigung e.V.,
Vol. 28, Gesellschaft zur Förderung der Abwassertechnik e.V., Bonn, pp.
857-868.
Schoder, G.S. 1975. SicherungsmaBnahmen beim Betrieb einer geordneten Grofl-
deponie (Security Measures for the Operation a Large, Controlled Deposit).
In: Müll und Abfall, Vol. 7, No. 1, pp. 4-14.
Schrammeck, E. 1973. Gewässerbeeinträchtigung durch Deponien und Abfall-
lap.erstätten - Empfehlungen zur laufenden Überwachung - (Water Pollution
by Deposits and Waste Dumps - Recommendations for Permanent Control -) .
In: Cas- und Wasserfach, Vol. 114, No. 5, pp. 214-217.
Schwüle, F. 1971. Die Migration von Mineralöl in porösen Medien (The Migration
of Petroleum Products in Porous Media). In: Gas- und Wasserfach, Vol. 112,
No. 9, pp. 465-472.
Siedlungsverband Ruhrkohlenbezirk. 1971. Untersuchung über die zukünftige
Abfallbeseitigung Im Ruhrgeblet (Future Waste Disposal in the Ruhr Area).
Schriftenreihe des Siedlungsverbandes Ruhrkohlenbezirk, Essen, Vol. 43,
71 p.
Sieker, F. 1969. Konzeption zur Analyse und Synthese des Abflußverhaltens
natürlicher Wasserläufe ("Hydrologisches Mosaik") angewendet in einer
Untorsuchung über die Belastung natürlicher Wasserläufe durch Autobahn-
abf Lusse (Conception for Analysis and Synthesis of River Discharge
Behaviour ("Hydrological Mosaik") as Applied in an Investigation of the
Pollution of Natural Receiving Streams by Runoff from Highways), Disser-
tation, Fakultät für Bauwesen, Technische Universität Hannover, 95 p.
Stalmann, V., Schoss, HD., Chilla, L. 1974. Häufigkeit von Starkregen im
Emscher- und Lippegebiet (Frequency of Heavy Rainfall in the Emscher and
Lippe River Basins). Eigenverlag Emschergenossenschaft-Lippeverband,
Essen, 77 p.
Stalmann, V. 1976. Starkregenhäufigkeit im Emscher- und Lippegebiet
(Frequency of Heavy Rainfall in the Emscher and Lippe River Basins). In:
Wasser und Boden, Vol. 28, No. 1, pp. 3-6.
Stegmann, R., Knoch, J. 1975. Sickerwasserqualität von Rottedeponien
(Quality of Seepage Water from Rotting Deposits). In: Müll und Abfall,
Vol. 7, No. 2, pp. 36-43.
Stegmann, R. 1974. Qualität von Sickerwasser aus Mülldeponien (Quality
of Seepage Water from Deposits). In: Sickerwasser aus Mülldeponien,
Schriftenreihe der Städtereinigung KG, Edelhoff, Iserlohn, No. 2,
pp. 22-43.
Stcgmann, R., Knoch, J. 1974. Restkonzentrationen von biologisch gereinig-
tem Müllsickerwasser und Versuche zur Nachbehandlung mit chemisch-
physikalischen Verfahren (Residual Concentrations of Biologically Treated
Seepage Water from Waste Deposits, and Experiments in Chemico-Physical
After-Treatment). In: Reinigung von Sickerwasser aus Mülldeponien,
Schriftenreihe der Stadtreinigung KG, Edelhoff, Iserlohn, No. 3, pp.
49-63.
104
Technische Universität Braunschweig. 1974. Wasserhaushalt und Bodennutzung
(Water Management and Land Use). Ziele und Programmedes Sonderforschungs-
progrannies 150, Schriftenreihe des Sonderforschungsbereiches 150, Vol. 1,
132 p.
Tessendorf, H. 1973. Verbrauchsentwicklung und Bedarfsdeckung im Versorgungs-
raum der Berliner Wasserwerke in den 70er Jahren (Development of Demand and
Supply in the Sphere of Influence of the Berlin Waterworks in the Seventies).
In: Gas- und Wasserfach, Vol. 114, No. 1, pp. 1-9.
Thews, J.D. 1971. Summenwirkung von Verunreinigungen in Trinkwassergebieten
(Cumulative Effects of Pollution in Drinking Water Protection Zones). In:
Tiemann, K.H. 1972. Die Auswirkungen des Straßenverkehrs auf Boden, Pflanzen
und Wasser (Effects of Road Traffic on Soil, Plants and Water). In: Zeit-
schrift für Kulturtechnik und Flurbereinigung, Vol. 13, pp. 90-108.
Weil, L., Quentin, K.-E., Rönicke, G. 1973. Pestizidspiegel des Luftstaubes
in der Bundesrepublik Deutschland (pesticide Level of Airborne Dust in the Federal
Republic of Germany). In: Mitteilung VIII, Deutsche Forschungsgemeinschaft,
Kommission zur Erforschung der luftverunreinigenden Stoffe, Bonn, 23 p.
Wilderer, P., Hartmann, L. 1972. Untersuchung über Menge und Abbaubarkeit
von Sickerwasserinhaltsstoffen aus einer Mülldeponie (Research into
Quantity and Degradability of Substances Contained in Seepage Water from
a Waste Deposit). In: Müll und Abfall, Vol. 4, No. 3, pp. 82-87.
Wittenberg, H. 1974. Der Einflufl zunehmender Bebauung auf den HochwasserabfluS
(Effects of Progressive Area Built-up on Flood Runoff). Mitteilungen des
Institutes Wasserbau III, Universität Karlsruhe, Vol. 4, 109 p.
Wolfkehl, 0., Boye, E. 1966. Einwirkung abgelagerter Müllasche und Müllkom-
postes auf das Grundwasser (Effects of Ash and Compost Deposits on Ground-
water). In: Schweizerische Bauzeitung, Vol. 84, No. 3, pp. 61-63.
105
Section 3 Models
Introduction
The future tasks of urban water management and other public activities
will be affected by problems arising from growing population density and
technological development (Hahn, 1972). In the realm of the utilization of
existing water resources by man and of the protection of water quality, this
requires optimal application of available technology and knowledge. Toward this
end, and in addition to other methods, water management is making recourse to
simulation and mathematical models. This Section is a summary of features of
various precipitation analysis and simulation models, surface runoff and sewer
system models, receiving stream models, related models, and on presently
foreseeable developments in modeling. In spite of advances in computer equipment
and modern methods of computation, it is impossible, on account of the regionally
limited character of actual tasks, and of the diversified conditions along a
watercourse, to develop uniform and universally applicable solutions (Lecher, 1968).
The tendency to aim at standard solutions cannot lead to adequate results. Every
water management problem calls for individual adaptation of techniques and methods,
including the use of operations research methods.
In order to obtain a general view of available modern methods, surveys

such as that of mathematical river basin models (Cembrowicz, Hahn, Plate, Schultz,
1975) and the list of surface runoff and transport models and suitable distinguishing
characteristics presently being compiled by the Institute for Water Management,
Hydrology and Agricultural Hydraulic Engineering of Hanover Technical University
(Verworn, 1976), have been organized more thoroughly than ever before. Some surveys
include a comparison of various programs, while others are merely lists.
Unless man succeeds in using these planning methods to control the effects
of urbanization, there is a danger that the natural environment to which he has
adapted will be destroyed and give way to a hostile environment (Plate, 1976).
Figure 4 presents a schematic survey and interrelation of simulation

models.
106
PRECIPITATION
RAIN-ROUTING QUALITY
VIA OF SURFACE
SURFACE RUNOFF SURFACE " RUNOFF
MODEL
SYSTEM RUNOFF
1 1
i
ROUTING OF SURFACE POLLUTANT ROUTING

SEWERAGE
RUNOFF AND OF SURFACE RUNOFF • NETWORK
DRY-WEATHER FLOW AND DRY-WEATHER MODEL
FLOW THROUGH
THROUGH TRANSPORT
TRANSPORT SYSTEM
SYSTEM
1
' '
TOTAL-RUNOFF POLLUTANT ROUTING

RECEIVING
ROUTING THROUGH OF TOTAL RUNOFF ' WATER
RECEIVING WATER THROUGH RECEIVING MODEL
SYSTEM WATER SYSTEM
FIGURE 4 - SCHEMATIC SURVEY AND INTERRELATION OF

SIMULATION MODELS
107
Precipitation Analysis Models
Precipitation falling on an urban area is part of the load a combined or

stonnwater sewer system has to accommodate. Because total inflow and its distribution
are of great importance for the dimensioning of sewers (Neumann, 1974), engineering
offices and university institutes active in this field have devoted much attention
to precipitation analysis. The large number of models available (inter alia: Euler,
1975; Keser, 1973; Sieker, 1975) cannot be described here because of space limitations.
Instead, brief mention is made of two models by way of example.
The intensities of, and the elapsed time between, two successive rainfalls
are essential factors in the calculation of sewer systems. In the "DC-REGDEF" model
of the firm Dorsch-Consult of Munich (Geiger, 1975), precipitation records for
specific events are divided into five-minute intervals and subjected to a regression
analysis.
A rainfall assessment program has been developed by the firm Lautrich and
Pécher of Düsseldorf, for the city of Düsseldorf, Kittelbach storm water discharge,
a special version of Program SBV in the Table at the end of this Section
(Landeshauptstade Düsseldorf, 1975). This model allows local influences on
stormwater runoff behavior to be taken into account, and permits determination of
the outflow from stormwater basins without detailed calculation of the sewerage
network.
Surface Runoff and Transport Models
"Comprehensive models" for surface runoff and wastewater flow through

sewers are individually listed in a table at the end of this Section in which the
prominent characteristics of the models are also indicated. This table was compiled
on the basis of information made available through correspondence with engineering
offices and university institutes. The replies varied in the extent of
inclusiveness and are therefore not entirely comparable. In all cases, differentiation
is made between surface runoff models and sewerage models. A surface runoff model
supplies the input values for a sewerage model. The main differences between the
various surface runoff models consist of the allowance made for losses such as
wetting loss, depression loss, infiltration loss and evaporation loss, and in the
structure of the rainfall event as based on block rains or hyetographs. There are
also differences in the allowance made for the influence of the terrain (Pécher,
1971).
In conformity with actual needs in the Federal Republic of Germany, all

models were developed for the recalculation of existing or planned networks, so as
to obtain basic information required for the improvement of inadequate drainage
systems, or for decisions on planning alternatives.
108
The sewerage models fall within two different groups. The first group is
based on the differential equation of St. Venant (KLym, Königer, Mevius, Vogel, 1972;
Kocks, 1974), which are simplified for solution according to requirements. The
second group uses the "rational" method (Imhoff, Irahoff, 1976; Ribbeck, 1968) and
other parametric procedures, with various modifications made relative to the
traditional method. The question of backwater and branching (Maas, 1975; Königer,
1972) is also solved in different ways, in some cases by way of exact computation,
in others by means of simplified formulations. The programs are applicable to urban
areas with combined or separate sewer systems. Some programs take quality parameters
such as BOD5 and temperature into account. However, because few measurements for the
calibration of the models are available at this time, the values involved are only
estimates.
The program QQS of the firm Dorsch-Consult of Munich is based on a

combination of unit hydrograph procedures for the calculation of surface runoff,
with an approximate solution of St. Venant*s differential equation for the computation
of runoff in sewer systems (Geiger, 1976). It not only simulates individual
precipitation events but also performs the requisite computations for a rainfall
record covering up to 20 years. The more important results are: annual and monthly
frequencies and duration curves of runoff and pollution load at any point of a sewer
system. Figure 5 is an overall flow chart for the QQS model.
Model HVM of the firm Dorsch-Consult, Munich, has been operational since
1966. Since then, it has been used to calculate sewer system performance for more
than 40 towns and cities, in Germany, Switzerland, England, Canada and the U.S.A.
Model SESIM of the firm Kocks, Düsseldorf, was developed from the Storm
Water Management Model and the Water Resources Engineers quality model QUAL II from
the U.S. It comprises modules for surface runoff, sewer system and receiving stream,
and can take into account up to 3000 drainage areas, sections and junctions, and up to
500 receiving stream sectors. SESIM was used for the general drainage scheme of
Landau/Palatinate, the motorway drainage system of Berlin, the drainage scheme for'
the industrial region of Surabaya/Indonesia, the Hamburg combined sewer system, and
the flood control system of Freiburg/Breisgau. This program also proceeds from
St. Venant*s differential equation and uses an approximation solution obtained via
very small time steps.
In the Sonnenburg program of the engineering office R. Sonnenburg,

Hungen, using the differential equation of St. Venant, the continuity equation and
the energy equation are transformed into a differential form. It was used for the
sewer systems of numerous cities and towns in the Federal Republic of Germany
(Duisburg, Heidelberg, Constance, Plettenberg, Iserlohn, Ulm, Waiblingen,
Bielefeld).
Model HSC was developed jointly by the firms Holfelder of Freiburg

(surface runoff model) and Sogreah of Grenoble, France (sewer system model). This
109
bil
OÍS
in
|U
ui
<
S
z o
o
3
<
s UJ
O
_l
a. O
bl 3 3
O
X
a: n
in a
1- < ¡ï <
o
o
o
o
Ü tr
o
fe
»-
3 <
a:
< ¿2 fe < <
o x
o
Q
&
o z o
o <t
•- .1-.
<
(T
3 UJ
3
O >
O
•
1 il m
H
O
UJ
s <r
Jz o
3
O ¡i
*
<
O
n >ki
00
o
u
1s IPITATl
A .1 <
o
EXAMII
ZATIO
AND
_ ^
" ^§1 3
Pi
kL
110
program, too, proceeds from St. Venant's differential equation. Quality parameters
with and without exponential degradation are taken into account. The model was used
at Seine-St. Denis, France, and Villingen/Schwenningen, Federal Republic of Germany.
Model INKA, of the engineering office TEAMPLAN Ingenieurgesellschaft in

Erftstadt takes surface runoff, transport in the sewer system, and water quality
and pollution load into account. It was used for the general drainage plans of
the cities Mönchengladbach, Remscheid and Heddesheim. Calculations for the
southern areas of the city of Cologne are scheduled for 1976. A special feature
is its use of the CROSS method for the calculation of branchings.
The engineering office Lautrich and Pecher of Dlisseldorf uses two programs
named FLUT and SBV. The FLUT program offers the possibility of simultaneously
taking into account 20 different rainfall events, allowing the determination of
diversified flood curves for individual or successive rain detention basins in the
sewer system. It is based on the Rational Method and was used for the sewer systems
of the cities of Düsseldorf, Cologne, Oberhausen, Siegen, Hattingen, Bonn, Hürth
and Grünstadt.
Program SBV determines the annual pollution loads discharged from sewerage
outlets into a receiving stream. In addition, it simulates, via annual rainfall
duration and annual rainfall frequency of an urban drainage area, the runoff rates
and specific content components occurring in sewer systems. To keep the calculation
within reasonable limits in spite of the large number of rainfall events involved,
the runoff conditions in a sewer system are represented by fictitious runoff curves.
It is based on the Rational Method. The program was used at Hagen, Alzey, Lahnstein,
Nierstein, Hamberg/Ohm, and for the systems of the Saar and Mittlere Wiesbach waste
water associations.
Model KANIL of the firm Abwassertechnik GmbH of Essen is Che only one
written in computer language PL/1. It is based on the Rational Method. An extension
of the formulations allows approximation to St. Venant*s differential equation. The
calculation of compound networks comprising combined and separate sewer systems is
possible. So far, the program has been used in 14 German cities, including Duisburg,
Mönchengladbach, Aachen, Gladbeck, Alsdorf and Brühl.
Model HHK of the engineering office of Prof. Dr. Billib, Hanover, was
developed at Hanover Technical University (Keser, 1973). It is based on measurements
obtained through numerous hydrological investigations in existing sewerage networks
involving the most diversified system and area characteristics (elementary method).
The model was used in Hamburg, Hanover, Hildesheim Bad Dürkheim, Salzgitter, St.
Ingbert and Rohrbach.
To resolve difficulties arising In connection with the examination of

such calculations (Ribbeck, 1968; Treunert in: Haendel, 1973), the Kuratorium für
Wasser und Kulturbauwesen, Bonn, has organized a committee to deal with these
problems.
Ill
An examination program for checking the calculation of sewer systems
was developed by the Landesanstalt für Wasser und Abfall Nordrhein-Westfalen
(North-Rhine Westphalian State Institute for Water and Waste), at Düsseldorf,
in cooperation with the firm Abwassertechnik Essen GmbH. So far, it has been
used for checking the calculation of about 30 sewer systems.
Receiving Stream Models
In the Federal Republic of Germany the demand for drinking water and
process water is met, to a considerable extent, from surface waters. A number of
mathematical surface water models has therefore been developed by university
institutes, water management agencies and engineering offices. These models
accommodate water quantity or water quality or both. On account of the extensive
computations required, the programs in the last category allow only for handling
a limited reach of a watercourse. Only a few examples of receiving stream models
will be presented.
The Niers model, developed at the Leichtweiss Institute of Braunschweig

Technical University (Rother, 1974), serves for the determination of the influence
of civil engineering measures on discharge such as, for example, detention basins
and their operation. So far, the model has been used in five river basins with
a total area of 6300-sq.km.
During the years 1972-1974, the Hydraulic Engineering Institute III of

Karlsruhe University developed a program which is based on the use of two linear
storage cascades for impervious and non-impervious areas (Wittenberg, 1976). This
double-cascade model for built-up areas allows for the optimization of five flood
discharge parameters. It was used in four densely built-up areas falling within
the jurisdiction of the Emschergenossenschaft (Water Management Association).
Model NECKAR, developed in three different versions by the Urban Water

Management Institute of Karlsruhe University, simulates the processes affecting
water quality. It covers biochemical degradation, sedimentation, phototrophic
assimilation, the oxygen consumption of the bottom sludge in a watercourse, and
the quality parameters BOD5, BODjj, NO3, 0 2 . It is calibrated on the basis of
parameter measurements in the watercourse concerned, and gives a longitudinal view
of actual river pollution. The model was used for the rivers Neckar and Murg.
A similar subject is covered by the model SAUERSTOFFHAUSHALT (Oxygen

Budget), which was developed by the Bavarian State Institute for Water Management
(Wolf, 1971). Here, the variation of the oxygen concentration in the watercourse
is computed, taking into account water temperature, sedimentation, oxygen consumption
of the bottom sludge in consequence of the oxidation of nitrogen compounds, and
other influences.
The biocenosis model of the Baden-Württemberg State Institute for

Environmental Hygiene aims especially at the interrelationships in the natural
self-purification of rivers (Boas, 1975). It describes processes taking place in
a watercourse and, contrary to Streeter-Phelps, it also analyses individual
processes. As far as that is possible in a program, the aquatic nutrient cycle is
simulated, and photosynthesis and weather effects are also taken into account.
112
A model for variegated nutrient studies for forecasting purposes and which
also covers the influence of urbanization was developed by the Geographical Institute
of Cologne University (Rump, Symader, Herrmann, 1976).
Model computations between the various programs showed that the trends
for oxygen concentration are uniform but that the individual values may differ
considerably. Here, further work to improve such models is certainly necessary.
Because the growing energy demand in the Federal Republic of Germany calls for the
erection of power stations at the more important rivers, a uniform procedure for
the computation of heat inflow into rivers was elaborated as early as 1970. The
water management association concerned (Länderarbeitsgemeinschaft Wasser, 1971)
published a guide entitled "Grundlagen für die Beurteilung der Wärmebelastungen
von Gewässern" (Fundamentals of Heat Pollution Assessment) which now serves as a
base for uniform heat load computation. The requisite programs were developed by
the State Institute for the Environment in Karlsruhe (Flinspach, Fleig, 1972;
Flinspach, Wdrner, 1976).
Model QQS of the firm Dorsch-Consult, Munich, deals especially with the
effects of urban sewer systems on the water quality of receiving streams. To keep
the extent of computation work within reasonable limits, only river reaches such
as are encountered along urban areas are covered. The loads coming from the QQS
model mentioned earlier are followed up through a sewer system of any branching
pattern and are superimposed. Besides single computations, 20-year series of
receiving stream pollution can be simulated, and the corresponding receiving stream
flow and initial pollution can also be taken into account.
In a similar way, the SESIM-Vorflutermodell (SESIM Receiving Stream Model)

of the firm F. H. Kocks KG, Düsseldorf, which was developed from the QUAL II model
of the U.S., is used either alone or in conjunction with the surface runoff and
sewer system models.
Within the framework of this Technical Memorandum, the Regionalverfahren

(Regional Procedure) model of the Emschergenossenschaft (Water Management
Association) occupies a special position. With the aid of long-term gauge
measurements in the Emscher and Lippe river catchments, the parameters for frequency
curves and their relation to drainage areas, built-up portions, etc., were determined.
On the basis of these values, graphs were developed which now allow the determination
of design floods for specific areas without measurements and without the use of a
computer. So far, this method has been used only for the 5-sq.km. to 150-sq.km.
drainage areas of tributaries in the Emscher and Lippe catchments.
Related Models
Besides the aforementioned models, which cover the effects of urbanization

on the water budget directly, there is a number of other models which are indirectly
related to the same problem. They include, among others:
- Models for the optimization of reservoir operation, e.g., the model for
the optimization of reservoir operation In the Ruhr catchment, as
developed by the Leichtweiss Institute for Hydraulic Engineering of
Braunschweig Technical University in cooperation with the Ruhrtalsperrenverein
in Essen (Maniak, Renz, 1975).
113
- Flood protection models, e.g., the model for the optimization of flood
protection measures which is being used in Lower Saxony, and which was
developed by the Institute for Water Management, Hydrology and Agricultural
Hydraulic Engineering of Hanover Technical University in cooperation with
the engineering office of Prof. Dr. Billib, Hanover (Hommel, Kleeberg,
Schreiber, 1976).
- Models covering the groundwater budget, noted earlier.
Future Development
Besides the continued development of existing models and their improved

adaptation to actual requirements, new ways of solving problems are-being tested.
Attempts are being made to simulate the complex interrelationships between urban
influences and natural processes in one single model. At Karlsruhe University, a
model is being developed which is intended to give an overall view of the effects
of urbanization on the hydrological cycle (Plate, 1976). To trace these effects
to their physical causes, the interrelationships between the water cycle and the
energy cycle as influenced by urbanization must also be taken into account.
The combination of water management factors and other anthropogenetic

influences in urban environments are the subject of a model being developed at
Dortmund University (Werner, Haendel et al„, 1975). The environmental influences
on man are simulated to obtain quality parameters for specific areas.
114
i
a ai 1 Ci
o d O 130)
l-l b i-tCOalbH
fe •o o OUH3 3
oto) fHeo m
o uuuoití •Huoao
bigOMg 0)
01 tí.
u 3W4-IÖH uuaioxi
u duo au
UdC>ou
3
O isssã
uihUh
30V«ISHH
MB^SUO
•
r-l
h •H
0)tí O tí
«O a O
IM •H
4-XU4-1 O O O O O
su
T3S) 3 Mu) o
tí tí Ci
o
ci
o
Ci
o
ci
o
C Ci C!
CU J33H c
33
OH
bH O-M
43
Ufl CBoo
¿S
C no
O 3
•H O
u tí b
OÍU o X. tí
r-tH •H
U OD uo
•H
¡t O
C¡
al
M
at
o
e
O
tí
a)
>>
>U
iHSt
UU
o
c
o
tí
o
tí
O
tí
c/a
otí Or*
a.
Ol 01 CL
0) <;«
bH
•o a. •HO
a
O TJ0I
en a
•u
c
CO
tí
•H
Cb
•HCL
ä
»
H
o
to
tH
<d eta)
m U-l
 rs
§ä
lu CL m
<ua
OU
<ura
OU u
ce
•ua
ou
ai
O a
o
u •H S 01 tí Ci >> 3H C rH o •H
O boi COI Ci 01 rH cai tí
•tí ntu >> CM OH U OH
fe (a.
0 0) a) •HO •HO O 9Se •HO u
C u
C
> O CM Uri 4M ai UH 01 m
o w •H
.Q«>
i-ii-t
i-iat
OU
1-W-l
aM-l
TUJ
b
•H
•oa
T)3b cou
T^
u
a u
m
flu) tí PO) bai •O
a n oo
XOO boi C C
z OO OU ato alo Ci •HCb MO
OH tío >o >o •H IUHCL o o
o >o o o
X X¡ ca Ä
CO
c eu-« OJrH C b b mlO b S c
uo aies HC« O o o £St* ou •H o "u
I CM
OJU J=¡
a
b o 3 «b
ooJBUOIat
oaMoioo)
HUrfH
bSJüa)
oat aio
•H
•utí
«a
C M Ja
•Hal O . ce
T W
ou
m
b c ouM.
C!
Cal
•HH
«On
bHC
•H
•H O
U O <H
aJçiXIl
. &
c a
e
o
18
Ort m
b
uoXb b
oloajaioa)
uoetria)
oianuoi
Uai
•Hb
a M a)
bS -b
a) al
bb
oouajal
acBai
ab
bol
J>! «1
oob •H Ö C M
•OH
CM O
atb - H
> a B -tí
w UCL
KM
00
O son a)b •
> 0 mo)
•Col «-O»
OßO
•HO
OXbO
/¿COO ¿¿H
U3 b
alai-««) ¿£ >U
OOI
iHCH
rHObOMala)
C»H<U3p*MU
u
.M-l al
o UU M uaioo l»H Ob oFöu oal >» u OOJC J3CD sagcFHCcI uoiou
_oa> aibcbon aibealrt O H OUI OÖ jCbCb OBai aftí C0)Ban>H0) Ooaioi
HI O 3b •a es ai o ù O K « catgue or>a.
b •HOOfr, ai oiai rHrt> CHO rHOHCi
o
CíOt .c oaaubu ooeaat
ad .aorte 38 n o s CL xu-xu CMOS Sacu-Mbai JiBaW
l-i
0) • •
ß
X B„
cue
uo
u
bei
U B„
O S OB u aio) JC
0)3 OO as a CO 0) JC CL
OJO
oo
MO) boi a)
a
01 ai oc « £ ai >i ai u a)
o
tí r-iOt-ln îs ts •Hoiaio C •H b
an
•fe tdfi 3 coC 3 !>> >-OaioOJ >> atu >. 3 00
Caro rHCSOtd J3
poro
&
o
Muoid
3UUÖ 3UUC
baiabo
3OC0U3 •eo
a>H IMS
¿£
•Hcrjsal 0)01 Olal^,
•tíOQH •Htdsa oaia>3 ÍTJC
•tíOUH j3tH(0M >c
4J •o
altíoi 0)
iHOJS u
oao E0 ai a 00 ai U) •H m m 00
0) 0) 01 01 ai ai B ai ai 01 O
UOH >> >> uo
I-40C5 >> •H
iH
ci o
as
ãr 3 3*
M-l
r-l M-l
0) O
•a c
£ ã
IH
O
0)
O
1
fO
<4-t
j w s
H
a S ú g g g
nt
¡5
3
ca
co
g >
a
S5
H
3
3 o
CO
CO
Cd
ca
U
en
*
•H
fe
>
(O
ca
s
S
115
for 1 rainfall
between Storms
O
u
from Water Balance
determination
•M -
Infiltration
recovery of
>u o c OO a
0B3O OU M
yes
yes
OT>H BMB
o >H u OMM 'UU o
tí uaOM •HO oo B
1 II I M I 1
oooru OOM
event
X H O MaOM
J33H
O M M O .
UOOM r-O
3-I03
OOU> ao «H Bucr
Moau MBO
00 O-HU
B u
tí •H •C
Pervious Areas
J3 •H • H TJ
a a, o Ma OB ÖOQ
 •H MOIS >6 3tíC •H O
OM
U U rH
Pervious Areas Pervious Areas Storm.Runoff
M M rH pa- O C«
U
O
00
o
M
3
ja - C
MM3
MM
UO 38 B u
MU
MB 3U
.CM
uoo
u o" OCE 0 3 Mtí
u
u
•a •Htí u
•HUO a
MOESOB
•HU
MVI u - •ria •MM O
3?
BM
a >> MUkHB OM •HCB
UOp
U M H M O 4M
J3 uo tí tH M M MM H U
TJ
U o ^BSSB OH» Buu eu U>U
M H d«c
M urne OMM O H
SI
C
U
•H
BU
UM
u
M
•BMHB3
U I H H C
•oou B33 -HOO 85 UM
r-r->>.
O O
tí o >oo •HOO* •9 " H MM
•H
•H 3
tí3
•HO- ïflattn £&* jeaa BOU
MM
OS
BO
OO
Buu
OHB
JEO OM« •HO UU rHBO
Other Rainfall
OBM •M TJ
«Ml •M 030
B tí
MUH a O MOO a O
aoa tí a
Infiltration on _ Losses pn
«M •O tíU OH u
tí oco o O a Otí tíd >vc BOU tí 5- «
O •H3B -H tí tí OHO o OB O
M U tí o HO o M O H M I •H g O U -H rHO -H
U H evo •H U - H
•H •O-M OH •H OH O OVO* B U •H U
tía MOOOM OMH UU OU ooa uco artMO MUH
Ma -H3U 000 B " oou 0*0 M
•HMC O MB Mtí O H rH min Ba M
U U 4 Ü M u tía •HO M 3 M
uoHao. •MO •UO MUMa MOM
a ua uarj tí MU •HO MU UUMHOM -HD -O
coa •HH00UM o u BBO nunnui
ovo BUVH> OHtJ o 3U UMa SO uoooMU OOMH OSO ucmMa
i-iQOHO 3QO tí aM uoa au OOOtíCO UMU
OTJH «MBU •O » M O ooo>o
tí UUO
tí B y O - M J U B •M 3
B0 H O H
o G O
ya-t O •H O
•H T M ) a •H O •H •H o O -H
a cu a U •H -H U U •u a •H Ba u
o M M M o u M
•o M 3 3
O >U 3 3 B u 3U 3
S aiac
tH o* COUH cr cr B O M MB cr
tí u OUCM-4H u o O -H 3 O u
fHUHO U •H UMMH •H O cr
O I M
MouBu 0 H C 4JUH>VM tí B Ullil o B
-nao M S o MUU M M o o *rW o
U 3-rH
MMMHU
MOUOM
u
ou
UM
M3
u
M
o
•HOHUMB
MBMB
M3UHMM
-*
u
M
u
M
•HOU
MBO
M30
B
o
u
sa u
M
O H M > ao OCT
3CU
>MUUOM S •S r>MU
M
o
U M
MM
BO s
Bu oo
torm Runoff from
"M OO
u
Areas
OM MM B UU
UU tí OO -H • H TJ •M |C
O D. o B OB BOO •H •M3 >6
3Bfi •H O OM
M U M u •MM rH Ba
U 00 M O MM MU o 3U
O o MB
Împervious
3 UU QBE MB
u u u cr §B •HU U - •HO« 03 O
a TJ MM OU •MM i M TJ
•HC u UMH
>» U O OM uoq MO BU Uù+i
M
M
O
x¡ £ tí
o
u
U *
uoa
an>
rH U
OHO Buu
MMM
coe:
iHO O H
U U
MU
BCC
U U
S
UM M«B OMM •HOO Buu
tí u C3 u UUM B33 BO
•oou • HJU 1 OO OHB
-H •H T W •HB3 •§£ UU rHßO
rH C JÍO T>HU
3 3g, ox 03
03U
a
Impervious Areas
r-« UCU •MO

m a
Other Rainfall
M MM o •H
•H • O •na •OM «H •oo
U •u H ao CM an BOB Bo
OC OH
U MB
BOO OHO OH
O
U
tí
53 •H3
U
T*J
• H
m OOM
OHUO.
-H - H IM H U
COM
M M on M U tío
rH UUM BBM
U rH ac r-IO
M OH aco «ri
aa MOa MOa
•HMB • OH
-H
ara •UOHU o MU3CL U M
12a
tía •HUC UOrl tí
M C« MUUO
CUMO
MU
OO UOrl •HMM o 30UU B BO uoo
orj •HO aoMa O O rH
OH •HMOO CU« tí UM
o OO oo->
tí«« OC B • O M « OU
TtOO ua WO
a •HOD •OM
CP u u
oo
»a
MUM a a
•O
o u ss
h M
UOM
rHB<
O O
tí
O
tí
O M
U
u
a
h*
••ss
o
B
O
tí
O
B
O
B
OO
T H M TJ3M
<H JOH •H OH00
títí w •o XUO
H O B •HCM
de -H •MHO.
MO
H «U UU
O O
TJ
£
•M • m
o U
M S
• «M M
E U 00 •§ U] U >
sa
M 3 ar
ar M
o H
on tí m
Ol i
116
o
U ta
ce MOI
•HOJ O «JO
O. a o u U-l
doeo u u M 4
o •-> ca 4 C al BX>» »M
•HMU M Kl OM ta u •HW O
UOC •Hol O 4 •OH CD CD U
id C u U UU •o O ai 01 MC
MC04 c C -H4 >. » OSM B OO
tlh£ •1 ta XI» u >> >, OH
•nau u u •ao M 30tC BU
-HÏ ca «u o •H M U T*t4
LMOI c C •ï a. oaM •a
an o o S c>*> «a
M o u 4»«
3 •H •H4* 44
cuca ca CD
oc CD 01 ai
--«o 3 > xi a . X»
(WH .-1 U 14 (0 •J
u
UCJ
O. 3 0 3
MHI U
3
•HJ», Ï5
XI •
coi
•HH
ecu •OM N XIOM •H -H •-t C XI U-IO
ata C O M 4 «-»-ï«) cab al f-H I-l tH 4 -i ai 014 >>
B >, >»
C
ni
alai
•uca
34MinM4B
OOualaKw
-HQHOUM-H
34toMoi
oaouo
4 4 4 II440IU
M M 0 3
•au 4 S 4
viccwa O U r-4 •4
»WO Mata4CM-i M4amal •"•40IBO •H
•rllj ali irr1!*! --IC04B
OCJ n tuwwcw
><a4ca-«'aw >coro4ta 4CDC0OH ^c
* H
•a H
CÏ •a •H
cao d • B >M
,-C
OCX
01 Ol
M
4
U .g
Oro
•H - H
«
OO
c >i
a >l
O U ca 3 'S 1 « C>M U
«C>U>cc
• »
•H« 3 C • H JÍ -04 MOI 303 e
ofM »C C 4C HIO 4-H C ai c O eoiai «MU 4C
M3 OO oo CO rHQ C3 O u o m •HM un CO
atea •*4H f-tW ahH 3UM « O H au>H uo CMU OH
.Cr. EU EU >u oao •>HDOJ HBU B óo ca A * >U
uai alai ata) 4 cecu UM4 OaM ><i: M coca 4
UM
3*
C3
g&
•3
UO"
M4-1
01 M O
•C0I3
ooeo"
EC3
oaur
oo*
>H4
aiuai
CCM
Je •3
ucr
ca •x r
a ai •OOI cool «Po caoo« •X>ai JHJO0 ai ai a . o»o
CCOM cao
CD
09 m B
ca C u U o
M C c o ca 4 -H u
Ol Ol o o E OfH U C
•H •H j= CO U*>U ai B ra ai o 4C
u >
Ol >u >+J u ai 3H4 u o ai t-icai o BO
3 OS JM JD4 l-H .OCOU alun .onça B OH
u •H w-4 COM >> C3 HCU ->> •H4M >u
4 3 mo •aia. Oa)4 caual 4
cil
Cu •H
o »1
(M a
O-H
ooo
M-H
couE
«xo
EB3
oa>o*
caMu
oaïc
•3
ucr
(U X» X» u4 wuo U>ai ox>H LOO
fects
3 ca
<d B
li
•O
> •W ai Ol •u
SB M U4
4> 4> u CD C03 ca CD u
u 3 C o ai o" ai ai O e
o »C c 4C B C90 B 4B
u oo oo CO >» U >. >> BO
dl •*-«r4 -r4r4 a>H aiu OH
3 Bu EU >u ^ C >u
A! atol 44 4 a« 4
•J C3 •3 EB •3
ca >C &
*oai
UO*
caal
O0I ucr
ia -oo u> cao
3 uai oo 4-HUÜO Jt B
o TUS -»<JC CD o « B J20
r-l OEM Ü0H C O o. C O i B u IM
ÜM -H0IC ---wc "OO M M O -H ai - o C •HU
rHlJtí fJH •oo ai -H U UUH ai «M Ï4
cu a•i H
a aMMH «u BOI B •U 4 Onu -H HU rH
4
-*a >.
an
O u>»
ai U4 3
cnSMcr
•H 4
>H3
-o
4
O
C
•HCO-
>. O O
IM } ^41 $ ^4J ucr u ô BU - ooa) •HOO" M 3 hOM
M ' CUCUB UOLlUC •Hai •Hrt-ii 4>HCC eu ai 4 ai 00 M 4MUC
3 oo oco 3 S O H C300 u tau OB U300
ca • i.iai. ico.i 'ioo
i. moco
incOH C>. CM3U aiU-tH aiuBu -HOU rH eacM
ai
B i H M M il i
oiai. 1 3 1 m
6 i n n il n i
44H3M
•HOO
UM
•HUOO
U4BB
*3 M > m
U44
•H ovo
04BC
if^>e
aiu4 Zi
M E
003U
Ecr-o
ai C3CH 3 C30-I 3 CO B3M4 •C33 EBC4 E4B ai M OOBH
M ¡»oHoMcr OB OOVM uootr OOMM 3M0I C O HMOP
Cu - foHohtr
OOU-NOO« •aaiuroo« oo u«rp< cauaMi C » ai e OCUJM
JS
M
x: X •Mal
JÍC je
.
o
-ao • O ü
MH XWI
E IH
3 4cc
4 4cc4 xipua CU M
u u li UHO •MM301
ai oto OKI ca •BH3MUDB •H 4ca>> a
ai . XI OOUM
2 c C Ol an 3«t: ooo C OKU
-H00O •Hötrun o. •HBOOO •H C O C «a a t o co >>
C >»
e « M O
iH otfcpi «COM 4 4 coaiuc 4MHH •Haaou ai* 4 4 04 O
01 MHM WHOM X •HO B 4 M •HMCU ¿ C -H XlCM >HaM
e aiOHO MC9MHMO MOOO OH03 aïoo UMO
c
m ?ss
C>u
OHO
;>H S
e?w
oHcai
iH
rH
iinmi-HBi i
ucouEou
MaK-ioai4
0I3B0I**
UU4MO
M0IMW-I
couaai
4o ca
MOOOQ
XCS
CD4U
«Mai
ac3
>UC0O
CM«H
.c
CJ WOG U04B 4 4E4UCOCL 4BOOU EDC 44BU
JUSOS

CD
> 5 S3
55 ca er SS 3 ca ca lu ca
s
117
a Ä ••o p
ai P
•a - H dm
uai
CP •C
H m
c -u 1-1 CUU M aio pHMUH
O Stí o Bo», pHODcd
a -H
mu U-IMOPH •HMOpHH m
COPO
<ud-H
OHM
733010.
OIHOId
7330100
m
0J m
01 m
ai -Z
TJc*Ha)
m
Ol m
ai
m
ai
m
ai
PHCU 010) ai ai cupedp ÏN
3M> umBO P m B a m >» p».
>» P H C d >. >» >»
3 aid 3 cuPai 330H
i"*
o o
OCP
Bump
OCPC3
EOarjHH
ar>H
BCPÏN
O pH 0«>+*d OOOC
tu
o«>o
uaicao oaimo> oooed
m
Xi •PC
>.
O. M CM«
•O ol ) !| [ [ Wl «d O
C h ai d e n oíd p P H O p CP
« oo 3 tue 3ai C C«e d pHPd >% o 3«
aiEo aiE -H •H rHdcdtN •H M
COPM
no man-f 0)0!
" H MO)
O O C»H C UJ ri« X!
p-*P r-l o apm M-4d -H p cd ai 73 a .
P«wo
H>, xMia. XiOl a. ol?
a. • O O H pH •HM
o -C c«
3WS u O >• :*. o >. ai copa) Ol H 03 «M
B 01 _°
DB
O IP
OU
ME
aiai
d
cd
d-4-KO
« « w d
e
cd
P O d >
3mo M
ato
DP
o
a.
auJ
au
00
aïoP
O 00 CJ p BMH cupHoia) C03 0CMC
u cu MOH Mm H M MOW) M BOPOP M*-lc M OI73H
4J 0>M 00>m ooooc OO >. 4JP U
e/J ( o>»
Ma o
UH «WCQÛ "o
H OCOCH cp a . C
cd aja V>
caca.
mmo.
C •e m c c C i
O •Et 01 «H o1 II m O •H o i
•H •öo
Ol O H aio O OC 1 1 •H SO co
p CV4U OH d CO P O ai p acN
pH et 3 o
Mm«
CCH-KB ctJH cd 3H cd c CHMO Xl P C M
«
p X W 3 .CP PH MP 3XÍ «
3 P O P
mod
3 USO" nsO cd C CTMcd
c r 3Pd >w •p*C
O. O Ol XI aw-tc 01 o ai
CH «H CP
• «
0IO ai> O omo
P C
E O •HOI
O oa>. oa X! PB >> C »M 4s
CJ CP
dcOH C
ded
•H cdM
pH a.
cd
cao
cdaic
p
•H
Su»
« 3 P H
O
•H
otBMmP
OHOOHC
01 O 3 O
M
p O.« M Coad 3 O H pal •HOOH30.
00 •Kid •H3 Ol P 00 ai d cd 0 0 73 3UOH
P Q H •H 3 m
O Ö H MHMOrlJ
m PO s 73« d o >oon
Vi a-ttJ OH 3 OOH M dai p H O C • o«
HM O« -HP«
O OH pH OMm • H a M-IO-H 730 OMOMC3
p sud
3 CO O
•o PCOH c O H cd
3d P H C «
>. o (ÍP oseooer
Ul M-pHU UHH casa. •H4-IM •Hm «pcdPoai
> w u x; o
C m . C
OhH O
ai J3I « •H
Ol
m •rIMCU
POd pHOdm c
•H JJ X!
OM Ccdd s3 cdsoid d M ua C
EXP aiocd rHC M m P H ai c C ai C -CMC O
«i 1 m to
c -c P OO O o o 3 « P H
u Xi oo 3 aim
•omo OUI m
P
VU 0) 01 01 ai o p>H uai Med HPH a) -p+H • H .cdo -»M •il M 1>P
OH Mm
un cd oo xi P H P O MoPOl 4JO>HmOOU PouJmPaio
vu >* >> >, CM M C C C«JC o C3CC ecc ddCCOMO)
cota ÖH0I aioM MHcai aroaM ecBHocdc aidapHaMC
PM •HC0P 3Pm p~o o a . P C u a •HoS
mcopcd Pcooaicaoa)
O P H M M H M C ¿JcdPoMOC
cd cd ai aiaicd CPCOHH aioaH al« oie; ai«P>«PG aitíaicsvico
cdmatPa. MOMO »lOhu lUMOXnu MOMOOHO
Un3 mua
d cd
m O cau
txts •H cd« 73
MJ >» XJ 00 73 ai
1do
-M O P * ) x i ca CO M m
OP •HO« pH 01 m
01 a
ai •5 SP •H a
ai O al
B« 9H> • O SS
>, >» 3
>. C PS
Bau
3013
mE
OrH
ISfë
333C M-tai
QUO O M O H •HM
>n
« C
1 XII co m u oo oo O
Vida I m Pdm C 73 73 •H
o5e H-PH5C a. M
ai ai
C
•H P
ai cu d o « C
d CHOO •PW300 ded 73 d
•SPSS •H a PC
«
OdoH d
•tw
m
J
P
CO -OH
oie « P
*J X !
cd
o_ •41 O
d
3
73P
• He CC
P
OHP
•HU
to«
OEM
•Hdo
PO
M H -a
•H
H •H300H
•OMoaa
« C
•H O
3 73
oud
S3
mai
OOO
•ptU «
M
OOH
Pcdm
MP "«tí O cd -Hcdd 01 - H 410« 3MBm cUP X*m
013 o»ac MPP O >OJ C
cd o do oa oucdc oooo otoo
3 Bo •adcd-J PH tXEjJBo 0IC3 73d>> aimaio •PO
•HS Qtd
(DOIS
do
cd» ' S B
cd
1 SfiStuB -^33
MOM
ctMP
OMH
>3PP 7317301
OH
«P
ÄÖ
73 3 P
ao ptd« Ot&MM« M4M 73ua> 00 COH P H M « 01
o m CmMH
MPal M«aiai MBPP0I ai S P COB MX!QO oimoo 1073
•HtOM -H3WH m aiMmmM P M m cd d o
OMS
O a»P3d août ce OOO
01QP 0IOUP M COMOP Cd OrH P3a> CHOO 01M U M OXU
3oimm O COHOO 0I373O 73 O « 0 0 oo
33« aiao3m 3oa oo «Coo
ca u IM •H H3 oo
C m •a -Cm ai o 3 c d 3 -41 m aie alai
aie
MmP • H 01 COh-tOI > OH » OM CUP ai OH OH
m
01 o 4Jum >, H U H U « ot>
aluo 00 X ! mu «c a «C >c x> *e
CM013 «M O •HUU maiCd cd UH COHOO PO •PO
>» C3H0M mo u
• M > M M O I aid MH0IC aiMcopai C3
•HHP
COLO
•aiHHiJ
H>0«
•HH>è)
OIMHM
•HouPm
cttH30Kd T JHMHO
•HOiaiMH ¡8
Cl M >s
eu
P T O
CoH >s
CM
3DOO JOTO MHOMQ 0)3>OP 3ta al MM
aló ata
ai
Runoff
•a
¡8
£3
CM tu 09
O u
a) J
S X,
CH H ^
B M
3
C/3
<y
S á
¡S
g g CO
b>
O
g >
ca > O a m S
*
OT or X M á en u¡ PS PU en
s
118
CO
tí 00
o Vi
i-ICO • a)
4J-I 0) 4-1
Uli Vi al
«tí ü ¡* r-ltí
aití a «lo
PS« o O 01 o O »
ai o o i-»H 1 o i
.tí tí tí tí tí CO tí 4J4J 1 tí i
5<J >. 0rl tí«
u • o>n>
H -rl
aro
•rltí tí*
r-frl XI
4-r-IO
OM
al OM •d
otíai xai
& Con ayo ai
1 3 00
u i
tí O rH -r>
aitío a* al C
Bccoetí o o 01 o o O o 1 o O
tí tí tí tí tí tí 1 tí •
• O
• o u o al >.
«ale JC
cous) o
oo al • •ao
tí .cao al nçj
•HW «tí
-rtrlIOl 01 •Seal
3« 3XOB UM 3
-H3W • C ai a •o
Stí
al
tíEa*J
o o s°
.-"O, o O
DrIUO
O V>rl r-*J
•nn
4-IO
1
1
o
o
•H
i
>C •rr-ltíal tí •ou tí tí •rltíO-l t
i-U u«iai> A H H 4JO03 oo. tu
•H OHOtrW UU Orkaol ai« 01
r-ltí anona. «OO omevi >tí Ä >
• H >V**B0I atí-H >v-a)o tí» 4Jn
3 •oaJataiu •o«n>. O Vi -o
Ol cy aiaoua a) D U C 04J »o
•a
at tí
tí ai «rl r-IC
O « *o
>H tí tí T»rl
o 4-U-I 0) o
•rl
4J4-I
•HOtíB o o O o tí«l 1 o
fHOMtí tí tí tí
«J tí tí
1
tí
Î
u «n o o tí«
u 301 4J 3 aoo
3 O U al
u tí U •oa
oo XO)
OJ M MU OrO
> tí
U o
•rl tí « o tí 4J -rl
r-4 OVltí 01 o Ol 4J
al iri -rl 01 0) •101 Vi
a)
3
•Halaj
na
oo -
S a
o
4J
VI
VI r^
-a •O O
a cr g O «0>fi «u • O 4JtS
a •4-1
M-l
tí rH
3U3 « c
o
14-1 -rl
a. M3 o
r«M 1
tí Orí
•
<utí| o fl-l •veno 01 tí
u O tí al O U Vi OW •rtO 1 o* * ^ o •
tí o a) 0) •«rW 00 •> « - r H
«I aleo •H tí Vi 4-1 coo o 3 4J014J
3 tí
•u aioal 4J O U al oo a Cvm
al « Ö U M-l - H tí -SÃS .Orí
i-tal « u 4J «| 01 4JCCH 4J «aivi
S •H
•M g
O
r-IU-tO
tí r-IC
4-1
al
3 O
.-1 tí
4-1
«0 OÖJÖ
coiaro
4JO - SBal
a -onoi
O •OHalo.
o
O
tí CO OUOp
tío o b
4J
0)
•H O
•Ö o al 3-4JIC
•MO*fal ü axon
cq m o a .
>> tí tí 1
u U
-rl ai O O 01 1
«J iH a Ti -rl • O rH Vi
a
8
al
fio S
01
4-1
al
4J
VI
•OOI
•o
•a a
01 JS
o
O
& au
uu
B Vi
4-1 es°§
aMakD
4-1 00
01
4J
«1
•a «tí
tí Orí
h u I-I tí 01 tí
<D ara o al 01 Vi 4JCWU 4J 1 O. "*lO i
.C QflJ
c o o 01 aloVKM 01 1 ai aiVtrl i
4J uai •H tí 4-1 a . olatí 3 4J4M-1
al r-iatai 4J O al ot> g 01 ai cual
al
-We
«H3 o 3 XMäi ^3
Coa
3-ic:
•H
4J
aj
""I
ai
4-1
a
r-trl O C 4J
3C01U4J
vra » t o O]
-JSP
a -OViai
>,
Vi -HaM 4J a al ooiai3al O oKau
•OIH ai o ? X DO-HO i PQ OOOCL
o m
tí Vi
tí
•rl
•a O
TI
al - » O
VI
ai •Oaieaai Ití
al •o m 4-1 •nosai JÍCO •o ai
U ai VI i-tHoai 4J0M ai je
tí 4J m
ao O O-HViOI TltíO •a utí
XD tí an O. atoou 3o<a tí Orí
4J3 01 «0 atcoo oo ai tí
•HU 3 0 3 « Vi ai a . oí xol • o. «.«o i
r-tH u e aleo 01 i-rom » Vi OÍ Vi 1 01 mVlrl i
au-> •rl a
O 4J 4J jooioc ai oo 3 4J0U-I
atacai
onsc
run
ü al 4-iuai ai- cual
35 •Si 01
aicaoo
i-iai o
gao
o 4J 4JQ.+I mnala.
ou a •Hrl
r-«o a) VH-n) a -onoi
c
•tfO
o
n
O0I
mai
al
?
ESA
tnajoâ
«rM
034J
o
A
oxdu
aoo«
H
ë
S
O-l 01 pa
01 OH
aro am
j
S X
1HO s 3
§8 vití m ú g Ol u g > «
O n
ss 53 g ti
i l-l * ca ca
s CO
1
119
c
•9 a.
UH
o C» •oai 3
«U OO •
ueo •HO
e>
«H S
on JM U O
OU Olatai rH
.c
*g
0»-l
M U
d«M 0
al
ai
oo ai
3
•
0
US
rH
O 1 O 1
BO 0 0 uaM CM • Ci •
•H
4J a) >» g«H
>:•. 3>U jam J5B
•HU jOUC U013 boat 0«
OM •HH »Kl aion 3b
O H «H -1K a lai 000
aid coatö «ocio COU ho
0I3C ai o l a OOC .CU
P43 >ott OtJH v a
or
tJCi •a 1
ai S
uo
4>H u 0
atu ou rH
DM ai ai iH
i-i OC»
(KO
CB
t>
m
ai
en
a)
0
ai
<a
Ol
m
Ol
• n e ai
00311
•Sib 1 0 1
•»•4H ou .COO • Ci 1
sen >> >* >> >» >> IH ai 003
•H o-m-i 3H
«IS nouai OO.
OO ' U U T3 M
U-4 i n
P*M SëSè
01 a
oca •O
O
SE
H O •« 0
a ai
a,
qjjj O -HU ai
POO UUOI uou CO
ç aïoo
X !
c c w ai 00
CXS-IWG
0
(3
0
0
0
c¡
0
0
0 I
•
ai 1
1
UOI
U3B<no.
S nue
U3UH
c >.
•HOI 3T3UÇJ
•H» 0-13 ai 1 1 HOB
KU cMooai rHH 01
SOI 30UÖ5 03BH
u-iaaia)in 0000
an
m
C 1
O 4J M
>HIXIS U
uuciai C«
TlUriQ O O O O. 0 0 0 1 0 1
H 4 W Ci 0 Ol 0 ci ci (3
at u p a a »4
3aSai eus
o u u 0
C H OH
1-1 -HU
0
•H Ci
CO U (3 0
UOI O 0 •H >.
tH
0H EtOH U Ci
<uu 3 OU Ol 0
BH «-Mot 0 0 h 0 0 1 O 1
4-H .fc 0 c¡ Ci • (3 1
•»H MSU «
O.
c >»
u
1IU «OO Ol •H
Hol t-wai U U
tabu
conc
t-*w O. G
oj
0 3
0
•H
oou
cj 0
•H
coc
•HCO «M -H w*> •HC3
>i 01
umao
i-ia»H
UdU
OO -
CBC
•33 ato
T*ri
•HCM •"OC OMUO
o>
•H UU
HOM a ai ai 0 0 0 CM • O t
• H O mon B
c
e
• H C OÖ
O IH
U G «• 1 Ci t
pi m
aw H O B Ü Ä3Ä S3
OXM
» M « UHH3 •MI Oh
at r-l B UHOCO U } U 000
ai CCCM voatao xo
0!
OHOU
«SB MD
HT« «a
eOU
COO 00
•H« • t3
UH B -H
30 C 0
c •Hua G
OMH •OB3 r-r£3 O
1% •HOU •to •H
* O
U M
0 eauu •HH 1
ue
1
i*3 0 Ci «>U 0 c O HH UU 1 BM • •
¿MO iH •HCOH CM O CM)
» 0 «M
uue
-HHçu UODO •HU003
r-W
•IM 00
w-ioScî
^-ffl#HH
Oi 1101
c9coc3h
S3
on
U-KJCTM
-HO M M
SC 3 btXOa >p*o 000 HWOO
OH 1-1 3Ot0Ml X« 3HUO
auava 1 O S BU
r-l
•u
•H
—
0 O
•0 C
3
á ai
1
M-l a w
0 0
J
u 2:
M
•H
g 3
ta S 3 3
g CO 0
g > M
« C? g
ta ca -1
S
SS
en M
2 s 01
X
tu CO
120
•o Ä • CO ••»»
•a o coco O
«utí tí T*
.tí
O c
o •oo
tí
•riri coo
CO «MO •ri r-l cd o Ori Otí JC « cou titi
*4ri O O. _ x> MU OO *u «cd O«
•u M •030 OQ •HM •riCU -
M
QUI
Ctí
33
tí o. E
Ot -ri
•o
Ol M - CCOO rO
ÄC
•rt UUrd B
Pc,
e
O
O
I-l M0B3 cd«M
Ci
Míe
MO ti
•ri
U • JC « 00 OMri lè-t
u •
OU
• o «
»rialau Cd
uva «o
-ri
cdoco •OOCO o •cd 00 efl •C .noa 003 O . •rit] r-l * 3 rH -rlMO 5
JOUB tí r-l u o 30 T-fOU tí o . Ol «3 ococd aoriM*
3u0 Ho
«l>rl 53 CM
•ri-l 300
cOC
cou OCO
TMrH
•ritícd
UO uItí
o
CM)
•riB»!
M «r»
•riSriOM
ola
cfl2 TJ •9°° cotí •riUO CO O
Ori otiacS
OriO Oco (B u a
COQ. OOO • •HM «on tíOC •duo
r>C0C MocdcT
OHM
au
•riU «
UE^>
cotí
o
oco
u
a?
Uoo a • õXo u
SOA
•ri
UO
•riM
-HO«)
O 00
ja
4JM
rHd
4-lCOri
a. « -
•u « E M
Tl
(U
•aa ÖH d-io ouo cdou •d o OMco - H O 3
coife
4Jß O*-*
oco
o>o •riOO-l
• •o
J400 00 B •< « ca c
*-4 JX) OC ooo) <ou tí
«<
•u •ri« o
DiVO
cd 4WH ISO
•OOJ3
UM o >«>
•riOO •OC TOO •a on oaiorsu Tl
nd "JÜ cd cou OStí SOO •rrl ••OC o u •ODMMceri +J
ad
HO
•ri 3 M
Ococc)
Q>H
1| | non
m cuo Uoteu
OCO
AO»
•qcoc
JOOri
cotíM
ou .
Oricd
CDrIO
u-Gt
•ri «
rHCtí
AS
•ri«
OMcd3O&0
U ¿ON
c
o
oedeo VCLt
•ou «tíB •MO •ricdO 4-1 M U cooc QÔ *JTOD&-#ri o
MOCO •Dtl 3UC0 xocd ooo CUVC
S<w 333 ""
o
oa-i
ss- oebci «•H atra
o
4-WtO
«MJ
O
•dtoal
B-i • H CO O CO
•riO 44-0-1 r H r-MJ
HK çdOri O JOOO
IU O o 0 o o o • H B M O MC0 o •riOO o o
oou:>Buo CO - Õ
Con
Real
C ti C ti c tí
ouc o o o
tí
COUOM
ti 0
03OUM0O 030cd
ridOOrUe QOO»-!
CO cri M M
a • O o
o CO •o •4-1 CO <H
m
tri 0rl o « u c
T* o
tí
MO O C
o Ö0J M o •o X3 Ori cdoeo u o •tooo
•o -ried
XU
o o
c
o u 3tí
uo
iririco •MU
cd
uoo
OMri
•ri
-ri
CO CO uoe
cdMri
tí uco one
3
03 •riri ooo r-l-l TMtí •riOC
31
«So OOJ •ou 03>,
ao.
•riria
tío MK0 •OB-4
ç>ric
rHÉS o»H
Ocdri r-l O H
o o ocd oo oh>> rH«!-!
B
u can COM
-OD. •rico titm S «o tina HOO cd M o .
o u
M u 1
3-M cd
3 OOtí o
J3 tí 31
u ovio
MUH U 0>US
a •HU iO •J CO CO i-fribriO M MV CO
o CD O H 03 O
ca
CO
fu
« o o « U e n o- Oo
T*f* ri
•oou
J1U
a o
CO
SHd
O ti >» S tí
>> >. •UOB?tíCBri •KJ >. >. a
rS
S •ri tíOOOUM "Otí
3 friHO
X>
rtH COM-trCM 003
O HOU
lane
TH-I •rim • u trim

< xtu
tí
o m
tri OH r«
O
o.
o
s gs
o
CO as
E-w <0 CO
m
tit o)
O,
u
CO o •ri > . •o
o
Ä
Ci O
"S"
CM uqoco o>e
a
«H
•i
*H . u
X co
-riri
BU u
O
CO -riOco
•riH
cd
MO
« 4 1
•4-0-1 v o
OriMO
u
•ri
B
•AU
tí>.
l
O 5° >> >> CO •uoo •ri
Ci •oo
S g. B
CD
8 S O uo
•riri o
cS ^8°
Uri 00
•ri "flog
Ort
O «m
M •ri
u
3 BM tí 00 - C m u
10
3" -^
. 4J •Ico U C •triO •OO tí>Ori >.
rri «aed
•ri
•o tí Qri T» •ri 0J •Ori CO-rHU
•ri INDU
M O TV> ex m O B 3 OU .-»-«33 _"3
Ol
g
O
OH
ÜS¡
p.
eo
olí
tí « N
•i «a.
CE"
o
•ri
•
O
IO •SiS
" ^S* •MM
OM3
OPri COMO
«I3-IO
E Brio.
OBB
Mrio
•H-KJ
r-l -tf BBo •oou C
o
r-l ra
O tíOI O •o •oo
tí owed •i •riM « « CMC
e•t B •riU
o •O •
8«Ç «M O O U
o
u 5 •00030 u Joou. •Va«)«
tí £? & e uo 3 c? r-ILLJ Uri «JOE
M co- •J
m Mu
3 •tí
ri o Sttí •d ooo Tjalri
•riOO O . TCO O
•ri - 1 MO
OC B cd o u OHUd -ri»S3«>U
O <*J • H O S tí «W «O «M 03 «CMO OB QO.-0
MOUTJO m MtílO-l ue caonSoooD
-rie ce
o i 3*001 -H
OriMO
«M3-H MrKOUCca
H r-r-l 5SSS3 t-i coHriH-1 r»e -OMOM OU/V>C9ri3
tí
S 8
cotí
30
3d
OO
oVKa
tíuol
•riettí o
u
Ü
U0-4
CO Ol CO CO
T*
u
iri-l
m
M U o. O. «otí O ta CO CO
CW
i-4W
3ri
•O O H « O OCO
O o r-m
O
m
•riM
3 e
•rim >» >, a o o
cß XJO i -sEiíSÊ % ^. r>>
>. >.
Ori •riOG CKUUO uo.
o u
Otí
cao CMS
OOU
a coaKoa CU
•rico
•riri
ace Runoff
of Model
ENBURG
£
1U l-l
S ai « á 3 g ia o O > J¿
Of 2 H H
<t 3
g 5= o O «
55 Ol
or H a t»
m s ai
B
121
o M
•H • u
MO O -m
O« •4 -ri O O CM CM
rH
cu m (•>
Ud- r4 rH • a\
»o <o CO 31 ^ M U 2 M
SS BC •
«rH
M
rH
«M
a
« «J •
jan >H
M o -i
HMH
onM
UXH
•nM a «M 0
ckr
i£ •to»
u H Uu
MO o O Oo «•• u ^3 O •a 0 • ¡EM
o
ut:
H01
o v e
53CO0-1I
HJUHtí
EHUC
S°2 E?3
C
sta
<ou
•HÖH
c §MC CrH
twcBS J>
o «o
OMH
0 <o
OMH
•M
O M>
M «O
O
A •OH «COMO COMO COMO an
O a
NU . OCOOOC ouo
uxoca •a
a Pauo p<uo
MJ
«M
ïua C 550C ssgg MUH Or-M
O U 3c
•Ma r. o « o •8CM «9Ca
* Se
lo8
caja 13b -JTJIU M 03
83B 3% S? 1*
•MU MOM
de MOO Ut
tí is M IoO w ao
S5*
r-OIM
OC
•CO 53°
«rr> ¡k2uO 03O-I O TtO T C O tfUKfO
1 M i 1*7»
• •3UD
MriuMI
OIO
Mao M«n> OMn
M I
aao
Oan
.03.
IUO
• m
ÖBBO
«O
•33
«ear
•31
oooo o MCO
UCO
DO*
«HI
MCO
uao
3H
« 0 *I
IvllO
otHOn
CKHMI
OU
s r>T«r>,
8*H Sä t-UQ r-uaj ¿3§£& ri»
HOT«)
Oa
M« •HM
•00 :>" • sT • > >r«*
.uo W U H « w
>
rH H o a >
n
>
M >
M >
n
33 a
ou •o
O c =5 2 ss zoo
3aB
I i 1 1 1
(M
oS •J
o
§
u
tOlJ
MOO
COU
tvnm
Is
KSU
MOO
cou
W - a
H
M
o
Pu
r-l
*v»
tí
«
Fl
M
o
fc
dio
0c5
coc
(«-^1
H
M
o
«1
£4
M
0
Ut
H
M
0
fa
•H
M
0
U)
CO
0
-oeo
HOU o c
mud rH 0
#u
-^
boo
ouB
SxMO
rH
coa-O GO » in
•M
CO s
CM
vi
CQO
e
•
Tt
<o
r*
- U Oo
roo o • m
u
hQH •-»HO
O rH
r-O
r-l
-^
rH
^
m
coo
M
s ta M
0
OM3
uoo*. o
Bt> CO
aoN o o^» a
3
O OO
vO-^
O *
lOD
•0 a
c
.0
=UO
O.PÍ
-4on €rU
PS O
Ti
«OIO
:oo s •
B
l-O S» »o se •
so
oc
CM
MN
Uri
SjOMM
5ioo ¡§3 K O
M n
IOM
l-M
o
>
ASO
CÕ3 «oo
-ri
Ol
on
il
r-CM
•o •Q H
c ta
"aos «cH cH
O 1 v^
•u o Oo
H Oo
H o o o
o o 0 0 0
U 3 •H •HM
MO
•HM •H T4 •ri So. •ri T4 •H *4
T «
u uo u u u O CU M M M M
C« u uo. •uo. u u uoa u u u
o u
o o Or*>> u
•
^ § Brl
ss e
a
a -rlUM
g § B
s
•H rM o 1 •
oo rH I M »jCi a
«o
(WU CU eo A m M .4JIC >>
-HC3 OS • 4
u <H <H «M
o « «o
Cioiaua OHHQO «Ha
ooa na o J5C tthH <HMH •MO •wo
oqcacac
o" e> uo
OOO
g
7S3
P4
oaou
UOrO
•HOMO
a«
•OTJ SÄ
U3H U>-i
loi
M3H
IIOOU0 1
0>M>MpM
QI-^KV-OV-SJCO >HKKJU
se
MO
OOCOXJ
ufn>ai
aOHOa
O H
33
ICI
rH 1
•H-v
1
vi m • ¿* «HC 1 rH •>. •a
•o ^ M oO O»I o rH» -ri «MO J< 0 4J
*-
o OHU
•• • - » • 1
•HMC o .
•u o
r-l oo o
O •
oa OH
33
-ri
tu ss.
gji
•H
c o u a. •OO
-r|C o o
o
>H
K
r-l
HM
^ .
3.
MM J »0H
30OJ
u MOO UO •
Si o
•h ou
ÄC u «a » ai
•03M |»||L 1
A*J o. • O MI u>. •ss
00
M•c
rS C OoC •0
3
rHO.
ou
•MM
•won
oo o
•UHO
«ooO C
.e o
0>U
•-HO á
•o
OU
UH
oo
>rlo>o
OO-I
r-tSO
rHO
oc
u «
-r*Orl
ouu
O -H
rHMO
OM
« 0
•HOM
0 00
"rf ^ ¿
M IOC «H «HO •rMI HM rtO OOO CU
« O « Ti
•ouo rH
•o
• ! • • OOC 0*0
u o >a
U «*4 •ri " U «u • •OH OMM rH •oa •rOO •HUM
••
•H c CU
•oo -
aaUO
OXJM ¿s
XM
«HO
•M»
•H>
•H
•H-l
•UHU
J3COI
Ma
M > I
MO
u 0
300U
•H3
Uoa
H •HOC • 1 1 »i
ÍA- o o»o ss oo »oo- H M j ooa
Pi BUMQ
u m i
.COHC
ocoo
uoou
oao
*Ot*4J 5¿
Mi
coe
SHU li
uuÇ
MUM
tj^-d
Bcfi
Ul
ao
»M
rHÄu
«USO»
ooo>
oSu5
u n .
OOM
»>CL
•
« 9.-S !>.
a 9 3 •1
ucu
o •HO r-Ml MOO OSOrl
o o X
U d«
ror Mes sag
rVOHH
s! 5
au r» J< M
°S2 •o
44 O
a o
>>
•
•HCSUri
o
o o«oa.o
>««O-IO Si (MC OMO o »s: a o
c roS 0
•rHO
IS T>H
OilUa«
¡XD00 s rng.
¿38 o
••HOU. C10
BOOM
B
•HUCMO
•Oí UPO
OOOBo
A )
o
UHÜ
QftO
II
o u
MU3
pao.
o od
Tta
o o
arte
ctii
JM4aWo
ou
03
•0
M
c J30«3H
H * SSt
QOOOU
«JOBO
l UTO« Bou OM> 5o2¡ •UCC MO Bau c
u MOCM MOCMO B OCOC
O OOrl KCOM 03 O0OHO
M «MOO. oaonoa Or«J oca
Or« « a OCÕO •oo
COO • M oouou
•M
Orí o «
o ou J
«o X
OH
«MO
x¡ ;S 3
s M
g > ^
55 ar HC
52
a¡
SE
CO
M
u
CO
CflPtf 8 ë r-l 3 ca ca
ë en
s
122
References
Abraham, C , Lyons, T.C., Schulze, K.-W., Stottmann, W. 1976. Zur instationären

Berechnung von Abwasserabfluß und Abwasserbeschaffenheit In städtischen Ent-
wässerungssystemen (Dynamic Computation of Sewage Runoff and Quality in Urban
Sewerage Systems). In: Korrespondenz Abwasser, Vol. 23, No. 3, pp. 59-65.
Annen, G. 1969. Das Zeitbeiwertverfahren zur Hochwasserberechnung (The Rational
Method for the Computation of Flood Discharge). In: Gas- und Wasserfach,
Vol. 110, No. 12, pp. 317-320.
Annen, G., Schoss, H.D. 1972. Das Einheitsganglinienverfahren bei stark bebauten
Gebieten (The Unit Hydrograph Method as Applied to Densely Built-up Areas).
In: Gas- und Wasserfach, Vol. 113, No. 2, pp. 78-81.
Billib, .11., Draschoff, R., Keser, J., Koehler, G., Sieker, F. 1972. Der Ein-
fluß zukünftiger Bebauung auf das Abflußverhalten eines natürlichen Einzugs-
gebietes - Beispiel für die Anwendung eines mathematisch-hydrologischen
Modells - (The Influence of Future Built-up Areas on the Runoff from a Natural
Catchment - Example for the Application of a Mathematical Hydrological Model).
In: Mitteilungen aus dem Institut für Wasserwirtschaft, Hydrologie und land-
wirtschaftlichen Wasserbau, Technische Universität,Hannover, Vol. 24, pp.
279-386.
Boes, M. 1975. Ein biozönotisches Modell zur Simulation des Sauerstoffhaushaltes
von Fließgewässern (A Biocenosis Model for the Simulation of the Oxygen
Balance in Rivers). In: Gas- und Wasserfach, Vol. 116, No. 9, pp. 399-407.
Böhnke, B. 1965. Die rechnerische Erfassung des Sauerstoffhaushaltes eines
Gewässers (Mathematical Determination of the Oxygen Balance of a River).
In: Veröffentlichungen des Institutes für Siedlungswasserwirtschaft der
Technischen Hochschule, Hannover, Vol.20, 55 p.
Cembrowicz, R.G., Hahn, H., Plate, E.J., Schultz, G.A. 1975. Mathematische
Flußgebietsmodelle (Mathematical River-Modelling). Bundesministerium des
Innern, Bonn, 351 p.
Deininger, R.A. 1966. über die Planung eines wirtschaftlich optimalen Systems
von Kläranlagen(Planning of an Economically Optimal System of Sewage
Treatment Plants). In: Gas- und Wasserfach, Vol. 107, No. 22, pp. 608-611.
Euler, G. 1975. Die Simulation der Niederschlagsaufteilung für Hochwasser-
abfluß-Modelle (Simulation of Rainfall Breakdown for Flood Modelling).
Technischer Bericht aus dem Institut für Hydraulik und Hydrologie, Tech-
nische Hochschule, Darmstadt, Vol.14, 157 p.
Draschoff, R. 1972. Häufigkeitsahalyse langfristiger Niederschlagsbeobach-
tungen als Grundlage der Abflußstatistik kleiner Vorfluter (Frequency
Analysis of Long-term Rainfall Observations as a Basis of Small River
Runoff Statistics). In: Mitteilungen aus dem Institut für Wasserwirtschaft,
Hydrologie und landwirtschaftlichen Wasserbau, Technische Universität,
Fleig, H. 1973. Gewässerkundliche Probleme im Zusammenhang mit der Aufstel-
lung von Wärmelastplänen (Hydrological Problems Involved in the Drafting
of Heat Load Plans). In: Deutsche Gewässerkundliche Mitteilungen, Sonderheft,
pp. 61-63.
123
Flinspach, D., Fleig, H. 1972. Die Aufstellung von Wärmelastplanen durch
rechnerische Simulation der Wärmeaustauschvorgänge - dargestellt am
Beispiel des Wärmelastplanes Rhein (Drafting of Heat Load Plans by
Mathematical Simulation of Heat Exchange - Demonstrated by the Heat
Load Plan for the River Rhine). In: Wasserwirtschaft, Vol. 62, No. 5,
pp. 127-133.
Flinspach, D., Wörner, Ü. 1976. Studie über das Abkühlungsverhalten erwärmter
FlieBgewässer (Study of Heat Exchange in Warmed Rivers). Ministerium für
Ernährung, Landwirtschaft und Umwelt Baden-Württemberg, Stuttgart, Vol. 5.
Geiger, W.F. 1975. Urban Runoff Pollution Derived from Long-Time Simulation.
National Symposium on Urban Hydrology and Sediment Control, University
of Kentucky, Lexington, July 28-31, 1975.
Geiger, W.F. 1976. Hydrologie Lumping Techniques and Detailed Flow Routing
Combined for Urban Runoff Simulation. ASCE, 34th Annual Speciality
Conference, Perdue 1976.
Geiger, W.F., LaBella, S.A., McDonald, G.C. 1976. Overflow Abatement
Alternatives Selected by Combining Long-term and Single-event Simulations.
National Symposium on Urban Hydrology and Sediment Control, University
of Kentucky, Lexington, July 26-29, Session A, pp. 72-80.
Günther, W. 1970. Optimierungsmodelle zur Planung und Betreibung von Anlage-
systemen zur Abwasserbehandlung in Flußgebieten (Optimizing Models for the
Design and Operation of Sewage Treatment Systems in River Basins).
In: Wasserwirtschaft-Wassertechnik, Vol. 20, No. 3, pp. 81-84.
Haendel, B. 1973. Der Berechnungsregen als Grundlage von Kanalnetzberech-
nungen (Design Rainfall as a Basis for the Calculation of Sewerage Networks).
In: Methoden der Kanalnetzberechnung. Dortmunder Arbeitstagung, Eigcnverlag
Institut für Umweltschutz und Umweltgüteplanung, Universität, Dortmund,
pp. 4-11.
Haendel, B. 1974. Die Möglichkeiten zur Ermittlung maximaler Regenabfluß-
mengen in der Kanalnetzberechnung und deren Zuverlässigkeit (Possibilities
of Determining Maximum Storm Water Runoff for the Design of Sewerage
Systems, and Its Reliability). In: Gas- und Wasserfach, Vol. 115, No. 1,
pp. 28-30.
Hahn, H.H. 1972. Operations Research und ¿eine Anwendung in der Siedlungs-
wasserwirtschaft (Operations Research and its Application in Urban Water
Management). Wasser und Abwasser in Forschung und Praxis, Vol. 5, Erich
Schmidt Verlag, Bielefeld, 340 p.
Homme1, H., Kleeberg, H.-B., Schreiber, P. 1976. Ablaufmodel1 für System-
simulation (Runoff Model for the Simulation of Systems). In: Wasserwirtschaft,
Vol. 66, No. 7/8, pp. 216-220.
Huthmann, G. 1976. Mathematisches Modell zur optimalen Regelung des Sauer-
stoffhaushältes eines Fließgewässers (Mathematical Model for the Optimal
Regulation of the Oxygen Balance in a River). Bundesanstalt für Gewässer-
kunde, Koblenz, 63 p.
Karpe, H.-J. 1969. Zur Dokumentation des EntScheidungsprozesses in der
Siedlungswasserwirtschaft (Documentation on Decision-Making Processes in
Urban Water Management). In: Gas- und Wasserfach, Vol'. 110, No. 6, pp.
145-147.
124
Keser, J. 1973. Folgerungen aus dem AbfluBverhalten befestigter Flachen für
die Bemessung von Regenwasserkanälen - Elementenverfahrcn (Conclusions
from Runoff Behaviour In Built-up Areas for the Design of Stormwater
Drainage Systems - Element Method). Mittellungen aus dem Institut für
Wasserwirtschaft, Hydrologie und landwirtschaftlichen Wasserbau, Technische
Universität, Hannover, Vol. 27, pp. 3-179.
Kleeberg, H.-B., Koch, H.-J., Schreiber, P. 1974. Ein ProRrammsystem zur Be-
schreibung des lang- und kurzfristigen AbfluSgeschehens in einem Flußgebiet
(A Computer Program System for the Description of the Long-term and Short-
term Runoff Regime in a River Basin). Mitteilungen aus dem Institut für
Wasserwirtschaft, Hydrologie und landwirtschaftlichen Wasserbau, Technische
Universität,Hannover, Vol. 31, pp. 65-76.
Klym, H., Königer, W., Mevius, F., Vogel, G. 1972. Die elektronische Berechnung
von Kanalnetzen (The Computerized Design of Sewerage Systems). Wasserwirt-
schaftliches Kolloquium, "Computer-Methoden in der Wasserwirtschaft",
ETH Zürich, Februar 1972, Eigenverlag Dorsch Consult, München, 71 p.
Kocks, F.H. 1974. Berechnung von städtischen Kanalisationsnetzen mit Hilfe von
Siniulationsmodellen (Calculation of Urban Sewerage Networks with the Aid
of Simulation Models). Eigenverlag F.H. Kocks KG, Düsseldorf, 75 p.
Künigcr, W. 1972. Anwendung eines komplexen Abflußmodells in der Stadtentwässe-
rung (Use of Complex Runoff Models in Urban Drainage). Symposium, Elektro-
nische Berechnung von Rohr- und Gerinneströmungen, Nürburg, Oktober 1972,
Lehrstuhl für Wasserbau und Wasserwirtschaft, Technische Universität,
München, 27 p.
Königer, W., Klym, H. 1972. Nichtlineare hydrologische Modelle in der Stadt-
entwässerung (Non-Linear Mathematical Models in Urban Drainage). In: Cas-
und Wasserfach, Vol. 113, No. 9, pp. 430-435.
KorLe, .î.W., Vielhaber, H. 1967. Untersuchung und Berechnung vernaschter Wasser-
versorgungsnetze (Calculation and Design of Water Supply Networks). Eigen-
verlag, Technische Hochschule, Aachen, 2 vol., 158 p. + 34 Figs. + 16 Tab.
Länderarbeitsgemeinschaft Wasser (LAWA). 1971. Grundlagen für die Beurteilung
der Wärnebelastungen von Gewässern (Fundamentals of Heat Pollution Assess-
ment). Eigenverlag der LAWA, Mainz, 88 p.
Landeshauptstadt Düsseldorf. 1975. Hochwasserentlastung Kittelbach - Regen-
auswertung (Stormwater Discharge Kittelbach - Rainfall Evaluation). Firma
Lautrich & Pecher, Consulting Engineers, Düsseldorf, 42 p.
Lecher, K. 1968. Die Simulation als Teil der Unternehmensforschung (Operations
Research) zur optimalen Lösung komplexer Wasserwirtschaftssysteme (Simulation
as a Part of Operations Research for the Optimal Solution of Complex Water
Management Systems). Mitteilungen aus dem Institut für Wasserwirtschaft, Hydrologie
und landwirtschaftlichen Wasserbau, Technische Universität, Hannover,
Vol. 13, pp. 6-150.
Liebscher,'H.-J. 1970. Hochwasserwahrscheinlichkeit (Flood Probability). In:
Deutsche Gewässerkundliche Mitteilungen, Vol. 14, No. 4, pp. 103-110.
125
Maas, S. 1975. Eine instationäre Kanalsimulation :*.m Vergleich mit dem Zeitbei-
wertverfahren (Dynamic Simulation of Runoff Processes in Sewerage Systems
as Compared to the Rational Method). In: Korrespondenz Abwasser, Vol. 22,
No. 7, pp. 203-209.
Maniak, U. 1973. Die Abschätzung von anthropogenen AbfluBverschärfungen mit
Hilfe von mathematischen Abflußmodellen (Assessment of Anthropogenic Influences
with the Aid of Mathematical Discharge Models). In: Deutsche Gewässerkund-
liche Mitteilungen, Sonderheft, p. 35.
Maniak, U., Renz, F.W. 1975. Stochastlsches Abflußmodell zur Optimierung ¿es
Talsperrenbetriebes im Ruhreinzugsgebiet (Stochastic Runoff Models for
Optimizing the Operation of Reservoirs in the Ruhr Area). In: Wasserwirt-
schaft, Vol. 65, No. I, pp. 22-23.
Marr, G. 1975. Analyse und Simulation des Niederschlagsabflusses in städtischen
Gebieten (Analysis and Simulation of Rainfall Runoff in Urban Areas). In:
Berichte aus Wasserwirtschaft und Gesundheitsingenieurwesen, Technische
Universität, München, Vol. 11, pp. 170-268.
Marr, G. 1976. A Comparison between Urban Runoff Models and Measured Data by
Means of Transfer Functions. Second International Symposium of Stochastic
Hydraulics, Lund, Sweden, Paper 19, 10 p.
Martin, G. 1975. Die Hydrologie der Regenentlastung an Überlaufbauwerken in
Mischkanalisationen (Hydrology of Rainwater Discharge as Applied to Overflow
Structures of Combined Sewerage Systems). In: Mitteilungen aus dem Institut
für Wasserwirtschaft, Hydrologie und landwirtschaftlichen Wasserbau, Tech-
nische Universität, Hannover, Vol. 34, pp. 11-176.
Mosonyi, E. 1970. Methoden zur Berechnung kritischer Hochwasserabflüsse
(Methods for the Computation of Critical Flood Discharges). In: Die Wasser-
wirtschaft, Vol. 60, No. 1/2, pp. 13-24.
Neumann, W. 1974. Die Anwendung von Modellen in der Stadtentwässerung (Applica-
tion of Models in Urban Drainage). Berichte aus dem Institut für Wasser-
wirtschaft und Gesundheitsingenieurwesen, Technische Universität, München,
Vol. 7, pp. 35-59.
Pecher, R. 1971. Das Zeltabflußverfahren zur Bemessung von Kanalnetzen (The
KaLional Method for the Design of Sewerage Networks). In: Gas- und Wasserfach,
Vol. 112, No. 6, pp. 312-319.
Pécher, R. 1973. Die praktische Bemessung von Kanalnetzen und Regenrückhalte-
becken mit dem zeitlich veränderlichen Abflußbeiwert (Practical Design of
Sewerage Networks and Retention Basins with the Aid of a Time-Variable
Runoff Coefficient). In: Korrespondenz Abwasser, Vol. 20, No. 2, pp. 29-39.
Plate, E.J. 1976. Auswirkung der Urbanisierung auf den Wasserhaushalt (Effects
of Urbanization on the Hydrological Cycle). In: Wasserwirtschaft, Vol. 66,
Na. 1/2, pp. 7-14.
Ribbeck, W. 1968. Beitrag zur Prüfung von Kanalisationsentwürfen mit Hilfe
elektronischer Digitalrechner (Examination of Sewerage Scheine Calculations
with the Aid of Digital Computers). In: Wasser und Boden, Vol. 20, No. 8,
pp. 231-235.
126
Rather, K..-H. 1974. Der. Einfluß von Veränderungen Im AbfluBsystem auf den
Hocliwasserablauf In kleinen Einzugsgebieten (Influence of Changes in the
Runoff System on the Flood Discharge of Small Catchment Areas). Schriften-
reihe des Sonderforschungsbereiches 150, Technische Universität, Braunschweig,
Vol. 2, 222 p.
Rump, H.-H., Symader, W., Herrmann, R. 1976. Mathematical Modelling of Water
Quality in Small Rivers (Nutrients, Pesticides and other Chemical Properties).
In: Catena, Vol. 3, Giessen, pp. 1-16.
Schmidtke, R. 1975. Entscheidungsmodelle In der Wasserwirtschaft (Mathematical
Alternative Models in Water Management). In: Wasser und Boden, Vol. 27, pp.
195-197.
Schröder, R.C.M. 1976. Auswirkungen der Urbanisierung auf das Abflußregime eines
kleinen Flußgebietes (Effects of Urbanization on the Runoff Regime of a Small
River Basin). In: Wasserwirtschaft, Vol. 66, No. 1/2, pp. 24-29.
Schulze, K.-W., Stottmann, W. 1975. Elektronische Kanalnetzberechnung - Stadt-
entwässerungs-Simulationsmodell (SESIM) (Computerized Calculation of Sewerage
Systems - Urban Drainage System Simulation Model (SESIM)). In: Grundlagen der
Abwasser- und der Abfallpraxis, 3-KA/C 5.3.4, No. 1, 7 p.
Siekdr, F. 1971. Hydrologische Bestandsaufnahme eines sanierungsbedürftigen
Kanalnetzes mittels Hochrechnung zeitlich begrenzter Messungen (Hydrological
Inventory of a Sewerage Network for Restoration Purposes by Way of Computer
Prediction of Temporary Values). In: Gas- und Wasserfach, Vol. 112, No. 10,
pp. 4«i9-i04.
SIvker, F. 1975. Verfahren zur Ermittlung eines Berechnungsregens mit örtlich
uai zeitlich ungleichmäßiger Verteilung (Method for the Determination of a
Design Storm with Disproportionate Distribution in Space and Time). In:
Wasserwirtschaft, Vol. 65, No. 11, pp. 293-296.
Sosnowski, P. 1971. Optimierung der Parameter hydrologischer Modelle
(Optimization of the Parameters of Hydrological Models). In: Wasserwirtschaft-
Wassertechnik, Vol. 21, No. 8, pp. 265-270.
Steinlein, H., Wolf, P. 1974. Die Nutzbarmachung mathematischer Modelle für den
Cevasserschutz (Use of Mathematical Models for Water Pollution Control).
In: Wasser und Abwasser - Bau Intern, No. 4, pp. 108-113.
Täubcrc, U. 1974. Wasserwirtschaftliche Systemanalyse des Wärmehaushalts von
Flüssen (Water Management Systems Analysis of the Temperature Balance of
Mvers). Dissertation, Technische Hochschule, Darmstadt, D 17, 107 p.
Verworn, W. 1976. Zur Klassifizierung von Kanalnetzberechnungsverfahren
(Classification of Calculation Methods for Sewerage Systems). In: Korrespondenz
Abwasser, Vol. 23, No. 6, pp. 178-181.
Warg, G. 1968. Die Ganglinien-Volumen-Methode für elektronische Kanalnetzbe-
rechnung (The Hydrograph-Volume-Method for Electronic Calculation of Sewerage
Systems). In: Wasser und Boden, Vol. 20, No. 8, pp. 235-237.
Warg, G., Kubat, G. 1969. Elektronische Berechnung von Kanalisationsnetzen,
gezeigt aa Beispiel Basel (Computerized Calculation of Sewerage Networks,
Demonstrated by the Example of Basel). In: Schweizerische Bauzeitung,
Vol. 87, No. 47, pp. 1-12.
127
Werner, G., Haendel, B., Heinz, I., Kreisel, W., Peter, G., Schuch, P.-C,
Schulz, V. 1975. Umweltbelastungsmodell einer Großstadtregion - dargestellt
ara Beispiel der Stadt Dortmund (BELADO) (Environmental Pollution Control
Model oí a Metropolitan Area, Demonstrated by the Example of Dortmund
(BELADO)). Beiträge zur Umweltgestaltung, Vol. B 10, Erich Schmidt Verlag,
Berlin, 45 p.
Wittenberg, H. 1974. Der ElnfluB zunehmender Bebauung auf den HochwasserabfluB
(Effects of Increasing Bullding Activity on Flood Discharge). Mittellungen
Institut Wasserbau III, Universität, Karlsruhe, Vol. 4, 113 p.
Wittenberg, H. 1976. Ein Prognoseverfahren für den Hochwasserabfluß bei zu-
nehmender Bebauung des Einzugsgebietes (A Method for the Prediction of
Urbanization Effects on Flood Runoff). In: Wasserwirtschaft, Vol. 66, No.
1/2, pp. 64-69.
Wolf, P. 1971. Die Berücksichtigung neuerer Erkenntnisse in Sauerstoffhaus-
haltsrechnungen für Fließgewässer (New Techniques for the Calculation of
the Oxygen Balance of Rivers). In: Gas- und Wasserfach, Vol. 112, No. 4/5,
pp.. ,200-203, 250-254.
Wolf, P. 1973. Gewässerkundliche Umlandsaufnahme und -analyse als Grundlage
für Sauerstoffhaushaltsbetrachtungen (Hydrological Inventory and Analysis
of Surroundings as a Basis for the' Calculation of the Oxygen Balance).
In: Wasserwirtschaft, Vol. 63, No. 9, pp. 273-277.
128
research in India
S. Ramaseshan
Department of Civil Engineering
Indian Institute of Technology
Kanpur, U . P . , 208016 India
and
P. B . S. Sarma
Water Technology Center
Indian Agricultural Research Institute
N e w Delhi, 110012 India
Report Background
A questionnaire was sent to a number of organizations in India involved in urban
hydrologie design, research and planning. This report is based on the limited number
of replies received and available publications. We are thankful to persons who have
responded to our questionnaire.
Because of the low priority for urban hydrology in India, and scanty information
dealing with mathematical models and urban hydrologie research, the scope of this
report is extremely limited.
Extent and Character of Urbanization
Urban areas (towns) in India are distinguished from rural areas by the
following:^} all places with a municipality, corporation, cantonment or notified
town area; a minimum population of 5000 with a density of not less than 1000 persons
per square mile and three-fourths or more of the working population engaged in work
outside of agriculture; and any other place which is considered by appropriate
authorities to possess pronounced urban characteristics and amenities. Also, the
1971 census identified "standard urban areas" which included rural and urban
populations within a fixed area with a good chance of becoming fully urbanized in
two or three decades, as well as "urban agglomerations".
Some of the salient features of metropolitan and urban growth in India
according to the 1971 census (provisional figures) are shown in Table 1.
The 1971 urban population of India was around 110 million persons, which
corresponded to 13% of the total population. The urban portion varied from 7% for
Himachal Pradesh to 31% for Maharashtra. These are much smaller than similar
131
figures for developed nations, such as 80% for the U.K. and 70% for the U.S.A.
Calcutta, Bombay and Delhi are 3 of the 25 largest urban agglomerations of the world.
Of the 2,921 towns in India, 340 had a population of more than 50,000, Table 1. By
1973-1974, water supply facilities served a population of only 83 million in 1,505
towns and sewerage facilities were available to only 39 million people in 190
towns.(2}
TABLE 1 - SOME CHARACTERISTICS OF THE URBAN POPULATION OF INDIA
NUMBER OF CITIES POPUIATION

POPULATION IN RANGE IN RANGE,
RANGE MILLIONS
1971 1961 1951 OF PERSONS
1,000,000 or more 142 113 81 57.0

50,000 to 1,000,000 198 138 102 13.2
20,000 to 50,000 617 484 353 18.9
10,000 to 20,000 931 748 630 13.1
5,000 to 10,000 756 760 1,158 5.7
Subtotal 2,644 2,243 2,324 107.9
less than 5,000 277 218 599 0.9
Total 2,921 2,461 2,923 108.8
Source; Chandra Sekhar, A., 1971, Provisional population totals. Census

of India-1971, Series A-Paper 1 of 1971, Supplement, Registrar
General of India, New Delhi.
According to the figures in Table 1, the number of towns with a population of
more than 20,000 has increased from 536 in 1951 through 735 in 1961 to 957 in 1971.
The population of towns with a population of more than 20,000 constituted about 82%
of the total 1971 urban population. The total population of India is expected to
increase from 547 million in 1971 to 1,000 million by around the year 2000. The
rate of growth of the urban population was 26% over the 1951-1961 decade, 38% over
132
the 1961-1971 decade, and is expected to be increasing currently at a higher rate.
Census figures for the past few decades indicate that population migration has been
generally from rural areas to large towns with populations exceeding 20,000, but
favoring the larger towns with populations exceeding 50,000. Hence, it seems
desirable to consider urban hydrology in India with particular reference to towns
having a population of 50,000 and more, even though smaller towns cannot be ignored.
Role of Urban Drainage
Because urban areas in India also overlap rural and green-belt areas the
population density in parts of some urban areas may be quite low. For example, in
certain areas of New Delhi the population-density is as low as 37 persons per
hectare.^-' Furthermore, except perhaps in the case of Bombay City, cities in India
have a small proportion of built-up area to total area and hence urban drainage
problems in India as in most other parts of the developing world are quite different
from the "concrete jungles" of the developed nations.
In a developing country, the priorities for economic development and investment
are for food, shelter, clothing, health and education. Urban drainage is generally
not taken into consideration except when it affects significantly any of the above
factors, particularly as a part of the more general problem of flooding of urban
areas. As almost all important cities of India are on the banks of rivers and are
subject to flooding, drainage of urban areas and riverine flood control are generally
interlinked. Recent proposals for urban drainage improvements such as in the Delhi
Metropolitan area,(-*' Lucknow, Patna,^^ etc., can be attributed to such a linkage.
Because of financial limitations and because urban drainage problems constitute
"negative goods," very little attention has been paid in India to urban drainage.
The following report hence summarizes the information that could be collected from
133
a limited number of organizations in India and poses a number of problems that are
considered relevant for a developing country like India in the areas of urban
hydrologie research and mathematical modeling.
134
Section 2 Urban drainage design in India
Current Design Practice
Urban hydrologie design in India is characterized by its extreme variability
in principles and practice. Generally, when sewerage is designed urban drainage is
also provided for as part of a combined system, for example in the towns of Uttar
Pradesh. Otherwise, separate sewer systems are designed and provided with little
provision for drainage, for example the Calcutta Metropolitan System^) with a
capacity corresponding to a "two month" recurrence interval rainfall.
The design of rural drains, or drains in rural parts of urban areas, are often
approached empirically. Thus, the Delhi Administration has improved the rural
drainage system of Alipur Block for a discharge of 0.078 nP/kra^ and in Khanjawala
Block for 0.056 m 3 /km 2 .( 3 ' Discharge for medium sized drains (larger than 2,000 ha)
from rural areas of the Najafgarh drain^ 6 ' and Shahdara drain' 7 ' of Delhi and of
North Bihar are estimated as 0.112 nrVkm 2 .
The "rational formula" is generally used in India, with a deterministic
interpretation,(°) to estimate the design peak flow in an urban watershed. Drains
are sized for the design flow under a "flowing-full" condition. The three factors
affecting a design flow in the use of the rational formula are respectively the
coefficient of runoff, the rainfall duration defined by the inlet and routing times
or the time of concentration, and the frequency of the design rainfall. There is
no uniformity in the estimation of these factors in India.
Coefficient of Runoff
The coefficient of runoff or the runoff factor defines the ratio of runoff to
rainfall and is usually estimated on an empirical basis. Values used vary from
135
around 15% for predominantly agricultural areas to 60% or 70% in the case of densely
paved and hilly areas. Recommendations differ widely, Table 2, and they are much
larger than the values deduced from observations in IndiaO) and elsewhere. (9)
TABLE 2. RUNOFF FACTORS FOR URBAN AREAS
RUNOFF FACTORS, PER CENT

EXAMPLE
POPULATION DENSITY, PERSONS PER HECTARE
Í370 371 to 618 >618
Delhi Master Plan^3) 35% 45% 60%

6
Najafgarh DrainC ) 45% 60% 60%
Shahdara Drain'7' 60% reduced to 407. because of sandy soil.
Uttar Pradesh 50% . 50% 50%
Bihar<4) 40% to 60% depending on built-up area.
Patna(^) 50% to 70% depending on built-up area.
5
Calcutta^ ) Composite values as function of duration
of storm and percentage of impervious
area.
Rainfall Duration
Because rainfall intensity decreases with increasing duration, the duration of
a design rainfall becomes an important factor. The associated duration varies from
reach to reach within a drainage area. When surface drainages are used, the
rainfall duration is assumed to be equal to the time of concentration for the basin,
which is generally estimated by empirical equations. When lined channels are used,
the time of concentration may be very much in error. The response time of sewered
urban catchments is small,(*0) aruj ¿ n sucn cases the empirical equations should be
used with care. Alternatively, an inlet time of 15 to 30 minutes and a travel
velocity of 0.6 to 0.9 meters per second along the drainage network are generally
136
recommended. Sometimes the duration is specified arbitrarily, for example 6 hours
in Uttar Pradesh and 1 hour for Delhi. .
Rural drains are designed for temporary flooding and thus storage effects are
accounted for implicitly. In India rural drains are generally designed for a three
day rainfall. Because of the large storage influences, the runoff is assumed to
occur more or less uniformly over a period of drainage or disposal which in turn
depends on the nature of the crop grown and its tolerance for submersion or flooding.
Thus, a drainage duration of three days is used in rural areas of Delhi where maize
and wheat are grown, whereas the duration is generally seven, days in paddy-growing
areas and ten days in West Bengal. The runoff from rural areas is hence much
smaller than from urban areas. When an urban drainage system includes both urban
and rural areas, the respective contributions are added to resolve the design peak
flow.
Frequency of Rainfall
The frequency of rainfall for rural and urban areas is generally adopted as
five years and two years respectively, Table 3. However, where improvement of an
existing drainage system would be costly, the recurrence interval adopted can be as
low as two months, ae in the case of Calcutta where an improvement to a three month
frequency capacity would have been 707» more costly than for a two month frequency
capacity.'-^ Flooding occurs only in the three monsoon months, with the result that
the actual return period for the design flood is only a fortnight. Similarly, in
Delhi many closed drains had been designed for a two year frequency and because of
the high cost of improvement they were not improved to the five year frequency adopted
for trunk drains. These experiences indicate that a combined system in India is very
costly. In view of this, design standards for Delhi and Bihar recommend only open
drains, lined if necessary, in urban drainage systems.
;
137
TABLE 3. FREQUENCY OF DESIGN RAINFALL
RAINFALL FREQUENCY
EXAMPLE
Urban Drainage Structures
Najafgarhi6) and ShahdaraC7)

Drains 5 Yr
Patna(^) 5 Yr
DelhiA-^ larger drains
(area, 20 hectares) 5 Yr 10 Yrs
Small drains 2 Yr
Uttar Pradesh 2 Yr 5 Yr
15
Gauhati^ ) 6 months
5
CalcuttaC ) 2 months 1.25 times drain
capacity
Structures and Pumping Systems
Structures across the drainage system are usually designed for a higher
capacity than that of the drains, Table 3. This is to avoid reconstruction costs
when the drainage capacity is increased later. The outfall generally discharges to
a natural drain or river. Because during the monsoon season the receiving river may
be in flood, sluice gates and pumps are generally provided to prevent backflow and
flooding. The capacity of the units is generally comparable to that of the drains.
Normally, pumping is required only when the river level is high. The conditional
recurrence interval for a critical flood with respect to the pumping capacity is
much larger than the recurrence interval with respect to the capacity of the drain.
Thus, in Lucknow a drain designed for a two year flood has a capacity of 18.56 nr/sec.
Since the Gomti River into which the drain outfalls has high stages only in the
latter half of the monsoon period, the pumping capacity provided is only 17.16 m3/ sec.,
and even this has a conditional recurrence interval of 30 years, which is comparable
to that used in the design of levees. It is necessary to investigate the consistency
138
of design between the drainage system, pumping systems and the river flood control
system.
Limitations of the Rational Method
The limitations of the rational method are well known. If the design is based
on peak flow alone, "the rational method is as good an arbitrary procedure as any,
considering the primitive state-of-the-art of urban hydrology".(Ü) But complex
stormwater drainages which involve diversions from one catchment to another, flood
storage, off-channel storage, permanent storage, totally impervious catchments,
pumping installations, land use planning, and control of water quality including
silting require knowledge of the flood hydrograph rather than peak flow alone. Then
the rational method is not applicable, and unit hydrographs and urban drainage
simulation models are needed for rational and economic design of urban drainage
systems.' -^J
Water Quality
Only recently has there been a realization in India of the importance of stream
and groundwater quality. As population and industrialization increase, water quality
simulation models may be adopted for estimation and control of stream water quality
problems. Urban drainage and sewage systems are major sources of such pollution.
At present, very little attention is being paid to the simulation of water quality
and perhaps none at all in connection with urban drainage systems.
Yet, water quality problems are important. For example, tannery wastes create
problems for the rayon industry in Kanpur. In addition to such problems of
industrial wastes and their treatment, there are two specific problems related to
water quality in urban drainage design. Wastewater sewerage connections are generally
made in Indian towns to less than 25% of the number of households and hence the
sewer system is overdesigned. This leads to the accumulation of silt and sediments
139
in the sewer. When storm runoff occurs such sewers are partially choked, leading to
underutilization of their capacity and flooding. Furthermore, the low-lying areas
prone to frequent flooding are often encroached upon by the poorest section of the
population, and are covered with sprawling slum areas with a high density of
population and meager civic amenities. Failure to provide an adequate urban drainage
system seriously affects the life of these people and exposes them to potential
health hazards. Thus, urban drainage systems are also linked with the overall problem
of slum abolition, resettlement and urban redevelopment.
140
Section 3 Field research
Introduction
To reiterate, India had an urban population of around 110 million out of a
total population of 550 million in 1971. This population was distributed over 2,921
towns of which nearly 1,000 had a population above 20,000 and 142 above 100,000.
Yet, because of economic factors, urban drainage has a low priority and very few of
the towns have adequate drainage systems. Hence the extent and depth of studies
in the area of urban hydrology are very limited. This Section briefly reviews the
status of urban catchment research in India on the basis of very limited available
information.
Precipitation
Urban drainage design requires depth-area-duration-frequency relationships
characterizing storms in urban areas, particularly for small durations. Based on 5
to 23 years of data from 50 self-recording raingage stations generally located in
towns, the India Meteorological Department has derived depth-duration-frequency
relationships for these stations for 15 minutes, 30 minutes, 45 minutes, one hour and
larger durations. On the basis of these results, generalized maps covering India have
also been prepared, for example, the 15-minute 5-year precipitation map for India shown
in Figure 1.(13) These maps were tested with data from Damodar Valley and a number
of stations maintained by the Research Designs and Standards Organization, Ministry
of Railways, and found to be satisfactory. They may be used for estimation of short
duration, extreme precipitation values for towns where self-recording raingage data
are not available. The India Meteorological Department proposes to revise these maps
as more data in terms of space and time coverage become available.
141
FIGURE 1- 5-YEAR, 15-MINUTE RAINFALL IN MILLIMETERS (13)
142
The India Meteorological Department is also studying the following characteristics
of short duration, intensive precipitation:
depth-area relationships;
depth-duration relationships for different frequencies;
critical time distribution of precipitation within a storm; and
relationship between fixed clock times and equivalent duration maximum

precipitations, for example 1-hour and 60-minute precipitations or 1-day
and 1440-minute values.
Studies have also been made in academic institutions and metropolitan planning
organizations concerning intensive, short duration rainfalls. For example, studies
have been made of the location of the peak rainfall intensity in storm patterns'*•*'
and synthetic storm patterns,VJ-'->' and a frequency analysis has been made of storm
rainfall.C16) it is expected that the pursuit of such studies will lead to a
better understanding of intense, short duration precipitation in India and hence to
reliable design aids in terms of charts and formulae for rational urban hydrologie
design.
Dense urban precipitation networks are a rarity in India. The Flood Control
Wing of the Delhi Administration has set up a network of 14 self-recording raingage
stations, Figure 2. These along with Palam and Safdarjung raingage stations maintained
by the India Meteorological Department and non-self-recording raingage stations in
and around Delhi constitute a network specifically installed to serve urban drainage
design. Based on the results of this network other urban precipitation networks may
be set up later.
Runoff
India has one of the oldest stream-gaging networks in the world. But there has
been a very great paucity of reliable data for the small drains of urban areas.
143
While river stage measurements and other data concerning flooding are available,
these pertain to the high flood levels in the rivers to which the drains discharge
and generally not to the discharge in the drain itself. Realizing the importance of
discharge data, attempts are being made to measure discharges in urban drains, for
example since 1975 in Delhi. The Experts Committee on drainage, floods and river
training for Delhi^^ has recommended that regular gage observations with permanent
double gages for stage and surface slope, and discharge observations with current
meters, be made daily in all drains, and that observations be made of peak values
in all trunk drains and the major collector drains. Further, the Committee
designated the major collector drain and five urban and five rural drains to be
gaged hourly (Figure 2). This stream-gaging network is being planned in terms of
rational formula criteria. The use of continuous stage recorders would be necessary
if unit hydrographs and urban drainage simulation are to be adopted for use in
design.
Some of the drainage systems have variable discharge pumping stations with
automatic discharge recorders. Records from the pumping stations may be used in the
modeling and analysis of existing urban drainage systems^'' and the effect of
storage on discharge. They may not be useful in new designs because the storage
equations needed for new systems cannot be derived from existing records.
Water Quality
No consistent water quality data are available for urban drainage. Because
water pollution is a serious problem only during low-flow seasons, it is the general
practice to conduct stream-quality surveys only in the non-monsoon season. Proposals
for stream quality monitoring also involve measurement only in the dry season or
"one-shot" measurement rather than continuous recording. For example, 52 water-quality
parameters for 16 drains in the Delhi metropolitan area were measured only once
in 1976.
144
145
Hence, it may be seen that urban hydrological research in India is just
beginning to take shape either as an offshoot of other programs or on the basis of
individual problems. While there seems to be some planning for collection of
precipitation data, discharge data collection is lagging and water quality data are
almost nonexistent. There seems to be a need for determining the type of data
needed for urban drainage design in India and other developing countries, for
designing and operating supportive short-terra and long-term data networks.
146
Section 4 Hydrologie modelling
Introduction
In India, urban hydrologie modeling is in a beginning stage both with respect
to adaptation and application of existing models developed in India or outside.
Hence, briefly reviewed in this Section are some recent efforts in the development
and application of hydrologie models in India.
Rational Formula
The rational formula is widely used in India for urban drainage design, as
noted in Section 2„ Some attempts have been made to measure the peak flow from a
drain and compare it with the peak storm intensity to derive the coefficient of
runoff. For csxample, in Delhi a number of storms were gaged for peak flow in 1975
and the resulting range of coefficients of runoff was used to determine the design
value.''' By choosing drains from essentially urban and agricultural areas, design
coefficients of runoff were correlated to urbanization, such as in terms of population
density. Thus, only a deterministic and not a probabilistic approach to the rational
formula was adopted.
Mathematical Model
A nonlinear hydrologie model of the form
has been developed' ^' for the storage in the combined sewer and drainage system of
Calcutta town (Figure 3), by relating the effective rainfall during a storm derived
from the recorded precipitation data to the record of pumpage from the stormwater
pumps, using a procedure similar to that of Prasad.t-"*' The parameter N was nearly
constant but the parameters K^ and K2 were not constant and so they were correlated
147
LEGEND:
— • Existing combined sewers
• Existing pumping station
Tributary area boundary
FIGURE 3- TOWN SYSTEM CATCHMENT OF CALCUTTA
148
by regression analysis in terms of storm characteristics such as total rainfall
excess, the duration of rainfall excess, and the time distribution of rainfall
excess in terms of the time to centroid and a shape factor. The results indicated
that the hydraulic capacity of the system is very inadequate, leading to frequent
flooding of streets. This agrees with the fact that the design capacity provided
corresponds to a 2-month recurrence interval.
Simulation of Drainage Systems
The hydrologie model ILLUDAS has been implemented and used for analysis and
design of urban drainage systems in India.(19) other models may also be implemented
in order that a model appropriate to the Indian environment may be selected for
urban drainage design.
149
Section 5 Conclusions
India has 3 of the 25 largest metropolitan areas of the world and had around
1,000 towns with a population of above 20,000 in 1971. The number of towns and the
urban population are increasing rapidly leading to the growth of still larger
cities.
Urban hydrologie problems of India, as in other developing countries, differ
from those of developed nations in several important respects. They include
lateral rather than vertical development;
limited amounts of paved areas;
intimate interaction between urban drainage and flood control;
preference for open drains over closed ones;
limited availability of continuous records of precipitation, streamflow

and water quality;
low fiscal priority for drainage investment;
limited numbers of sewer connections and hence silting of combined sewers;
high cost for construction and modification of combined sewer systems; and
limited capacity for financial investment.
Design of urban drainage systems in India is based on the "rational formula"
using arbitrary assumptions concerning the duration and frequency of rainfall and
the coefficient of runoff. The use of the rational formula may be justified by
the lack of adequate continuous records of precipitation and streamflow. Yet,
there is a vital need for rationalization and standardization of design procedures
based on engineering and economic considerations.
Very often the design of urban drainage systems involves consideration of
150
flood storage, permanent storage, off-channel storage, inter-drainage diversions,
pumping installations and silting of drains. This requires a knowledge of flood
hydrographs rather than only flood peaks. Awareness of this important fact has
recently (1975) resulted in the initiation of programs to collect self-recorded
or continuous data concerning discharge in the drains and in the use of
mathematical models for analyzing and designing urban drainage systems.
Simulation models are useful in the analysis of complex drainage systems
where storage, pumping, silting and quality control are involved, and hence in
the economic design of complex drainage systems. The type of drainage system
simulation model that is suited to urban areas in developing nations needs to be
identified, and computer programs suitable for applications need to be developed.
Other mathematical models may also need investigation.
REFERENCES
1. Natarajan, D., 1972. Indian Census through a hundred years. Census Centenary
Monograph No. 2., Office of Registrar General, Ministry of
Home Affairs, New Delhi.
2. City and Town Planning Organisation, Planning Commission, Govt, of India, 1976,
Personal Communication.
3. Master Plan Organisation, 1976. Outline of the Master Plan of Drainage in the
Union Territory of Delhi, Report of the Experts Coraraitte ,
Flood Control Department, Delhi Administration, Delhi.
4. Lahiri, P. K. and Ghosh, P. F. W;, 1976. Recommended design standards for

storm water drainage schemes, soil sewerage schemes and
sewage treatment works. Public Health Engineering Department,
Govt, of Bihar, Patna.
5. Master Plan for water supply, sewerage and drainage, Calcutta Metropolitan
District (1966-2001), 1965. Report of Engineering consortium
of M/s. Metcalf & Eddy, Ltd., Boston, Mass., and M/s.
Engineering Science Inc., Arcadia, California, U.S.A.
6. Flood Control Wing, 1974. Revised estimate of remodeling and lining of Najafgarh
drain, Delhi Administration, Delhi.
151
7. Flood Control Wing, 1975. Project for Storm Water Drainage of Shahdara area
(Alternate Schema), Vol. I, Delhi Administration, Delhi.
8. Aitken, A. P., 1973. Hydrologie investigation and design in urban areas - a

review. Tech. Paper No. 5, Australian Water Resources Council,
Australian Govt. Publishing Service, Canberra.
9. Schaake, J. C , Jr., Geyer, J. C , and Knapp, J. W.,. 1967. Experimental

evaluation of the rational method. Journal of the Hydraulics
Division, ASCE, 93 (Hy 6): 353-370.
10. Sarma, P. B. S., 1970. Effect of urbanization on runoff from small watersheds.
PhD Thesis, Purdue University, Lafayette, Indiana.
11. McPherson, M. B., 1974. Better design of stormwater drainage systems. APWA
Reporter, 41 (2): 7-9.
12. Sarma, P. B. S., 1973. Trends in urban hydrologie design - a review. Journal
of the Institution of Engineers (India), 50 (CI4) : 1-7-190.
13. Harihara Ayyar, P. S., and Tripathi, N., 1974. Rainfall frequency maps of India.
Meteorological Monograph Hydrology No. 6/1974, India
Meteorological Department, New Delhi.
14. Bandhyopadhyaya, M., and Bose, B., 1971. Determination of the location of peak
rainfall intensity in the synthetic storm pattern. Scientific
Report No. 71-2, Fluid Mechanics and Fluid Machinery Section,
Mech. Engrg. Dept., Jadhavpur University, Calcutta, India.
15. Bandhyopadhyay, M., 1972. Synthetic storm pattern and runoff for Gauhati,
India. Journal of the Hydraulics Division, ASCE, 98
(HY5): 845-857.
16. Raman, V., and Bandhyopadhyay, M., 1969. Frequency analysis of rainfall
intensities for Calcutta. Journal of the Sanitary Engineering
Division, ASCE, 95 (SA6): 1013-1030.
17. Bandhyopadhyay, M., and Bose, B., 1975. A nonlinear hydrologie model for
Calcutta Town System. Proceedings, National Symposium on
Urban Hydrology and Sediment Control, Report UKY BU109,
Univ. of Kentucky, Lexington, Ky., U.S.A.: 219-226.
18. Prasad, R., 1967. A nonlinear hydrologie system response model. Journal of
the Hydraulics Division, ASCE, 93 (HY4): 201-221.
19. Kirpal Singh and Ramaseshan, S., 1975. Simulation of urban hydrologie systems
using ILLUDAS. Unpublished report, Dept. of Civil Engrg.,
Indian Institute of Tech., Kanpur.
152
research in the Netherlands
F. C. Zuidema
Rijksdienst voor de IJsselmeerpoIders
Lelystad - Smedinghuis
Netherlands
In spite of a continuous growth in the population of the Netherlands

(13 million in 1970 with 15.6 million expected in 2000) and an increasing
population density (384 inhabitants per square km in 1970, 403 in 1975, 434 in
1985, and 463 in 2000), development in urban water resources research has been
rather tardy. However, there is an enormous diversity among urban hydrological
problems, which are solved adequately. For example, while only a few urban
catchment studies are going on, mathematical models are used for different goals
and at different levels: models for rainfall-runoff, water quality, comprehensive
urban water and water resources management.
Many institutions in the Netherlands are involved fully or partly in

hydrological research, but only a few are active in urban aspects.
Most of the contributions to this Technical Memorandum were acquired in

the form of reactions to a letter of request from the Netherlands National Committee
for the I.H.P. to about 35 experts and organizations. The writer hopes that other
experts will make suggestions and offer additional information for a subsequent
revision of the report. A more extensive version would lead to a better and more
balanced review of these aspects of urban hydrological research. Finally, several
contributors to the present report will present papers containing more detailed
information during the I.H.P. International Symposium on the Effects of Urbanisation
and Industrialisation on the Hydrological Regime and on Water Quality (Amsterdam,
October, 1977) and during the immediately following International Workshop in
Amsterdam on the Impact of Urbanisation and Industrialisation on Regional and
National Water Planning and Management.
Hopefully, all efforts mentioned in this introduction may result in a

better insight on urban hydrological processes and resultant improved practical
applications, and will promote the interests of a wide variety of users.
Acknowledgment
The writer served as the editor of information provided by a number of

contributors. He highly values these vital contributions. He also wishes to
express his gratitude to Mr. M. B. McPherson for the final editing of this report.
155
Section 2 Urban catchment research in the Netherlands
Current research on urban catchments is described in this Section.

Attention is paid to instrumentation, data collection, data processing and data
analysis. Because good equipment and an adequate data collection system are
essential for obtaining good results, no pains should be spared in the conduct of
these parts of the research. Only after some years of extensive effort can one
expect to arrive at useful conclusions. Time is needed to compensate for years
of winter periods without sufficient rainstorms, surveillance of very sensitive
electronic equipment for the collection of hydrological data, malfunctions in
data logging systems, etc.
In the analysis phase of a study, good information on conditions during

the period of data collection is needed. Unexpected values can regularly be
encountered which, in some cases, might be explained by special circumstances.
For this reason, attention is given in this Section to these different aspects of
urban catchment research.
Urban catchment research in the Netherlands is limited to the basins in

the new town of Lelystad in the vicinity of Amsterdam and in the village Neede in
the eastern part of the country, together with study areas along roads in
Lelystad.
URBAN CATCHMENT HYDROLOGY IN LELYSTAD

(J. A. van den Berg, IJsselmeerpolders Development Authority, Lelystad)
I am the daughter of Earth and Water,

And the nursling of the Sky;
I pass through the pores of the ocean and shores;
I change, but I cannot die.
For after the rain when with never a stain
The pavilion of Heaven is bare,
And che winds and sunbeams with their convex gleams
Build up the blue dome of air,
I silently laugh at my own cenotaph,
And out of the caverns of rain,
Like a child from the womb, like a ghost from the tomb,
I arise and unbuild it again.
Shelley, The Cloud
156
Introduction and Description
A considerable part of the territory of the Netherlands consists of

polders, Figure 1, which are areas with a controlled water level. The newest
ones were gained by the partial drainage of the Zuyderzee, Figure 2. This project
originates in the Zuyderzee Act of June, 1918, and consists of the construction
of an enclosing dam and the reclamation of five polders, with the dam and four
of the polders completed at the present time.
157
/ ^
( «Sr 1
J
^
£
) -
;
s ¡¡¡
O
^^
V - Si»
«L * S|i
o m.*
w
^v. r^sîî"\ I
o ' • V \ *Ï§ A: £
£—v JBjI
—v
'
s\
v*
y-,
^
• ^ S ^
• -
V. "
\ry s
J
^ - ^ ^ Z ,^
V
* _
fe
r u«
^ —™ ^ * * ^ ^
i
.»
l l l
J
AKE
*^'
UJ:
ft:
%: •^L
X . • *-
*•
.ï-
158
Use of the newly reclaimed land is changing with shifts in social
demands. The use of the first polder, which was developed before World War II,
was as follows: 87% farm land, 1% residential and 3% woods and nature reserves.
The approximate corresponding figures for the Southern Flevoland polder, which is
still in development, are 50%, 18% and 25%, respectively.
It is a known fact that the hydrologie cycle of an urban area differs

from that of arable land and forests, the main difference being that a paved area
yields a high percentage of surface runoff and has a small storage capacity. This
difference can be illustrated by the discharge from a paved area (storm water
drains of a separate system) versus that from an unpaved area (subsurface
drainage) which may be regarded as being representative of grassed land, Figure 3.
The quick response of- the paved area to precipitation is important not only in
the design of urban drainage systems but also for the water management of the whole
polder because of the growing percentage of housing and industrial areas. The
period of decisive rainfall differs too, because in the urban area it is fixed by
the heavy storms during summer, while for arable land the critical period occurs
during winter, when the soil is saturated.
Because the new town of Lelystad has been planned to grow from the
present population of 15,000 to 100,000 inhabitants by the year 2000 and the
building of the new town of Almere (125,000 to 250,000 inhabitants) will start very
soon, it is important to know the inflow and outflow hydrographs resulting from the
precipitation on an urban area. Therefore, some catchment areas in Lelystad have
been equipped with instruments, where precipitation, storm water discharge,
subsurface drainage discharge and groundwater level are recorded continuously,
while the soil moisture content of the unsaturated zone is measured periodically.
The purposes of the hydrologie research in Lelystad are: to study the

relation between rainfall and urban runoff; to refine the design criteria for storm
water drains, the subsurface drainage system and the system of open drains; and
to study the consequences of urbanization on the water management of a polder.
Almere and Lelystad have systems of open drains providing primarily for
the drainage and storage of the discharge of precipitation and seepage water. The
water level in the open drains is above that of the surrounding polder-water and
is controlled by adjustable weirs.
Equipment and Data Collection
To study the urban hydrologie system, different kinds of catchment areas

have been under investigation since 1969:
a housing area of 2-ha (covered area 44%);
a parking lot of 0.7-ha (paved area 99%);
a shopping and office centre of 2.2-ha (covered area 95%);
an unimproved building site of 4-ha; and
two flat roofs with areas of 250-m^ and 350-m^ respectively.
At the above sites, precipitation is measured by means of three ground-
level raingauges with an orifice of 3846.5-cm2. The funnel is connected with a
subsurface reservoir, which has a volume corresponding to a 100-ram depth of
precipitation.
159
rainfall il-sec. ha I
FIGURE 3- COMPARATIVE
HYDROGRAPHS
rainsewer discharge ll'sec/hal
Hyetograph and outflow hydrographs

of the Etonnwater drains (paved
area, 0.88-ha) and of the
subsurface drainage system
(unpaved area, 1.12-ha) for a
catchment area in Lelystad.
The rainstorm of August 3 had an
intensity of 45ramwithin 45
minutes.
subsurface drainage discharge U / s e c / h a I

0.50
0.40
160
Discharge of storm sewer systems is measured with a rectangular Thomson
V-notch weir. The V-notch is installed near the outlet to the open drains.
The subsurface drainage system consists of a number of parallel drainpipes

connected with each other by a drain collector. An electromagnetic flowmeter is
installed near the outlet to the open drains.
The soil moisture content of the unsaturated zone in the unpaved areas is
measured periodically by a neutron scattering device.
With the exception of soil moisture content, all the elements are
continuously measured and recorded in a central recording station, Figure 4.
The measuring system is subdivided into:
a. Input subsystem. A stage indicator transduces level of rain gauge, stage

at the Thomson V-notch, and groundwater level. A potentiometer is
activated by a float, so that a change of water level is converted into
a voltage swing. An electromagnetic flowmeter is used for measurement
of the discharge of the subsurface drainage system. This type of
measurement is necessary because it deals with small quantities having a
low potential energy. Here, the discharge is proportional to an electric
current which is converted to a voltage swing by a fixed resistance.
In addition, a digital input (position of a switch) is used for checking
the downstream situation of the Thomson V-notch (free fall or submerging).
b. Digital voltmeter. The voltmeter converts the analog inputs representing
physical quantities into digital input suitable for computer use. A
scanner interrogates the different input channels and connects them
sequentially to the voltmeter.
c. Time base generator. It enables the performance of any command with a
stated time relationship.
d. Controller. The controller commands the input and output subsystems and
has possibilities for internal data reduction and incorporation of
programmed decisions.
e. Output subsystem. The controller outputs the data to a tape punch or
t tape cartridge and printer.
Data registration takes place according to the principle of significant

changes in the value of a hydrological element. The controller compares every
measurement with the last value that was output. If the difference between both
exceeds a certain level the measurement is recorded. Each record is composed of a
date-time group, the channel number and the value of that hydrological variable.
From the start of the measurements in 1969, a Philips data logger was the
controlling device, Figure 4(A). This data logger was developed and constructed
for this special purpose. The system was comprised completely of hardware and not
programmable. Output was available on paper tape and print. Since November, 1976,
a Hewlett-Packard 9825-A programmable calculator has been used, Figure 4(B). Output
is also available on tape cartridge. The control function is established by software.
This calculator also has a capability for further data handling and screening.
161
_ puncher
/ input printer
a.d.c.
unit control „ • • 1
printer
scanner i i
analog
input input '
unit
, 6 red print
< memory comparator
control
, i , i
input
unit '
\
t
scanner
controller time base
drive
m
i
digital sub
input scanner
analog input
is ) -e scanner
~I digital
voltmeter
event
(a.d.c.)
rx
interface
bus
printer
\
digital multi puncher 1
sense *
input programmer
card
,
' H P 9825 A calculator with

time tape cartridge drive
and display
base
input controller output
FIGURE 4-LELYSTAD DATA LOGGING SYSTEM

(A m Original System; B « New System, November, 1976)
162
Data Handling
The recorded data are directly suited for further computer operation. The
data are sorted according to channel number and relegated to monthly files.
Superfluous and erroneous data are eliminated as much as possible by a computer
program. In order to visualize the data, a plot for each variable is made. In a
second operation the time series are transformed according to the nature of the
variable (e.g., precipitation, groundwater level) and the type of transducer, with
the help of control measurements made by hand and consideration of catchment
parameters.
After this operation, the data are available in the following form: for
precipitation, a cumulative time series starting with zero in the beginning of each
month, with registration in units of 0.1-ram rainfall; for storm water discharge, a
time series in units of 0.5-mm height relative to the bottom of the Thomson V-notch,
with heights smaller than 15-mm deleted; for subsurface drainage discharge, a time
series with mean discharges per 3 hours in 0.01-mm depth; and for groundwater level,
a time series with mean level per 3 hours in cm relative to a fixed level.
A Synopsis of Some Results
A systematic elaboration of all data has not yet been carried out.
However, some interesting rainstorms have been analysed from which the following
results can be shown.
For the three ground level raingages in the catchment areas (at distances
of 140-, 260- and 340-m from each other) a high correlation exists between both the
total amounts of rainstorms (Van den Berg, 1973) and the amounts for each time
interval of 5 minutes and longer (Figures 5 and 6 ) .
The measured values of discharges and groundwater levels are essentially

continuous in time, i.e., in going from a high level to a lower one, all intermediate
values are passed through. Perhaps the precipitation need not be continuous. The
phenomenon whereby the probability that a variable (continuous or discrete) equals
or exceeds a certain value depends not only on that value but also on preceding
values is called persistence or autocorrelation. The closer in time the successive
values of a variable are sampled, the greater will be the autocorrelation between
them (Van den Berg, 1976 A ) . A statistic indicating the rate of persistence between
the terms of a time series is the coefficient of autocorrelation, r, , which equals
the covariance between each term and the k following term divided by the variance.
To give some idea of the persistence in time series of precipitation and

storm water discharge, in Figure 7 a correlogram is shown with provisional figures
for k • 1 to 200. It concerns the median of r, for rainfall and storm water discharge
intensities calculated per 30 seconds. Only wet 30-second periods are considered.
The visualized figures in Figure 7 may be compared with the value of the 5 per cent
probability level of a theoretical distribution of autocorrelation coefficients
derived by random sampling from a universe having a true autocorrelation of zero.
Of all samples used for the configuration of figure 6 the sample size n exceeds 100,
while the 5 per cent significance point amounts to 0.154 for n =• 100 (Ezekiel and
Fox, 1967), graphed in Figure 7.
163
Runoff from the Paved Area
A part of the precipitation on the paved area does not run into the storm
water drains and is considered as lost precipitation. For the period 1969-1975
runoff coefficients have been calculated (i.e., the runoff volume compared to the
rainfall on the paved area) for the housing area and the parking lot (Van den Berg,
1976 B ) . A picture of 1973-1975 data is given by the graph of the cumulative
distribution of the relative frequencies of runoff coefficients, Figure 8. This
graph has been made from measurements in the housing area with a runoff volume of
at least 5-mm.
The Null-Hypothesis, holding that the frequencies of the runoff

coefficients are normally distributed, is not refuted by the chi-square test, either
raingauge 3
rainfall in mm
15
,<
17 /
• •
/
/
/
q / •
*
*
y •
H
y
*
* limit oí 95 °o confidence interval
3 v
. y = 1.004 x + 0.02
• Í • •
n . 113
•V
7r r = 0.96
n
0
A 3 6 9 12 15
rainfall in m m
raingauge 2
FIGURE 5-CORRELATION BETWEEN TWO RAINGAUGES

Correlation between the amounts of 113 rainstorms measured
with two groundlevel raingauges at a distance of 140 m
(wind direction between south and west).
164
•
r
t
ït
L oo r-»
165
k
in
- _ _ . _ prtciprtitnn
tdiirgi
na
IM;
04 1
\
I
\
\
nv \
\
\ n.100
S
--___
' — •
.111
10 70 40 100 150 200
k
FIGURE 7- CORRELOGRAM
Correlograra with the coefficient of autocorrelation r^ of the order
k for intensities per 30 seconds (median of 9 samples). The dashed
horizontal line shows the 5% probability level of a theoretical
distribution of autocorrelation coefficients derived by random
sampling from a universe having a true autocorrelation of zero.
.„, relative frequency ( n = 64 I
9Í1
fln
7fl
fin
Si)
4(1
in
?n
in
n
10 20 30 40 50 60 70 80 90 100'.
runoH coefficient
FIGURE 8-CUMULATIVE DISTRIBUTION
Cumulative distribution of the relative frequencies of the runoff
coefficients of a housing area in Lelystad. The data are from 64
rainstorms with a runoff volume of 5-mm or more, during the
period 1973-1975.
166
for the housing area or for the parking lot. The F-test shows that the standard
deviation of both samples does not differ significantly, and the Student test
indicates the same for the means of the two samples. So it may be supposed that
the frequency distributions of both catchments belong to the same population.
Best estimates of the parameters of this population are a mean of 66.6% and a
standard deviation of 10.05. Supposing that these samples of respectively 64 and
51 rainstorms are representative, it can be calculated easily that, for instance,
95 per cent of all rainstorms (with a runoff volume of at least 5-mm) will have a
runoff coefficient which is smaller than 83.1%.
Applying modeling techniques, the net precipitation rate is an important

part of the input. So a method is evolved which seems to give a rather useful
distribution of the losses during a rainstorm. The storrawater drains in the
catchment areas are permanently filled with water for the greater part (prism
storage) in order to have a quick response when it starts to rain, or in other
words to be able to study the inflow hydrographs (Van den Berg, 1974).
As Figure 9 shows, the inflow hydrographs as well as the outflow

hydrograph can be simulated by the Nash model, i.e., a cascade of n linear
reservoirs (n not necessarily being an integer). The fitting of the model is
expressed by the model efficiency after Nash (Nash and Sutcliffe, 1970):
2 2
F F
2 ° -
R
—
F
o
2 2 2 - 2
where F - £(q' - q) , F 0 =» E(q - q) , q and q1 are respectively the measured and
calculated discharge during the same time interval At, and q is the mean measured
value.
Discharge of the Subsurface Drainage System and the Groundwater Level in the
Unpaved Area
Automatically recorded data for such items have to be corrected with the
help of control data, frequently as measured by hand. Concerning the groundwater
level data, each year the piezometer tubes are water leveled several times to be
able to eliminate soil shrinkage. The groundwater level has to be controlled
because skiddings in the mechanical transmission would give rise to the
appearance of sudden jumps in the level of the float.
The measurements of the discharge of the subsurface drainage systems by

an electro-magnetic flowmeter have to be adapted because of changes in the converting
factor. The factor is calculated from the regression of the control data with the
recordings. The precipitation of iron compounds gives also rise to problems causing
an increasing (i.e., time dependent) resistance to the flow into and through the
drainpipes and through the throat of the flowmeter. Only a long series of adjusted
data will make it possible to analyse the different processes. Figure 10 shows a
graph of the groundwater level and the subsurface drainage discharge in relation to
the precipitation for two catchment areas, viz. the parking lot and the housing area.
The different response to the precipitation has to be explained in the first place
167
by a difference in the storage capacity of the soil in the two areas. 97 per cent
of the surface of the parking lot is paved. The soil below the pavement consists
of compressed sand, while the upper layer of the unpaved part of the housing area
(56%) consists of a loamy soil with a thickness of 0.5- to 1-m. The drainage
system is at the same level.
mm/mm
Housing area Parkinglot

inflow inflow
Í14
}
i
ft /< \
f i
It
11
It
i it
»V / i
\i
V \\
n? i
' 't
1
s f » «
«»
•
i
* \
TI 1
i
1
^»,
n ... •A . . ^ _
>n •^0
¿
1
\ ^
I I I '
m.acnrprf
i l
outflow outflow
0.4 calculated
il
t\ ft I
ft/
[t
/ 1 il
\l
f.J
¡1
\ 1 /'
1 1 If \\
0.2 V t
.'
1 •
s
1
•
J \
t
•
•
\N
*-
20
**"i —,
40
J 60 80 100
^
120
\
20
v
40 60 80 10U
^—'
120
FIGURE 9-COMPARISON OF RESULTS

Graphs of the net precipitation, the inflow calculated by
means of the Nash model or via the continuity equation
("measured") and the calculated and measured outflow of
the stormwater drains of a parking lot and a housing area
at Lelystad (8 August 1970).
168
rainfall in m m
70
BO
50
40
in
70
j***
in ¡J—
groundwater table ( cm )
1
36
30
24
—*
y
18 _ y
f
/
• Paikinglot t
í
r*""""""1 r—— Housing area f'

y
_3
subsurface drainage discharge (l/min )

11?
•
IRR y
J
y
y
14 4
y
1
1
9R . » - — - •
7?
4R
74
n
S 10 15 20 25 30 31
days
FIGURE 10-GROUNDWATER RESPONSE
Rainfall and the response to it by the groundwater table and the
subsurface drainage discharge of two different catchment areas
(January 1970). The reference level ("zero") of the groundwater
table is 160-cra below the soil surface in the housing area and
100-cm below soil surface in the parking lot«
169
References
Ezekiel, M. and K. A. Fox (1967) Methods of correlation and regression analyses:

Wiley, New York.
Nash, J. E. and J. V. Sutcliffe (1970) River flow forecasting through conceptual

models. Part 1 - A discussion of principles. J. Hydroloy 10, pp. 282-
290.
Van den Berg, J. A. (1973) Neorslagmetingen in een stedelijk gebied (in Dutch)
(i.e. "Measurements of precipitation in an urban area"), Flevobericht
no. 88, Usselmeerpolders Development Authority, Lelystad.
Van den Berg, J. A. (1974) Enige aspecten van de afvoerhydrologie van Stedelijke
gebieden (in Dutch with an English summary) (i.e. "Some aspects of urban
runoff") H 2 0 (7) nr. 23.
Van den Berg, J. A. (1976 A) Data analysis and system modelling in urban catchment
areas (in the new town Lelystad, the Netherlands), Hydrological Sciences
Bulletin, XXI, 1.
Van den Berg, J. A. (1976 B) Progress report, june 1976, Usselmeerpolders

Development Authority, Lelystad.
170
CATCHMENT STUDIES IN URBAN HYDROLOGY (NEEDE)
(F. Slijkoord, Heidemaatschappij, Arnhem)
Study Area
2
The catchment area chosen is the village Neede, ¿bout 2-km in size,
with an average imperviousness of 30% and about 10,000 inhabitants. The village
sewer system for combined wastewater and stormwater runoff has four drain points:
a wastewater treatment plant (maximum capacity, 0.110-m /s); and three outfalls
with different thresholds. The village is situated partly in a flat area, partly
on a hill.
Ins trumentat ion
The precipitation in the catchment area is measured in three standard

Hellman rain gauges (funnel area 0.02-m , height 1.2-mi and two rain gauges with
a funnel at the bottom (funnel area 0.02-m and 0.25-m , respectively).
The discharge of the wastewater treatment plant is measured in a Venturi

flume, and that of the outfalls by means of HL-flumes in the receiving open
channels. Upstream and downstream water levels of the flumes are recorded.
Additional information, particularly as to fluctuations of the storage, is obtained
by measuring the water level at several locations in the deepest parts of the sewer
system.
Nine measuring points are connected into a multi-point recorder (central

data-logger). The principle of this measurement is to record the magnitude of an
electrical resistance, which has to be identical with that of the water level. In
this way, the results for two of the rain gauges, three of the discharges and all
storage-levels are recorded. The remaining measuring instruments have automatic
recorders. In the early stages of the investigations, all recordings have been
continuous, mainly to check whether the instruments were reliable or not. The
analog recordings are converted separately to digital ones on paper tape, which
can be fed into a computer.
Results
At this stage of the investigations the following provisional information

can be given on some estimated parameters of the runoff process: translation time
(peak of precipitation to peak of runoff), about 20 minutes; reaction time
(assuming a first order reaction), about 15 minutes; and average runoff coefficients
ranging from 0 to 0.7. On the basis of tentative figures, however, the conclusion
seems justified that the average number of discharges estimated at the outfalls is
not in fair agreement with the number predicted according to current standards.
Future measurements will also include development of a pollutograph at the outfalls.
A synopsis of the results of these studies will be given at the

International Workshop on the Impact of Urbanization and Industrialization on
Regional and National Water Planning and Management, Amsterdam, October, 1977.
Attention will also be given to the relation of the runoff analysis of urban areas
to the receiving waters in a rural area (a river basin of a polder).
171
SOME ASPECTS OF THE RUNOFF FROM ROADS
(B. R, Voortraan, IJsselmeerpolders Development Authority, Lelystad)
Introduction
When roads are missing curbs and gutters, rainwater flows over their
edges into a ditch or trench. Besides the surface runoff there might exist a
subsurface drainage system. Roads especially, being built upon soils consisting
of silt-loam, peat or clay, need a drained blanket. In order to determine the
quantitative aspects of the surface and subsurface discharge from roads, a number
of catchment plots have beeir instrumented. A description is given below of the
catchment plots in the new town of Lelystad and some results are discussed.
The Discharge of a Subsurface Drainage System of a Road
Roads built upon silt-loam or clay soils need subsurface drainage. As

a rule, only the foundation blanket is constructed with drainpipes. According to
the width of the road, a drainage system is constructed on both sides or only one
side of the road. To answer the question of which part of the rainfall upon roads
is discharged by blanket drainage, a catchment-plot was installed along the road
Houtribdreef in the new town of Lelystad.
Figure 11 shows the cross-section of the road and verges (receiving

channel berms and median strip). The two drainpipes are connected over a length
of 118-m of one traffic lane. Before the data collected can be analyzed, the
question of the extent of the area drained by one unit of drain length has to be
resolved. Assuming that the watershed in question extends from the centre of the
middle verge, and that half of the ditch verge will drain directly to the ditch
and not to the drain pipe, the drained area per unit length (F) will be between
the centre of the middle verge and the centre of the ditch verge, so:
F » k middle verge + lane + \ ditch verge (Case A ) .
One can imagine that a part of the runoff from the road does not infiltrate into
the soil of the ditch verge and will reach the ditch as surface runoff or through
the subsoil. This will be the case at rainstorms with higher amounts and higher
rates of precipitation. Then, as a rough indication for the drained area the
following may hold:
F » \ middle verge + 2/3 lane + \ ditch verge (Case B ) .
From measurements of the groundwater level in the middle verge during rainy periods
between August 5th and September 10th, 1973, it has been determined that the
groundwater level did not rise. So, no contribution of this area to the discharge
of the subsurface drainage system can be expected. This means that the surface of
F has to be adapted to:
F » road + \ ditch verge (Case A 1 ) ; and

F = 2/3 road + \ ditch verge (Case B').
172
Table 1 shows some calculations of the discharge of a blanket drain, based on the
above-mentioned definitions of F. From this table it can be concluded that half
of the precipitation on roads and verges will be discharged by a subsurface
drainage system (blanket drain). Under circumstances in which direct runoff to
the ditch can be expected, the portion carried by the blanket drainage will be as
high as 70 per cent.
The positive influence of a subsurface drainage system of roads is

illustrated in Figure 12: a nearly constant deep groundwater level in the blanket
compared with the groundwater level in the centre of the ditch verge, and a sharp
reaction of the drain discharge to precipitation.
TABLE 1 - EXAMPLE OF RESULTS, SUBSURFACE DRAINAGE

Precipitation and discharge of a blanket drain (length 118 m') of a road
in Lelystad, during four rainy periods in 1973. The measured discharge has been
converted into depth (mm) and the runoff coefficient (R.C.) for some definitions
of the drained area F.
Discharge Discharge for Various

Dates, Precipitation, of Values of Area F
1973 mm Blanket»
Drain, m mm R.C. mm R.C.
Case Case
A' B»
5-19 August 41 19.5 0.40 0.53
16 21.5
30 August to
10 September 30 17.7 0.49 0.65
14.6 19.5
Case Case
A B
10-16 October 25 21.6 13.0 0.52 15.8 0.68
16-31 October 60 58.5 34.9 0.58 42.7 0.72
173
o
to
I-
<
UJ
<
<
a:
LU
m
i
<
o
«I
u.
o
z
o
I-
o
UJ
to
i
(O
to
o
ce
o
UJ
CD
174
precipitation ( m m / 2 4 hours: 08-08 o'clock)
H 3 a
] 1
i
o lepth in cm
i. 1 b!. -tóid;
70
40
fin
80
100
170
140
1B0
——— groundwater level in centre of ditch verge (north)
180 groundwater level in blanket of road (north)
170
110
100
1
1(1
i
80
I
70
I
RO
I i
sn J
40 1 IW •
10
\ 1
70
\ 1
\ \
in
n,,, „,_,,.,.,.., ,,»-», • '1 ' - ^ t 1 '

20/7 25/7 30/7 5 / 8 ' 10/8 15/8 2 0 / 8 2 5 / 8 3 0 / 8 5/9 10/9 15/9 2 0 / 9 2 5 / 9 30/91973
date
FIGURE 12-FIELD OBSERVATIONS

Precipitation, groundwater level and discharge of subsurface drainage in a
catchment plot along the Houtribdreef, Lelystad, for the period July-
September 1973.
175
Surface Runoff from Roads
Three small catchment plots have been instrumented to investigate the

surface runoff from an asphalt road. The plots were located in the town of
Lelystad and consisted of the following roads, Figure 13:
Houtribdreef, a two-lane road with lane widths of 8-m. A middle verge

separates the lanes; both sides of the blanket are drained; and
the slope has a one-way direction down to the 4.5-m wide ditch
verges.
Zuigerplasdreef, an 8-m wide lane. Both sides of the blanket are
drained with a two-sided slope down to the 5.5-ra (west) and 9.5-m
(east) wide verges.
Runderweg, an 8-m wide lane. There is no subsurface drainage; and the
slope is two-sided down to the 5.5-m wide verge.
Both sides of the road were equipped as catchment plots. Gutters were
constructed in the verges at increasing distances from the side of the road in
order to catch the water. Each gutter had an outflow into a reservoir of 100 or
200 litres. Generally, the content of the reservoirs was measured after each
rain storm. Each gutter had a length of 3-m with a top width of 7-cra, constructed
at distances of 1, 2, 3, 4 and 5 metres from the side of the road. Optimal catch
conditions were ensured by means of a strip of aluminum (5-cm wide) at the top of
the gutter. This strip was put into the sod, approximately 1-cm below the surface.
A total of 108 runoff measurements were made in 1973 and 1974. From a comparative
point of view, 94 showers were taken in consideration, with a total precipitation
of 853.5-mm, ranging from 0.7-mm to 34.8-mm. The precipitation was measured by
means of a ground level rain gauge. These values were recorded continuously by a
data logging system. The periodical observations of groundwater level and runoff
were not automatically recorded.
Relation Between Rainfall Intensity and Surface Runoff
In order to study the relation between rainfall intensity and surface

runoff, the data were subdivided into six classes of rainfall intensity, from
which the corresponding runoff coefficient for each gutter was determined. In
converting the runoff volume to a depth, the area F which is drained by a gutter
has to be known. This area consists of a part of the verge, the interval between
the road and the gutter and of a part of the road. For a road sloping sideways
unilaterally, the total width of the road is assumed to be drained by the gutter;
and for a road sloping sidewards bilaterally, only half of the road has been taken
into account.
Comparing the results for different distances to the road it seems that
the surface runoff does not always decrease proportionally with a greater distance
from the road. Combining the results of each two succeeding gutters makes this
relation more clear, Table 2. This table gives the runoff coefficients for the
combination of gutters 1 + 2 , 3 + 4 and for 5 alone. Runoff coefficients are
indicated for a gutter respectively 2, 4 and 5 metres from the roadside. Also
given in the ¿able are the amount of rainfall and the number of rainstorms from
176
which the runoff coefficients were calculated. For each class of rainfall intensity,
the entries in Table 2 show a relation of decreasing runoff coefficient with greater
distance to the roadside.
The data in the rainfall intensity classes of 20-25 mm and higher contain
some inaccurate measurements. This has been caused by two factors. First, all
rainstorms between two measurements of the runoff volume have been considered as
one rainstorm, while the interval between two measurements varied between a few
hours and several days. Second, it appeared that the volume of the collectors was
sometimes too small to contain all the water discharged by surface runoff. Clearly,
both facts influence mostly the measurements for the higher classes. The division
of the data according to Table 2 shows rather large differences between the figures
of several classes, for instance gutter 2 compared with gutter 5. Therefore, the
dispersion is expressed as the quotient of the standard deviation and the arithmetic
mean, the coefficient of variation V. This coefficient is rather great for gutter 5
and for some other cases. The null-hypothesis, holding that the mean values of the
different classes of rainfall intensities do not differ significantly from each
other, was tested via analysis of variance. From the F-test it appeared that most
differences are significant for classes 0-5 mm to 10-15 mm inclusive. Figure 14
graphically presents the runoff coefficients for the six classes of rainfall
intensity. It shows that the good relation between the runoff coefficients and
the mean rainfall intensities existing in the lower three classes is not present
in the higher classes (more than 15-mm), which may be the result of the factors
mentioned above. However, the data give an impression of the mean surface runoff
coefficient for different rainfall intensities. For rainstorms up to 15-mm
inclusive, the runoff coefficient will be smaller than 0.5, 0.2 and 0.05 at a
distance to the roadside of respectively 2, 4 and 5 metres. Probably the runoff
coefficient will not exceed a value of 0.1 at a distance of 5 metres from the road,
even for rainstorm intensities higher than 15-mm.
Relation Between Rate of Rainfall and the Surface Runoff
Besides the mean runoff coefficients, Figure 14 shows the average value
of three storms with the highest rainfall rate within the six classes. The values
are approximately 0.7, 0.35 and 0.2 for storms of up to 15-mm inclusive, at a
distance to the roadside of respectively 2, 4 and 5 metres. The same tendency is
present: a considerable increase of the runoff coefficient up to rainstorms of
15-mm inclusive. For higher rainfalls there is no significant relation.
Factors Affecting the Surface Runoff of Road and Verge
Topography, depression storage and wind and spray effects by traffic may
influence the behaviour of the runoff process. Although the study areas are in a
flat area, there is a microrelief, for example traffic ruts in which rainwater will
be collected temporarily. Depression storage will be larger if the grass-covered
side of the verge along the road is somewhat higher than the side of the road cover.
These circumstances will cause a certain conveyance of the runoff instead of a
uniform surface flow of water per unit of road length. The spray effect by traffic
during and after rainstorms cannot be quantified.
177
north
north
'" "i'i'i
DC Houtribdrecf
~)
Runderweg
j'i'i'i'i'i
^ TTTTTir
south
south
north
ndervveg
.i. gutter
Situation study-areas
south
Tïï
Üi EL
no 5—!f
ja. -Gl
MK
RI
Hi-
(S- -side of ditch
ll " tr*
{ — 3 m — no 4 j! n "
ii
- 3 m — no3 n 'i
II pipe verge
(.—3m—' no 2 j|
E f-—3m—— no 1
Jgutter
ï E
3m—
•side of road
J !L_
road
FIGURE 13-ROAD SURFACE RUNOFF STUDY INSTALLATIONS
178
r~ r» m tn vo CM vO IO CT* en en 00 vo en en
M in 10 r- O r-l
r-l vO O m vO O N v O vo 0 r-»
00 *
• r-4 > 0 • 0• •
0 r-l • r-4•CM• O • r-l• iH* O • O• CM• O • rH• r-l•
•tf cu •
en 4J en
00 r- en VO CM CM r- r~ m CM CM CM CM
1 0 m • CM CM i-l iH O O CM rH
CM
O tn rH 0 CM O O
r- H co u
0 • •
eu O• 0• •
0 O• O •O • O • O •O • O • O •O • O • O •O •
«tf • H <r 00 co tn on>* CM <* CM vO CM
CM CM r-l r-4 CO IO Cl r-4 VO CM rH vO
* <r en 0
m CM > O• 0• «
0 O • O •O • O • O •O • O• O •O • O • O• rH•
en
1 n 1-4
m c* • 1-1 en CM CM tn m vo m m rH 00 CM f.
<N CM CM r-l CM r-l O tn r-l O CM rH O
u en rH O • • •
•
PÎ O» O • O• O • O •O • O • O •O • O • O •O • O O O
r-l m r-4 r-4 -tf r- tn en en r-
* •tf m tn r-i m CM cn en r~ en m 00 N^-O
m i-i > O • 0• •
0 O• O «O • O • O •O • O • O• rH• O • O• iH•
! CM vu •
01 CM
vO • O CT> -tf 10 tn CM <t CM CM <fr 0 0 CM <t CM
A CM i-l CM r-l CM r-l O en rH O • * rH O en iH O
u m m • ai
>, g • • • O •O • O• O •O « • O •O • • O •O •
ens it;
g 0 O O O O O
« Xi
en •a
lE
-i
g
•H
c
u O 4-1 1-4 vO vu vO vO 00 r- r-4 CM m
0)
a en en
H 4-> a CM co CM r~. 00 O CM IO O ^t m CM en r~ r^ tí
C/3 u
* o
i-l O 1-1 vo• > O• 0 • 0• O • O• r-4• O • O• r-l
» O • O• rH• O • O• rH• •H
rH CM 4-1
1 0 r- •S vO ed
14-1 0 r-l CM vO CM 00 CM r- r-H r-4 co en rH •Í C A H
m
1-1 ki CJ i-l Ü• <± i-lOO ( M H O •H
c en tn t-4 en r H - o en 0 0
•H
Ol
aj u • 0 • 0• »
0 O • O •O • O • O •O • O • O •O • • O •O • >
A4 SA O
"g r-l
ai
•a
I4-4
Z3 -a
O 30
r-l
cd
IO H m CM r~ r-4 tn r*
en N kO CM m co 00 •a
Ol Ol CO N vO CM CO m r~. CM vO O CM e
* m m m
m
1-4 10 O > 0 • 0• 0
• O • O• r-H• O • O• 1-4• O • O• CM• O • rH• CM• cd
0 r-l •
1 r-l CM ai
0 O CM m vo r^ CM -i vo en <f en
O•
Xi
1-4 CM CO CO r-l 1-1 1-1 0 en rH O <T CM 0 CO rH O
•
od O • 0• 0
• O • O •O • O• O •O • O• O «O • O • O •O • M-l
O
C
eu
•H
m 1—1 10 r-- <n co H ri vO 00 CM CM j-i
vo co N fs<f OO vO rH
rs<t H CO rH O O
vD * m 3
O • > 0 • 0• 0
• r-l • r-4• C •M O • rH• CM• O• rH
• m
* 0• • •
CM in er
1-1 CM vO
1 CO m m en r-4 vo m 0 x:
u ci CM r-l !-t O O O O O en rH O CM O O
Pi 0 • O • O• O• O •O • O O O O • O •O • O • O •O •
00 vO <t tn r-» en C M vO • * CM m
r- CM H neo rH en CM rH O CM 00 CM 00
* co • • • • • •
0 > 0 • O • •r-l • • •
CM tn CM rH • • •
CM en rH CM CM O r l H ed
m en
1
CM
• •H
u
0 00
rH • en co r » O -tf CM C M -tf r-- 00 CM rH CM CM I-.
1-1 •H O r-4 O O rH O O iH O O cd
u iH O O >
•
OS 0 • 0• 0
• O» O •O • • • •
O O O O • O *O • O • O •O • U-l
o
u
C
U-l eu
ai •H
t-l <M E S E E B E E B E E ai E E B
U E B B u
ai O 7 ai CM «3- CM -sl- CM -j- m CJÛl •tí CM <f IO CM -d- m
u ai
3
f
<2
ai
00
CO
• H
lt-4
C M
•0 (U
ai
rH
«4-1
ni •0 e« a)
4J •ri xi x: Xi x¡ oo td
eu
4J 3 co z C co Ï5 •H H S
3 3
0 0Í
es)
179
2 m from the side of the road
O1
0.6
_1J1
i r
S:N S^l EW
'¿¿¿l¿¿¿ \//sï//>i///. V//Af//#//A y//A//AsM \///l//sV/A
H H R Z H R Z H R Z H R HZ R
4 m from the side of the road
0.5
0.4
0.3
0.2
0.1 JLjj
NJS7¡ÍEM
0 ¡MM
H R Z
vfcwf///¡
H R Z H R Z H R Z H R Z H R Z
r c. 5 m from the side of the road

0.5
N north
S -south
0.4
~E • east
W -west
0.3
0.2
W
0.1 „ 1
•NEW SNJW S NEW SN™ 1 1
R Z R Z R Z R Z M R Z
0-5 5-10 15-20 20-25 25-35

rainfall intensity (mmI
FIGURE 14-SURFACE RUNOFF COEFFICIENTS

Surface runoff coefficients (R.C.) for 6 classes of rainfall intensity (mm) at
distances to the side of the road of 2-, 4- and 5-m. Mean values of 94 storms in
1973 and 1974 (shaded) and average values of three storms with highest rainfall
rate (black). The three roads in Lelystad with catchment plots are:
H = Houtribdreef R • Runderweg Z = Zuigerplasdreef
180
Evaluation of Results
The collected data are not sufficient for evaluating a water balance of
the system of "roads and verges". In spite of certain failings in instrumentation
and some non-clarified circumstances, such as local concentration of runoff, vague
watersheds and a sometimes too limited volume of gutters and reservoirs, the
following may be concluded from this catchment studyi
surface runoff coefficients of roads will not exceed 0.2 for verges up to
5-m in width and 0.7 for verges up to 2-m in width;
the contribution of the groundwater flow to the runoff process (direct to
the ditch or via a subsurface drainage system) exceeds the surface
runoff at verges of more than 2-m; and application of subsurface drainage
of the verges, besides the blanket- drainage, may be a useful alternative
for ditches and open drains along roads.
181
Section 3 Urban hydrology modelling in the Netherlands
Urban hydrological models developed and used in the Netherlands can be

divided into three broad categories after McPherson (1975):
planning models, used in macro-scale applications, such as for

metropolitan or city-wide master plans;
design-analysis models, hydraulically more sophisticated and thus more
detailed tools; and
operations models, likely to be more application-specific than planning
or design models.
Although McPherson restricts these categories of mathematical models to "the

simulation of urban rainfall-runoff and rainfall-runoff-quality," it can be stated
that the same division may be used for some models developed in the Netherlands,
in which social elements are introduced in addition to pure hydrological (natural)
and operational (artificial) elements.
In this section are briefly described studies on mathematical modeling

and their applications. At this moment the review is not complete. Included is
a description of a water resources management system that can be considered as a
comprehensive urban water model.
Reference
M. B. McPherson (1975), Urban Hydrological Modeling and Catchment Research in the

U.S.A., Technical Memorandum No. IHP-1, ASCE.
MATHEMATICAL MODELS IN URBAN HYDROLOGY

(F. Slijkoord, Heidemaatschappij, Arnhem)
To evaluate the "transport capacity" of newly designed and/or existing

drainage control systems it is of vital importance that runoff for a network to be
studied be simulated in relation to (normative) rainfall. For this simulation,
Heidemaatschappij has developed two different computer programs which can be
classified as "design-analysis" models. The principle of the first program is
based on steady flow in closed (and in open) conduits. By means of this program,
the water level for every node of the network under study can be calculated, i.e.,
the response of a rainfall of constant, normative intensity. To assess the "polluting
capacity" of a sewerage network (the pollution being caused by discharges from the
outfalls) a comparison can be made between the actually available and the normative
storage. The second computer program deals with non-steady flow in open channels.
With this model the runoff can be simulated for a time-dependent input at any point
of the boundary of the system studied. For the type of problems indicated this
input is rainfall. The response is calculated for a series of matching intensities
and durations of rainfall. Chosen a priori as normative input is that input which
182
gives the most critical response. The program will be extended so that it can be
used for closed conduits as well as the calculation of transport of local pollutants
into and through a system. To determine the validity of these models for practical
usage a verification study will be needed. In order to verify the tentative criteria
for sewer systems, a test program has been launched to measure rainfall and runoff
of a village. With the help of these measurements an attempt will be made to
derive the parameters by which rainfall can be converted into runoff. In recent
years considerable experience has been gained with this type of work with reference
to rural areas.
STATIC AND DYNAMIC SIMULATION OF WATER TRANSPORT IN A COMPLEX NETWORK SYSTEM

(Twente University of Technology Enschede; Department of Chemical Engineering)
Increasing cultural eutrophication is threatening the unique ecosystem

of the water basin of Vollenhove. For management decisions, a knowledge of the
transport and fate of phosphates would be required. A study of the water transport
has been presented by Lijklema and Van Straten (1975). Dynamic simulations, based
on a nodes and links model, show that considerable oscillations in flows and water
levels occur, induced mainly by the intermittent operation of the pumping station.
The dynamic behaviour reflects the natural frequencies of the system. The results
were compared with those obtained from a static model. It was concluded that
static models tend to underestimate the degree of mixing in the network. The
research program for 1976 included a water transport model for the whole water
basin of Vollenhove, field tests of the model and studies on the influence of wind,
Lijklema, Van Straten and Nissink (1976). Other research subjects are related to
model studies on oxygen balances of small rivers, on uptake of oxygen by sediments
and to sediment-water interactions.
References
Lijklema, L. and G. Van Straten (1975). Static and dynamic simulation of water
transport in a complex network system. In: Modeling and Simulation of
Water Resources Systems, G. C. Vansteenkiste (Ed.), North Holland
Publishing Company.
Lijklema, L., G. Van Straten and H. Nissink (1976). Simulation of the Water
Movement in a Polder District. In: Systems and Models in Air and Water
Pollution, Institute of Measurement and Control, London.
DIGITAL MODELING OF NON-STEADY FLOW IN OPEN CHANNELS

(J. E. G. Bouman and E„ Schultz, IJsselraeerpolders Development Authority, Lelystad)
A digital model has been developed to facilitate the design of open

watercourses in the new towns of Lelystad and Alraere in the Lake IJssel Polders.
In Lelystad the application is restricted to the design of watercourses, culverts
and weirs within the urban area. The application of the model to Almere will be
somewhat broader as a result of the plan of development. Alraere fundamentally
consists of five urban areas surrounded by small rural areas, so mixed runoff from
urban and rural areas has to be taken into account.
183
In modeling the given situation, the system of watercourses is transformed
into a system of branches and nodes. In the branches the equation of momentum is
used and in the nodes the equation of the conservation of mass. The unknown variables
are the discharges in the branches and the water levels in the nodes at the time
level t + At. To solve the equations a finite difference scheme is used according to
the implicit method, as described by Allersraa (1973) and Vrengdenhil (1973). All
equations are transformed into linear relationships, which makes it possible to use
matrix-vector operations. The unknown water levels in the nodes are expressed as
a function of discharges in the branches. Together with fitted boundary conditions,
the linear equations for the branches are solved for each time step at the time
level t + e.At (0.5< 6 < 1 ) .
After solving the equations, which gives only the discharges in the
branches, the water levels in the nodes are calculated from these discharges. The
equations for the branches can be adjusted in such a way that culverts and weirs
can be taken into account. At the nodes there is a possibility to simulate a
fixed water level, a water level-time relationship or a pumping station. The
stability of the computation scheme has proven to be satisfactory, especially in
rather complicated systems of open watercourses with rapidly changing flow.
References
Allersma, E. (1973). Hydraulics of open water management. Proceedings and

Informations nr. 18 of the Committee for Hydrological Research T.N.O.
Vrengdenhil, C. B. (1973). Computational methods for channel flow. Proceedings

and Informations nr. 18 of the Committee for Hydrological Research T.N.O.
HYDRODYNAMIC LINEAR DISTRIBUTED MODEL OF SURFACE RUNOFF

(Th. J. van de Nés, Provincial Waterboard, Arnhem)
Recent technological progress has stimulated a hydrodynamic approach

to the surface runoff problem. Cheng Lung Chen and Ven Te Chow (1968) formulated
a mathematical model which describes the mechanics of surface runoff of a
catchment by treating the watershed as a nonlinear distributed system, solved by
the method of characteristics which, however, for complex systems results in an
extremely lengthy computer program. Bravo et al. (1970) used as basic elements
for a linear distributed model a linearized version of the complete dynamic
equations, introduced into the MIT catchment model. This model has some
disadvantages, because water movement is restricted to flows in one direction and
the analytical solutions are very complex. For these reasons Van de Nés (1977)
simplified the dynamic equation and leads it, after linearization, to a diffusion
type equation as proposed by Scho'nfeld (1948), Hayami (1951), Daubert (1965) and
Harley (1967). As part of a model for the rainfall-runoff relation of a catchment,
the hydrodynamic linear distributed model of surface runoff without internal
boundary conditions consists of a cascade of linear conceptual elements.
The model simulates a complex drainage system as a network of overland

flow and channel flow elements. These elements obey the one-dimensional equations
for unsteady flow in a channel. The catchment is subdivided into smaller elements
184
according to Bravo et al., connected with each other by the channel sections.
Approximating the complex geometric characteristics in this way, five elementary
types of problems can be selected: overland flow; fully lateral inflow; partial
lateral inflow; tributary inflow; and upstream inflow. It is assumed that the
catchment can be modeled by a suitable combination of these five elements and
that one element cascades into the other so that interactions between the elements
are excluded.
For the mathematical formulation of the five linear elements, two basic
problems have to be solved. First, the problem of the partial inflow into an
infinite channel reach, which is emphasized by Van de Nés (1976). Second, the
problem of the upstream inflow in a semi-infinite channel, solved by Harley. The
analytical solutions of the diffusion type equations for suitable boundary
conditions yield the impulse response functions, which characterize the operations
of the elements. Special attention is given to the application of the techniques
of linear system analysis, such as moments and spectra. These techniques produce
information on the relative importance of the various conceptual elements.
Consequently it is possible to decide on the necessary detail in the variation in
time and space of both the inflow and structure of the drainage model. Results
obtained by using the linear model have been compared with results of a more exact
non-linear model and have been encouraging.
Conclusions
A complex distributed-surface runoff system can be composed of the linear

elements described. Because the whole complex system is linear and time variant
the principle of superposition applies.
The necessary complexity of a distributed model of surface runoff can

be determined by techniques of linear system analysis. Physical information about
the surface runoff system should be available.
The techniques of linear system analysis show the effect of the system
parameters P*(the dimensionless length parameter), Q (the characteristic time of
the system) and R (the dimensionless inflow length parameter expressing the
spatial distribution of inflow) on the behaviour of each conceptual element in
relation to a given input. In this way the filtering effect of the system can be
determined.
The maximum sampling interval of the input depends on the filtering

effect of the system and can be derived from the amplitude density of the
impulse response.
Comparison of the linear solutions with the complete non-linear

solutions for the various conceptual elements shows that with a proper choice of
the reference discharge or water depth a reasonable agreement can be obtained.
The efficiency coefficient R E » 90%.
The identification of the system parameters P, Q and R can be made in

two ways. First, from the physical characteristics of the surface runoff system,
or second, by methods of systems analysis, if the input and output data are known.
185
References
Bravo, C. A., Harley, B. M., Perkins, F. E. and Eagleson, P. S. (1970). A linear

distributed model of catchment runoff. M.I.T., Hydrodynamics Laboratory
Technical Report No. 123.
Chen, C. L., and Chow, V. T. (1968).. Hydrodynamics of mathematically simulated

surface runoff. Department of Civil Engineering, University of Illinois,
Hydraulic Engineering Series No. 18.
Daubert, A. (1965) Quelques aspects de la propagation des crues. La Houille

Blanche, No. 3, p. 341-346.
Harley, B. M. (1967). Linear routing in uniform open channels. M. Eng. Science

Thesis, National University of Ireland, Dept. of Civil Engineering.
Hayami, S. (1951). On the propagation of floodwaves, disaster prevention. Res.

Inst. Bull. 1, Kyoto University.
Nes, Th.J. van de (1973). Linear analysis of a physically based model of a

distributed surface runoff system. Doctoral Thesis, Agricultural
University, Wageningen, The Netherlands.
Nes, Th.J. van de (1977). Linear analysis of a physically based model of a

distributed surface runoff system. Journal of- Hydrology.
Schönfeld, J. C. (1948). Voortplanting en verzwakking van hoogwatergolven op

een rivier. Ingenieur, B., Jan. p. 1-7.
RAINFALL-RUNOFF OBSERVATIONS IN URBAN AREAS, PHYSICAL-MATHEMATICAL ASPECTS

(Department of Hydraulics and Catchment Hydrology, Agricultural University, Wageningen)
In 1967 a joint program with the IJsselmeer Polders Development Authority

was started on the instrumentation of urban catchments in the built-up area of
Lelystad (Kraijenhoff Van de Leur and Zuidema, 1969). Weir boxes, flumes and
recording devices for measuring rainfall, storm flow and groundwater flow were
designed and calibrated in the Hydraulics Laboratory. The department's staff
developed computer programs for quality control and processing of the Lelystad
data logger (Figure 4) output (Progress Reports, 1967-1972).
Following changes in personnel the cooperation was interrupted. The

emphasis of the department's experiments had to be shifted towards new experiments
involving a tilting flume with a rainfall simulator, sprinkled-roof experiments
and sprinkler-induced runoff from various paved areas (Zondervan and others, 1975
and 1976).
Theoretical studies resulted in simple linear models for rainfall-surface

runoff relations (Nes and Hendriks, 1971).
New fitting procedures for modeling non-linear catchment responses are

being developed and attention is being given to the generation of short interval
rainfall sequences.
186
References
Kraijenhoff van de Leur, D. A., and F. C. Zuidema (1969). Een onderzoek naar de
relatie tussen neerslag en afvoer in het stedelijk gebied van Lelystad
(in Dutch with English summary) "A study on the relation of rainfall and
runoff in an urban area at Lelystad," H 2 0 (4), nr. 4.
Progress Reports of the Institute for Hydraulics and Catchment Hydrology, Wageningen
University, 1967, 1968, 1969, 1970, 1971/72.
Zondervan, J. G., and A. Dommerholt (1975). Transformation of rainfall into sewer

inflow on a certain type of inlet area (in Dutch). Laboratory Note nr. 33.
Zondervan, J. G., and R. C. Koole. Suggestions for a runoff experiment in a

suburban catchment area (in Dutch). Laboratory Note January 1976.
Zondervan, J. G., and A. Dommerholt (1976). Transformation of rainfall into sewer

inflow on various types of paved areas (in Dutch). Laboratory Note nr. 38.
Nes, Th.J. van de, and M. H. Hendriks (1971). Analysis of a linear distributed
model of surface runoff (in English). Laboratory Report nr. 1.
MATHEMATICAL MODELING OF OVERFLOW FREQUENCY, QUANTITATIVE AND QUALITATIVE ASPECTS

(D. H. V. Consulting Engineers, Amersfoort)
In the Netherlands, combined sewer systems are often used for draining
wastewater and rainwater. The usual capacity of the systems is such that precipitation
with a continuous intensity of 60-litres per second per hectare can be drained off.
Furthermore, the capacity and the location are chosen to ensure unimpeded drainage
of wastewater. Most of the land in the Netherlands is gently sloping or nearly
horizontal, and the groundwater table is high and soil conditions are often poor.
Deep excavations would lead to unacceptably high investments and moreover, the level
of the receiving body of water is often only slightly below the surface of the land
to be drained. Thus, the available potential height is small, which means that
the diameter of pipes must be relatively large. Another consequence is that water
must be pumped to the water purification plant while,.for financial and technical
reasons relating to the purification process, the capacity of the pumps is limited
to 2 to 5 times the capacity required for the discharge of wastewater.
Rainwater which drains from impervious surfaces in excess of the

capacity of the system, and which cannot be evacuated by the pumps, overflows into
the surface waters via weirs or emergency outlets. In flat areas the sewer system
is located entirely below the crest of the weirs. This means that the dimensions
of these pipes can be calculated with the aid of formulae developed for permanent
flow in completely filled pipes. The number of times the rainwater mixed with
wastewater overflows, i.e., the overflow frequency, can be calculated with the aid
of a mathematical model in which the drainage system is reduced to a reservoir
with an overflow, the capacity of the reservoir being equal to that of the drainage
system situated below the crest of the weir. During rainshowers, the pump
extracting water from the reservoir is deemed to be working at maximum capacity.
187
The mean theoretical overflow frequency can be calculated by introducing the data
relating to the quantity and duration of the rainshowers observed over several
decades. The overflow frequency calculated in this manner is used to evaluate and
compare combined sewer systems.
Because the overflow frequency gives no information concerning the

degree of pollution, and even the separate rainsewer systems cannot safeguard
surface water totally against pollution, models have been constructed which enable
calculation of the pollution of the surface water in relation to the drainage
system being used. The models take account of the following aspects: the load of
wastewater; the sedimentation of solid wastes during dry spells; the duration of
the dry spells; the duration, quantity and subsequent pollution of the rainwater;
the disturbance of the sediment; the dimensions of the drainage system; the
capacity of the pumps; and faulty connections.
The mean discharge of wastes and extreme shock loads are calculated by
means of the models. In this manner it is possible to arrive at a more indicative
evaluation of the effects of the technical solutions chosen to meet the problem of
transporting wastewater than is possible if the evaluation is based solely on the
overflow frequency.
STORMWATER CHARACTERISTICS OF SOME ROTTERDAM SEWERAGE DISTRICTS

(G. T. M. van der Kroon, Rotterdam Municipal Works)
Investigations into the factors that influence the stormwater quantities

in a combined sewer system, as in Rotterdam, are important because of two reasons:
in the past, the costs of research were in poor accordance with the investment
costs either of wastewater treatment plants (w.t.p.'s), or of the implementation
of the requirements by officials of w.t.p.'s; and the model computation which has
been used in Rotterdam in 1971 for the establishment of the design criteria with
respect to the preliminary plan of a central w.t.p. has increased the cost of the
project considerably in comparison with cost estimates for former phases of the
plan. For instance, the cost of connecting the old quarters of Rotterdam to a
w.t.p. was estimated at one hundred million guilders in 1971, comprising the costs
of new force mains to connect the existing lift stations to the w.t.p., upgrading
lift stations (13 stations), reconstruction of sewers, etc. This explains why
urban drainage systems deserve more interest and study. To be successful, this
study must be done on a full technical scale. It is expected that as a result of
this, many important data which may be of benefit to the cost will be collected
and at any rate will yield rational solutions to the items of the comprehensive
plan with respect to sewage treatment.
Design criteria for sewer systems in the Netherlands, and especially in

the polders of the coastal region of this country, where many pumping systems are
applied in polder catchments, are governed by a model called "stormwater overflow
frequency model". This model, which is generally used for pumping systems in flat
areas, has been specially developed to establish some design criteria for the
connection of urban catchments to wastewater treatment plants. The pumping system
of Rotterdam needs the concept of stormwater overflow pumps to the river to
maintain a low stormwater overflow frequency to the surface water inland of the
embankment of the river and to maintain a reasonable stormwater pumping rate to
188
the wastewater treatment plant. The stormwater overflow pumping system has the
following advantages: a reduction of the pollution of the inland surface water
(i.e., canals) in the centre of the city; a reduction of the total quantity of
dilution water needed for maintaining the quality of the water in the canals;
and a reduction of the stormwater pumping rate to the wastewater treatment plant.
The present stormwater overflow frequency model gives a prediction of the
stormwater overflow frequency value which is not in agreement with the actual
value. An alteration of this model has been elaborated which gives results that
are more in agreement with actuality.
Reference
Kroon, G. T. M. van der (1976). Stormwater characteristics of some Rotterdam

sewerage districts (paper to be presented at the IHP International
Symposium, Amsterdam, October,.1977).
STUDIES OF WATER RESOURCES MANAGEMENT IN GELDERLAND

(Committee for the Study of Water Resources Management in the Gelderland Province,
c/o Provincial Waterboard, Amhem)
Purpose
Since 1971 a Committee for the Study of Water Resources Management in

the province Gelderland has been operating according to the following instruction:
"creation of a scientific base for an optimal management of the surface and ground
waters available in Gelderland with regard to their quantity and quality".
Because of the fact that water has numerous functions in our society,
water resources management cannot be uncoupled from developments in the society.
One of the fundamental problems in good water resources management is the fact
that the natural water supply does not evolve synchronously with variations in
water demands, but is bounded by time and place. The policy of the Government will
be one of achieving a balance between needs and supply to minimize social demands.
Therefore, it is necessary to determine the importance of the social interests.
However, many social interests are not weighable or are very difficult to assess,
which complicates decision-making progress. Because of this fact, it is felt
necessary to integrate knowledge and the development of new methods for the benefit
of decision making and planning of water resources management. Attempts have been
made to solve these problems with the aid of a systems theory approach, where the
theory concerning hierarchical problems clearly offered perspectives (Colenbrander,
1976; Van de Nés, 1976). In this approach, the water resources management system
is divided into three types of elements, viz., social, natural and artificial
elements, and into a number of levels (sub-systems) that are studied in different
degrees of detail, mostly byraultidisciplinaryteams, with the aid of mathematical
models (Figure 15). In this way, the results of measures to be taken can be
predicted, so that alternative plans can be valued and mutually compared. This
procedure includes the possibility of dividing the complex decision process over
various levels, so that it can be solved better. The application of a planning
189
INTEGRAL WATER
TU RES.MAN.MODEL
4 STRATUM
INTEGRAL APPROACH
V
I QUALITT MODEL FOR
i QUANTITT MODEL
un GROUND AKD SURFACE FOR GROCXD AND
3 STRATUM WATER I SURFACE WATER
INPUT SOCIAL
ELEMENTS
V7-_-_-_"
r MODELS W "-""r:r:::::::.--.--"
HOBELS ÕT THE MubkLt oí vm
GROUND AKD CHEMICO-BIOLOG
SURFACE WATER lUALITY OF QUALITY OF
FLOW SURFACE WATER GROUND WATER
KD
2 STRATUM
COHERENCE NAT.
AND ARTIFICIAL
ELEMENTS
INFORMATION
í> CONDITIONS
FIGURE 15-HIERARCHY OF THE WATER RESOURCES MANAGEMENT SYSTEM
190
procedure within the decision process enables attainment of an integral water
management plan. As a result, recommendations will follow with regard to a
rational organization for water resources management.
Conclusions of the First Phase of the Study
Development and testing of a number of mathematical (demand and supply)

models for the benefit of the quantity and quality management of surface water
and groundwater in Gelderland leads to the following conclusions:
the physical and social consequences of various measures in a number of

cases can be predicted with reasonable reliability;
the short-term and long-term measures for the management subsystems (water
supply, water level control and water treatment) can now be better
adjusted to one another;
further development of an integral water resources management model must
facilitate economic evaluation of water resources plans, and this
integral model fundamentally forms the scientific base for the
selection procedure for long-term planning;
the degree of detail of the required data with respect to time and place
can now be indicated better, with the result that improved directives
can be drawn up for the collection and processing of those data;
application of the above-mentioned methods has led to a number of concrete
studies for the Achterhoek, a part of the province of Gelderland; and
the manner in which the results of the investigation can be evaluated within
the organizational structure of water resources management has led to a
better insight on the desired structure for the process of decision
making.
Future Plans
To make the methods operational and to profit from the results obtained
so far, this research will be continued with special regard to the following points:
the mathematical (demand and supply) models of quantity and quality management
have to be made generally applicable, with a number of partial processes
better formulated for that purpose;
the way to couple the various submodels needs further study;
the development of a mathematical model concerning the quality of groundwater
is desirable;
the quantification of the relations between certain measures in water
resources management and social consequences needs further research;
the integral (economical) water management must be elaborated, especially
concerning the total integration of various subinterests (nature,
landscape, recreation, etc.); and
the availability of the developed methods have to be improved for the benefit
of potential users.
191
Further Information on the Water Resources Management System
First Stratum. Figure 15 describes the hierarchy of the water resources

management system. In the first stratum, the one with the basic elements, we can
distinguish: social elements representing the demands for water for various
purposes (population, industry, agriculture, nature, outdoor recreation);
natural elements representing the various natural water supplies with their
physical, chemical and biological characteristics (such as rainfall, evaporation,
soil moisture, groundwater, surface water); and artificial elements representing
technical and administrative measures (such as water supply, water level control
and water treatment). The social elements are responsible for the demand for
water and the amount of wastewater for a certain area, which is highly dependent
on the economic development within the area. So, two econometric prognosis models
for future demand are found to be suitable tools: extrapolation models, based on
a continuation of a series of developments from the past towards the future
(however, such an extrapolation yields only imperfect information); and
determinant models, in which various factors that affect the demand are taken
into account. The future demand can be influenced in two ways: by thé
development of needs, and by the extent to which the needs will be met (permits,
pricing policy).
Second Stratum. At this level, the coherence between the natural

elements is being studied with the object of establishing the interactions of
measures of the various management subsyste:ns. Thus, one can study the mutual
coherence of such measures. A number of models have been developed for the
determination of the flow of ground and surface water (quantity models), and for
the description of the oxygen dynamics of brooks (chemico-biological quality
models). A coupling of the models enables the integration of quantity and quality
aspects. The description of the processes is based as much as possible on physical
laws, and supplemented with data from empirical knowledge. In so doing, assurance
can be attained that the parameters have a direct physical meaning and that the
model can be applied for a large number of situations.
Models for Flow of Ground and Surface Water
A model was developed by de Laat, van den Akker and van de Nés (1975),
in which the subsystems that have been described within the first stratum, such
as rainfall, evaporation, soil moisture, ground and surface water, are now
described in their mutual coherence, Figure 16. The model can be applied for
every random non-homogeneous area. It consists of a net of squares with an
arbitrarily chosen magnitude. The model describes, for every random point of
time, the condition of the various subsystems at the nodal points of the network.
The relation between groundwater on the one hand, and soil moisture and surface
water on the other, is established at the nodal points of the network.
Geohydrological data on the deep subsoil, soil physical data on the

upper soil, and data on the vegetation, must be input at each square of the grid.
Every arbitrary operation,of groundwater extraction, infiltration and brook level
regulation has effects on the groundwater level, the actual evaporation and the
runoff in the remaining nodal points, as functions of time and place. These
effects can be calculated per nodal point with the aid of the model.
192
To illustrate an effect of groundwater extraction (approximately 2
million m3/year) of the pumping station "Het Klooster," in Figure 17 are plotted
measured and calculated groundwater levels for a significant point within the
area as a function of time.
With the help of a rainfall-runoff model (Kraijenhoff van de Leur, 1973)

it is possible to establish which part of the rainfall is discharged rapidly. The
model consists of a fast and a slow component. As a result, an empirical relation
between groundwater level and runoff can be derived, while also peak rates of
drainage can be predicted from rainfall.
SURFACE RUNOFF ^ 0, [SURFACE R U N O F F
SYSTEM
"V (NOT M F U . T R A T M Q PRECIPITATION)
ii
p (PHËCIPITATIONI^ V
OlRlMOFF,
RIVERS
I IMFLTRATION)
1
SOI. MOISTURE
SYSTEM
J
GROUND WATER
SYSTEM
SUB SURFACE
RUNOFF
Ob|BASE
SYSTEM
O T H E R CUMATOLDGlCAL ^
DATA
E (EVAPORA (ON)
FIGURE 16- HYDROLOGICAL MODEL FOR A BASIN
APRIL' MEI 'JUNI* JULI ' AUG ' SEPT' OCT ' NOY ' DEC ' JAN F E B ' MAART
1973
FIGURE 17-GROUNDWATER MODEL RESULTS

Comparison of the Calculated Head (—) with Observed Values (")
193
Models for Biochemical Quality of Surface Water
It is difficult to give a good estimate of the reaction of surface water

to any type of pollution, because of the simultaneous occurrence of biological,
chemical, physical and hydraulic processes. In describing in a model the very
factors that affect the water quality, the problem arises that water quality
cannot be characterized by one specific variable, because various decomposable
and non-decomposable substances finally determine the water quality. Important
variables in evaluating this quality are the concentration of dissolved oxygen,
and the pollution that consumes oxygen on biological oxidation by micro-organisms
(Biochemical Oxygen Demand, BOD). Besides the model for oxygen, a model for BOD
must be developed. In the model (de Boer, 1974), a river is divided into a number
of arbitrary sections, thus enabling calculation of the oxygen content and the BOD
as average values per section for a series of arbitrarily chosen, but equal, time
steps. The model is also suitable for describing concentration changes of non-
decomposable substances.
In addition to climatological data, also needed are hydrological data

such as the runoff, water depth and profiles of the river, and physico-chemical
coefficients must be known for each section of the river. The respiration and
photosynthèses of algae is not yet included in this model, being restricted to
situations in which these may be neglected. Bannink (1975) developed another
model that describes the oxygen balance for a given standing crop of algae. At
the moment, the models of de Boer and Bannink are being integrated.
A coupling of the quantity and quality models will lead to an integration

of both phenomena, which may bring us after incorporation of the social elements
(third stratum) to an integral water resources management model (fourth stratum),
Figure 15.
References
Bannink, B. A. (1975). Een nuraeriek model ter beschrijving van de zuurstöfhuishouding

in de Oude IJssel, waarin met name de invloed van fotosynthese op deze
huishouding is gevat. (A numerical model for the description of the
oxygen dynamics in the River Oude IJssel, with special emphasis on the
influence of photosynthesis on this dynamics). Technische Hogeschool
Twente, afdeling Chemische Technologie, Enschede, the Netherlands (in
Dutch).
Boer, B. de (1974). Simulatiemodel van de zuurstöfhuishouding in beken en

rivieren. (Simulation model of the oxygen dynamics in brooks and rivers).
Technische Hogeschool Twente, afdeling Chemische Technologie, Enschede,
the Netherlands (in Dutch).
Boer, B. de (1976). A moving cell simulation model of the dynamics of dissolved

oxygen and related quality variables in rivers and brooks. In: System
Simulation in Water Resources, éd. G. C. Vansteenkiste. North Holland
Publishing Company, Amsterdam, the Netherlands.
Colenbrander, H. J. (1976). Systems Theory as an approach for regional Water

Resources Management. Proceedings and Information nr. 22 of the Committee
for Hydrological Research T.N.O.
194
Comraissie Bestudering Waterhuishouding Gelderland (1975). Modelonderzoek 1971-
1974 ten behoeve van de waterhuishouding in Gelderland. Deel 1:
Onderzoek in relatie tot beleidsvoorbereiding. (Model investigations
1971-1974 on behalf of the water resources management in Gelderland.
Part 1: Investigations in relation to policy preparation). Arnhem,
Commissie Bestudering Waterhuishouding Gelderland (1976). Modelonderzoek 1971-

1974 ten behoeve van de waterhuishouding in Gelderland. Deel 2:
Grondslagen. (Model investigations 1971-1974 on behalf of the water
resources management in Gelderland. Part 2: Basic studies) Arnhem,
Kraijenhoff van de Leur, D. A, (1973). Een elektrisch analogon voor het neerslag-
afvoermodel Wageningen en enkele toepassingen. (An electrical analog
for the rainfall-runoff model Wageningen and some applications).
Landbouwhogeschool, afdeling Hydraulica-afvoerhydrologie, nota 25 en 31,
Wageningen, the Netherlands (in Dutch).
Laat, P. J. M. de, Akker, C. van den and Nes, Th.J. van de (1975). Consequences
of groundwater extraction on évapotranspiration and saturated and
unsaturated flow. International symposium on the application of
mathematical models in hydrology and water resources systems, Bratislava,
Czechoslovakia.
Nes, Th.J. van de (1976). The Structure of the decision making process within the
Water Resources Management System. Proceedings and Informations nr. 22
of the Committee for Hydrological Research T.N.O.
WATER MANAGEMENT MODEL FOR THE NETHERLANDS

(T. A. Sprong, Rijkswaterstaat, Service for Water Management, The Hague)
The Situation
Figure 18 shows the situation in the Netherlands, where the Rhine in an

average year accounts for 707. of the total available water quantity including the
precipitation. Water balances have been drawn up for both the northern and southern
part of the country for a fictitious dry summer and a dry June in the year 2000.
The conclusion was that the total quantity of water available would be sufficient
in the future, but that additional infrastruetural facilities would be necessary
to safeguard supplies at all times and places.
Since a shortage may occur during short periods in a summer, the balances
must be determined for shorter periods. Chosen was a 10-day period, with three
such periods in a month. Also necessary was particularization according to place.
For this, 33 points which are focal water distribution points have been specified,
Figure 19. To be able to determine at what point and in what period the shortages
will occur, requirements in, for example, the year 2000 have to be set against the
sources of supply, which can be obtained from historical hydrological data. So,
water balances will have to be drawn up for 33 points for a series of years (e.g.,
40 years), with the decade as the balance period. This, however, means calculating
47,520 water balances (36x40x33) for one variant. Therefore, a digital computer
model has been devised for this purpose.
195
*>
N
fi. YSSEL
BARRIER OAM
LAKE]
NORTH SEAfCAMAL
Sf ,*
/AMSTERDAM
RHH
THE HAGUEj
NEW WATERWAY,
ROTTERDAM
HARINGVLIET SLUICES LEK
VAAL
SCHELDT-RHtNC CANAL
•#**•• •
a Ü
FIGURE 18-MAP OF THE NETHERLANDS INDICATING MAIN RIVERS AND CANALS
196
N \
- û*\,?
O
/ i
29 / f
AMSTERDAM
THE HAGUEx
24
II
I >26
7Î
|2 ¿at/m ...
t&*.\
n\
33
\ \
4 •
FIGURE 19-NETWORK FOR THE WATER MANAGEMENT MODEL OF THE NETHERLANDS
197
The Model
The model is a balance model, its purpose being to assist in the

determination of water management policy. This involves a comparison of sources
of supply and requirements and the testing of infrastructure measures. For
these purposes the computer program should be as flexible as possible in order
to serve a variety of purposes. It was necessary for the water management
infrastructure (the network), that all che data relevant to it become input data
in the model. For this purpose the network (system of rivers, canals and lakes)
is schematicized into a system of nodes joined together by links. A water balance
is calculated at each node. At any node water can be distributed (between links),
stored, abstracted or added. If certain constraints are not met (for example,
minimum rates of flow), the time, place and requirement not complied with are
identified.
Further Development of the Model
Due to the fact that much of the Netherlands lies below mean sea level,
salinification is an important factor in water management. This salinification is
caused by seepage of saline groundwater into the deep polders, by leakage at locks
which communicate with waterways containing brackish water, via estuaries and
river mouths (for example the New Waterway) and, furthermore, via the Rhine River.
The severe salinification and pollution of this river represents a tremendous
problem for the Netherlands. The intention is therefore to use the model not only
for drawing up water balances but also for the preparation of associated salt
balances. The consequences of measures to reduce salt pollution can then be
studied. Salt (chloride ion) is admittedly only one of the relevant quality
parameters, but in the special circumstances prevailing in the Netherlands it is
certainly one of the most important. Further, factors influencing the choice of
salt are that being a conservative substance it is relatively simple to use in
such calculations and a good deal of data on it is available. The incorporation
of other than conservative quality parameters in this model, notably those
relating to domestic and industrial effluents, would seen to be impossible for the
time being. The processes involved are generally speaking very complicated, and
certainly the flow time can no longer be ignored. Such processes can therefore
be better studied in separate models. The water management model described must
be regarded as one of the tools that are available for planning the optimal use of
the raw water.
Reference
Sprong, T. A. (1976). Water management model for the Netherlands. ECE Seminar
on long-term planning of water management, Bulgaria, May 1976.
198
Urban hyetiological modelling and catchment
research in Norway
Nils Roar Selthun

Hydrological Division
Norwegian Water Resources and Electricity Board
P . O . Box 5091, Majorstua
Oslo 3, Norway
This report has been prepared under a project on urban catchment research,
Project PRA 4.2, financed by the Norwegian Ministry of Environmental Affairs and
administered by the Hydrological Division of the Norwegian Water Resources and
Electricity Board.
Until the last few years, hydrological research on urban catchments has been
virtually non-existent in Norway. This is partly due to a general orientation of
the resources of Norwegian hydrology towards satisfying the pressing needs for
hydrological data for hydropower development. This has focused interest on large
and medium-sized catchments.
The design of urban drainage systems has relied on practices developed abroad.
As a consequence, runoff coefficients are used that were originally developed for
climatic and hydrological conditions quite different from those found in Norway.
The result has usually been overdimensioned systems.
During the last decade, the rapidly increasing investments in drainage systems
have brought attention to the adverse economics of poor design methods. In 1971
a six-year governmental research program for purification of wastewater was
launched by the Norwegian Ministry of Environmental Affairs. The program consists
of some 40 subprojects, of which the following are related to urban hydrology:
Project 4.1, Collection of rainfall intensity data. Administered by
the Norwegian Meteorological Institute, Hydrometeorological Division.
Project 4.2, The effect of urbanization on runoff from small catchments.
Norwegian Water Resources and Electricity Board, Hydrological Division.
Users report in preparation.
201
Project 4.5, Design of retention basins and storm overflows. The
Norwegian River and Harbour Laboratory, Users report is in Norwegian.

(1)
Project 4.6, System analysis of urban water systems. The Norwegian
Institute of Water Research. Users report is in Norwegian.' '
Project 4.7, Investigations of urban runoff composition. The Norwegian
Institute of Water Research. Users report (in Norwegian) in press.
Project 4.10, Preparation of storm data (for planning use). The
Norwegian Institute of Water Research. Users report is in Norwegian.(3)
A manual on instrumentation and data collection in urban water systems is

being prepared through cooperation between Projects 4.1, 4.2 and 1.1 (Composition
of sewage).
This report will mainly describe activities and some results of Projects 4.2,
4.6 and 4.7.
References
1. Mosevoll, G.: "Regnvannsoverlöp og fordröyningsbasseng". PRA, users report
no. 2, ISBN 82-90180-01-2, 69 pp., Oslo, 1975.
2. Lindholm, 0.: "Systemanalyse av avlöpsanlegg". PRA, users report no. 1, ISBN
82-90130-00-4, 57 pp., Oslo, 1975.
3. Lindholm, 0.: "Valg av modellregn". PRA, users report no. 6, ISBN 82-90180-05-5,
16 pp., Oslo, 1975.
202
Section 2 Urban catchment research in Norway
The first systematic urban catchment investigations in Norway are being
carried out as part of two projects in the Norwegian research program for
purification of wastewater: Project 4.2, the effect of urbanization on runoff from
small catchments; and Project 4.7, investigations of urban runoff composition.
Project 4.2
The project is run by the Norwegian Water Resources and Electricity Board
with some cooperation with local municipalities. Field studies and data collection
were started in 1972 in two types of research catchments. One type consisted of
paired rural catchments, one of which was expected to be urbanized during the
project period with the other to be left untouched as a control catchment. We
shall refer to these as "A-basins". The other type was single catchments in urban
areas, termed "B-basins". Observations included in the program have been
precipitation (time resolution, 1-minute), discharge (5-minute interval), soil
water content (observation interval half-monthly or monthly, in 20-cm intervals of
depth down to 2-m, A-basins only), ground water level (weekly, A-basins only), and
evaporation from sunken pans (daily, A-basins only). In addition, two A-basins
have been equipped with automatic recording climatic stations, logging data every
30 minutes on magnetic tape. Detailed descriptions of instruments and observation
routines are available.'•*•'
Precipitation measurements. Fuess siphon pluviographs with a chart paper
speed of 20-mm/hour are used. All registrations are digitalized on standard
8-track paper tape by means of a chart reader and stored on magnetic tape. Since
203
1974 the basins have also been equipped with a Plumatic, a Norwegian tipping-bucket
rainfall recorder. This recorder registers rainfall increments of 0.2-mm on
continuously running magnetic tape.
Discharge. Used are standardized channel weirs of a type designed by the
Division of Hydraulic Engineering at the University of Trondheim.'2' The weir
consists of a rectangular inlet section ensuring a standardized rating curve,
and a V-notch weir with a rectangular upper section. The project has used the
original weir configuration in open channels and has modified it somewhat for use
in manholes. Water level is measured by means of a float and changes in water
level since the last time interval are recorded every 5 minutes on standard 5-track
paper or plastic tape. The data logger is the Norwegian SIGUM LP II, and the
output is directly computer-compatible.
Data Processing. All registrations of precipitation (from both types of
instruments) and water level are stored in a compact and standardized file
structure on magnetic tape after thorough quality checking. The files are
operated through a library of programs for data processing, retrieval, tabulating,
plotting and analysis. A flow chart describing the data processing routine is
included in Figure l.w)
Catchments. The project is operating five sets of paired A-basins and seven
B-basins. The locations of these 17 basins are shown in Figure 2 (Nos. 1-17). A
summary of some basin characteristics is given in Table 1. The catchments are
described in more detail elsewhere. (1) All have separate surface water drainage
systems.
204
UI
>
UI
oc
UI
<
2£ a
n < o
s
ö °"
5 5 d
CO Ü £ i¿
z
CO
O
<
«p 2 ce ce
H Ul
m
il CA
»- (0 o
3 o Z
O Lu
< Z 2 s. z o
LISH
AND
í° -Z O u. • tr
• > Ul
<o p Ul u
OC X CL
Ul Ò.
2
S*2
5g
H U.
LT
o
HI Ul
a
ui
LU
o
Ul Ul
X Z)
o
fc Û.
I oÈ ce
s < <
<S>
° h 3
< §
< 2 <S)
< LU
S z O
o o O
LT
Ul o o
CL
RTE
a.Ul <
5? Ul
<
<
O 2
o
< (0
Ul u 1 Q
<O
Ul
Ul
a. X I
s <
APE
CO oc O < <
o
< «• Ul
< è s ui
CO t-
S
LU
•p i CO
Ul oc a
Ul cr
< 15 CO <
o
a. z
O
O
<
a D.
1t- £ 8
_l t-
O CD
<
a. < LL.
T< _
T ~
Z I
_! O
ï z
o
Z
M
CO
Ul
< O o
S < <
O
>- 5 o
> o 2 t oc
< <
ce ui < a.
ui <
ui 2
(0 u¡ s
co (^
O ui
oc
205
NORWAY
•hS^
11)11 /Uilkm
—1 ' • • á _
FIGURE 2-URBAN CATCHMENTS INVESTIGATED BY PRA PROJECTS 4.2 AND 4.7

(PRA 4.2: NOS. 1 TO 17) (PRA 4.7: NOS. 14 TO 22)
206
Project status. Daily mean values of precipitation and discharge, graphs and
tables with 5-minute time resolution of significant storm events and seasonal
variations of other variables are published in biannual data publications.

(4)
Some preliminary data analyses have been presented.^5' A statistical analysis of
runoff coefficients is in preparation. An analysis of the effects of urbanization
on catchment water balance has been hampered by a general delay in the urban
development of the A-basins compared with the original schedules. None are
finished yet, and this reduces the value of the data collected in these basins for
urban hydrological applications. As a consequence, the project period may be
extended past 1976. But the data from the A-basins indicates that extensive
overland flow from grassed areas is rare in Norway, at least for the recurrence
intervals used in drainage system design. This should permit application of the
original Road Research Laboratory method assumption that, from a design standpoint,
only paved areas contribute to storm runoff. This assumption will be tested in the
model verification study.
Project 4.7
This project has been run by the Norwegian Institute of Water Research (NIVA)
in cooperation with the Division of Hydraulic Engineering at the University of
Trondheim and the town authorities of Oslo and Berum. The project period was from
late 1974 to early 1976.
Data have been collected from nine catchments, five with separate drainage
systems and four with combined systems. Four of the catchments were B-basins,
where quantity data were collected by Project 4.2. At the rest of the catchments
precipitation and discharge were measured in various ways. Water samples have been
collected manually, with 5 to 20 samples taken during storm events and additional
207
Industry
X
1
> <D
r-4 i-l T> X X X X X X X X X
o 00 - H iH x
XI tí S W
V •H Cd 0)
d t o u< ctí
d
cd rH
•H Cd.
h
d) 4-1
•-ï d
O. >, d)
Res id
Multi
Famil
X X X X X X X X X X X X
at ion
, ,
od
1 1 1 1 i • i 1 1 1 1 i 1 i 1
en en es CM CM CM CM CM cn en en <t cn •d- CM -Í
fÜ r-- r^ r-. —
r r«. I-. r~- r-- r-- r» r» p* r» p» r- •*
r- r-.
 o\ Oi
to F 4 iH rH r-l i-i i-i r-l rH rH rH rH rH rH r-l l-i r-i rH rH
J3
O
uau a\ en «* <f C7\ <J\ m IO rH rH 0\ P^ lO O 00 O I-»
112
112
141
101
TOT
' u o i C3B3TdToaaä CM CM o O cj\ (Ti CM 00 sr

ao 00 vO 00 p~ r- p-. 00
vO
UBS«
TTTM i
, i
, i vO o lO r^ en en 00
' s s a iísnoTAaaduii en CM CM CTi en rH
>*
en o r-> CJ> m CTi CM vO IO m •* VO -3- a\ CM vO r~
BU, 'B3.IV m
<j\
<í
00
en VO vO CT> •tf
vO
VO
vO
CM
CM
00
rH
00 en
CM
•*
CM
o» r» vO o->
rH
en en en en en
rH
**adA"i o u U O u
< < < < <! < < <
« PQ PQ « « « pq
kogen
Vika
iva
len
ard XI
00
XI cd
cd 00 ja cd "4J3 rH eu co
S
cd
XI
I-i
01
TI
W
-a 3
B d
m
4J
co .y CO o
fi cd
cd rH
X
XI
cd
(U rH
td
•H
rH
n
rH
¡S i-H •JO cd rH XI (U iH •H CO >
•rl CO 4J u
4J ta u Ai
i
Orm
0 u p rH r u :o 3 • eu a) rH
H
z > S â d CO > o
>
and
XI
4J
d
•I-I •o 3
ans
r-l rl ta
cd
a
0 B
a X
•H *>-l
co
atd •H
eu
•H HH 0) •H
o
•rl <fl
CU
00 m
XI
d
XI
5 CO
4J
CO o
XI
d
ia
d d
cd
d
cd
.2 CO •H
to rH
to
o
IH
á S c/a C/1
a £ a < O H
*
o• rH CM en < • in vO r^ 00 Oi o rH CM en
-* m VO
rH rH rH rH rH
rH rH
z rH
208
samples taken during dry weather flow. Samples were analyzed for the following
constituents and properties:
All samples — Less frequently —
chemical oxygen demand (COD) biochemical oxygen demand (BOD7)
volatile suspended solids (VSS) oil
suspended solids (SS) alkalinity
total phosphorous (TOT F) conductivity
total nitrogen (TOT tf) PH
lead (Pb) copper (Cu)
zinc (Zn)
cadmium (Cd)
orthophosphate
Leakage of wastewater into storm water drains was measured by tracer methods
in two catchments with separate systems, one of which was 20 years old and the
other one quite new. Leakage averaged 0.1% of wastewater in the new systems, but
exceeded 10% in the old one.
Locations of the catchments are shown in Figure 2. The catchments are
described in Table 2 and greater detail is given elsewhere.'"' The data are
presented and analyzed in a report. (**' Surface water contributions and dry
weather flows are treated separately to make results for combined and separate
systems comparable. Concentrations of pollutants are correlated with discharge,
time from start of storm, time since last rainfall, precipitation over last five
days and with concentrations of other pollutants (constituents). Total annual loads
are analyzed against catchment characteristics such as percentage impervious area,
person equivalents and catchment slope* Annual loads are shown in Table 3. Surface
water loads in combined systems have been compared with those of separate systems
209
Industry
X X
al
Ol « tí
CO f-l 0
9 cd
•rl
i-i
4J
T3 O 3 X X X X X X
tí h 4J
cd 4) -rl '
•J
i (O
O tí
«
tí O H
•H
4J r-l
CO cd
1-1 i-I
4-1
£ tí
rl H0) r^l f
CUl
X X X X X X
0 00 -rl - H
tí 8 «
T)
nne
•H PM
CO 3 0)
«¡
r-l
cd cd
•-ï •rl
Pw ru 4J
r-l tí
O . >> 41
Famil
Multi
Res id
X X X X X X
aad sauatBAXnBa en o en en
i
CN
en r-l •* CN
uosaad I AãTéy3ã
CN
r-l rH
en en
uoT3Btndoj
. . mm
o o o o
'U0"J3B3TaT03íá
UB3H 3 en
ao
ri- 00 00
o o
ri
CT>
r» ! > •
r^ r^ o\
o m
e1n r-l
% i». en co a o r~ a\ 1
o\ en r-l en o
rH vO O
'ss3UsnolAaadmi -* o
CM
A o\ CN NO fi> en o o en o
cd • C-.• • ey>• • •
oi cd
U X!
a\ «o
en en rH
r-l
CN
en
en
m•
l"1
•
cu
I-I o•
00
I-I CN en
<
t Ol
Q. co CO co CO co
o o u o
H
>>
kken
kken
cd
•rl
01 CU
> o M
1
cd OJ r-l i-i i-i A! M
•o
4J
XI
cd cd
2 u cd rH r-l 1-1 1-1 01 rH ai
u co u JA > > r-l J¿ u
CO a co ed cd rH r-l co J>¡ Ol cd
01 a 01 r-l 0 O 1-1 3
> o > m :£ co CO CQ
«
u
3
00
>,
u
1-1 TJ
iH M
cd a
i-i O
D. 0) i-I •a Oí O
•H 43 U-l ai ai S
O •o 01 4J
a co S
1-1 tí O cd
o •o
CO u
O g01 rH
CO cd
u
o O
3 r-l
O H O
tí
CO
* « tí
* «* in v£> r~ 00 o\ o r-l CS
0 CN CN CN
55•
210
<r -d- vO m o -d- -d- -d-
3 00 CM o o CM m -tf i-H
Ü • • • * l • • • •
o o o o o o O o
ï
*CLUi C "ï m
m
en CM CN
o
vO
00
•d- CO 00
¿J N • • • • l • • • •
ca CN o o O o o CM o
3
<U m CM o rH
O J3 00 rH rH rH vO CN CM r-i
CO Q* • • • • i • • • •
U-l i-H O O O O O r-\ O
u
3
CO Z CN CT. en o CM CM
• • • • i 1 1 • •
s H -3- O r-- -Cl- r-l 00
•H O rH •-H CM •-<
H
CI)
J=
-^ PH
00 "1 vO r
~ - in en vO T-{ CM rH
M H •
« • • • • m • •
O CN r-4 O o O m ir, V0 u-i
H a H rH
•H
3 U
U
en en
CO
1-1
vO
00 en m
CN i l
r-» en
00 rH
en o o
spo
> en rH \o -d- CM
tu
c U! en en o m 00 CN r- t'-
ni eu en m -ci- vO
-ce rH o m vO en vO en
w u c/> en en f^ -d- en 00 m
H rH rH rH rH rH
H
i-H
cd O r» 00 t>- o I-. oo en o
Cd rH CM
O
3 a H o t-< C0 en i-» -cí-
C o r-. rH 1-1 i-H CM m -d-en en
C u rH rH
<
r-.
518
131
187
c/l Q
O 1 1
'
1 i
•
PQ
en
en
eu
C
en
3 O m E
O r« CI en 00 r^ e1 O
n r~{ en r-^ eu
•H B-î os -d- en rH en KO 1 u
> o O en
)J CM rH >,
(U en
a XI
He eu
tí
•H
„ CM CN
0
ô r- en o O en O
O
•
O
i
cd o • « • • • •
cu cd en t'- v£> rH rH
en CM en in en 00
o O
u £. en en en r-. rH
< en rH CM
u
ems and
j|
CU
a W w en en o en O U o
>. •
H CM
eu u
u en
kken
c 3 >»
ika
00 en •
(!) •H
eu > O « eu CU
eu
4J
1
O
ara
E XI XI tí
cd
Z U
•i-i
eu rc-di rH •rl
rH
j¿ ¿¿
•rl •rl
u
eu
ca
rH
•rl 5
•u en u M > > rH J¿ XI a. •u
en Q . en et) n) rH rH en . * eu eu o
eu ex eu rH Ss O O •rl 3 4J
ed
en tí
> O > PQ :o en W « Oí
O
O
II II
rH en 1
*
o• •et m v£> —
r 00 o
en CM f-i
CM
CM
CM
z
211
and.have been found to be generally higher in combined systems, probably due to
washout of sedimented sewage from pipes during storms.
Pollution of surface water has been compared with that in sewage, and although
total loads in sewage are one order of magnitude greater than those in surface water,
loads in surface water often exceed loads in sewage during storms.
References
1. FRÀ project 4.2: "Introduksjon av maleprogram og malefelter." Project report
no. 1, NVE, Oslo 1974. 140 pp., English summary.
2. Tveit, J.: "Insulation against ice at measuring weirs". Nordic Hydrology, 5,
pp. 32-49, 1974.
3. Hetager, S. E.: "Presentasjon av forskningsprosjekt PRA 4.2: Urbaniseringens
innvirkning pa avlb'psforholdene i sraa nedbb'rfelt". ("Presentation of a
Norwegian research project: The effect of urbanization on runoff from small
catchments"). In PRA/IHD: "Kvantitativ urban hydrologi," a Nordic symposium
on urban hydrology, pp. 65-77, Oslo 1975.
4. PRA project 4.2: "Dataoversikt 1972 - 1974." Project report nó. 2, NVE,
Oslo 1975. 197 pp., English summary.
5. Roald, L. A., and S. E. Hetager: "Mulige effekter av urbaniseringen pá
avrenningsforlo'pet i Moss - Oslo omrâdet". ("Effects of urbanization on
runoff from small catchments in the Oslo-Moss region"). In PRA/IHD:
"Kvantitativ urban hydrologi," a Nordic symposium on urban hydrology, pp.
132-145, Oslo 1975. *
6. PRA project 4.7: "Forurensning i overvann." Report 0-57/74, NIVA, Oslo
1976. 55 pp. + data appendix.
212
Section 3 Urban hydrological modelling in Norway
Research on urban hydrological models is only a few years old in Norway. The
main achievement in this field has been the development of user-oriented models
for system analysis of urban water networks and wastewater treatment plants at
the Norwegian Institute for Water Research (the NIVA models). Systematic testing
of urban surface runoff models was started in 1976 by Project 4.2 (the effect of
urbanization on runoff from small catchments, described in Section 2) on data
collected by the project over the past four years.
The NIVA models
This model package has been developed over the last five years at the
Norwegian Institute for Water Research, and consists of two models: a network
model, handling combined sewer and separate storm sewer systems, and a wastewater
treatment plant model. The models have been described in several papers by
Lindholm,C1»2»-*»^/ and are, according to the nomenclature used in the first report
in this special IHP series,^-*' combined planning and design/analysis models.
The following description is based on Lindholm^-' and the users manual^"'
(in Norwegian),
The network model. This model can be used for
dimensioning of conduits;
dimensioning of retention basin volumes;
computation of hydrographs and pollutographs within a system and at outlets, and
flow and pollution at overflows;
computation of water levels in manholes and varying energy gradients under
surcharged conditions;
213
calculation of pump sump levels;
calculation of construction costs for a design network; and
simulation of annual or seasonal flow and pollution characteristics.
According to Lindholm:^' "The overland flow is presently simulated by means of
a runoff coefficient which may vary with time, time of entry, and different
overland storage functions (for the temporary storage on the surface). The routing
in sewer lines is a storage routing technique similar to that used by the modified
Road Research Laboratory (RRL) method.^''
"A maximum of 10 pumping stations may be included. The stations may have
two levels of discharge switching on and off at different levels in the pumping
sump, and the program keeps track of the water level in the pumping sump at each
time interval.
"A retention basin may be included at every node, as well as storm overflows.
All constituents are regarded as conservative when they pass through the network.
A relationship between concentration of the constituent and the time after the
start of rainfall is given as input. This function is used to compute the
pollutographs discharging from each subcatchment into the sewer network. In the
sewer network, a routing technique computes pollutographs in accordance with a
plug flow. From the retention basins, however, the program always computes the
pollutographs in accordance to the complete mixing concept. The sewer network
model consists of 6000 FORTRAN statements.
"In order to simulate a whole year's rain activity, there are two possibilities:
(1) to actually simulate a whole year's recorded rainfall, minute for minute; and
(2) to distribute the recorded rainfalls into 5-10 different base rainfalls, and
then compute the average rain intensity, average duration, and frequency per year
of each base rainfall." The latter alternative is usually chosen when the model is
214
used for planning purposes, for obvious economic reasons. A computer program for
converting actual rainfall intensity data as stored on magnetic tape at the
Norwegian Meteorological Institute, Hydrometeorological Division, into base
rainfall of this kind is described by Lindholm' ' (users report in Norwegian).
The hydraulic properties of the network model, and its performance compared
with other routing techniques, have been tested by Sjöberg.^ '
The wastewater treatment model. This model is matched to the network model.
Quoting Lindholm:'^' "The main objective of this model is to study the
performance of a sewage treatment plant receiving both sewage and rainfall runoff
over a given period of a whole year. The total program configuration, i.e., the
combination of network model and treatment plant model, can be used to analyze
the performance of a combined or a separate sewer system in a given area.
"The hydrographs and pollutographs are routed through the different unit
operations. At present the following units may be chosen: primary clarifier,
aeration tank, secondary clarifiers, flocculation and coagulation units, retention
basin, and overflow regulators. Only BOD^ and phosphorous are included at present.
The time interval between each computation may be chosen.
"An optimizing subroutine is implemented in the treatment plant model. When
a total amount of discharge of pollutants to the receiving waters is fixed, the
subroutine may search the least-cost combination of storage facilities and unit
processes. The method is based on a least-cost gradient technique. The wastewater
treatment plant model consists of 3000 FORTRAN statements.
"The subroutine may optimize the treatment/storage facilities for a constant
flow, a runoff from a single storm, or the ever changing influent for a complete
year."
The network and wastewater treatment models have been used for the system
215
analysis of several planned projects, the largest being the Oslofjord project,
planned to serve a population of 600,000 person equivalents and an area of
100-kra .^ ' The models have been leased to several Scandinavian consulting
engineers.
Model Studies at Project 4.2
This project includes the testing of urban hydrological models on Norwegian
data, i.e., data collected at the research catchments run by the project. The two
models that are being tested are the RUNOFF block (quantity part) of S W M M , ( ^
reprogramraed to accommodate metric units, and a somewhat revised version of
ILLUDAS.<12)
Preliminary results show that SWMM is working very well, even when the
default catchment parameters are used. This is partly due to the small runoff
contribution from grassed areas in Norwegian storm events.
The original version of ILLUDAS gave estimates that were too low for
subcatchment times of concentration for paved areas for all realistic choices of
gutter depth, and consequently overestimated peak flows. This part of the model
was replaced by a uniform sheet flow computation of concentration time, based on
the maximum 5-rainute rainfall intensity involved. This seems to give a satisfactory
fit to flow peaks, but the linear nature of the computation of paved area runoff in
ILLUDAS will necessarily give a poorer fit at lower flows. The main reasons for
testing the model are that it is easily adapted for design of networks of moderate
size on desk-top computers, and that its routing technique is essentially the same
as that of the NIVA network model.
The SWMM RUNOFF block will, after testing, be implemented as an option for
surface runoff computation in the NIVA network model.
216
1. Lindholm, 0. G.: "Factors Affecting the Performance of Combined versus
Separate Sewer Systems". 7th IAWPR conference, Paris, France, Sept. 9-13, 1974.
2. Lindholm, 0. G.: "Modeling of Wastewater Disposal Systems". Models for
Environmental Pollution Control, Deininger, ed., Ann Arbor Science Publishers,
Inc., Ann Arbor, Michigan, 1973.
3. Lindholm, 0. G.: "A Pollutional Analysis of the Combined Sewer System".
Thesis presented to the University of Trondheim in partial fulfillment of
the requirements for the degree of Doctor of Engineering, 1974.
4. Lindholm, 0. G.: "Pollutional Analysis of Combined Sewer Systems". Journal
of the Environmental Engineering Division, ASCE, Vol. 102, No. EE2, Proc.
Paper 12030, April, 1976, pp. 301-312.
5. McPherson, M. B.: "Urban Hydrological Modeling and Catchment Research in
the U.S.A." ASCE Urban Water Resources Research Program Technical Memorandum
No. IHP-1, November, 1975.
6. Lindholm, 0. G.: "Systemanalyse av avlöpsanlegg". PRA, user's report no. 1,
ISBN 82-90180-00-4, 57 pp., Oslo 1975.
7. Watkins, L. H.: The Design of Urban Sewer Systems". Road Research Technical
Paper No. 55, Her Majesty's Stationary Office, London 1962.
8. Lindholm, 0. G.: "Valg av modellregn". PRA, user's report no. 6, ISBN
82-90180-05-5, 16 pp., Oslo 1975.
9. Sjo'berg, A.: "Beräkning av icke-stasjonära flodesforlopp i reglerade vattendrag
och dagvattensystem." ("Calculation of Unsteady Flows in Regulated Rivers and
Storm Sewer Systems"). Division of Hydraulic Engineering, Bulletin no. 87,
Department of Civil Engineering, Chalmers University of Technology,
Gothenburg 1976.
217
10. Balraer, P„, J. Glomnes, 0. Lindholm and N. Saltveit: "Management of Urban
Runoff and Wastewater in the Oslofjord Area". Nordic Hydrological Conference
1976, pp. V-37 to V-50, Reykjavik 1976.
11. University of Florida: "Storm Water Management Model User's Manual, Version
II". EPA-670/2-75-017, U.S. EPA, Cincinnati, Ohio 45268, 350 pp., 1975.
12. Terstriep, M. L. and J. B. Stall: "The Illinois Urban Drainage Area
Simulator, ILLUDAS". Illinois State Water Survey, Bulletin No. 58, Urbana,
Illinois 61801, 90 pp., 1974.
Abbreviations Used in This Report
PRA: Research program for purification of wastewater administered by the Norwegian
Ministry of Environmental Affairs, 1970-1976.
NVE: The Norwegian Water Resources and Electricity Board.
NIVA: The Norwegian Institute for Water Research.
Address List for Norwegian Institutions Referred to in This Report
The Norwegian Institute for Water Research,
P.O. Box 233, Blindern, Oslo 3.
The Norwegian Meteorological Institute,
P.O. Box 320, Blindern, Oslo 3.
The Norwegian River and Harbour Laboratory,
Klebuvn 153, N-7000, Trondheim.
The Norwegian Water Resources and Electricity Board,
P.O. Box 5091, Majorstua, Oslo 3.
University of Trondheim,
The Norwegian Institute of Technology, N-7000 Trondheim.
218
Urban runoff research in Poland
Pawel Blaszczyk
Research Institute on Environmental Development
ul. Krzywickiego 9
02-078 Warsaw, Poland
Introduction
Since 1972, the Research Program of Urban Sewerage and Drainage Schemes
has been conducted by the Research Institute on Environmental Development in
cooperation with the Institute for Water Supply and Water Constructions at the
Technical University in Warsaw. The main objective of the Program is the
verification of assumptions and methods for planning and designing storrawater
and combined sewer schemes in large urban areas. The Program is financed by
the Ministry of Administration, Land Economy and Environmental Protection.
Specific objectives of the Program include definition of:
- the relations between rainfall and the runoff from urban combined
sewer and stormwater schemes, as well as the pollution content of
precipitation and runoff;
- the influence of the storage capacity of drainage networks and flow-
balancing tanks on the diminution of pollution loads transported by
stormwater to receiving waters; and
- the influence of short and undetermined rainfalls on the quality of
receiving waters.
Earlier research in Warsaw on the influence of combined sewerage
schemes on the Vistula River,v*»î as well as a survey of the area conducted as
part of a general project for development of sewerage and drainage in Warsaw,

(3,4)
clearly showed the need for investigation of the relations between rainfall and
the runoff from drainage catchments in order to meet the requirements for
planning and designing sewerage and drainage schemes. At the same time, previous
221
research' » ' had proven that the configuration of a town and the distribution of
its buildings have an important influence on the occurrence of zones in the area
with differing probabilities of rainfall of the same intensity. Consideration of
only these preliminary findings emphasized their great importance in the design
óf economical drainage schemes adapted to urban requirements^'' that would not
unduly pollute receiving waters.(^>9)
Research on the Relations Between Rainfall and the Runoff in Urban Drainage
Networks
Studies on the relation between rainfall and the runoff in urban
drainage networks were carried out at three drainage catchments in Warsaw: two
separate storm sewer systems having catchment areas of 23-ha and 280-ha,
Figure 1, and a combined sewer system with a catchment area of 600-ha. Values
of the ratio of impervious area to total catchment area are 0.4, 0.5 and 0.35,
respectively. Rainfall measurements are made by means of eight graphic rain
gauges and outflows are indicated by means of four level recorders installed in
specially made manholes. The clock mechanisms of all instruments are precisely
synchronized and are checked every day.
Studies of the quality of stormwater transported by drainage networks
are carried out simultaneously with observations of rainfall and runoff in the
drainage network. An automatic apparatus for taking water quality samples is
coupled to an apparatus that registers the level of water in an interceptor or
trunk sewer, enabling the taking of samples at intervals of a minimum of every
ten minutes. To carry out observations on relatively large drainage catchments,
it was necessary to develop a model of the passage of a storm over a catchment,
and only then to compare the registered rainfall with the associated runoff in
the drainage network. This was made possible by the use of a network of graphic
rain gauges.
222
FIGURE 1- SEPARATE STORM SEWER SYSTEM STUDY AREA, WARSAW
223
The results of the research will be the determination of detailed
relations between rainfall and runoff and its pollution, but not restricted
only to the drainage schemes examined. A prognostic method will thus be developed
for estimating the quantity of stormwaters and the pollution loads drained off
with them, with the method being based on observations of rains for specific
types of drainage catchments.v*0,11,12; Prognosis will cover not only a single,
one-time, process but also the characteristics of occurrences of different kinds
during a year.
Research on the Influence of the Storage Capacity in Drainage Networks and
Flow-Balancing Tanks on the Diminution of Pollution Loads Transported by Runoff
to Receiving Waters
Research on the possibility of reducing the amount of pollution load
entering receiving waters, as a result of runoff storage in networks or in
special balancing tanks, and conveying stored water to urban sewage treatment
works after rain has stopped, has been carried out in connection with a combined
drainage scheme in Warsaw.

(11-14)
Sewerage catchments ranged in size from 35-ha
to 2200-ha and were in addition to the three sewerage catchments mentioned
earlier where investigations of the relations between rainfall and runoff are
carried out. Models of drainage networks for catchments of different size,
shape and density of network were utilized.
The research also included the modeling of theoretical sewerage and
drainage schemes. The results proved that storage of rainwater in the network,
with subsequent conveyance to sewage treatment works after rain has stopped, is
the most appropriate method for protection of receiving waters against pollution,
both for combined sewerage and for separate storm sewerage.(•"'
224
Increasing the level of storrawater flows, in connection with the
reconstruction of interceptor schemes which transport storm or combined waters to
sewage treatment works, is a solution with no essential influence on the standard
of land drainage.
Judicious location of balancing tanks in a network can greatly reduce
the load on such tanks.
Research on the Influence of Short Undetermined Rainfalls on the Quality of
Receiving Waters
An attempt was made to estimate the influence of short undetermined
rainfalls on receiving waters, using theoretical model studies of changes in the
oxygen content in a river resulting from biochemical decay of pollution drained
into the river.(*•£*) For study purposes, an appropriate model of occurrences was
programmed on a computer in order to estimate the content of oxygen and BOD,
varying over time, in particular sections of the river along its banks. Included
was the estimation of the influence of initial conditions such as: loads
entering receiving waters from runoff and rainfalls; distribution of the load in
receiving waters over time and distance; and the influence of decay on the
quality of waters during a decline in pollution load. The model, currently
under development, is of the dynamic type.
It is assumed that the result of work carried out in this field will
be the elaboration of a method for estimation of the influence of rainfalls
on receiving waters and identification of the associated criteria required.
A report presenting all research accomplishments for the years 1972-
1976 within the Research Program of Urban Sewerage and Drainage Schemes is
expected to be issued at the end of 1976. It is assumed that the Program will
be continued in the following years.
225
References
1. Collective work of the Water Supply and Sewage Department at The Engineering
College in Warsaw. The influence of storm sewers in the Warsaw sewerage
scheme on the Vistula River. Warsaw, 1965.
2. Blaszczyk, P., Nowakowska-Blaszczyk, A., Roman, M. The influence of storm
sewers on sanitary conditions of receiving waters. Edited by PZiTS, Warsaw,
1968.
3. Blaszczyk, P. Problems of disposal of rain waters from the centers of towns.
Edited by Town-Plánning Committee, PZiTS, Zakopane, 1967.
4. Blaszczyk, P., Fidala-Szope, M,, Sawicka-Siarkiewicz, H. The method of
analysis of the performance of an existing sewerage network. Edited by
PZiTS, Wroclaw, 1971.
5. Budziszewska, E. From studies on the distribution of atmospheric falls in
the area of Greater Warsaw. Part I, Geophysical Magazine, V.11/1966, No. 4.
Part II - as above, V.12/1967, No. 3 and 4.
6. Lorenc, H. Normal storm falls in the area of Greater Warsaw and the
probability of their occurrence. PIHM, Warsaw, 1971.
7. Nowakowska-Blaszczyk, A., Blaszczyk, P. Problems of disposal of rain
waters in view of the progress in the country's urbanization. Gas, water
and sanitary technique, No.5/1969.
8. Milaszewski, R., Osuch-Pajdzinska, E. Characteristic of pollution of
sewers resulting from rains. Gas, water and sanitary technique, No. 4/1972.
9. Blaszczyk, P. The influence of sewage schemes on the pollution of receiving
waters. Exchange of experiences in the field of outflow and outfall
technique. Stuttgart University, Seminar, 1975.
10. Fidala-Szope, M. Pollution of drains from rain sewerage. The works of the
Institute of Environmental Development, Warsaw, 1976.
226
11. Sawicka-Siarkiewicz, H. Quantitative characteristics of falls from combined
sewerage by storm outfall. Ibid.
12. Sawicka-Siarkiewicz, H. Pollution of drains from combined sewerage. Ibid.
13. Osuch-Pajdzinska, E. The analysis of possibilities to reduce pollution
falls contained in rain sewers drained off by rain and combined sewerage.
Dissertation - Engineering College in Warsaw.
14. Osmulaka-Mróz, B. The influence of rainfalls from urban sewerage on surface
water. The works of the Institute of Environmental Development, Warsaw,
1976.
227
Urban hyetiological modelling and catchment
research in Sweden
Gunnar Lindh
Department of Water Resources Engineering
Lund Institute of Technology
University of Lund
Fack 725, S-220, 07, Lund, Sweden
Background
Within the framework of IHD activities involving studies of representative and
experimental basins, two Swedish basins were chosen to permit the inclusion of research
on effects of urbanization, namely the Verka and Värpinge basins, described in Section 2.
These studies have been supported by funds from the Swedish National Science Research
Council. Urban hydrology has been identified as an important field of research in a 1974
special report on the long-term program for hydrological research in Sweden.
In 1972 the Swedish Council for Building Research initiated the formation of a
special Program Group for the purpose of promoting research and development in the
urban hydrological domain. The goal of the Program Group is to "increase the
knowledge in the geohydrological domain with special respect to an economical and
environmental optimization of construction activities and, moreover, to transfer
this knowledge to decision makers of the construction process."'^.) Especially
recognized is the need for knowledge and methods to minimize disturbances in the
natural water balance, avoid deleterious effects of groundwater level variation
and more effectively manage urban stormwater runoff. Represented in the Program
Group are technical universities, city authorities, consulting firms and the Swedish
Council for Building Research. The Council mainly supports research projects that
will yield results useful in engineering practice, which in many ways is opposite
to the emphasis of the Natural Science Research Council. These divergent views
of project relevance in terms of expected results have caused problems, especially
wtien difficulties have arisen in matching goals of specific projects with the
231
concerns of one or the other of these two research councils.
A recent national report,

(3)
prepared by Chalmers University of Technology
and Lund Institute of Technology for the Swedish Council for Building Research,
with the help of a national review panel, had as its objectives: the description
of problems in the management of water in urban areas and estimation of costs for
their alleviation; and the specification of a continuation in urban hydrological
research in the context of a program meeting both immediate and long-term needs.
The remainder of this Section is devoted to a summarization of certain findings
from that report.
Water Budget in Urban Areas
Water budget models can be formulated with respect to both an outer system
(the "natural" cycle encompassing urban areas) and an inner system (the conveyance
and distribution of water for uses within urban areas).
Discussed here first is an outer system water balance for urban areas in
Sweden from the perspective of national annual totals. The total national urbanized
area is 4024-knr (as of 1970), having an estimated average annual precipitation of
700-mm. Evapotranspiration is estimated at 450-ram per year from permeable surfaces
and evaporation is ignored for impermeable surfaces. Of the groundwater leaking
into sewer systems, one-third is assumed to enter stormwater systems and the other
two-thirds the wastewater systems. Also assumed is that no surface runoff is
generated from permeable areas and that precipitation falling on them enters the
ground and is mostly conveyed via sewer systems and eventually reaches receiving
water bodies. Paved and other impervious surfaces are taken as 30 per cent of the
total urbanized area. The resulting general urban area water budget for Sweden is
shown in Figure 1, covering the outer system or "natural" cycle. The volumes
indicated (in millions of cubic meters per year) should be regarded as rough
approximations.
232
EVAPOTRANSPIRATION
1270
IMPERMEABLE PERMEABLE
SURFACES SURFACES
VITOOJ.
VIA
COMBINED (Eo_ J==( GROUNDWATER,
HOUSE DRAINS, ETC.
GROUNDWATER
SEWERS
VIA STORM SEWERS 7 8 o \ WATf üg' N G
FIGURE 1 - GENERAL URBAN AREA WATER BUDGET FOR SWEDEN,

OUTER SYSTEM. (39)
920
HOUSEHOLDS
>
WATER Vwîfps-
W
SUPPLY 945>WORKS
*J^° 170.
SOURCES - > INDUSTRIES
110 *> PUBLIC USES

107
H 800
^ GROUNDWATER
m ^"^
VIA
/ 340 COMBINED
SEWERS
USE IN W A T E R W O R K S 38
I a»
COMBINED
TREATMENT SEWER
PLANTS
OVERFLOWS
35
RECEIVING WATERS
FIGURE 2- GENERAL URBAN AREA WATER BUDGET FOR SWEDEN,

INNER SYSTEM. ( 3 9 )
233
Based on data from the Swedish Water and Sewerage Association, a general urban
area water budget for Sweden has also been estimated for the inner system, for the
conveyance and distribution of water for uses within urban areas, shown in Figure 2.
As for Figure 1, the volumes shown for Figure 2 are in millions of cubic meters per
year, and should be regarded as rough approximations. Of the total length of
stormwater and combined sewers, it has been estimated that 30 per cent is combined,
partly based on the assumption that about two per cent of the average annual
wastewater volume circumvents treatment at sewage purification plants via combined
sewer overflows entering receiving waters.
The total urban water budget can be described as the sum of the outer and inner
systems, Figure 1 plus Figure 2, where the sole connections would be "via combined
sewers" (340xl0°-nr/year) and "receiving waters".
Pollution Load due to Urbanization
The theoretical pollution load of effluents from local wastewater treatment
plants in urban areas (including connected small industries) imposed on the
watercourses and coastal waters of the country in January 1975 amounted to 33000-tons
of BOD7 per annum, 3000-tons phosphorous per annum and 17000-tons of nitrogen per
annum. These estimates are based on characteristic amounts of pollution of 70-g of
BOD7, 3.2-g of phosphorous and 11-g of nitrogen per person per day; and the
theoretical purification capacity of treatment plants with regard to each of these
constituents.
Using average values of 80-g of B0D7/m3, 3-g of P/m3 and 10-g of N/m3 multiplied
by the total combined sewer overflow volume of 35x10 -nr in Figure 2, the following
values of pollution load are obtained: 2800-tons of BOD^/year, 105-tons of P/year
and 350-tons of N/year. Corresponding rough estimates for stormwater pollution,
assuming an average pollution content of 15-g of BODy/m , 0.1-g of P/nr and 1.5-g
234
of N/nr* applied to a total water volume of 840x10 -m3 per annum are: 12600-tons of
BOD7 per annum, 84-tons of phosphorous per annum, and 1260-tons of nitrogen per
annum.
The preceding pollution load estimates should not be regarded as highly
reliable because of uncertainty on all components, particularly for combined sewer
overflows and stormwater discharges. Regardless, these estimates do give an
indication of the relative magnitude of loadings. In summary, the estimated total
pollution loads on the receiving waters of Sweden, in tons per annum, are:
Pollutant Wastewater Combined Sewer Overflows Stormwater Discharges Sums

BOD7 33000 3000 12006 48000
P 3000 100 100 3200
N 17000 400 1200 18600
Community Costs for Water Management in Urban Areas
Total operating and financing costs for urban water management were estimated
at 2000-million Swedish Crowns for 1974, that is, 5.5-million Crowns per day. The
figures below, in million Swedish Crowns per year, provide more details:
Inves tment
Cost of Interest in New
Operation Costs Construction
Water works | \ 160

> 400 \ 400
Water pipes ) J 330
Sewage purification plants \ } 480

\ 600 ? 600
Sewage pipes J J 620
To these costs should be added those from damages to buildings, etc., caused by
subsidence, which have not yet been evaluated.
Research Aims
The general aim of national research efforts is to increase knowledge on urban
hydrological matters concerning the water balance in urban areas in order to improve
235
methods for water management in urban areas so that natural resources — water, land
and vegetation — are used more constructively in urban planning. Accordingly,
three aims for research can be formulated, as follows —
Increased knowledge of the water budget of urban areas. In order to reach this
goal a comprehension of the hydrological processes in urban areas is required. This
is not the case only for the quantitative part but also for the qualitative part of
hydrological processes. Knowledge must be acquired so that the water cycle within
urban areas can be described quantitatively and qualitatively.
Improved methods of water management within urban areas. Water is needed for
a variety of uses and after use it is conducted via wastewater sewers to sewage
treatment plants and thence to receiving waters. The transport of precipitation
water (storrawater) within urban areas causes problems concerning effects on
receiving waters and interruptions in treatment plant operation. Different methods
are needed, that will provide possibilities for improving water management from a
construction-technical as well as an environmental point of view, methods such as
flow modulation storage, percolation storage, etc. Additional knowledge is required
also in connection with the maintenance and clearing of pipelines, and improvements
in the purification of water and the operation of wastewater treatment plants.
Use of natural resources constructively in planning. Water, land and vegetation
within urban areas and within future urban areas constitute resources which must be
judiciously used.^' Areas suitable for treating storrawater or vegetation needing a
lot of water are natural resources which, through suitable planning of construction,
can be used in other ways than merely as foundations. Water uses must also be
considered during construction planning, so the protection of resources is not
jeopardized.
236
Organization of Urban Hydrological Research
Urban hydrology is a borderland between scientific and applied technical research
to which, in addition to the previously mentioned Research Councils, various authorities
and organizations must contribute. A co-ordinating group for the research has been
proposed.(3) Efforts are currently being made to form this group (September, 1976).
The members of the group will represent research councils, consultants, organizations
and authorities as well as universities.
References
1. Future Swedish Hydrological Research (in Swedish with summary in English) -

Swedish Natural Science Research Council, Stockholm 1974.
2. Status report Nr 4 (in Swedish) - Program group for geohydrological research,
Swedish Council for Building Research, Gothenburg 1974.
3. Carlsson, L., and Falk, J.: "Urban hydrology — review of research needs" (in
Swedish, to be republished in English), Division of Water Supply and Sewerage,
Chalmers University of Technology, Publication B 76/6, and Department of Water
Resources Engineering, University of Lund, Bulletin No. 56, Series A, Gothenburg
and Lund, 1976.
4. Ahlgren, N. et al.: Water in Human Settlements — A Contribution to the United

Nations Conference on Human Settlements, Ministry of Housing and Physical
Planning, Stockholm, 80 pp., 1976.
237
Section 2 Urban catchment research in Sweden
Royal Institute of Technology, Stockholm
The representative basin "Verka11, noted in Section 1, is situated about 35-km
north of Stockholm, Figure 3, and covers some 116-km^ of which 447» is forest and
52% is open country.'1' The countryside is flat and ranges from 3-m to 70-m above
sea-level. The area is drained by the Verka river which connects Lake Fysingen
and Lake Mälaren. The basin consists of two main parts, one part west of Lake
Fysingen which is partly affected by urbanization and the other part east of the
lake which part has not been developed. Within the eastern part a small area, the
Markim Valley to the northeast, has been chosen for intensive study of certain
factors (soil moisture, groundwater, chemistry, etc.). Besides this there are
20-subareas, at the outlet of each of which are stations for measuring runoff. These
subareas have specific characteristics with respect to size, shape, position,
topography, geology, land use and so on.
The Verka basin is very well equipped with observation instruments with which
a series of records have been obtained of runoff, precipitation, groundwater
fluctuations, soil water content, water quality, etc. Studies at the Verka catchment
were started in 1965. These observations make the Verka basin very suitable as a
reference area for urban hydrological studies. The influence of urbanization on
hydrological processes is to be studied by comparing the Verka subbasins with a
number of urbanized nearby basins. Potential areas are in the adjacent basin of the
Märsta river (Figure 3). This basin can be divided into five subbasins, comparable
in size and natural hydrological characteristics to subbasins of the Verka basin
and subject to different degrees of urbanization. In the Märsta basin are located
238
FIGURE 3 - URBANIZED AREAS NEAR THE VERKA BASIN.
239
a highway, the Arlanda airport and an old community center. The airport, which
' serves Stockholm, is at present subject to enlargement and a new center is to be
constructed in the near future.
Lund Institute of Technology/University of Lund
At the Department of Water Resources Engineering, different scales of urban
catchment studies have been undertaken. A small project involved the use of a rain
simulator in the Hydraulic Laboratory.'2' In a second version of the simulator^'
the runoff area was 6-ra x 4-ra. We are aware of the fact that such rain simulator
devices have been very much critized. But remembering that they only represent
themselves and that generalizations are difficult to make they indeed may be used
to deliver data for a test of different theoretical approaches aiming at the
calculation of the runoff process.
Next, there has been a study of the runoff process at a small parking lot.
(4)
This study reveals that applying a simple linear model taking into account the
relationship between runoff and surface storage does not result in acceptable
prognoses. Instead, a non-linear model seems to be sufficiently adequate. A model
based on the kinematic wave theory may give an acceptable prognosis for the runoff
hydrograph. However it was observed that there was an interdependancy between
Manning's coefficient and rain intensity, which relationship ought to be studied in
more detail. The runoff study necessitated the design of a special apparatus for
recording runoff quantity in a gutter inlet, Figure 4.
In order to study the variation of soil moisture content, special studies have
been made at two experimental 100-m2 runoff plats.(*) The two areas have been
isolated from the surrounding soil, Figure 5. Runoff and soil moisture variations
have been observed. After one year of observation the two areas have been partly
240
wave height from
meter raingage
control unit
with clock
tape
puncher
FIGURE 4 - APPARATUS FOR RUNOFF MEASUREMENTS IN A

GUTTER INLET.
241
8-m 8-m
'/Vrí///,
A ^
ri
o
rain
gauge
JHî. •5
E
i
CM
•1
gauge
(• = soil moisture tubes)
FIGURE 5- MAP OF RUNOFF PLATS.
242
"urbanized" by covering them with impermeable sheets in such a way that at one of
the areas the upper part was covered whereas at the other the lower part was covered
by the sheets, Figure 5. What could be observed was that percolation increased by
300% when the area observed was lying downstream from a covered area. The soil water
content in the permeable part downstream from the upper covered part increased to
field capacity at the border zone between the covered and non-covered areas. A few
meters downstream from this zone the effect could not be observed. After the
areas were covered the soil moisture content seemed to be constant. The conclusion
from this experiment was that it is possible to infiltrate the runoff from the
upper impervious part, in an effective balance between évapotranspiration and
infiltration. There are three other experimental areas where urbanization is
being studied. One is the experimental basin "V'árpinge"'6' belonging to the IHD
series of representative and experimental basins in Sweden, as noted in Section 1.
Another is the experimental field at "Sturup"'''' which is the main airport in South
Sweden, The Värpinge basin is about 3-km , is equipped with a meteorological
station and was chosen because it is expected that it will be developed for urban
use later on. However, this plan was postponed, mainly because of strong public
opposition against its implementation. The area of the Sturup basin is about
40-km and the investigation of this basin is especially concerned with soil water
and groundwater variations resulting from the pavement of distinct parts of the
area as required by airport traffic. The report of this study will be available
soon.
As a complement to the rural basin Värpinge, which has not yet been urbanized,
a catchment in Klostergarden in the southwestern part of Lund was chosen for study.
The purpose of this investigation is to compare the parameters of the respective
water balances for the rural and urban catchments and to study the urban runoff
243
process. The Klostergarden catchment is dominated by six to eight story dwellings.
The catchment area is 13.9-ha, of which 48% consists of impermeable surfaces such as
roads, parking-lots and roofs, Figure 6. Field measurements were started in June 1976.
Rainfall. For the recording of rainfall intensity, Figure 6, an instrument
which operates on the "tipping bucket" principle has been constructed. In order to
assure good measurement accuracy, the raingage has been equipped with a 3000-cra^
catchment surface. The volume of water resulting in a tipping is dependent on the
rainfall intensity. Thus conversion from the recorded number of tippings to
intensity must be made via a calibration curve. On the average, for the three gauges
involved, each tipping is equal to 0.02-mm of rain.
Runoff. Runoff is measured at the outlet of the basin and at four impermeable
subareas within the basin, Figure 6. The gauges are similar to that in Figure 4.
A vertical pipe filled with an electrolyte (0.5% NaCl) is connected to the water
in an inlet; and for the catchment gauge is connected to the storm drain through
a rubber membrane. Two parallel platinum wires are mounted in the vertical pipe.
By measuring the electrical conductivity of the wires in the electrolyte the water
level can be obtained. Because the pipes are placed upstream from V-notch weirs,
the runoff is determined by a rating curve. To increase the reliability of the
system, a supplemental mechanical recorder has been installed at the catchment
outlet.
244
50 100
i_ _L
Legend
Storm drains larger than 225-mm 4>.
® Raingages
Runoff.gauges for paved subaraas:

2
1 - area approximately 326-ra
2 - area approximately 81-nr
3 - area approximately 306-ra
o
4 - area approximately 277-m
u Runoff gauge for the whole area of 13900-m .
FIGURE 6- KLOSTERGÂRDEN STUDY AREA,
245
Central Recording. In order to transfer signals from the measuring units to
a tape puncher, a special control unit has been developed. This unit consists of
a clock, a digital voltmeter and a pulse counter. The recording (tape punching)
starts automatically when an initial pulse from a raingage enters the control unit.
The time interval between each punching is set at one minute. That is, each minute
the voltage of the runoff gauges and the number of tippings for each raingage are
punched. Termination of recording is determined by the elapsed time since the
last pulse from any raingage was received, the limit being one hour. As all the
instruments are connected to the same clock, Figure 7, there is an absolute
synchronization of recording time for rainfall and runoff. The central unit where
recording takes place is situated in the celler of a building within the catchment.
All cables connected to the gauges are placed in the storm drains.
Water sampling. During the first months of recording, water samples have been
taken during one storm.
In addition to these experimental investigations, two studies have been made
of the literature on urban runoff,(°»9)
Chalmers University of Technology, Gothenburg
As part of the research project "Urban Storm Water Pollution Sources" at the
Department of Water Supply and Sewerage, Chalmers University of Technology, rainfall
and runoff in four catchment areas in Gothenburg are recorded. One of these is
the Bergsjb'n area,'10' located about 15-km east of the city center. Bergsjb'n is
a residential area of 15.4-ha with 3-6 story dwellings, Figure 8. The impermeable
portions are about 407. of the total area and the remaining 60% is occupied by lawns,
forest and other permeable areas. The impervious areas are rather steep, with an
average slope of about 30%. The drainage system is of the separate storm sewer type
with pipe diameters varying from 225-mm in the system to 800-mm at the outlet.
246
g
1
CO
CD
2
>
8
LU
cr Lu
ö
u_
h-
CO
u. >-
o en
o
UJ
o
o
8 J3x3|duinuj
¡3 o
CO Lu
Q
o or
jajaumoA 104161p
z od
> a:
LU LU
O
UJ 1-
or
Lu C/)
LL
V 9) 0) o
Lü
tr
CD
CO
I-
2
LU
s
y y y
v—
CED
LU tO 1/1
0)
?
Lü2 .Eo> i
TS? *- r ai
Us
247
Rainfall and Outlet Runoff. Rainfall and runoff are measured by electronic
gauges and the electric signals are transmitted via telephone lines to a central
recording station at the University, Figure 9. The rainfall gauge is of the
tipping-bucket type with one tip per 0.05-mm of rain, and is located near the center
of the catchment. Catchment flow is measured by an ultrasonic water-level gauge at
a V-notch weir at the outlet pipe.
Central Recording. Rainfall and runoff outputs are presented partly as a
printed page and partly as a paper tape at the central recording station at the
University, Figure 9. The recording frequency is once every hour in dry weather and
once every minute in wet weather. Signals from the four field stations in the
Gothenburg area, one of which is Bergsjb'n, are recorded independently at the central
recording station at the university.
Water Quality Sampling at the Outlet. Composite, flow-proportional water
samples are automatically taken during rainfall. The sampling is entirely controlled
by the flow gauge. It is also possible to take time-proportional samples (up to 30
samples per rainfall).
Runoff from Subareas. An effective study of the runoff process requires
detailed analysis of subcatchments that are approximately uniform from a hydrological
point of view. During one summer, runoff from four different subareas in Bergsjön
was also measured. The subareas were an asphalt paved area, a concrete paved area,
a roof and a composite area, ranging in size from 400-nr to 8000-m . Water flow was
measured by a mechanical/electronic gauge at V-notch weirs. All flow devices were
placed in manholes or catchbasins. The water level signals from all four subareas
were transmitted to a local central data log where all data was recorded on a
magnetic tape. The gauging frequency was once per 30 seconds during rainfall and
once per 10 minutes during dry periods. Examples of the weir installations are
given in Figure 10.
248
FIGURE 8- BERGSJÖN CATCHMENT AREA.
LEVEL FLOW V/F TRANSMIHER

^ -> CONVERTER * * ->
GAUGE TRANSFORMER
LEVEL SAMPLER
~> SWITCH 9>
SCANNER
RECEIVER F/V DIGITAL TELETYPE

• ^
CONVERTER
^ VOLTMETER -£
^¿_
TIMER
FIGURE 9- DATA TRANSMITTING SYSTEM.
249
(a). Manhole (b). Catchbasin
FIGURE 10- FLOW-MEASURING INSTALLATIONS.
The subárea flow gauging was undertaken by the Division of Hydraulics. Runoff
hydrographs for two short, intensive storms are given in Figure 11.
Subárea Water Quality Sampling. Water sampling was done for the roof area.
The sampling was time-proportional and 12 samples could be taken per rainfall. The
sampler started automatically when the flow increased.
The Division of Hydraulics has also performed some studies of urban runoff in
the city of Linköping in collaboration with city authorities. Moreover, some
studies are being carried out at the city of Halmstad.
A characteristic feature of contemporary urban runoff studies is the increasing
consciousness of the importance of the water quality composition of urban storm
runoff. The Division of Water Supply and Sewerage at the Chalmers University of
Technology has made a series of studies of this problem. One of the most important
contributions was made in 1974,' ' and fundamental studies continue to be made.^ »"'
Investigations have centered on the Bergsjb'n area, described above, and on an
industrial area in Gothenburg. In each of these areas, field stations for measuring
rainfall and runoff and for water quality sampling were installed. Rainfall and
250
Discharge
(l/s -ha
4
RAIN 25.09.1974
Concrete surface
Roof surface
Composite area
^•Time
(min)
Discharge
(l/sha
30.09.1974
Asphalt surface
— — — Roof surface
— — — Composite area
—~* Concrete surface
Time
(min)
FIGURE 11- HYDROGRAPHS FOR BERGSJÖN SUBAREAS,
251
runoff were continuously recorded and all water sampling was flow-proportional.
While the investigation is still in progress, it may be mentioned that in 1973 one
sample was taken every sixth or twelfth minute during runoff and in 1974 all samples
were composited. A total of 82 runoff events in the residential area and 35 runoff
events in the industrial area were studied. The physical-chemical analyses carried
out were mainly for lead, copper, zinc, iron, KMnO^ demand, suspended solids, total
solids, residue on ignition, total phosphorous, sulphate, pH and conductivity. A
total of 12500 laboratory analyses were performed during the investigation
described.
From the results obtained it can be mentioned that for the residential area the
first part of the runoff history is the most polluted. Roughly speaking, one may
say that the first third of the runoff volume contains 44% of the total pollution
amount whereas the last third contains only about 23% of the total pollution amount.
This indicates that one certainly has to take care of the first part of the runoff
quality if the polluted water is to be treated or stored. But it should be
emphasized that this statement is not true for the runoff from the industrial area.
It was found in that instance that the first third contained about 30% of the total
pollution amount and the last third about 34%, so the conclusion to be drawn is that
the pollutant is rather equally distributed over time. Furthermore, the observations
from the industrial area show much larger standard deviations.
A study of the seasonal variation of some components has yielded different
results. Pb, Cu and Zn appear in small quantities during spring and early summer
but in big quantities during autumn and winter. For Fe, no seasonal variation has
been observed. As to suspended material, there is no general variation for the
residential area but there is a very evident variation for the industrial area.
Because the investigations are not yet complete, it is difficult to draw any
definite conclusions,
252
The Division of Water Supply and Sewerage has also made an extensive study of
injection through wells in confined aquifers. Experimental studies have been
carried out within an area in the Gothenburg region. In this area, because of
several tunnels constructed in the bed rock, a lowering of the water level in soil
as well as in bed rock has resulted. According to infiltration test results, there
seems to be a very rapid response of the piezometric head within the aquifer. Some
decrease of the already attained piezometric head was observed when the infiltration
test was extended to about 86 days. This tendency of decreasing head with time
during infiltration is supposed to depend on clogging in the infiltration well.
University of Lulea
The University of Luleâ is the youngest of the Technical Universities in Sweden.
The Division of Water Resources Engineering has just started some urban hydrological
research with the goal of considering the special problem encountered within the
northern part of Sweden. There, snowmelt constitutes an essential part of the
hydrological cycle and therefore the influence of snowmelt on the quantity and quality
of urban runoff will be taken into special consideration.
Research Work Conducted Outside the Technical Universities
The initial emphasis in IHP Subproject 7.1, to which this Section is a
Swedish contribution, focuses on studies of urban catchments. Even though true urban
catchment studies are carried out mainly at the technical universities, some consulting
firms are involved in more special urban catchment problems. Examples are full
scale investigations concerning urban storm water infiltration, developing a system
for taking care of urban runoff, and storage as well as infiltration of runoff from
roofs.
253
References
1. Nilsscm, L Y: "Verka representative basin" - Department of Land Improvement and

Drainage, School of Surveying, Royal Institute of Technology, Report 3:1, b-d,
Stockholm 1973.
2. Lindh, G, Niemczynowicz, J and Falk, J: "A physical model of runoff" -

Proceedings of the "Nordisk hydrologisk konferens", Sandefjord, Norway 1972,
3. Gottschalk, L and Niemczynowicz, J: "Surface runoff from impermeable surfaces -

a model study" (in Swedish with short summary in English) - "Kvantitativ urban
hydrologi", Nordiskt symposium, Sarpsborg, Norway 1975.
4. Falk, J and Niemczynowicz, J: "Runoff from impermeable surfaces" — Tenth

Anniversary, Papers on Research in Progress, pp. 66-81, Department of Water
Resources Engineering, University of Lund, Bulletin No. 55, Series A, Lund,
July,, 1976.
5. Falk, J and Niemczynowicz, J: "Effects of urbanization on the water budget -

a runoff-plot study" (in Swedish with short summary in English) - "Kvantitativ
urban hydrologi", Nordiskt symposium, Sarpsborg, Norway 1975.
6. Falkenmark, M (ed): "Hydrological data - Norden", Representative basins -

National committees for the International Hydrological Decade in Denmark,
Finland, Iceland, Norway and Sweden, 1972.
7. de Maré, L: "Water balance inventory of the Sturup airport" (in Swedish) -

Department of Water Resources Engineering, University of Lund, Bulletin Serie
B, No. 29, 1973.
8. Lindh, G and de Maré, L: "Storm-water runoff" (in Swedish) - Department of

Water Resources Engineering, University of Lund, Bulletin Serie A, No. 9,
Lund 1972.
9. Lindh, G: "Methods for calculation of urban storm-water runoff" (in Swedish) -

Department of Water Resources Engineering, University of Lund, Bulletin Serie A,
No. 24, Lund 1973.
10. Lyngfelt, S: "Urban runoff studies in Bergsjcm, Gothenburg" (in Swedish with
short summary in English) - "Kvantitativ urban hydrologi", Nordiskt symposium,
Sarpsborg, Norway 1975.
11. Lisper, P: "On the composition of storm water and its variations" (in Swedish
with summary in English) - Chalmers University of Technology 1974.
12. Malmquist, P-A and Svensson, G: "Urban storm water quality" (in Swedish with
summary in English) - Division of Water Supply and Sewerage, Chalmers University
of Technology, Publication B 75:3, Gb'teborg 1975.
13. Svensson, G: "The composition of storm water, effects of urbanization" (in
Swedish with short summary in English) - "Kvantitativ urban hydrologi", Nordiskt
symposium, Sarpsborg, Norway 1975.
254
Section 3 Urban hydrological modelling in Sweden
Introduction
As noted in the preceding Sections, Swedish activity in urban hydrology field
research is of rather recent date. This is likewise true for research on urban
hydrology mathematical modeling, the theme of this Section. Studies are mainly
going on at the Chalmers University of Technology (Gothenburg) and at the Lund
Institute of Technology (University of Lund). Some work is also carried on by
other research groups. Financial support is in some cases provided by the Swedish
Council for Building Research.
We may of course regard urban storm runoff as a closed problem and analyze it
as a whole. Because of the complexity of the problem this is not an approach
especially well adapted to its purpose. The generally applied method of anaylsis
is to divide the problem into at least three parts. These parts are clearly
distinguishable from a phenomenological point of view. The first part may be
described as an entirely hydrological process, where runoff from pervious or
impervious surfaces is calculated with due regard to the hydrological water balance
equation. The second part of the runoff process may be regarded as starting when
storm runoff enters gutter inlets and the like, a hydraulic process which may be
analyzed by routing inlet hydrographs through the network system conveying water
beneath the ground. Due consideration must be given to storage effects and other
factors. The last part of the runoff process may be considered as the treatment of
polluted stormwater before discharging it to a receiving water body. This points
out the essential problem included in storm water runoff, namely to regard this
process not only as one depending on quantity but in fact still more on quality.
255
When describing the runoff process as one that could be divided into several
parts, there could be added the mathematical-physical picture of the process input
to the runoff history, the precipitation process with its distribution in time and
space. However, when considered, the precipitation process is sometimes included in
the part involving surface runoff to inlets and the like, the first part noted
above.
Chalmers University of Technology
An extensive project supported by the Swedish Council for Building Research
is being conducted by the Division of Hydraulics. The research work is divided into
two parts, one accounting for precipitation, infiltration, surface storage, surface
runoff and gutter flow,^1' and the other analyzing flow in the pipes of the network
system.(^' The first-mentioned part of the model has some resemblance to the
so-called "University of Cincinnati model". Thus the infiltration process is
accounted for by the well known Horton formula. Depression storage is considered
but evaporation and interception are neglected in these urban studies. Surface flow
is calculated via a simple kinematic wave theory by using Manning's formula, and
storage effects are considered whereby an empirical relationship between water depth
and mean storage is used. Gutter flow is calculated starting from the equation of
continuity.
The model has been tested on data from the research area "Bergsjo'n" in
Gothenburg, described in Section 2. In Figure 12, calculated and observed runoff
hydrographs are compared for two different hyetographs. In these calculations a
simplified version of the more accurate model for pipe flow (see below) was used,
assuming that the runoff hydrograph will not change its shape from one gutter inlet
to another.
256
CO CO
se
t-< en
co co
E Ë
m n
ri —
il —
ii m
II II
t- • *
CM CO
S E Ë o o
E Ë Ë Il II
Oí O 05 JZ X¡ x: r:
^H O
*,w^.^,*^. a. n,
en a
OJ
E Ë t-,
a- II Il II
E Ë bo bfi
o n
CO i-H
u u
. , XI • n Q ^
s- to oo >>
£
AN
7 -o
V
>
Q)
í-1 Il II
JZ
X) T )
O
t- 3 ->
eu (D
a
put
D X) T;) >
» *?? Ë ti u (0
COM PUTE
PHS, BER
fc 3 m
°ö
0) a> a h J2
o
U O
5" 01
3
Ë
o °
e
3
r~4
r—1
O
> o
«- O
> t«-i
c!
r. O Q <
• ^H
c A; •Z. (T
a 3 cfl
» • •
PÍ « a) < o
O
m ce
X Q
" ^ ÛL >-
S~ X}TSU3}UT UTBy. <L X
_i—.—: ". ^j
rr
(0 Li_
33JEUSSTQ 8 o Li_
O
v-
iii 4¿
V _)
X cr
n
FALL
RVE
7> LU
(/)
< m
cr o
CM
0J
•
UJ
£T
3
oo O
o
I li-
r—
O
I
ro
, i" ni
aSanuosiQ
257
The routing process in a network system is considered more thoroughly in the
"CTH-Storm Sewer Routing Model DAGVL-A,"(2) an implicit, linear type model. With
this model it is possible to calculate flows and water levels in selected sections
in a converging pipe network. The basic equations used are the Saint-Venant equations.
Flow in a full pipe section is simulated by assuming the existence of a narrow slot
introduced at the upper part of the pipe section, the slot width being a function
of the velocity of sound in the conduit.'-*) The calculations are initiated by
dividing each pipe branch into a series of reaches. For each reach, equations of
continuity and momentum are transformed into a finite form by a box scheme for the
momentum equation and a centered scheme for the continuity equation. The coefficient
matrix which describes the flow condition in the linear system of equations has a
block-diagonal form which means that the system of equations may be solved by using
the so-called "double-sweep method". ^»^•' From the point of view of numerical
stability, the time step for computation may be arbitrarily chosen.
In its present form, the solution does not permit beginning simulation with an
initially dry pipe network. A certain amount of base flow must be assumed. It has
been observed that in the case of a small baseflow in conjunction with a hydraulic
jump, and with large changes of flow at a full- pipe section, strong fluctuations may
occur in calculated flows and water levels. Such fluctuations, which may make the
calculations go wrong, are eliminated by the use of a "dissipative interface".^'
A simplified model DAGVL-B, which assumes free outflow conditions at the
downstream section of each pipe, has also been developed. This model does not take
backwater effects into account but may nevertheless be quite accurate as far as
flow rates are concerned.
Unfortunately, there are very few existing field data at hand which could be
used for comparing theoretical and experimental results. In order to improve on
258
this regretable situation a test reach composed of a plastic tube of 85-m length
and with a diameter of 0.105-m has been installed in the Hydraulic Laboratory.
However, only a few test runs have been made. One of these is shown in Figure 13.
Again with financial support from the same source, the Division of Water Supply
and Sewerage has made some preliminary attempts to develop a model for water quality.
Moreover the Division of Hydraulics has started some work on the mechanism of
material transport in a pipe network with the aim of learning more about wastewater
transport.
Lund Institute of Technology (University of Lund)
The Department of Water Resources Engineering has approached the runoff problem
in another way. The idea was to start from very small areas and then calculate the
total runoff history of larger composite areas by time-dependent integration of
the runoff histories from the small areas.
Before describing the research work that deals directly with runoff, it may be
pointed out that there have been some preliminary studies on mathematical modeling
õf precipitation processes'"»'* with objectives for instance of finding what
distribution functions could be most appropriate. But these studies must also be
seen as a first attempt to consider the problem of what shall be meant by a
"determining rain". The study will be continued on a mathematical-physical basis
by analyzing the effect of two or more consecutive rain events compared to one
single rain event that might have a bigger peak intensity. This study was
financially supported by the Swedish Natural Science Research Council.
Returning to the runoff studies, some fundamental investigations have been
performed using a laboratory model.' » ' This laboratory model includes a rain
simulator mounted over a small area 6-m x 4-m in size. Initial studies showed that
the runoff history could be determined from three parameters, namely: the initial
259
Calculated flows: At «0.2 min
A A X =4.7-5.7 m
* s ix
x
* •
#
A *
â
• X
A • X
_ ^ ^ . A • „ ^ - > ^ _ * A * •
à • _ X
T Tim* in minutes
Calculated and measured waterdepths

At =0.2 min A X =4.7-5.7 m
Time In minutes
Depth Calculated and measured waterdepths

in c m At =0.4 min A x =0.4 m
6 -
4 ••
Time in minutes
M e a s u r e d values : solid line

Calculated values: A x=2&5m (distance from
ï xX=*755m t h e ^ P°int>
Full capacity of the pipe = & . 3 1/s
Diameter of the pipe - 10.5 c m
FIGURE 13- LABORATORY TEST, COMPARISON BETWEEN

MEASURED AND CALCULATED WATER DEPTHS.
260
storage (a characteristic of the surface); the maximum storage (depending only on
the rain intensity, assuming that the duration of the rain is longer than the time
of concentration); and the maximum value of runoff. Some mathematical theories of
runoff were tested and compared with experimental results. Thus, the kinematic wave
theory and linear as well as non-linear reservoir theory were applied. Moreover,
some lumped models were investigated. In Figure 14 theoretical and experimental
runoff hydrographs are shown. The conclusion that can be drawn is that results from
the use of the kinematic wave theory in this case are in good agreement with the
observations. It goes without saying that the results of observations may not
allow generalization due to the special character of the rain simulator.
At the Department of Water Resources Engineering there has also been performed
a study concerned with the runoff from a parking lot,'!") where a new flow-measuring
device was field-tested for the first time (Figure 4). The runoff process was
analyzed by means of a simple linear model as well as a non-linear model. Also,
the kinematic wave theory was again used. Figure 15 shows a comparison between
calculated and observed runoff histories. The non-linear theory and the kinematic
wave theory were used in these examples to calculate the hydrographs. It appears
that the results based on theoretical approaches show rather good agreement with
observed data. A special study was also made in order to analyze mathematically
the relationship between runoff and storage.
STEGA-Group
STEGA is the name of a research group established in 1966. The members of this
group have focused their interest on "ground water research in urban areas with
practical application". The group has presented a computer model for calculation
of the ground water balance.(I1/ The work is carried out with financial support
from the Swedish Council for Building Research. The model considers the flow
261
QI5S
Concrete Surface.
Observed Hydrographs ( )
Compared with Calculated (- -),
Kinematic Wave Theory.
Painted Concrete Surface.

Observed Hydrograph ( )
Compared with Calculated ( ),
Linear Theory.
Tme.min.
Painted Concrete Surface.

Observed Hydrograph ( )
Compared with Calculated ( ),
Non-Linear Theory.
Trot min.
FIGURE 14- RAIN SIMULATOR HYDROGRAPHS.

E
E
1
0
E
a
FIGURE 15- PARKING LOT HYETOGRAPHS AND HYDROGRAPHS.
263
between several groundwater reservoirs. In the model those quantities of water which
move during a certain time interval between different parts of the system are
estimated. Such water transfers cause water level changes in the groundwater
reservoirs. Pressure differences between reservoirs and the hydraulic resistance
between the reservoirs then determine the flow process during the time interval
considered next. As a result, the model indicates the groundwater levels in the
different reservoirs. The model can be used for prognoses. Thus the effects of land
development may be foreseen by this model, the most apparent instance being the
lowering of the groundwater level.
Concluding Remarks
The research described above pictures a very intensive period of urban
mathematical work. However, the financial support for this research is rather modest.
There is a need for more experimental data by means of which the usefulness of
theoretical models could be tested. Without a sufficient amount of data it is
impossible to determine the different parameters which constitute important
ingredients of such models. It is also not possible to test models on independent
data, which could make it possible to compare the efficiencies of different model
approaches.
References
1. Arnell, V: "Runoff Model for the Analysis of Storm Water Runoff in an Urban
Watershed". (In Swedish with short English summary). - "Kvantitativ urban
hydrologi", Nordic Symposium in Sarpsborg, Norway 1975.
2. Sjo'berg, A: "CTH - Storm Sewer Routing Model DAGVL-A". (In Swedish with short
English summary). - "Kvantitativ urban hydrologi", Nordic Symposium in
Sarpsborg, Norway 1975.
3. Cunge, R J A and Wegener, M: "Integration numérique des équations d'écoulement

de Barré de Saint-Venant par un schéma implicite de differences finies". -
La Houille Blanche, 1964:1.
264
4. Abbot, M B and Ionescu, F: "On the Numerical Computation of Nearly Horizontal
Flows". - Journ. of Hydraulic Research, 1967:5.
5. Abbot, M: "Continuous Flows, Discontinuous Flows and Numerical Analysis"» -

Journ. of Hydraulic Research, Vol. 12, No. 4, 1974.
6. Gottschalk, L and de ¡Mare, L: "Analysis of Precipitation Observations" (in

Swedish). - Department of Water Resources Engineering, Bulletin Series A No. 21,
Lund 1973.
7. Gottschalk, L and Falk, J: "A Stochastic Approach to Analysis and Generation

of Precipitation Series" - a paper presented at the Nordic Hydrologie Conference
in Reykjavik, 1976.
8.. Gottschalk, L and Niemczynowicz, J: "Runoff from Impervious Surfaces" (in

Swedish). - Department of Water Resources Engineering, Bulletin Series A No. 27,
Lund 1974.
9. Gottschalk, L and Niemczynowicz, J: "Surface Runoff from Impermeable Surfaces"

(in Swedish with short English summary). - "Kvantitativ urban hydrologi",
Nordic Symposium in Sarpsborg, Norway 1975.
10. Falk, J and Niemczynowicz, J: "Runoff from Impermeable Surfaces" - Tenth

Anniversary Papers on Research in Progress, pp. 66-81, Department of Water
Resources Engineering, University of Lund, Bulletin No. 55, Series A, Lund,
July, 1976.
11. Lindskoug, N-E and Nilsson, L-Y: "Ground Water and Urban Planning, Report
from STEGA 1966-73" (in Swedish with English summary). - Statens Institut for
Byggnadsforskning, Report R 20:1974, Stockholm.
265
[A.] S C . 77/XXI.16/A
ISBN 92-3-101555-9
In this series
1 Perennial ice and snow masses. A guide for compilation and assemblage
of data for a world inventory.
2 Seasonal snow cover. A guide for measurement, compilation and
assemblage of data.
3 Variations of existing glaciers. A guide to international practices for
their measurement.
4 Antarctic glaciology in the International Hydrological Decade.
5 Combined heat, ice and water balances at selected glacier basins. A
guide for compilation and assemblage of data for glacier mass balance
measurements.
6 Texbooks on hydrology—analyses and synoptic tables of contents of
selected textbooks.
7 Scientific framework of world water balance.
8 Flood studies—an international guide for collection and processing of
data.
9 Guide to world inventory of sea, lake, and river ice.
10 Curricula and syllabi in hydrology.
11 Teaching aids in hydrology.
12 Ecology of water weeds in the neotropics.
13 T h e teaching of hydrology.
14 Legends for geohydrochemical maps.
15 Research on urban hydrology, vol. 1.
16 Research on urban hydrology, vol. 2.
17 Hydrological problems arising from the development of energy.
18 Urban hydrological modelling and catchment research, international
summary.
19 Remote sensing of snow and ice.
20 Predicting effects of power plant once-through cooling on aquatic
systems.
21 Research on urban hydrology Vol. 3
International Hydrological
Programme
Research
on urban hydrology
Volume 3
1979 follow-up reports from
eleven countries
Proceedings of two special

sessions, Spring Annual Meeting,
American Geophysical Union,
Washington, D . C . , 28 M a y 1979
Co-sponsored by the A G U
Committee on Urban Hydrology,
the United States Committee on Scientific
Hydrology and Unesco
General editor: M . B . McPherson
The Unesco Press

The designations employed and the presentation of the material
do not imply the expression of any opinion whatsoever on the
part of Unesco concerning the legal status of any country of
territory, or of its authorities, or concerning the frontiers of
any country of territory.

Organization
Printed by Imprimerie J. Floch, Mayenne
I S B N 92-3-101984-8
© Unesco 1981
Printed in France
Preface
Although the total amount of water on earth is generally assumed to

have remained virtually constant, the rapid growth of population,
together with the extension of irrigated agriculture and industrial
development, are stressing the quantity and quality aspects of the
natural system. Because of the increasing problems, man has begun to
realize that he can no longer follow a "use and discard" philosophy
— either with water resources or any other natural resource. As a
result, the need for a consistent policy of rational management of
water resources has become evident.
Rational water management, however, should be founded upon a
thorough understanding of water availability and movement. Thus, as
a contribution to the solution of the world's water problems, Unesco,
in 1965, began the first world-wide programme of studies of the hydro-
logical cycle — the International Hydrological Decade (IHD). The
research programme was complemented by a major effort in the field of
hydrological education;and training. The activities undertaken
during the Decade proved to be of great interest and value to Member
States. By the end of that period, a majority of Unesco's Member
States had formed IHD National Committees to carry out relevant
national activities and to participate in regional and international
co-operation within the IHD programme.. The knowledge of the world's
water resources had substantially improved. Hydrology became widely
recognized as an independent professional option and facilities for
the training of hydrologists had been developed.
Conscious of the need to expand upon the efforts initiated during
the International Hydrological Decade and, following the recommen-
dations of Member States; Unesco, in. 1975, launched a new long-term
intergovernmental programme, the International Hydrological Programme
(IHP), to follow the Decade.
Although the IHP is basically a scientific and educational
programme, Unesco has been aware from the beginning of a need to
direct its activities toward the practical solutions of the world's
very real water resources problems. Accordingly, and in line with
the recommendations of the 1977 United Nations Water Conference, the
objectives of the International Hydrological Programme have been
gradually expanded in order to cover not only hydrological processes
considered in interrelationship with the environment and human activities,
but also the scientific aspects of multi-purpose utilization and
conservation of water resources to meet the needs of economic and
social development. Thus, while maintaining IHP's scientific concept,
the objectives have shifted perceptibly towards a multidisciplinary
approach to the assessment, planning, and rational management of water
resources.
As part of Unesco's contribution to the objectives of the IHP,
two publication series are issued: "Studies and Reports in Hydrology"
and "Technical Papers in Hydrology". In addition to these publications,
and in order to expedite exchange of information in the areas in which
it is most needed, works of a preliminary nature are issued in the form
of Technical Documents.
The "Technical Papers in Hydrology" series, to which this volume
belongs, is intended to provide a means for the exchange of information
on hydrological techniques and for the coordination of research and data
collection. Unesco uses this series as a means of bringing together and
making known the experience accumulated by hydrologists throughout the
world.
Contents
SECTION 1 - INTERNATIONAL SYMPOSIUM 11

Introduction 11
Background 11
International Symposium on Urban Hydrology 12
Recommendation to Unesco 12
Some General Observations 12
Acknowledgments . . 13
References 13
SECTION 2 - PROGRESS SINCE 1976 IN AUSTRALIA ' . . 15
Introduction ; IS
Research Programs 15
Applications 20
Conclusion 22
References . 22
SECTION 3 - PROGRESS SINCE 1976 IN CANADA 25
Introduction 25
Urban Catchment Research 25
Hydrological Modeling 26
Urban Runoff Controls 27
Implementation of Research Findings in Practice 28
Summary and Conclusions 31
References 31
SECTION 4 - PROGRESS SINCE 1976 IN THE UNITED KINGDOM 35
Introduction 35
Recent Urban Catchment Research 35
Recent Urban Hydrological Modeling 37
A Manual for Urban Storm Sewerage Design and Analysis 38
Table 1 - The Wallingford Hydrograph Method - Design Use ... 41
Figure 1 - 3x3 Matrix of Standard Runoff Hydrographs 44
Acknowledgments 45
References 45
SECTION 5 - PROGRESS SINCE 1976 IN THE FEDERAL REPUBLIC OF GERMANY ... 49
Introduction 49
Hydrological Investigations in Urban Drainage Areas 50
Models in Urban Hydrology \ 55
Figure 1 - Comparison of Computed Peak Discharges
for Catchment I . . 56
Summary and Recommendations . . . 61
SECTION 6 - PROGRESS SINCE 1976 IN SWEDEN 65
Introduction ' . . 65
Figure 1 - Research Administration 66
Urban Catchment Research in Swoden . 67
Figure 2 - Regional Distribution of z Values
in Southern Sweden 68
Figure 3 - Data Collection System for
the Raingauge Network in Lund 68
Figure 4 - Systemizing Detention Storage According to
Location in Sewage System and Examples of
Technical Performance . . . 71
Figure 5 - General Urban Area Water Budget for Sweden,
Inner and Outer Systems 72
Urban Hydrological Modeling in Sweden . . .73
Application of Research Results 74
References 74
SECTION 7 - PROGRESS SINCE 1976 IN FRANCE 79
Introduction 79
Urban Catchment Runoff Quantity Research . 79
Table 1 - Characteristics of Experimental Watersheds 80
Urban Catchment Runoff Quality Research . . 81
Runoff Quantity Modeling Research . 81
Runoff Quality Modeling Research 82
Practical Applications 82
Conclusions 82
References 83
SECTION 8 - PROGRESS SINCE 1976 IN NORWAY 87
Introduction 87
Urban Hydrological Modeling in Norway 90
Figure 1 - Unit Operations in Sludge Model . . . . . . 92
Figure 2 - Simulated Hydrographs at the Outlet
from a Subcatchment 93
from a Subcatchment ... 93
from the Whole Catchment 95
from the Whole Catchment 96
Abbreviations Used in This Paper . . . . . 97
Address List for Norwegian Institutions Referred to in This Paper . . 97
References .' 97
SECTION 9 - PROGRESS SINCE 1977 IN THE NETHERLANDS . . 99
Introduction 99
Figure 1 - Water Quality Systems Modeled, Lelystad 101
Figure 2 - Calculated Water Level Rises 105
Future Elaboration 109
References 109
SECTION 10 - PROGRESS SINGE 1976 IN POLAND
Introduction Ill
Supplements and Changes in the Formulas and Calculation
Algorithms Used in Poland in the Rational Method Ill
Method of Simulating Hydrographs of Runoff from
Rainwater Sewer Schemes Based on the "Black Box" Model 113
Figure 1 - Hydrograph Simulation Based on
the "Black Box" Model 114
Method of Simulating Pollutographs of Urban Runoff 113
Table 1 - Parameters for the F o r m u l a ^ = A Q Ç ,
Pollution Indices and General Suspensions 115
Figure 2 - Pollutograph Simulation 116
Principles of Forecasting Runoff Pollution to Meet
Requirements for Planning New Drainage Schemes 115
Research on the Influence of Urbanization on
Changes in Catchment Runoff Parameters 117
Models of Runoff from Urbanized Catchments . . . . 117
Models of Pollutant Propagation in a River 117
Figure 3 - Temporal and Spatial Distribution of BOD5 118
Figure 4 - Dissolved Oxygen in the River as
Affected by Stormwater Discharge 118
References . 119
SECTION 11 - PROGRESS SINCE 1977 IN INDIA 121
Editor's Note 121
Some New Developments 121
A Developing Nation Perspective . 121
The Institute of Hydrology 122
References . 123
SECTION 12 - PROGRESS SINCE 1975 IN THE U.S.A 125
Introduction 125
Areawide Planning 125
Catchment Research 126
r
Figure 1 - Framework for Data Collection . 127
Figure 2 - Typical Installation of the Urban
Hydrology Monitoring System 128
Catchment Modeling . . . . . 126
Urbanization Effects 129
Conclusions 130
References 130
SECTION 13 - SUMMARY OF SYMPOSIUM DISCUSSION 137
Section 1 International symposium
by M . B. McPherson*
Introduction
Because the complex interactions of human activity in concentrated

settlements with air, water and land must be taken into account, urban hydrology
is a distinctive branch of the broad field of hydrology. As opposed to conventional
hydrology, because urban development everywhere has been in continuous states of
expansion and flux, urban hydrology contends with the dimension of dynamic change.
Also, urban water management utilizes the social and biological sciences as well
as the physical sciences. The fact that the term urban hydrology gained currency
less than two decades ago reflects the lag in the recognition that America became
an urban nation over half a century ago. Since then, the term has been tacitly
expanded to include all urban water resource matters in, or interfacing with,
the hydrologie cycle, including water quality considerations. These remarks,
although addressed to a U.S. audience,^1) appear to be applicable by and large
to just about all nations around the world.
Background
Under the International Hydrological Programme (IHP), twelve national

reports were assembled on urban hydrological modeling and catchment research:
U.S.A., Australia, Canada, the United Kingdom and the U.S.S.R.¡(2) and the
Federal Republic of Germany, Sweden, France, Norway, the Netherlands, Poland and
India.(3) Urban drainage was featured. In an international summary of these
twelve reports,
(4,5)
three fundamental research objectives were identified with
regard to urban runoff: determination of the hydrological effects of urbanization;
development of measures that would offset the adverse effects and enhance the
assets of urban runoff; and resolution of improved tools of analysis for the
planning, design and operation of urban drainage systems.
Development of urban runoff tools of analysis was regarded as having
gone beyond the field-measurement base that supports their validity. Recognized
was a need everywhere for more field observations from representative and
experimental sewered catchments to improve the reliability of such tools of
analysis. Emphasized was an urgent need for a better understanding of the
underlying, fundamental processes involved in the accumulation and transport of
pollutants in urban catchments. This concern reflects the great interest in most
nations in recent years on pollution analysis, for the definition of sources of
pollutants and their environmental impacts, and resolution of means for effective
pollution abatement.(4,5)
A 1977 international symposium in Amsterdam(6) found little in the way
of new general fundamental characteristics of urban runoff and reached unanimity
principally on the need for greater knowledge of the basic processes underlying
the land accumulation and water transport of pollutants in urban areas. Then in
1978, modeling and related experiences were reported in 51 papers presented at the
first international conference devoted to urban storm drainage, held at the
University of Southampton, United Kingdom. (7)
*: Director, ASCE Urban Water Resources Research Program, Marblehead, Mass. , U.S.A.
11
International Symposium on Urban Hydrology
The report which follows is an assessment of what is new, what advances

have taken place and what progress has been made in eleven of the twelve nations
since their original reports,(2,3) spanning 1975-1977, were preparedJ These
updating reports were presented on May 28, 1979, at a full-day Symposium at the
Spring Annual Meeting of the American Geophysical Union (AGU), held in
Washington, D.C. Co-sponsors of the Symposium were the AGU Committee on Urban
Hydrology, the U.S. National Committee on Scientific Hydrology and Unesco. The
writer served as Symposium chairman.
Sections 2 through 12 are the written versions of the national updates.

The oral presentations at the Symposium were necessarily condensations of these
papers due to time limitations. Twenty minutes was allocated to each national
report, plus five minutes to answer questions. The national update presentations
were followed by a group discussion among the eleven speakers in a panel format,
and the Symposium closed with a general floor discussion. An attempt was made
to record all discussions, and an abridged summary of Symposium discussions
comprises Section 13.
Recommendation to Unesco
Immediately following the Symposium on May 28, 1979, the international

participants discussed any possible need for their future reinvolvement.
Recognized was that although over the next few years there will be various
international meetings dealing with urban hydrology, none of these are likely to
include progress reports of a national scope, as.at this Symposium. It was felt
that in about four years another international symposium might be fruitfully
convened, for the purpose of again reporting updates of national progress, but
hopefully with a greater number of nations participating. The participants
hereby respectfully bring this suggestion to the attention of Unesco, as a
possible item on the agenda of Che next phase of the IHP.
Some General Observations
The most obvious change reported for nearly all of the nations was a
greatly increased use of urban runoff mathematical models. As Professor Laurenson
of Australia put it, such models are no longer the playthings of researchers but
are becoming the stock in trade of practitioners at large. Mr. Marsalek óf
Canada emphasized that ultimately the objective function in such modeling must
be the receiving waters. Dr. Lowing of the U.K. and Dr. Massing of the F.R.G.
noted dissatisfactions that have been expressed with the design storm concept,
uncertainties voiced on some aspects of modeling and a new recognition of the
need to be able to simulate surcharging under certain circumstances. Mr. Falk
of Sweden could foresee great debates over the trade-off of further reduction
in wastewater loads versus reduction in pollution loads from urban runoff.
Dr. Desbordes of France pointed to a catchment in his country where there are
six internal measurement/sampling stations, in recognition of our poor mastery
of underlying processes, particularly for water quality. Mr. Skretteberg of
Norway and Mr. Zuidema of the Netherlands echoed a general trend of a
revitalization of urban hydrology research despite a threat of serious
curtailment in several countries only two or three years ago.
12
Acknowledgments
Assembly of these Proceedings was supported by a grant from the

National Science Foundation to ASCE. However, any opinions, findings, and
conclusions or recommendations expressed herein are those of the respective
writers and do not necessarily reflect the views of the National Science
Foundation. Dr. Edward H. Bryan is the NSF monitor for the grant to ASCE.
Mr. Harry C. Torno, a member of the AGU Committee on Urban Hydrology,

taped the entire floor discussion. Mrs. Richard Symmes, ASCE Program Secretary,
assisted Mr. Torno, subsequently transcribed the discussion tapes, and retyped
the edited papers.
Professor Jacques W. Delleur, Chairman of the AGU Committee on Urban

Hydrology, prepared in advance a comprehensive description of the Symposium
papers.(8)
Lastly, appreciation is due Mr. John P. Ritter, who supervised the
Symposium arrangements for AGU.
References
1. McPherson, M. B., Urban water resources, EOS_, 57(11), pp. 798-806, 1976.
2. Unesco, Research on urban hydrology, Vol. 1, Technical Papers in Hydrology 15.

185 pp., 1977.
3. Unesco, Research on urban hydrology, Vol. 2, Technical Papers in Hydrology 16.

265 pp., 1978.
4. McPherson, M. B., and F. C. Zuidema, Urban hydrological modeling and

catchment research: International summary, ASCE UWRR Program Tech. Memo.
No. IHP-13, Amer. Soc. Civil Engrs., 48 pp., NTIS: PB 280 754, November, 1977.

catchment research: International summary, Technical' Papers in Hydrology 18,
Unesco, 48 pp., 1978.
6. International Association of Hydrological Sciences, Proceedings Amsterdam

Symposium on Effects of Urbanization and Industrialization on the
Hydrological Regime and on Water Quality, IAHS-AISH Publication No. 123,
1977.
7. Helliwell, P. R., editor, Urban Storm Drainage, Pentech Press, Plymouth^

England, 728 pp., 1978.
8. "Meetings: Urban Hydrology Sessions at the 1979 Spring Meeting," EOS,

Transactions of the AGU, Vol. 60, No. 16, April 17, 1979.
13
Section 2 Progress since 1976 in Australia
by A. P. Aitken* and E. M . Laurenson**
Introduction.
In a 1976 report in this series, Aitken'*•' discussed urban hydrologi'cal

modeling and catchment research in Australia. Since then, steady progress has been
achieved in Australia in this field. The report which follows endeavors to bring
together, in summary form, the highlights of Australian progress in urban
hydrology since 1976.
It is apparent from this report that the high level of interest in

urban hydrology reported in 1976 has continued over the last few years. In the
areas of data collection, basic research and practical applications, there are
indications that the present high level of interest will continue for some years
to come. Although activity in both research and data collection has continued at
a reasonable level in recent years, probably the most notable advances have been
made in the wider practical application of recently developed modeling techniques.
This situation has been achieved as a result of a greater emphasis on post-
graduate training and research in this area, together with workshops and lectures
aimed at introducing practitioners to modern techniques.
As a background to this paper, it may be helpful to some readers to

note that, although Australia is sparsely populated, it is a highly urbanized
country. At the present time over 11 million people out of the total Australian
population of 14 million live in urban centres. With two-thirds of the population
in 12 cities it is not surprising that Australian cities present similar urban
hydrological problems to those in other, more highly populated, countries.
Another notable factor concerning Australian cities is that it has been the
practice, almost without exception, to separate the stormwater drainage system
from the wastewater sewer system. This situation is in direct contrast to that
existing in many of the older cities in the United Kingdom, Europe and the United
States of America.
Research Programs
No major, coordinated research program exists in Australia specifically

for urban hydrology (or urban water resources). Nevertheless, research activity
is widespread with several universities and other tertiary institutions actively
pursuing programs. These are mostly funded by the Australian Water Resources
Council, the Water Research Foundation of Australia or internally by the tertiary
institutions themselves. In addition to these formal research centres, some
government agencies and private consultants have also become involved in urban
hydrology research and are usually funded through their own revenue sources.
*: Senior Engineer, Hydrology, Snowy Mountains Engineering Corporation, Cooma

North, New South Wales, 2360, Australia.
'*: Professor of Civil Engineering, Monash University, Clayton, Victoria, 3168,

Australia.
15
The remainder of this sub-section contains short descriptions.of five
of the more recent research projects in Australia. Details of these have been
published in the references cited.
Rainfall-Runoff Modeling
In the 1976 Report, reference was made to the testing of several event
models using Australian rainfall-runoff data. Included in these were the well
known Road Research Laboratory Model, the Laurenson Runoff Routing Model (which
is now widely used in Australia for both urban and rural catchments) and several
models developed in the USA. The Rational Formula was also tested against data
observed on several gauged catchments. Since the 1976 Report, two more notable
studies have been carried out on event models:
(a) Further development by the Monash University group of the Laurenson Runoff
Routing Model (LRRM) to allow it to be used more satisfactorily on urban
catchments. Crouch and Mein(&) developed an empirical formula to enable
the time-lag parameter of the model to be estimated for urban catchments.
The strong influence of slope on time lag in urban catchments, which
contrasts with the situation on rural catchments, was demonstrated. They
also tested a loss model consisting of different initial losses on the
pervious and impervious areas followed by a constant continuing rate of
loss on the pervious area. This loss model, however, proved unsatisfactory.
Based partially on Crouch's results, Laurenson and Mein(H) proposed the
use of the same empirical formula for the time-lag parameter as is used
for rural catchments and included the effects of urbanization in the
computation of relative delay time for the individual model storages,
thereby providing a model suitable for partially urban and partially rural
catchments. They also used a loss model consisting of an initial loss
inversely proportional to the fraction impervious, followed by a constant
proportional loss, a loss model strongly suggested by Crouch's data.
(b) Blunden and Moodie(^) applied the auto-regressive moving average models of
the rainfall-runoff process in the manner developed by Young(19) to
estimate floods on five small urban and rural catchments. The models were
fitted using data sets for minor complex storm events, and the adjustment
of the data to obtain effective rainfall- inputs and constant catchment
response times was discussed in detail. The model fit obtained was
generally good, and it was shown that the model-generated impulse response
function (analogous to the unit hydrograph) was relatively insensitive to
model order. The model-generated impulse response function for one data
set was compared with a unit hydrograph derived using matrix inversion,
and a preliminary discussion of the relationships between the impulse
response functions and catchment characteristics was included.
Event-type models of the rainfall-runoff process, such as those

discussed above, either do not attempt to predict the volume of loss that will
occur in a given storm or require some empirical and often unreliable means of
predicting the initial antecedent wetness of the catchment. To enable the actual
antecedent condition for a storm to be estimated more reliably, a continuous
mathematical model for the rainfall-runoff process is required. Two recent
Australian developments in this area of modeling are:
(a) A model developed at Monash University by Wootten and Mein.(18) This

model simulates the soil moisture changes between major storms and thus
16
enables reasonable estimates to be made of the antecedent conditions prior
to a flood event. The infiltration component uses the Green-Ampt model,
extended to simulate redistribution, and the Road Research Laboratory
method is used for routing runoff from the impervious and pervious portions
of the catchment.
(b) The adaptation of the Australian Representative Basin Model (ARBM) to urban
catchments as described by Black and Aitken. (2) This model was originally
developed for rural catchments as part of the research program of the
Australian Water Resources Council. The model is of similar complexity to
that described above. Testing showed that, although it yielded reasonable
estimates of runoff, it used a considerable amount of computer time and is
therefore an expensive tool if used on a routine basis. The report by
Black and Aitken(2) includes a detailed manual describing all aspects of
the ARBM computer program.
Comprehensive Urban Water Model
As cities grow in size and population, the problems of supplying them

with water increase, and the quantities of runoff and wastewater to be disposed
of also increase. There is a growing feeling that the runoff and wastewater
of urban areas, previously thought of principally as waste material to be
disposed of, should be studied instead as a resource. In this light, urban
hydrology models should not be simply rainfall-runoff models as for rural areas
but should also include the water supply-sewerage sub-system and the interconnections
between this and the rainfall-runoff sub-system.
It was foreshadowed in the 1976 Report(l) that development of such

models in Australia was contemplated and Graham and Laurenson(S) at Monash
University have now reported on such a model. It is an areally lumped model
using a daily time step to predict monthly quantities of stormwater, wastewater,
and évapotranspiration from data on rainfall and water supply. The main
interconnections between the rainfall-runoff sub-system and the water supply-
wastewater sub-system are garden-watering and infiltration of stormwater into
wastewater sewers. Other minor interconnections also are modeled. Good aggregate
data on rainfall, water supply, and wastewater were available to test the model,
but lack of other data made reliable testing of the model difficult and forced
the adoption of an areally lumped model. Nevertheless, results were encouraging.
Retarding Basin Design
Increasing use is being made of retarding basins (detention storages

with uncontrolled pipe outlets) to reduce peak discharges through temporary
storage.of runoff. Many such basins have been designed on a rule-of-thumb basis
and the basis of some other designs is obscure. The development and use of
runoff-routing models for this purpose was described in the 1976 Report.(1)
While these applications have continued, as will be described, research at
Monash University has further developed the modeling approach and has produced
a rational and practical design procedure. (11,12)
With the storage-elevation relation known, the design problem is to

determine the number and diameter of pipes through the base of the dam, the
length and elevation of the overflow spillway near the crest of the dam, and
the elevation of the dam crest. The design must satisfy one of two criteria,
either that the peak outflow discharge be limited to a specified value or that
17
the peak water level be limited to a specified elevation. Two design floods
must be calculated: a basin design flood to determine the spillway crest
elevation and/or the pipe outlet dimensions; and a spillway design flood to
determine the dimensions of the overflow spillway. The Laurenson Runoff Routing
Model is used to calculate inflow design hydrographs for storms of various
durations, to route these through the basin with trial and error variation.of
pipe outlet and spillway dimensions, and to investigate the effects of the basin
on hydrographs downstream. Further studies of retarding basin behavior and
design have been carried out by Joy and Woodhouse.(10)
Water Quality Modeling
Although there exists an awareness of water quality problems in urban

areas in Australia, research into modeling of the rainfall-runoff-water quality
process has received surprisingly little attention. Furthermore, the well-known
rainfall-runoff-water quality models developed overseas (e.g., STORM and SWMM)
have not been tested in Australia. Perhaps one reason for this situation is the
lack of adequate data in some cities. Another possible reason is the fragmentation
of responsibility for water quantity and quality within government agencies.
Maybe, however, the real reason is that potential users of these models cannot see
how the results they yield will assist in solving water quality problems.
The New South Wales State Pollution Control Commission(16) is showing

considerable interest in pollution caused by urban runoff and has collected some
very useful data which could be used in modeling. Similarly, Cordery(^) has
carried out many measurements of pollutant loads in urban drains. These
measurements support evidence from earlier overseas studies which have shown that
urban stormwater is more heavily polluted, but contains far less nutrients, than
sewage which has undergone secondary treatment. Cordery's laboratory experiments
showed that, in consequence of this, greater water quality benefits could accrue
at less cost from a primary settling treatment of urban stormwater than from
tertiary treatment of sewage effluent. There appear to be few actual applications
of water quality models to urban catchments in Australia. One exception is
described by Moodie. (13)
He described the development and calibration of a
computer-based model of pollutant transport and decay for a fully urbanized
catchment. The model simulated the variations of biochemical oxygen demand and
dissolved oxygen along a watercourse and was used to evaluate various alternative
water quality improvement strategies. It was shown to be an efficient means of
determining cost-effective strategies and also was found to be a valuable means
by which the user could increase his knowledge of the complex behavior of
pollutants in an urban watercourse. Similar water quality modeling work has been
carried out in The University of New South Wales and in Melbourne University.
(17)
Rational Method Research
Despite the fact that the rational method has been used in urban
hydrology applications for about a century, surprisingly little research has been
conducted into,the proper values of time of concentration and runoff coefficient
for use with the method on urban catchments. Design values of these parameters
in Australia have been based on interpretations of formulas and data from
overseas, with an infusion of engineering experience, but there has been no
analysis of Australian urban catchment rainfall and runoff data.
The persistence of this situation was highlighted by the fact that the
Second (1977) Edition of the Institution of Engineers guide to practice,
18
"Australian Rainfall and Runoff,"(9) included the same poorly founded methods
for determination of time of concentration and runoff coefficient as in the
original 1958 edition. In an effort to correct this situation, Laurenson at
Monash University has commenced a project to analyze some of the available data
to yield sound information on these simple parameters.
Educational Aspects
The new developments in urban hydrology are specialized and therefore

only briefly lectured upon in undergraduate courses in engineering hydrology.
To update engineers and others working in this field in Australia, various
universities and other tertiary institutions have been very active in conducting
seminars and workshops and part-time Masters' courses. Meetings have been held
in most of the larger Australian cities and support for these meetings has been
most enthusiastic.
"Australian Rainfall and Runoff" Workshops
In 1977, the Institution of Engineers, Australia, issued a revised

"manual" on flood estimation in Australia entitled "Australian Rainfall and
Runoff: Flood Analysis and Design".™) Although the document covers both the
rural and urban scene, the greatest group of engineers and surveyors using the
document has interests primarily in the urban drainage sections. This document
has been presented to users around Australia with the aid of workshops conducted
by Universities and by the Institution of Engineers or other interested bodies.
Specialized Urban Hydrology Workshops
Specialized urban hydrology workshops were held at Monash University

in 1977 and the University of New South Wales in 1978. There appears to be a
continuing demand for this type of workshop. Both workshops included lectures
and practical sessions on computer models, including:
Road Research Laboratory Model (from the United Kingdom);

Stormwater Management Model (from the USA EPA);
Regional Stormwater Model
Australian Representative Basin Model;(2) an d
STORM (from the US Corps of Engineers).
Computer facilities were made available to participants at both workshops in
order that "hands on" experience of the models could be gained. The workshop
syllabuses covered stormwater drainage design, the estimation of floods in
urban creeks and rivers, urban water quality modeling, and legal and
administrative aspects. (In 1978 also, a more general urban hydrology workshop
was conducted at the Swinburne College of Technology). This type of continuing
education activity is popular with engineers in Australia, but universities and
colleges are only slowly developing their ability to mount such activities and
many more resources must be devoted to this work if the evident demands are to be
met.
Data Collection
Given in the 1976 Report(l) were details of the Australian data

collection program for urban catchments. In general, the size of the program
was satisfactory although the quality of some of the data was not good. Some
19
of the data were not readily available, which was also unfortunate. Since 1976,
several more urban catchments have been instrumented and, provided that funds
can be made available for analysis in the future, present progress can be
considered satisfactory.
Problems with Data Collection
Despite the satisfactory progress being made, several problems still

exist. The inadequate coverage of stations in the nation's largest city, Sydney,
is one of these. • In all cities, however, there are at least some stations where
the ratings are of doubtful reliability or non-existent, and where the routine
processing of charts lags behind data collection, sometimes by many years.
Availability of data from many stations is consequently poor. It is likely to
be several years before sufficient resources can be applied to solve these
problems and bring the data collection system to a uniformly high standard of
operation.
Real-Time Data Collection
The most exciting advance in data collection in Australia is that

which has taken place in Melbourne in the last few years. A network of rainfall
and water-level stations has been installed primarily to assist in the early
forecasting of floods in urban areas. Porter et al.(14) have described this
system, in which point measurements of rainfall and stream levels at remote sites
are transmitted every six minutes via dedicated lines to a central location where
a mini-computer processes and stores the data received on magnetic-tape cassettes
for subsequent analysis. The real-time nature of the data acquisition and the
immediate assessibility to this data enables early warning to be provided of
potential flooding of downstream urban areas in catchments draining through
metropolitan Melbourne, as well as providing several other advantages over
conventional data acquisition methods in hydrology. The gauge network (as of
February 1979) comprises 47 stations at remote sites. Twenty-eight of these
are dual stations recording both rainfall and water level, three record rainfall
only, and 16 record stream level only.
Water Quality Data Collection
Programs of sampling and analysis have been commenced by many agencies

in many cities. Special efforts have been made to try to coordinate the
activities of the various agencies and to avoid the wasteful overlap which had
seemed possible in the early stages. There is little published material on the
development of the water quality data collection programs and it is consequently
difficult to judge how well planned the activities are. Most data collection
organizations are still gaining initial experience and a clear picture is yet to
emerge, but a good deal of data is being collected.
Applications
While progress in data acquisition and modeling since 1976 has been
generally satisfactory, probably theroostsignificant advance has been that the
engineering profession, in general, is using more advanced techniques to solve
urban hydrological problems. This situation has been reached because of:
a recognition by engineers that the design of complex drainage

systems (e.g., those which include retarding basins) cannot be
20
achieved satisfactorily without better tools than, say, the
Rational Formula; and
continuing education of engineers and others (as discussed above)

combined with a greater familiarity with computer techniques.
Urban Flood Estimation, Drainage.Design, and Retarding Basin Design

with Event Models
In the 1976 Report, (1) reference was made to the Regional Stormwater
Model (RSWM) as described by Goyen and Aitken.
(7)
This model has now been used
to design drainage systems in about twenty different locations in Australia and
is currently being used for the design of an urban drainage system in Indonesia.
As pointed out in the 1976 Report, it and similar models were developed to cope
with design complexities introduced by installing retarding basins in urban
catchments.
In an effort to make event-type rainfall-runoff modeling a tool widely
available to practicing engineers, Laurenson developed and distributed a general
purpose runoff routing computer program called RORB, initially for rural
catchments. Laurenson and Mein'H^ subsequently developed this further for use
on urban catchments and for retarding basin design. It has found widespread use
in Australia since its release.
Since 1976, interest in the design of retarding basins in urban areas
has continued to grow. Two recent papers on the design of retarding basins have
been presented at hydrology symposia in Australia. These papers are by Mein and
Woodhouse
(12)
and Joy and Woodhouse. (10) The general purpose runoff routing
computer program RORB developed by Laurenson and MeinUl) has also been used for
some retarding basin designs.
In addition to the modeling applications discussed above, it is known
that the Road Research Laboratory Model has been used by some consultants in
Australia for design of stormwater drainage networks, although published
information is not available on these applications. A computer application of
the rational method to urban drainage design, facilitating the repetitive
calculations of times of concentration, rainfall intensity and runoff
coefficients, has also been used but not published.
Simulation with Continuous Models
The application of continuous models to urban catchments in Australia

to solve practical engineering problems appears to be just commencing. One
example, has been recently described by Black and Codner. (3) They found that the
use of continuous simulation models in urban drainage studies facilitated the
investigation of many different schemes and selection of the most suitable one.
An insight into the long-term behavior of the selected scheme may be obtained
by continuous simulation.
This application required the use of the RSWM and the Australian
Representative Basins Model (ARBM). The RSWM has been modified to operate as
either a discontinuous or a continuous model. When operating as a continuous
model, rainfall-excess input is provided using the ARBM, which is a continuous
rainfall-runoff model. As mentioned earlier, the ARBM has been modified to
simulate runoff from urban catchments, as described by Black and Aitken.(2)
21
Sample results from two studies were presented. In the first study,
the RSWM was used in preliminary investigations of alternative drainage
strategies for Middle Creek catchment in Albury-Wodonga. In the second study,
both models were used to simulate ten years of runoff from the proposed drainage
scheme for Baranduda Catchment. From the simulation, frequencies of occurrence
of depths of water in the proposed retarding basins and floodways were obtained.
Flood Forecasting
Devastating urban flooding in Brisbane and minor urban flooding in

Melbourne in 1974 led to concerted action in both cities to develop better
flood forecasting and warning services and generally better preparedness for
floods. In the Melbourne case, the sophisticated telemetered data collection
network, mentioned above, feeds a forecasting system based on the use of a
continuous rainfall-runoff model, which is an adaptation of the Monash Model. (15)
The Brisbane studies have involved retarding basin and stream improvement as well
as flood forecasting aspects.
Conclusion
This paper has reviewed progress in Australian urban hydrology research

and associated activities since the 1976 Report. (1). In general, it is believed
our progress has been steady. The most significant advance has probably been in
the widespread application of new techniques. The education of engineers and
others in recent developments appears to be succeeding, while data collection and
basic research are progressing at a satisfactory level.
References
1. Aitken, A. P., (1976), "Urban hydrological modeling and catchment research in

Australia," ASCE UWRR Program Technical Memorandum No. IHP-2, ASCE, New York,
N.Y., May 1976, NTIS No. PB .260 686.
2. Black, D. C , and Aitken, A. P., (1977), "Simulation of the Urban Runoff

Process," Australian Water Resources Council, Technical Paper No. 26.
3. Black, D. C , and Codner, G. P., (1979), "Investigation of urban drainage

strategies using simulation," Institution of Engineers, Australia, Hydrology
Symposium, Perth (in preparation).
4. Blunden, D. W., and Moodie, A. R., (1978), "Auto-regressive moving average

models of the rainfall-runoff process in small catchments," Institution of
Engineers, Australia, Hydrology Symposium, Canberra, pp. 62-69.
5. Cordery, I., (1976), "Evaluation and improvements of quality characteristics

of urban stormwater," Report No. 147, Water Research Laboratory, University
of New South Wales. (Summary in "Quality Characteristics of Urban Storm
Water in Sydney, Australia," by Ian Cordery, Water Resources Research,
Vol. 13, No. 1, pp. 197-202, February 1977).
6. Crouch, G. I., and Mein, R. G., (1978), "Application of the Laurenson runoff
routing model to urban areas," Institution of Engineers, Australia, Hydrology
Symposium, Canberra, pp. 70-74.
22
Goyen, A. G., and Aitken, A. P., (1976), "A regional storrawater drainage
model," Institution of Engineers, Australia, Hydrology Symposium, Sydney,
pp. 40-44.
Graham, G. S., and Laurenson, E. M., (1978), "Total urban water resources
modeling," Institution of Engineers, Australia, Hydrology Symposium, Canberra,
pp. 52-56.
Institution of Engineers, Australia, (1977), "Australian Rainfall and Runoff:

Flood Analysis and Design," The Institution, Canberra.
Joy, C. S., and Woodhouse, M. P., (1978), "Aspects of the analysis and design
of retarding basins," Institution of Engineers, Australia, Hydrology
Symposium, Canberra, pp. 85-89.
Laurenson, E. M., and Mein, R. G., (1978), "RORB Version 2 User Manual,"
Monash University, Department of Civil Engineering.
Mein, R. G., and Woodhouse, M. P., (1977), "Design of retarding basin

systems," The Institution of Engineers, Australia, Hydrology Symposium,
Brisbane, Preprints of Papers, pp. 141-145.
Moodie, A. R., (1975), "Computer modeling of water quality in an urban

watercourse," Institution of Engineers, Australia, Conference on Computers
in Engineering, Canberra, pp. 26-32.
Porter, J. W., et al., (1976), "Computer-operated telemetered gauge network

for flood forecasting and prediction," Institution of Engineers, Australia,
Conference on Computers in Engineering, Canberra, pp. 69-73.
Porter, J. W., and McMahon, T. A., (1976), "The Monash Model: User Manual
for Daily Program HYDROLOG," Report 2/76, Department of Civil Engineering,
Monash University, 41 p.
State Pollution Control Commission (New South Wales), (1977), "The quality
of Sydney's natural waterways in relation to its growth".
Vass, R. J., (1978), "Simulation of Water Quality Management in a River

System," Ph.D. Thesis, The University of Melbourne, Department of Civil
Engineering.
Wootton, R. M., and Mein, R. G., (1976), "Development of a continuous urban

rainfall-runoff model," Institution of Engineers, Australia, Hydrology
Symposium, Sydney, pp. 88-92.
Young, P; C., (1974), "Recursive approaches to time-series analysis,"

Bulletin Institution of Maths, and App., Vol. 10, pp. 209-224.
23
Section 3 Progress since 1976 in Canada
by J. Marsalek*
Introduction
The paper which follows is an updating of a 1976 national state-of-

the-art report on urban hydrological modeling and catchment research in Canada.^'
The information presented below is divided into four sections that deal with
urban catchment research, hydrological modeling, runoff controls, and implementation
of research results. While the period covered in the national report was
characterized by extensive research activities, the period covered in this paper
could be characterized as being one of implementation of research results in
practice.
Urban Catchment Research
In the 1976 national report,(*' 15 urban test catchments were described.

Of these 15 catchments, only five were monitored during the 1976-1979 period. A
rapid decline in the scope of field data collection was caused by a lack of funds.
Recent urban catchment studies dealt mostly with verification of runoff

models and characterization of pollutant loadings in urban runoff. In several
instances, the catchment studies also included field evaluation of selected runoff
controls and investigation of the effects of stormwater discharges on receiving
waters.
Data collection for model verifications was conducted in three urban

catchments(2-5) u n der the sponsorship of the Urban Drainage Subcommittee (UDS).
In all three catchments, both runoff quantity and quality were monitored.
Previously reported urban catchment studies dealt only with such basic
water quality parameters as BOD, COD, solids and nutrients. Changing environmental
concerns led to investigations of less common parameters including microbiological
parameters, polychlorinated biphenyls (PCB's), organochlorine pesticides and
heavy metals.
Microbiological studies of urban stormwater in several catchments found

strikingly high levels of microbial populations.w) These levels were similar to
those encountered in dilute sewage and, therefore, constituted a health hazard.
Observations of water quality in the Rideau River indicated increases in pollutant
concentrations due to urban stormwater runoff into the river.(°) Such increases
were largest for microbiological parameters.
PCB's, organochlorine pesticides and heavy metals were observed in

stormwater runoff from two urban catchments. (9) Significant portions of the
annual PCB's and pesticide loadings were found to be transported by solids.
Quality of runoff from two urban catchments and the effects of storage
on runoff quality were investigated in Winnipeg.(1") «jê data collection program
included the monitoring of precipitation, quality of inflows into and outflows
from storage reservoirs, plankton counts, observations of fish, sedimentation
rates, and surveys of aesthetic and recreational aspects.
*: National Water Research Institute, Burlington, Ontario, Canada.

25
Snowmelt runoff quantity and quality were observed in Toronto.(H/
Field samples of snowmelt runoff were subsequently used in a treatability study.
Hydrological Modeling
Urban hydrology is recognized as a distinct field of hydrology with its

own set of predictive models and analytical expressions. Various urban
hydrological models used in Canadian studies were discussed in detail in the 1976
national report.(1) In recent years, only minor refinements of the existing
models were undertaken. Additional research was conducted into selection of input
data for hydrological modeling.
In the preliminary analysis (planning) stage, relatively simple tools

are used to obtain an overall assessment of runoff problems and to provide
estimates of the effectiveness and costs of runoff management alternatives. A
planning model was developed by the Pollution from Land Use Activities Reference
Group (PLUARG) to simulate outputs of land-derived pollutants under selected
management scenarios, (12)
The model is based on a conceptualization of watersheds
as sets of identifiable units of specified land form and land use. These units
contribute pollution loads to the main river in series from headwaters to the
river mouth. To evaluate the contribution of individual units, one needs to
develop pollutant unit loads for various land forms and uses. For urban areas,
annual pollutant unit loads were estimated under a contract commissioned by the
Urban Drainage Subcommittee.(13) These loads were further expanded and modified
to meet the PLUARG requirements.^» 15 )
The probabilistic mathematical model developed by Charles Howard and
Associates, Ltd., was further refined.(1°) The model was used to determine
least-cost storage-treatment combinations for various levels of control over both
runoff quantity and quality. The modeling results were verified against results
produced by means of the STORM model.
Detailed modeling of urban runoff (design/analysis stage) was discussed
in the 1976 national report,(!) which also contained a recommended methodology for
application of urban hydrological models. The period from 1976 to 1979 was
characterized by a fairly extensive use of existing hydrological models in
practical applications.(17) During the same period, only limited attempts were
made to refine existing design/analysis models.
Model verifications were continued under the sponsorship of the Urban

Drainage Subcommittee. The Stormwater Management Model of the U.S. Environmental
Protection Agency was verified on three catchments and the ILLUDAS model of the
Illinois State Water Survey was verified on two catchments. w> 18) Although good
verification results were obtained for runoff quantity, many problems related to
the modeling of runoff quality remain unresolved. (6)
Modeling of soil infiltration and runoff from pervious areas was

studied at Queen's University.(19) A study of energy losses at sewer.pipe
junctions(20) produced data which could be used In pressure-flow routing through
sewer networks. A research study of simultaneous simulation of major/minor
drainage is currently under way at the University of Ottawa.(21)
Input data for detailed urban hydrological modeling were studied by a

number of researchers. Economic selection of an optimal design return period
for urban drainage networks was studied in some detail.
(22)
The selection of
26
the optimal return period resulted in the minimum total annual drainage costs
which had been defined as the sum of construction and failure costs.(23)
Rainfall inputs for urban hydrological modeling were studied

extens ively.
(24,25)
In several studies, runoff hydrographs were simulated for
selected actual storms, flood frequency curves were produced and, for selected
return periods, the flow peaks simulated for actual and synthetic storms were
compared„(26,27)
Synthetic design storms of the Chicago type were found unacceptable for
drainage design in the Municipality of Surrey, British Columbia.(28) Runoff
peaks of desired return periods were, therefore, derived from simulations for
actual storms and probability distributions of three parameters of the simulation
model used.
To assist model users, some further work was done to simplify the
implementation of hydrological models on computers. Toward this end, a model
package FAST-SWMM was developed.
(29)
The processes in the package solicit and
accept user-directed initial SWMM model input in a conversational free-format
model, solicit and accept design-oriented input data, check the validity of the
input, and return to the user terminal values of certain dependent parameters.
The main objectives of the package are to reduce the complexity of SWMM job
submissions, minimize user errors, and reduce the total turn-around times.
In another study, (18)
the ILLUDAS model was modified for implementation
on a minicomputer (a 16K bytes storage programmable calculator). Subsequent
model testing was done in the Malvern test catchment.
Urban Runoff Controls
Between 1976 and 1979, urban runoff controls were proposed for a
number of locations in Canada. Field evaluation of such controls has rarely
been undertaken.
The costs of controlling pollution loads in urban runoff were
investigated in two studies.
(13,30)
A study sponsored by the Urban Drainage
Subcommittee examined the costs in the Province of Ontario. (13)
Various degrees
of urban runoff pollution control would be achieved by means of optimal
combinations of storage and treatment. The control of runoff pollution would
be primarily based on the control of biochemical oxygen demand. The second
study(30)
dealt with the control of suspended solids in urban runoff and
produced estimates of control costs on a nationwide basis.
The effects of stormwater impoundments on runoff quality were studied
in Winnipeg.
(10)
Two urban areas served by ten interconnected impoundments were
studied. Water quality of urban runoff passing through the impoundments
improved substantially. The impoundments were found to have significant aesthetic
and recreational benefits, with water quality being suitable for secondary forms
of recreation. Appreciable savings in drainage capital costs over conventional,
all-conduit, systems were reported. The Winnipeg study also dealt with
maintenance of stormwater impoundments. The aspects investigated included
treatment for algae, water weeds control, mowing of surrounding grassed areas,
27
litter removal, sediment removal, routine inspections and water quality monitoring.
Fish surveys were undertaken in the Winnipeg study of stomiwater
impoundments. It was observed that such water bodies are less than ideal for
raising fish but can support a fish population. In another study,(31) t n e
tissue of fish caught in an urban storrawater impoundment was analyzed.
Observed concentrations of heavy metals (particularly lead) were strikingly
high.
The treatment of urban runoff was addressed in two recent studies.(H >32)
In one study, four fine-mesh, high-rate, commercial screening devices(32) were
evaluated under field conditions. The other study, which dealt with the
treatability of snowmelt runoff, found natural sedimentation ineffective. (11)
Chemical coagulation produced satisfactory results with respect to removals of
suspended solids, chemical oxygen demand, and total phosphorus.
Storm-tracking studies(33) a r e currently under way in Montreal. The

studies suggested the feasibility of implementing a computer-controlled operation
of the combined sewer system on the island of Montreal.
Implementation of Research Findings in Practice
Contents of this subsection are partly based on a 1978 paper.(6) The

research program described in the 1976 national report,(1) as updated herein,
identified problems with urban drainage, indicated methodologies for solving the
problems, provided technical alternatives for resolving the problems and
analytical methods to aid in choosing cost-effective alternatives. A government-
sponsored program to implement the findings.is under way with the goal of
resolving problems related to urban drainage. This program consists of a
technology transfer program, demonstration projects, river-basin studies
incorporating urban drainage assessments, development of a manual of practice
on urban drainage, and development of a policy on urban drainage for the Province
of Ontario. Through the initiative of some universities, other implementation
activities are taking place outside of the government-sponsored program.
Technology Transfer
Research findings and technical methodology have been transferred to

planners and designers of urban drainage systems through a series of conferences
and workshops, through publications, and through distribution of the storrawater
modeling package described in the first national report. (1)
In the Canada-Ontario Agreement Research Series, 16 reports related to

urban drainage have been released to date with an additional 14 to be released
subsequently. Published reports include the proceedings of a workshop on
stormwater management models(34) arwj the proceedings of a conference on modern
concepts in urban drainage.(35)
The storrawater models developed and studied as part of the Canada-

Ontario research program are available for a nominal fee through the Ontario
Ministry of Environment, and to date have been distributed widely across Canada.
The models are becoming part of the standard engineering practice in Canada,
indicating the relative success in both the training program and the distribution
of modeling technology. Model practitioners meet on a semi-annual basis at
"Stormwater Management Model Users' Group Meetings" organized jointly by the
US-EPA and Canadian institutions. Also, training workshops and courses dealing
with urban hydrological models are offered by several universities.
28
To facilitate implementation of new urban runoff management techniques
in practice, a cooperative project was started at the University of Ottawa.
(21)
The cooperators fund the project and advise on modeling applications and in
return receive information on updated or modified models.
Demonstration Projects
Studies which demonstrate new methodology have been carried out in
Ottawa, St. Thomas and Sarnia.
A study of the effects of controlled and uncontrolled stormwater

discharges from a proposed development with an ultimate population of 100,000
was carried out in the Ottawa area. The study report'3°} recommends a drainage
system that would incorporate a series of stormwater retention ponds, as well as
a number of site controls and management techniques.
In St. Thomas, Ontario, a technology demonstration study'") w a s

carried out to apply advanced stormwater management techniques to find cost-
effective solutions to stormwater problems. The first objective was to minimize
pollution loads to the receiving stream from stormwater discharges, combined
sewer overflows and treatment plant bypasses, compared to dry-weather treatment
options. At the same time, the objective was to minimize health hazards and
property damage from flooding of basements and other areas during storm events.
A similar study has been carried out in Sarnia, Ontario, with the
primary objective of reviewing the combined sewer separation program to halt
basement flooding. This study demonstrated that partial sewer separation would
be less costly than complete separation, while still providing adequate flooding
protection and reduction of pollution loadings.
River-Basin Studies
On-going and planned river-basin studies in the Grand River and Rideau
River in Ontario serve to implement past research findings and to continue with
applied research in the framework of a comprehensive river-basin study.
In order to evaluate non-point source problems in the Grand River

basin,
(38)
a monitoring program is underway to sample runoff events and to record
receiving water, quality continuously. A comprehensive, continuous water quality
model is being linked to a continuous urban runoff model that already incorporates
point source inputs, in order to evaluate the costs and effectiveness of urban
point and non-point controls. The results are expected to be available in 1980.
A study of the effects of urban runoff on the Rideau River in the
Regional Municipality of Ottawa-Carlton has been proposed. The major concern is
the bacterial pollution of bathing beaches on the Rideau River from combined
sewer overflows, and the potential pollution of bathing beaches from storm
drainage for a large proposed new urban development.(8/ Research-related aspects
of the proposed study include the evaluation of the treatment efficiency of an
existing large man-made stormwater treatment pond, the evaluation of street
sweeping efficiency as a control measure, and the development and application of
a continuous receiving water quality model for bacterial pollution.
29
Manual of Practice on Urban Drainage
In order to provide an overview of urban drainage problems and

solutions, a "Manual of Practice on Urban Drainage" was prepared.(39) ^he
manual is not a design handbook in the sense that it contains all the
information needed to plan and design urban drainage systems. Instead, it
provides guidance for the adoption of improved drainage practices for the
purposes of improving property protection, providing for environmental
protection and attaining economic efficiency. The manual contains chapters
dealing with urban drainage problems, planning, methods of computing runoff
quality and quantity, source controls, collection, storage and treatment
alternatives, administrative and legislative concerns, and a design example.
The manual is tailored to the legislative and planning framework in Ontario
and contains numerous Ontario examples of planning, design and operation of
urban drainage and quality control systems.
A similar, though less comprehensive, document was prepared by the

Province of Alberta.
Dflvelopment of a Policy on Urban Drainage for Ontario
In order to implement the concepts derived from research and as

presented in the Manual of Practice described above, a committee was formed
to develop a policy on urban drainage for the Province of Ontario.(40) The
proposed policy will cover the five specific areas described below:
1. Municipalities, in coordination with conservation authorities, will be

required to develop master drainage plans for all watersheds in their
boundaries.
2. Municipalities with sewage collection and treatment systems will be required

to formulate and implement a comprehensive pollution control strategy which
includes consideration of both wet-weather and dry-weather pollution sources.
3. New drainage systems for all development will be designed using the major-
minor concept, which recognizes the dual role of the drainage system to
provide for convenience during minor (high frequency) runoff events and to
minimize property damage and protect life during major (rare) runoff events.
4. Proponents of new urban development proposals will be required to indicate

the effects of the development on the watershed and to carry out mitigative
measures as required.
5. Proponents of new urban developments will be required to plan for and carry
out an erosion and sediment control program in the planning and construction
stages of development, and to follow up with an adequate maintenance program.
The policies are presently under discussion within government and with
the groups affected, such as municipal engineers, consulting engineers and
developers. Technical guidelines for each policy area have to be written and
approval agencies have to be designated in order to implement the policy. It is
felt by the policy committee that the proposals have an excellent chance of being
adopted by. government and accepted by all parties involved, because of the clear
environmental and flood control benefits, the existing legislation and approval
mechanisms which allow for ready implementation, the available technology, and
30
in view of the projected future capital expenditures on drainage and pollution
control works to service new development and to remedy existing problems.
Summary and Conclusions
Following the completion of the government-sponsored research program,

Canadian urban hydrology research entered a new phase. While the basic research
has been dramatically reduced and is continued by only a few agencies and
universities, a number of extensive applied research studies addressing local
problems are planned at present. Such integrated studies will involve both
field monitoring and computer simulations and will deal with all the principal
point and non-point sources of pollution including urban runoff. The proposed
studies deal not only with urban watersheds, but also with the receiving waters.
The first of these studies has just started in the Regional Municipality of
Ottawa-Carlton.
During the 1976-1979 period, most government research funds were spent
on the implementation of research results in practice. Towards this end, an
extensive technology transfer program was undertaken and an urban drainage manual
and policy documents were prepared. Other implementation activities were
undertaken by engineering companies and universities.
Further research is planned to advance the understanding of processes

affecting the quantity and quality of urban runoff, to advance design
methodologies for urban drainage, and to study the effects of urbanization on
receiving waters.
References
1. Marsalek, J., Urban hydrological modeling and catchment research in Canada,

ASCE UWRR Program Tech. Memo. No. IHP-3, Amer. Soc. Civil Engrs., 52 pp.,
NTIS: PB 262 068, June, 1976.
2. Gore & Storrie Ltd., Storrawater Management Model verification - Hamilton

test catchment, a draft report submitted to the Urban Drainage Subcommittee,
Toronto, 1978.
3. M. M. Dillon, Ltd., Stormwater Management Model verification, a draft report

submitted to the Urban Drainage Subcommittee, Toronto, 1978.
4. Marsalek, J., Malvern urban test catchment - Vol. II, Canada-Ontario

Agreement Res. Report Series, Ottawa, 1979 (in press).
5. Lorant, F. I., and C. Doherty, Verification and calibration of the Illinois

Urban Drainage Area Simulator (ILLUDAS), in Proceedings Stormwater
Management Model (SWMM) Users' Group Meeting, May 4-5, 1978, edited by H. C.
Torno, U.S. Envir. Prot. Agency. EPA-600/9-78-019, 244 pp., July, 1978.
6. Weatherbe, D. G. and J. Marsalek, Canadian research on non-point sources of

pollution with particular emphasis on urban runoff, a paper presented at
the Amer. Soc. Civil Engrs. convention, Chicago, October 16-20, 1978.
7. Dutka, B. J., A. A. Qureshi, I. Rybakowski and S. Tobin, Microbiological

characteristics of urban stormwater runoffs in central Ontario, Canada-Ontar
Agreement Res. Report No. 8?, 129 pp., Ottawa, 1978.
8. Laboratory Services, Env. Prot. Service, Stormwater monitoring study, Rideau
River, National capital area, 274 pp., MS report no. OR-24, Environment
Canada, Ottawa, December, 1978.
9. Marsalek, J., Polychlorinated biphenyls (PCB's) and organochlorine pesticides

in surface runoff from two urban test catchments, Nat. Water Res. Inst.,
unpublished report, 30 pp., Burlington, Ontario, July, 1978.
10. Chambers, G. M. and C. H. Tottle, Evaluation of stormwater impoundments in

Winnipeg, a draft report, 103 pp., the City of Winnipeg, December, 1978.
11. Kronis, H., Characterization and treatment of snowmelt runoff from an urban
area, Canada-Ontario Agreement Res. Report Series, Ottawa, 1979 (in press).
12. Johnson, M. G. et al., Management information base and overview modeling, a

PLUARG technical report, International Joint Commission, Windsor, 1978.
13. Sullivan, R. H., W. D. Hurst, T. M. Kipp, J. P. Heaney, W. C. Huber and S.

Nix, Evaluation of the magnitude and significance of pollution loadings
from urban stormwater runoff in Ontario, Canada-Ontario Agreement Res.
Report No. 81, 183 pp., Ottawa, 1978.
14. Waller, D. H. and Z. Novak, Municipal pollutant loadings to the Great Lakes
from Ontario communities, Canada-Ontario Agreement Res. Report Series,
Ottawa, 1979 (in press).
15. Marsalek, J., Pollution due to urban runoff: unit loads and abatement
measures, a PLUARG technical report, International Joint Commission, Windsor,
1978.
16. Flatt, P. E. and C. D. D. Howard, Preliminary screening procedure for economic

storage-treatment trade-offs in stormwater control, in Proceedings
International Symposium on Urban Hydrology, Hydraulics and Sediment Control,
edited by D. T. Kao, Univ. of Kentucky Report UKY BU 114, December, 1977.
17. Perks, A. R., The development of storm drainage modeling in Canada, in Urban
Storm Drainage, edited by P. R. Helliwell, pp. 362-378, Pentech Press,
London, 1978.
18. Bessette, Crevier, Parent, Tanguay & Associates Ltd., ILLUDAS model study,
a draft report to Supply and Services Canada, Ottawa, 1979.
19. Watt, W. E. (Dept. of Civil Engrg., Queen's University, Kingston, Ont.),

personal communication.
20. Townsend, R. D. and J. R. Prins, Performance of model storm sewer junctions,

J.Hydraul.Div., Amer.Soc.Civil Engrs., 104(HY1), pp. 99-104, 1978.
21. Wisner, P. E. (Dept. of Civil Engrg., Univ. of Ottawa, Ottawa), Personal

communication.
22. Adams, B. J. and M. A. Qazi, Optimization of urban drainage systems, in

Proceedings Second Urban Water Resources Workshop, Univ. of Toronto,
Toronto, March 14-16, 1979.
32
23. Kirby, D. C. W., Economic selection of optimal design return period for
urban storm drainage networks, B.A.Sc, thesis, Dept. of Civil Engrg., Univ.
of Toronto, Toronto, April, 1979.
24. Charles Howard & Associates, Ltd., Analysis and use of urban rainfall data
in Canada, a report to Environment Canada, August, 1978.
25. Watt, W. E. and J. Marsalek, What the practising urban hydrologist needs
from the hydrometeorologist, pp. 15-23, in Preprint Volume, Second
Conference on Hydrometeorology, Oct. 25-27, 1977, Toronto, Amer.
Meteorological S o c , Boston.
26. James F. MacLaren Ltd., Report on design storm selection, the City of
Winnipeg, 24 pp., May, 1978.
27. Marsalek, J., Research on the design storm concept, ASCE UWRR Program Tech.
Memo. No. 33, Amer. Soc. Civil Engrs., 28 pp., NTIS: PB 291 936,
September, 1978.
28. Russell, S. 0., F. I. Kenning and G. J. Sunnell, Estimating design flows

for urban drainage, J.Hydraul.Div., Amer.Soc.Civil Engrs., 105(HY1),
pp. 43-52, 1979.
29. James, W., A pre- and post-processing program package for the Stormwater
Management Model, in Proceedings Stormwater Management Model (SWMM) Users'
Group Meeting November 13-14, 1978, edited by H. C. Torno, U.S. Envir. Prot.
Agency EPA 600/9-79-003. 238 pp., November, 1978.
30. Charles Howard & Associates, Ltd., Stormwater pollution control costs in
Canada, a draft report to the Central Mortgage and Housing Corporation,
Ottawa, August, 1977.
31. G. Moolamuttil (Dept. of Biology, Univ. of Waterloo, Waterloo, Ontario),

Personal .communication.
32. Kronis, H. and F. A. Tonelli, Evaluation of high rate screening devices,

Canada-Ontario Agreement Res. Report Series, Ottawa, 1979 (in press).
33. Nguyen, V., M. B. McPherson and J. Rousselle, Feasibility of storm tracking

for automatic control of combined sewer systems, ASCE UWRR Program Tech.
Memo. No. 35, Amer. Soc. Civil Engrs., 29 pp., NTIS: PB 294 611, November,
1978.
34. Stormwater management workshop, Conference Proceedings No. 4, Canada-

Ontario Agreement Series, Ottawa, 1977.
35. Modern concepts in urban drainage, Conference Proceedings No. 5, Canada-

Ontario Agreement Series, Ottawa, 1978.
36. Gore & Storrie Ltd., Report on stormwater management for the south urban
community, a report to the Regional Municipality of Ottawa-Carlton,
October, 1977.
33
37. James F. MacLaren Ltd., Stonnwater management technology systems
demonstration in the City of St. Thomas, a report to the Central
Mortgage and Housing Corporation, May, 1978.
38. Hore, R. C. and R. C. Ostry, Grand River, Ontario, Summary pilot watershed
report, a PLUARG technical report, International Joint Commission, Windsor,
April, 1978.
39. Manual of practice on urban drainage, Canada-Ontario Agreement Res. Report

Series, Ottawa, 1979 (in press).
40. Policy on urban drainage for the Province of Ontario, a discussion draft
document, Toronto, to be released in late 1979.
Section 4 Progress since 1976 in the United Kingdom
by M . J. Lowing*
Introduction
Since the 1976 release of our national report^) there have been a
number of significant events in urban hydrology in the U.K. The Department of
Environment/National Water Council Working Party on the Hydraulic Design of
Storm Sewers has continued to steer production of a new design manual for storm
sewerage and the contributory research - principally by the'Hydraulics Research
Station and the Institute of Hydrology - has been completed. There was an
International Workshop on inlet hydrograph models held at the Institute of
Hydrology(2) and, also in April 1978, an International Conference on Urban
Storm Drainage at the University of Southarapton(3). These meetings provided
opportunities for preliminary discussion of the proposed procedure to be
included in the new manual as well as enabling researchers and engineers to view
their own problems in the wider context.
This report is in three parts. The first two parts are intended to be
simple revisions of the "catchment research" and "hydrological models" sections
of the original report(l). The third part is a summary account of the proposed
design and analysis package to be published in the new design manual and currently
under test by selected users.
Recent Urban Catchment Research
The Hydraulics Research Station evaluation of flow measurement methods

in rivers, open channels and sewers has now been published(^).
In pipe, by water level measurement. The Water Research Centre

commissioned a U.S.-based firm to monitor flows in three sewered catchments.
They insert an appropriately sized ring into the pipe and monitor the pressure
on the surface of a streamlined moulding at the bottom of the ring. Checks on
assumed friction coefficients, for conversion of level to flow, are made by
velocity measurements with a probe-mounted electromagnetic gauge. The maximum
velocity is found and multiplied by 0.85 to give the mean velocity.
Dilution gauging. The Institute of Hydrology has ended its development

work on an automatic dilution gauging apparatus (5). The apparatus comprises
triggered (by rising water level) upstream injection of lithium chloride tracer
through.a magnetic valve at the base of a constant head (Mariotte) vessel,
upstream sampling for background concentrations, downstream sampling into a
maximum of 48 vacuum bottles at 2%-minute intervals, and a magnetic tape logging
system recording water levels and the various sampling actions. It had been
hoped that the system would be simple and reliable enough for routine use in
calibrating level/discharge relationships at a number of sites but the results
were disappointing. Problems of reliability and contamination were encountered
and were solved only to be replaced by others. Sometimes the system produced
*: Institute of Hydrology, Wallingford, Oxon, 0X10 8BB, United Kingdom.
35
flow hydrographs which were in good agreement with downstream check flumes but,
at other times, agreement was poor and for no obvious reason. The main
conclusion was that reliable automatic dilution gauging in storm sewers could
be achieved only by the dedicated involvement of an experienced technician
applying a high level of preventive maintenance.
Other methods of flow measurement. The Water Research Centre is

carrying out field trials of a British velocity meter which utilizes the Doppler
effect. The aim is to calibrate the meter from downstream flume-measured flows
and then to use it when the flume is overtopped by high flows.
Rainfall Aspects
The Meteorological Office has completed its storm movement project

based on dense networks of recording raingauges (6). It seemed that the velocity
of rain cells was not predictable from wind velocity, and that it would be some
time before reliable predictions of the frequency distributions of speed and
direction of storms could be made. It was independently concluded that-storm
movement was not a significant factor in storm sewer design (although it could be
important in reproduction of some individual recorded events). In the space and
time scales of runoff from most U.K. sewered catchments, it seems satisfactory to
continue with the usual assumption of a stationary (i.e., areally uniform) rain
field.
The use of radar for warning of flood runoff from urban areas and in
guiding the operation of pumping stations in low-lying coastal areas is being
studied by the Water Research CentreC?). Developments in the use of weather
radars to provide practical short terra quantitative forecasts of rainfall have
been made by the Meteorological Office's Radar Research Laboratory(8).
Gauged Urban Catchments
The Water Research Centre (Medmenham Laboratory) has begun (August '78)
to monitor a storm sewer on the Chelmsley Wood catchment (1000 ha) in Birmingham.
This is adjacent to catchment No. 50 on the previously published map(l). Flow is
monitored by way of depth plus an assumed friction coefficient. Depth is
recorded at two-minute intervals on a Microdata logger. The WRC is also gauging
a combined sewer in Colne North Valley (500 ha) to the north of Burnley
(approximately midway between Nos 1 and 2 on the previously published map(l)).
A glass-fibre flume was inserted in an opened section of the sewer, and depth
is measured by sonar reflection. This is the site where the Doppler meter
mentioned above is to be investigated.
Small Sub-Catchments Gauged at Inlet Point to Pipe System
The Institute of Hydrology gully-meter* sites at Bracknell and

Stevenage provided data through the summers of 1976-78. A new site was
instrumented by the Institute in Wallingford in 1978 and will be continued.
Also, Trent Polytechnic has three experimental sites in Nottingham.
The study on the laboratory catchment at Imperial College was

completed(9). A selection of storms was applied to various combinations of
area (up to 48 m^) and slope of concrete surface to yield a total of 140
hydrographs. • ' •
*: in U.S. parlance, a gully is an inlet.
36
Water Quality Studies
The previously reported Trent Polytechnic (Nottingham) work led to the

development of a model of total suspended solids concentrations (10).
Noted is
the importance of runoff from unpaved areas iir producing abnormally high loads
after high intensity summer storms. Other workers at Nottingham have studied(H)
the links between the quality of the "first flush" runoff and that of the stored
water in the gully pots. Blaming the gully pot for acting as a concentrating
point for pollution due to human activity during dry periods, they remark on the
apparent advantage of having fewer gullies.
The work at Middlesex Polytechnic has continued(12) with increasing
emphasis being given to the monitoring of hydrocarbons (oils, etc.) and heavy
metals'"). The Water Research Centre (Stevenage Laboratory) work on the
Shephall catchment was reported(14) with some interestingly different conclusions
from those of Tucker and Mortimer(lO) regarding the significance of runoff from
unpaved areas and of the length of the antecedent dry period. The differences
emphasize the need for a much larger body of data from a variety of catchments
before it will be possible, reliably, to predict likely pollution characteristics
from those of the catchment and the storm in an unraonitored catchment.
Data Collection and Processing Systems
All the U.K. urban hydrological data collected or collated (from

earlier studies) between 1974 and 1979 by the Institute of Hydrology, and used
by them in model development and calibration, is described in detail by Makin
and KiddC 15 ).
Recent Urban Hydrological Modeling
The TRRL Method
Although the Transport and Road Research Laboratory (TRRL) method

itself has not been changed, the revised design guide (Ref. 13 of the original
report'-*-') includes new rainfall information and enables engineers to explore
the effects of alternative design storms. Others have sought to make the method
more easily used on small in-house computers. Martin and King,(16) f or example,
describe enhanced handling procedures for the input and output data.
Other Methods Developed in the U.K.
The method due to Sarginson and Nussey (Refs. 10, 16, and 17 in the
earlier report(l)) has been developed further. They have now reverted to the
concept of two unequal linear reservoirs to represent overland flow and pipeflow.
They suggest that the overland flow storage constant should be 20% of the pipeflow
storage constant and they present a prediction equation for the sum of the two
values in terms of catchment area and s lope(17).
As stated in the original report, (D a single linear reservoir was used

by the TRRL to model unpaved area runoff in tropical conditions. The full
reference is now available'18) and the work was also described at the
Southampton conference (19).
The models developed at the Institute of Hydrology comprise a

prediction equation for percentage runoff from the whole catchment,
(20,21) a
37
depression storage rood
ei(21,22)
and a non-linear reservoir model for the
above-ground routing(21,23,24). ihe calibration of the depression storage
and non-linear reservoir models was aided by inclusion of data from elsewhere
in Europe which were assembled at Wallingford for the purposes of an International
Workshop^). Various models of the aboveground process were examined by the
Workshop but the only clear conclusion was that the non-linear models were better,
as a group, than the linear ones in reproducing observed hydrographs. The concept
of a two or three minute time of entry, as currently used by the rational and TRKL
methods in the U.K., was shown to be the least acceptable of all the methods
examined. Calibration of the non-linear reservoir was also helped by the
laboratory catchment data, (9)
which gave results for very small areas.
A model for pipe surcharging has been completed at the Hydraulics
Research Station(25). Together with the hydrological models' mentioned above,
it forms part of the proposed design and simulation procedure described later in
this report.
The Water Research Centre (Stevenage) work on suspended solids
concentrât ion(14) was utilized by Price and Mance(26) to develop a runoff quality
model that was again linked to the hydrological and hydraulic models referred to
above.
The Urbanized Catchment
Unit hydrograph changes following urbanization of North London

catchments have been further described by Hal 1(27).
At the Institute of Hydrology, Packman has extended his studies to

consider partly urbanized areas^28)_ Also, using the Silk Stream catchment in
North London, he explored the possibilities of using unit hydrographs for
subcatchments and doing river routing by a Muskingum-Cunge procedure as described
by PriceC29).
The previous recommendations from the Institute of Hydrology (Ref. 28

in the earlier report^)) for predicting the broad effects of urbanization on the
mean annual flood, the T-year flood and the hydrograph shape, have been slightly
revised and formalized(30; as a contribution to a forthcoming publicátionOl) by
the Construction Industry Research and Information Association. The likely
contents of this publication have been previewed by Hall and others
(32).
A Manual for Urban Storm Sewerage Design and Analysis
The final part of this report outlines the new design and analysis
procedure to be recommended by the Department of Environment/National Water
Council Working Party on the Hydraulic Design of Storm Sewers. The Working Party
has been monitoring the relevant research since 1974 and the manual which
describes the procedure and which represents the culmination of the Committee's
efforts is in an advanced stage of drafting. The methods themselves are under-
going extensive testing before public release but their general structure is
known and can be described here.
Background of the New Procedure
For the past 15 years, the design and analysis of storm drainage
systems in the U.K. have been dominated by two methods: the long established
38
rational method; and the TRRL hydrograph method which was issued in 1963 after
a research programme lasting several years. The rational method only provides
a value of the peak discharge and takes no account of the variability of storm
rainfalls. The TRRL method provides a complete hydrograph of response to any
rainstorm. Both methods are relatively simple (although the TRRL method is
still sometimes misunderstood), inexpensive to run, and with comparatively few
input data requirements.
As engineers became familiar with the TRRL method, an awareness grew

that the basic principles of the model possessed flaws which, although not
always leading to incorrect answers, made the method unattractive to those who
were keen to see a better appreciation of physical processes. The problem was
to do this without making any new model more difficult or time consuming for the
engineer to use.
The move towards a new procedure gathered momentum in 1974 when the
above-mentioned Government Working Party was formed and took an immediate
interest in the research programmes of the Institute of Hydrology and the
Hydraulics Research Station. Better rainfall statistics had been produced by
the Meteorological Office for the Natural Environment Research Council's massive
Flood Studies Report, and others were busy with research into water quality models,
sewer construction costs, and flood damage assessment.
By April 1979 the contributory research had been completed and the
results collated to form the basis of a new procedure. This was first tested by
those users represented on the Working Party, which is currently (June 1979)
about to begin a period of testing by external users.
The new procedure takes account of surface storage and attenuation,

flood wave movement in pipes, flows in surcharged pipes, and the selection of
design conditions according to economic criteria of costs and benefits. But not
all those aspects are needed all the time; and the choice of method (and hence
design costs) must be tailored to the size and needs of the scheme. The
recommended procedure therefore contains the following options for the design of
pipe networks:
- a simple manual method, based on the rational formula but modified

in accordance with recent U.K. research;
- a computerized hydrograph method (the Wallingford hydrograph method)

with limited data requirements, incorporating separate models of the
above-ground and pipe-flow phases of urban runoff (this method will
be appropriate in the majority of applications);
- a computerized method incorporating the optimized design of pipe

depth and gradient; and
- a computerized method to examine the performance of an existing

system or proposed design when operating under surcharged conditions.
The second of these methods is described in some detail next because it is

central to the procedure as a whole.
39
The Wallingford Hydrograph Method of Sewer Design
The Working Party sees the Wallingford hydrograph method as a natural

successor to the TRRL method in that:
- the input data requirements are broadly similar;
- the new method is firmly based on hydrological data and uses

conventional hydraulic principles for pipe flow routing; and
- the combination of method and design conditions is intended to

produce a peak discharge with prescribed return period (existing
methods specify a return period for the rainfall).
The structure of the method as it would be used in design is outlined in Table 1.

The method can also be used to simulate existing systems and, by using the
pressurized flow equation when discharge exceeds pipe capacity, it can make a
partial allowance (it treats each pipe in isolation) for the effects of
surcharging. It is anticipated that normal use of the design procedure as a whole
would involve subsequent use of the fourth method listed above. Storms which are
more severe than the design storm would be input and the performance of the
designed system examined under the generally surcharged condition which would
obtain. The fourth method therefore includes a proper analysis of surcharged flow
involving interaction in the sequence of pipes (25). it also enables use of a
flood damage model^-33-' where, given data relating to the number, type, and ground-
floor levels of affected properties, it is possible to calculate the tangible
damage costs.
The third method can use either the basic hydrograph method or the
modified rational method but, in addition, allows the optimum pipe depth and
gradient to be designated. Extra input data are required to establish the
permitted range of manhole depths and cover and to estimate the construction
costs'-^). Th e cost calculations are made on the basis of a model developed at
the Transport and Road Research Laboratory(35)#
The effects of sewer ancillaries such as tanks, pumping stations, and

overflows can be simulated by the basic method.
Attention is next given to those aspects of the new procedure which

relate specifically to the twin themes of catchment research and hydrological
modeling which were linked in the title of the earlier national report^).
The Use of UK Catchment Data to Calibrate New Models of Urban Runoff
Reference is made in Table 1 to several research activities which

provided data and model calibration. From the hydrological point of view, the
two most significant are the percentage runoff study and the inlet hydrograph
studyt21).
The percentage runoff study was confined to several catchments providing

reliable flow data at the outfall. The catchments have been listed(l) and a
complete account of the data is available(15). 510 events from 17 catchments
contributed to the final form of a regression equation relating the percentage
runoff (PR) to proportion of impervious area (PIMP), soil type (SOIL), and
antecedent condition (UCWI):
40
TABLE 1 THE WALLINGFORD HYDROGRAPH METHOD - DESIGN USE
What the User Provides What the\ Method Does Background Research
as Input Data
Specifies design return
period for flows.
Looks up key rainfall Produces design storm Rainfall statistics based
statistics from profiles for 15, 30, 60 on the UK Flood Studies
published maps and and 120 minute Report (Ref..14 of
tables. durations. national report'1''); and
probability relationships
between rainfall quantity,
duration, profile, ante-
cedent conditions, and
peak discharge based on
simulation studies at The
Institute of Hydrology.(36)
Gives figures for Calculates, for the Multiple regression studies
proportion of design antecedent relating observed runoff
impervious area and condition (varies with volumes to catchment and
soil type (from annual average rainstorm characteristics(20>21)
published map). fall^ 36 )), a value for
average percentage run-
off (i.e. runoff
coefficient x 100%).
Estimates proportions Calculates individual
of paved, roof, and percentage runoff
pervious surfaces values for the three
within the catchment surface types.
as a whole.
Estimates proportions of Calculates depression Depression storage from
paved, roof, and storage and deducts it research findings of UK and
pervious surface within from beginning of design Swedish investigators
subcatchments and allo- storm profiles. (22).
cates each subcatchment Calculates the notional
to one of three slope contributing area of each
classes and one of three surface type within each
area classes. subcatchment.
For each of the nine Model research and develop-
slopes/area combinations ment at the Institute of
and for the special Hydrology^21).
pitched roofs case,
assigns an appropriate
value of the non-linear
reservoir storage
parameter, K, and cal-
culates a 'standard
hydrograph' (see
Figure 1).
For each subcatchment in
turn, calculates the
inlet hydrograph from the
standard hydrographs and
contributing area values.
(Continued)
41
TABLE 1 (Continued)
What the User Provides What the Method Does Background Research
as Input Data
Specifies the pipe con- Accomplishes pipe flow Method research and
nections, pipe lengths routing of the inlet development at the
and gradients, dry hydrographs by the Hydraulics Research
weather flow, pipe Muskingum method as Station(37.38)#
roughness and manhole modified by Cunge.
headloss coefficients.
In design of system Replaces a group of pipes Calibrated from
additions, to simulate a by an equivalent application of the full
number of subcatchments, 'sewered subcatchment' model(39).
optionally specifies use model.
of a simpler 'lumped' For each pipe in turn,
model. specifies the next
largest commercially
available diameter
above that needed to
accommodate predicted
peak discharge from the
worst of the four storm
durations.
42
PR = 0.829 PIMP + 25.0 SOIL + 0.078 UCWI - 20.7 .
Many combinations of the variables were investigated^^) before final calibration(21).

Note that storm size was not a significant factor in explaining variability of PR.
Note also that soil type was a significant variable (SOIL is a mapped variable for
the UK, having been of key importance to rural catchment runoff prediction as
described in the UK Flood Studies Report). This is of interest because it is often
contended that, in the UK, it is only runoff from impervious areas which is of
importance to urban drainage design. Whether the apparent importance of the soil
type should in fact be attributed to runoff from unpaved areas, to a difference
in infiltration rates through nominally 'impervious1 areas, or to infiltration/
exfiltration through sewer joints, must remain a matter for speculation. There
may, of course, be no cause and effect relationship at all, with both PR and SOIL
being correlated with some other factor. However, the inclusion as a significant
variable of the antecedent condition index supports the idea of pervious area
runoff being involved.
The inlet hydrograph study was based mainly on UK catchment data collected
specifically for the purpose, (15) although the International Workshop held at the
Institute of Hydrology(2) enabled a wider variety of European data to be used in
calibration and testing of various models. The Imperial College laboratory catchment
also provided dataw). The model chosen to describe the attenuation of a net
rainfall profile by the above-ground routing process is the single non-linear
reservoir:
S = kqn .
Studies showed*-2^' that k and n were highly interdependent when optimized on the
available data.
Events from 28 subcatchments contributed to the evaluation of optimum

parameter values. Because the global optimum of n was near to 0.67, and this
value has logical connections with simple hydraulic theory, it was fixed before the
derivation of k. Average k values for each subcatchment were then related to
catchment characteristics:
k = 0.051 SLOPE -0 - 23 PAPG0-23
where PAPG is the paved area per gully. Although this prediction equation enables
any combination of slope and area to be used, it was thought adequate, for design
or simulation purposes, to coarsen the discrimination of these variables and, as
explained in Table 1, all subcatchments are classified into one of nine slope/area
combinations. Figure 1 shows an example of a set of the nine 'standard hydrographs'
and the range of slopes and areas which are typical of UK conditions.
The model of surface-storage attenuation underlines the fairly obvious

fact that the rainfall profile is more attenuated before it enters the pipe system
when there are fewer gullies. Although the diameter of the smallest pipe size might
not be reduced below that specified for maintenance reasons, the point where there
is an increase to the next largest size could well be further down the system and"
necessary lengths of the smallest pipe size are thereby reduced.
Thus, there is another argument (to go with the one based on the action
of gully pots as pollution concentrators) for fewer gullies. American readers are
reminded that UK gullies are much smaller and more closely spaced than those in
the USA.
43
1=
N
E o
UJ O o
10
O o a
ui
< Au,i -
-JC5 5 o. CO
o. j g (3
< < o X
a > <
a. Q.
<
oc
a
o
E oc
o
o
Ï3 Q
CM >-
E X
o
O o O
CO
DC o
_ _ o LU
CL z
So 3 o ce
<
O < Q LU Û
ÍM > <
oc
< <
a
a z
LU
¡S
(4
E
W
E
•o
04
I CO
U.
oc
jO o
j ° a s
ui
S a. co
CO 2 _ O
< < O X
a. > <
10
* CO m
e CM
IO V CM i*
CO Q UJ O CM Q
UJ
a. A
UJ
V UJ
UJ UJ UJ I-
UJ a 3 a -
is »-
a.
ÛJ UJ H
< O -J O
IJO — CO UJ _l ZJ < Q
—
I _I < Q
U_ co > <
CO CO X ÛV g« UJ
UJ ¡ * ^ 0 oc
S N >< C3
LU
CL
O
c/>
44
Acknowledgments
The author would like to thank his colleagues and contacts for the
helpful information supplied for this summary compilation. It was prepared at
the Institute of Hydrology and supported by the UK Department of the Environment
under Contract No. DGR/480/38.
References
1. Lowing, M. J., 1976, Urban hydrological modelling and catchment research in

the United Kingdom, ASCE UWRR Program Tech. Memo. No. IHP-4.
2. Kidd, C. H. R., 1978b, Rainfall runoff processes over urban surfaces,

Proceedings of an International Workshop, Report 53, Institute of Hydrology,
Wallingford, UK.
3. Helliwell, P. R., (Ed.), 1978, Urban Storm Drainage, Proceedings of an

International Conference at Southampton University, Pentech Press, London.
4. Charlton, F., 1978, Measuring flow in open channels: a review of methods,

Report 75, CIRIA, 6 Storey's Gate, London.
5. Harvey, R. A., Kidd, C. H. R., and Lowing, M. J., 1979, Automatic dilution
gauging in storm sewers: final report, Institute of Hydrology, Wallingford,
UK (in press).
6. Shearman, R. J., 1977, The speed and direction of movement of storm rainfall
patterns, Unpublished report from Met 08b, Meteorological Office, Bracknell,
UK.
7. Cole, J. A., and Evans, G. P., 1978, Potential for using storm runoff warnings
in the operation of pumped sewers in coastal towns, Proc. Int. Conf. on Urban
Storm Drainage (Ed. by. Helliwell), April 1978, Southampton 679-685, Pentech
Press, Plymouth, UK.
8. Collier, C. G., and Browning, K. A., 1978, The short period rainfall
forecasting pilot project and its relevance to storm sewer operation, Proc.
of Int. Conf. on Urban Storm Drainage (Ed. by Helliwell), April 1978,
Southampton, 42-50, Pentech Press, Plymouth, UK.
9. Johnston, P. M., Wing, R. D., and Pavlov, R., 1978, Overland flow on urban
surfaces: Final Report to NERC (contract F60/C1/12), Imperial College of
Science and Technology, London.
10. Tucker, C. J. G., and Mortimer, G. H., 1978, The generation of suspended
solids load in urban stormwater, Proc. of Int. Conf. on Urban Storm Drainage
(Ed. by Helliwell), April 1978, Southampton, 695-704, Pentech Press, Plymouth,
UK.
11. Fletcher, I. J., Pratt, C. J., and Elliott, G. E. P., 1978,, An assessment of
the importance of roadside gully pots in determining the quality of storm
water runoff, Proc. Int. Conf. on Urban Storm Drainage (Ed. by Helliwell),
April 1978, Southampton, 586-602, Pentech Press, Plymouth, UK.
12. Ellis, J. B., 1976, Sediments and water quality of urban stormwater, Water
Services, 730-734.
45
13. Ellis, J. B., 1979, Heavy Metal Incidence in Urban Stormwater, in
Management of Heavy Metals in the Environment (Ed. by Perry).
14. Manee, G., and Harman, M. M. I., 1978, The quality of urban storm water
runoff, Proc. Int. Conf. on Urban Storm Drainage (Ed. by Helliwell),
15. Makin, I., and Kidd, C. H. R., 1979, Urban hydrology project: collection
and archive of UK hydrological data, Report 59, Institute of Hydrology,
Wallingford, UK.
16. Martin, C , and King, D., 1978, Improvements in the TRRL hydrograph program,
Proc. Int. Conf. on Urban Storm Drainage, (Ed. by Helliwell), April 1978,
17. Nussey, B. B., and Sarginson, E. J., 1978, A linear reservoir model for
urban runoff, Proc. Int. Conf. on Urban Storm Drainage (Ed. by Helliwell),
18. Watkins, L. H., 1976, The TRRL hydrograph method of urban sewer design
adapted for tropical conditions, Proc. Instn. Civ. Engrs., Part 2, 61,
539-566.
19. Watkins, L. H., and Fiddes, D., 1978, The design of surface water sewer
systems in the tropics, Proc. Int. Conf. on Urban Storm Drainage (Ed. by
Helliwell), April 1978, Southampton, 243-255, Pentech Press, Plymouth, UK.
20. Stoneham, S., and Kidd, C H. R., 1977, The prediction of runoff volume
from fully-sewered catchments, Report 41, Institute of Hydrology, Wallingford,
UK.
21. Kidd, C. H. R., and Lowing, M. J., 1979, The Wallingford Urban Subcatchment
model, Report 60, Institute of Hydrology, Wallingford, UK.
22. Falk, J., and Kidd, C: H. R., 1979, Depression storage on paved surfaces,
Journal of Hydrology, in preparation.
23. Kidd, C. H. R., 1976, A non-linear urban runoff model, Report 31, Institute
of Hydrology, Wallingford, UK.
24. Kidd, C\ H. R., 1978a, A calibrated model for the simulation of the inlet
hydrograph for fully sewered catchments, Proc. Int. Conf. on Urban Storm
Drainage (Ed. by Helliwell), April 1978, Southampton, 172-186, Pentech Press,
Plymouth, UK.
25. Bettess, R., Pitfield, R. A., and Price, R. K., 1978, A surcharging model
for storm sewer systems, Proc. Int. Conf. on Urban Storm Drainage (Ed. by
Helliwell), April 1978, Southampton, 306-316, Pentech Press, Plymouth, UK.
26. Price, R. K., and Manee, B., 1978, A suspended solids model for Storm Water
Runoff, Proc. Int. Conf. on Urban Drainage (Ed. by Helliwell), April 1978,
27. Hall, M. J., 1977, The effect of urbanisation on storm runoff from two
catchment areas in North London, Internat. Assoc, of Sei. Hydrol.,
Pubn. No. 108, 485-500.
46
28. Packman, J. C , 1978, Flood simulation in partly urbanised catchments,
Proc. Int. Conf. on Urban Drainage (Ed. by Helliwell), April 1978,
29. Price, R. K., 1977, FLOUT - A River Catchment Flood Model, Report IT68,
Hydraulics Research Station, Wallingford, UK.
30. Anonymous, 1979, Flood Studies Supplementary Note No. 5, Institute of

Hydrology, Wallingford, UK.
31. CIRIA, Design and Management of River Systems to Cater for Runoff from
Catchments with a Degree of Urbanisation, Construction Industry Research
and Information Association, London (in preparation).
32. Hall, M. J., Prus-Chacinski, T. M., and Riddell, K. J., 1978, Some aspects
of the design of stormwater balancing ponds for catchment areas subject to
urbanisation, Proc. Int. Conf. on Urban Drainage (Ed. by Helliwell), April
1978, Southampton, 421-433, Pentech Press, Plymouth, UK.
33. Chatterton, J. B. , and Penning-Rowsell, E. C , 1978, The benefits of urban

storm drainage: computer modelling and standardised assessment techniques,
34. Price, R. K., 1978, Design of storm sewers for minimum construction cost,
35. Farrar, D. M. , and Colyer, P. J., 1978, A procedure for calculating the
cost of stormwater sewer construction, Proc. Int. Conf. on Urban Drainage
(Ed. by Helliwell), April 1978, Southampton, 627-635, Pentech Press,
Plymouth, UK.
36. Kidd, C. H. R., and Packman, J. C , 1979, Selection of design storm and
antecedent condition for urban drainage design, Report 61, Institute of
Hydrology, Wallingford, UK.
37. Price, R. K., and Kidd, C. H. R., 1978, A design and simulation method for
storm sewers, Proc. Int. Conf. on Urban Drainage (Ed. by Helliwell), April
1978, Southampton, 327-337, Pentech Press, Plymouth, UK.
38. Bettess, R., and Price, R. K., 1976, Comparison of numerical methods for
routing flow along a pipe, Report IT162, Hydraulics Research Station,
Wallingford, UK.
39. Price, G. A., Packman, J. C , and Kidd, C. H. R., 1979, A simplified model
for sewered catchments, Report 62, Institute of Hydrology, Wallingford, UK.
47
Section 5 Progress since 1976 in the
Federal Republic of Germany
by Herbert Massing*
Introduction
As mentioned in the 1976 FRG Technical Memorandum,(1) about 40

institutions and experts contributed information to that memorandum, and a
small group of hydrologists cooperated in the final wording. After the 1976
Workshop at Lund, this group constituted itself as the FRG IHP Working Group
to continue the work in urban hydrology on the international and national levels.
The FRG IHP Working Group has set itself the following tasks -
Following up on the results from the IHF Workshops at Lund, Sweden, and
Zandvoort, Netherlands, by, for example:
- publication of a shortened German-language version of the Lund Workshop
paper ;
- cooperation of the IHP Working Group with the FRG MAB Working Group on
"Ecological Aspects of the Urban System"; and
- participation in the realisation of the proposal submitted to the
Director General of Unesco by the Zandvoort IHP Workshop.
Continuation of Unesco-IHP activities concerning the project "Urban Hydrology,"

with, for example:
- proposals concerning activities within the framework of the second phase
of the IHP, and to be submitted through the National IHP Committee of
the FRG; and
- cooperation in the existing international IHP group on urban hydrology,
including this progress report.
Dealing with questions of urban hydrology on the national level, for example:
- preparation of a leaflet on urban hydrology for city planners (non-
hydrologists); and
- listing of occasions on which urban administrations should procure data
and other information on the prevailing state of surface waters.
International activities, for example:

- exchange of information and expertise, similar to the May, 1979, SWMM
Meeting and Ecole Polytechnique Design Storm Seminar held in Montreal;
and
- impulses for bilateral and multilateral work.
The FRG IHP Working Group has proposed that, over the next several years, the
entire range of problems falling under the heading "Urban Hydrology" be especially
promoted at the national level within the framework of the second phase of the
IHP.
*: Chairman, Working Group on Effects of Urbanization and Industrialization on

the Hydròlogical Regime and on Water Quality of the IHP National Committee
of the FRG, Landesanstalt ftlr Wasser und Abfall, Nordrhein-Westfalen, D-4000
Dusseldorf, FRG.
49
Some of the projects and activities begun or completed since the
publication of the 1976 Technical Memorandum are presented below. Due to lack
of space or time, only brief description can be given. The following members
of the FRG IHP Working Group contributed to the report or assisted in drafting
it: W. F. Geiger, Munich; F. W. Renz, Essen; H. Schiller, Munich; and K. W.
Schulze, Hamburg.
Hydrological Investigations in Urban Drainage Areas
Water Management in Urban Areas and Their Surroundings, with Special

Attention to Social Fallows
The development of urban areas in the FRG is still in progress. A

continued growth of the population in densely populated areas and a population
decline in rural areas can be observed. In addition, there is a trend to move
out of the inner cities. All large cities have registered a population decline
during the last 10 to 15 years. People move to the urban fringe zones and their
agricultural surroundings, where urban problems of another kind arise. One of
these problems is the social fallows, i.e., farmland no longer tilled but not to
be confused with land temporarily lying fallow in the course of crop rotation.
(2)
Social fallows are a phenomenon which, ever since the 1950's, has spread widely
in the heavily industrialized countries of Central Europe; and their causes are
mostly of an economic and social nature.
The effects of urbanization and industrialization on the water cycle
and all its components are very pronounced, reaching far beyond the city limits
into the agricultural surroundings which, in turn, have a compensatory influence
in terms of climate, water management, ecology, and public health. The continued
areal expansion of the urban population imparts growing importance to the
compensatory function of the agricultural zones.
In recent years, a large number of investigations have been undertaken
on the effects of fallow areas on water management.(3) The results are still
incomplete, but already show that the effects are important (air movement,
temperature, precipitation, évapotranspiration, runoff, water quality, etc.). It
is an urgent necessity and in the best interests of the densely populated areas
that the function of the surroundings as a natural resource benefiting the city
in terras of water management be preserved.(4) This is first of all a question of
the exploration of urban drainage areas in the broadest sense of the word, but it
may also have to include the work required for the preservation of fallow areas
(mulching, clearing, planting, etc.).
Hydrological and Ecological Aspects of Dredging Pool Development and

Utilization
Intensified building and road construction activities in recent decades

have led to a considerably increased demand for gravel and sand. The extraction
of gravel and sand has become an important factor in the national economy, and
affects also the environmental conditions of settlement areas. (5)
At the end of
1975, there existed about 3,900 dredging pools in the FRG with a total water
surface of approximately 25,400 ha. Additional groundwater exposure at the rate
of 1,200 ha/a is expected.
The dredging pools usually resulting from gravel and sand extraction
affect the water budget and the ecosystem in a number of ways. The direct
50
contact with the groundwater during and after extraction influences runoff and
water quality not only in the extraction area but also in the surrounding aquifer.
The extent of these effects depends on climate, subsoil conditions, the possible
influence of surface inflow, and human influences. For example, the effect of
the dredging pool on the water balance of the extraction area is that of an
additional water diversion (long-term mean of excess evaporation is between 100
mm/a in the Alpine foreland and 240 mm/a in the Upper Rhine valley).(") The
constituents and energy budget of the dredging pool is similar to that of a
natural lake, with colonization, seasonal periodicity and aging or eutrophication.
A close interrelationship exists between the dredging pool and the surrounding
groundwater body.
On the other hand, after the termination of extraction activities,

dredging pools offer the possibility of subsequent utilization, and this may be
of considerable importance to trade and industry in neighboring settlement areas.
In principle, dredging pools should be as large and as deep as possible, because
a few large pools are better than a large number of small ones. The bank line
should be designed on an irregular pattern providing for ample diversity.
Dredging pools with less than a 2 m water depth should not be permitted for
limnological reasons (prevention of premature aging) and because they entail the
danger of "wild garbage dumping".
(6)
Regular inflow and throughflow of surface
waters and the introduction of waste water, even if biologically treated, cannot
be permitted. Use as a bathing lake, even though seasonally limited, is the most
intensive and in respect of structure and appearance the most demanding form of
utilization.
(5)
Water Supply of Urban Areas and Artificial Groundwater Recharge
Urban water supply, in particular public drinking water supply systems,
should be based on groundwater because, in most cases, it is of far higher and
more constant quality than surface water, and the overlying soil layers and
surface formations provide far better protection from unfavorable influences.
However, in most urban areas of the FRG this is not possible, because the ground-
water resources are insufficient to meet the high water demand per unit area. They
must, therefore, be increased by artificial recharge with water drawn from more
distant aquifers, or by recourse to the use of surface water. In North Rhine-
Westphalia, the most densely populated and most intensively industrialized of the
Federal States, the public water demand is met to the extent of 43% by natural
groundwater, of 44Ï by infiltrated surface water and induced streambed infiltration,
and of 137» by surface water (reservoirs).
The great importance attaching to artificial groundwater recharge in
urban areas was manifested by an impressive international symposium on this
subject. held from 14 to 18 May, 1979, at Dortmund (Ruhr Region). Besides the
competent FRG organizations, UNESCO, WHO, IAHS, WRC and other international
organizations participated in the symposium, and 524 experts from 30 countries
were in attendance. A typical example of the more than 60 papers delivered on
that occasion is the paper by L. Blasyw) o n the projected airport 20 km
northeast of Munich where permanent lowering of the groundwater level by way of
drain trenches is planned. The drain water will be diverted to the north boundary
of the airport where it will re-infiltrate the subsoil to counteract the
groundwater deficit in the Munich area.
51
Diversion of Altmtlhl and Danube River Water into the Regnitz and
Main Basin
Water diversion for the purpose of procuring public drinking water,

process water for industry, and dilution water for the heavily polluted surface
waters of urban areas is practiced in many parts of the FRG. The most important
of these measures is the diversion of Altmuhl and Danube river water into the
residential zones and industrial agglomeration areas along the rivers Regnitz
and Main.
In the drainage area of the Danube, water is in abundant supply.

Particularly attractive is the idea of diverting water from the Danube region
into the Main region to ensure a sufficient water supply for the conglomeration
areas on the rivers Regnitz (Nürnberg, Bamberg) and Main (Schweinfurt, Wiirzburg,
Aschaffenburg, Frankfurt). This is especially true in the area of Nurnberg-
Fürth-Erlangen, where irregressive tendencies in the structure of population and
industry have been experienced due to water becoming increasingly a constraining
factor(°) over the last decades. For many years now, work on a new waterway
interconnecting the rivers Rhine, Main and Danube, has been going on. At present,
the last uncompleted section of the Main-Danube Canal, which crosses the water
divide, is in the construction stage. The canal is scheduled to be completed
around 1985. The navigation canal will also be used for diverting Danube water
into the Regnitz/Main/Rhine river system.(9)
Water Resources Research in Urban Hydrology
Water resources research in the FRG is sponsored by a number of federal

and state agencies, research foundations and societies. Major funds for water
resources research are raised by the "Deutsche Forschungsgemeinschaft" (DFG), the
German Research Society. This society was established in 1949 and is a supra-
regional organization for the promotion of German science with complete autonomy
in all scientific matters. (10)
In general, funds for the German Research Society
are provided by the Federal Government and the eleven States in equal shares.
Priority research programs and projects have been promoted by the German
Research Society for research problems of such complex nature that their solution
does not seem possible through isolated individual accomplishment. For instance,
research on the effects of increasing urbanization and industrialization on the
environment cannot be realized without interdisciplinary cooperation. One of
these priority research programs is the "Sonderforschungsbereiche" (SFB), special
research areas at selected universities, promoted since 1967. The special
research areas involve scientists of different disciplines from universities and
other institutions for carrying out long-term coordinated work in specific fields
of research. Fields of research are proposed and pursued by the universities.
Characteristics of special research areas are interdisciplinary cooperation,
broad scope and continuity of research work. The implementation procedure
follows three basic steps: establishment of special research areas based upon
university recommendations; application by the universities for financial support
from the German Research Society; and review and selection of university
proposals by the German Research Society.
Problems directly related to urban hydrology are investigated within

the framework of two special research areas, SFB 81 and SFB 150, out of a total
of six such special research areas related to water resources that were funded
in recent years : (11,12)
52
SFB 81 - Qualitative and quantitative runoff in natural
and artificial waterways, with special
consideration given to alpine and subalpine
conditions; and
SFB 150 - Water conservation and land use.
SFB 150 was financed from 1973 to 1976, and was carried out by the
Technical University of Braunschweig. Research started within the framework of
the International Hydrological Decade and was continued under the program on
irrigation problems in arid regions. The representative value of individual
hydrologie measurements and of observation networks was investigated for
different climatological conditions. Special attention was paid to qualitative
aspects of runoff from rural areas under the impact of increasing urbanization.
SFB 81 was initiated in 1974, and is handled by different Bavarian

state authorities. Of the ten individual projects within this special research
area, the projects that directly pertain to urban hydrology are:^13) A - 3 ,
definition of precipitation properties to describe design storms for sewer system
planning; A-4, quantity and quality of storm runoff from urban areas; and A-6,
quality and quantity of infiltration from urbanized areas.
Project A-3 started in 1978, and is intended to define a statistically

significant design storm for sewer system planning, which will be variable in
space and time. To date, existing design storms have been analyzed and continuous
precipitation records for 27 years from the Munich area were digitized in
sufficient detail to allow for minute interpretation. At present, these
precipitation records are analyzed for frequency and duration of exceedence of
properties correlated with combined sewer overflow occurrences, whereby
precipitation and overflows are linked by a mathematical model.
Project A-4 started in 1974, and focused on a very dense sequence of

stormwater runoff quantity and quality data to allow stochastic data analysis and
mathematical description of the basic processes involved. Data have been collected
from two catchments since 1976. On a smaller catchment of 25 ha, mainly nutrients
contained in the surface runoff are investigated. The larger catchment of 542 ha
yields BOD, COD, TOC, settleable solids, nitrogen, and phosphorus data at a combined
sewer overflow. Data collection continues. To cope with the immense amount of data,
all measurements are directly recorded and checked by a computer system. Data
collection and mathematical modeling of the quantitative aspects of stormwater
runoff have now been completed. The mathematical formulation of the pollutant
washoff and transport processes is presently in the development stage. Parallel
to this effort, the collected data are being stochastically analyzed.
Project A-6 commenced only recently. It is intended to investigate the

chemical changes of infiltrated water to describe the quality of groundwater
recharge and flow within the urbanized area of Munich. Initially, groundwater
data are being collected for the 542 ha area mentioned above in connection with
project A-4.
Because interdisciplinary cooperation is of great importance in special

research areas, a colloquium on urban hydrology was convened. The two-day
meeting, sponsored by the German Research Society, (14) was held in Munich in '
January, 1978. It was intended to stimulate communication between different
research groups, governmental authorities and practicing engineers, and to assess
the present state of problem identification and research. The topic of the
53
meeting was restricted to storrawater runoff and 21 papers were delivered and
discussed.
Papers and discussion at the colloquium concerning data collection

showed that a large number of rainfall-runoff data has been collected in recent
years in the FRG. It became evident that data collection programs aré poorly
coordinated and do not allow comparative evaluation or collective use for
investigation of the basic processes involved. There is mainly a lack of data
for describing the influence of street cleaning, de-icing, traffic, and
construction on surface runoff pollution, and for defining the balance of
pollutant accumulation and decay due to storm runoff and the dry spells in
between.
Recent precipitation studies have concentrated on analysis of the

spatial variation of rainfall intensities. Aerial registration of rainfall
intensities with the aid of radar technology is presently under investigation.
In urbanized areas, particularly tall buildings cause disturbances in the
rainfall intensity images, the filtering of which has, as yet, not been
achieved sufficiently to allow practical use of radar.
Six papers at the colloquium dealt with mathematical modeling of the

rainfall-runoff process in urban areas. The variety of available models was not
considered as a problem, but rather as a necessary stage of development. Ongoing
research efforts, as described above in connection with the special research area
SFB 81, project A-4, aim at reducing the parameter-oriented relationships used
in modeling quality aspects. Further emphasis was placed on methods and require-
ments for the control of model results. Pertinent guidelines are in preparation.
The papers and discussion at the colloquium concerned with operational

control of urban drainage systems emphasized the importance of such control for
reducing receiving water loadings. First steps towards controlled sewer systems
have been made in the cities of Hamburg and Bremen, both located on the seacoast.
The city of Munich is presently also studying the possible advantages of sewer
system control. It was suggested that such measures should be considered as
alternatives in future planning studies elsewhere in the FRG.
The colloquium showed that differences in the terminology used, and the
reasoning followed by scientists, members of authorities and practicing engineers,
make it difficult to transfer knowledge and experience. It also became evident
that the findings from different research areas related to urban hydrology (i.e.,
hydromechanics, hydrology, meteorology, and operations research), are not
sufficiently utilized. The colloquium was well attended and was considered as a
successful first step to help overcome these deficiencies.
In the context of interdisciplinary research, a meeting on design,

optimization and operation of surface ponds, held in November, 1978, at Murnau,
Bavaria, should be mentioned. This meeting, also sponsored by the German Research
Society, was attended by Austrian, Swiss and German experts and touched on the
subject of urban hydrology. It was stated that one of the major difficulties in
dimensioning and operating water reservoirs is the unreliability of future water
consumption estimates. The applicability of recent studies on water consumption
was questioned, and it was suspected that the costs involved may well influence
the predicted water consumption figures.
54
Interest in urban hydrology research is still growing. It recently
received new momentum when the German Research Society decided to give priority
to sponsoring urban water research programs in the years to come.
Models in Urban Hydrology
Comparative Study of Methods for the Computation of Runoff in Urban

Drainage Areas
More and more, urban runoff calculations are being carried out with the
aid of digital computers. Calculation and results are checked by the ordering
institution and/or the water authority. Checking of so-called "urban runoff
simulation models" and of the resulting output is often particularly troublesome
and difficult. The aim of these methods is to reproduce the transformation of
rainfall into runoff true to nature. For that purpose, surface runoff and sewer
runoff are mostly calculated by separate model sections. Each section consists
of physically founded formulations. The mathematical solutions are often quite
sophisticated. Checking the results by hand is rarely possible.
A great number of different methods for calculation and computation

are applied. To gain knowledge about the qualification and applicability of
various methods, a comparative investigation of computer methods was carried out
in 1975/1976.Í1*,16) T n e institute for Water Management at the Technical
University of Hanover was appointed to conduct these investigations. A joint
working committee of the "Kuratorium für Wasser und Kulturbauwesen" (KWK)
supervised the study. Ten consulting engineers took part with their respective
methods. The calculations were made for two test areas. For both areas, data
for rainfall-runoff events were available by which the results of the calculations
could be checked by the committee. The consultant engineers had the possibility
of calibrating the parameter of one test of investigation with the measured
rainfall-runoff event. The applied methods included the so-called "rational"
method as well as the newly developed and more sophisticated methods which,
theoretically, allow more correct simulation of natural events. Together with
the calculations, the consultant engineers had to fill in a questionnaire in
which the principal features of the tested calculation method had to be described.
The following examinations were to be carried out: comparison of

results for two catchments; comparison of measured data with results; and
comparison of chosen parameters for fixed catchment characteristics. Furthermore,
the following questions were of interest:
Which methods will admit non-uniform precipitation distributions as inputs?

Is it possible to calibrate the parameters of the method for any corresponding
measurement data of precipitation and runoff?
Are the results improved if the computations are made with calibrated parameters?
Classification of Methods. Due to the differences in these methods, a

classification according to specific application was proposed. Distinction should
be made between hydrological methods with or without the possibility of calibrating
the parameters with measurement data, and hydrodynamic methods with or without
the possibility of calibration. The term "hydrologie" implies that transformations
from rainfall to runoff are made by means of fixed formulas or functions (rational
method, unit hydrograph). The routing in sewers is done by algebraic functions
or simple routing-storage procedures. Computations are characterized as
"hydrodynamic" if the procedure for routing sewer flow is based on the dynamic
55
equations. This characterization is valid even when the terms of acceleration
of the momentum equation are neglected (momentum equation - frictional equation).
The possibility of calibration implies that non-uniform rainfall distribution can
be accepted as an input function. Otherwise, model parameters would be calibrated
or tested insufficiently. All hydrodynamic methods are based on separate surface
and sewer models.
Results. For given rainfalls, maximum discharges and maximum water

levels for the manholes in the main sewers were to be calculated. For fixed
system points, hydrographs were to be computed. Furthermore, maximum velocities,
total runoff and overflow volumes and hydrographs were to be computed. Out of
the large number of results, one example is sufficient for relative comparison.
Results of computations by the respective methods of the ten consultant engineers
for this case resulted in a peak discharge at the sideflow weir of catchment
No. I that varied (for the uniform 30 min. rainfall which is the critical
rainfall duration for this catchment) from 51.2 cfs to 84.4 cfs (1.45 to
2.39 m3/s and (for the 60 min. hyetograph) from 57.2 cfs to 104.9 cfs (1.62 to
2.97 raVs). The variations from method to method are graphed in Figure 1. These
deviations amount to 165% and 183%, respectively, for the 30 min. and 60 min.
hyetographs. Deviations in total runoff volume amounted to 1437o and 154%. The
calculated overflow volumes to the receiving water varied for the 30 min.
rainfall from 2,163 m 3 to 3,034 m 3 and for the 60 min. hyetograph, from 2,492 m3
to 3,149 m3. The differences in the results are quite considerable.
o = 60 min. Hyetograph
100
o « 30 min. Uniform rainfall
distribution
5 10
NUMBER OF METHOD
FIGURE 1 - COMPARISON OF COMPUTED PEAK DISCHARGES FOR CATCHMENT I
Conclusions. Comparison of the results shows that there are serious

problems in the application of methods to compute runoff in urban drainage
systems. Partially, they are due to discrepancies between basic assumptions
of the methods, and to the respective field of application. For example, only
a model that can be calibrated on historic data may be expected to give
realistic results; and hydrological methods (i.e., transformation of rainfall
into discharge hydrographs) are usable only for uniform flow conditions (no
looping, backwater, etc.) whereas hydromechanical methods (solution of equations
of continuity and momentum) handle non-uniform or transient flow.
56
The differences in computed runoff volumes indicate that the deviation
in peak discharge does not depend primarily on model structure. The reasons may
be seen in the different surface loss assumptions for clearly defined catchment
characteristics. Consequently, the problem of urban sewer computations lies not
only in the modeling of the rainfall-runoff process but in the fixing of surface
losses or specific rainfall-runoff ratios. It is quite remarkable that information
about the quantitative aspect of surface losses is at present unsatisfactory.
This requires the intensified collection of rainfall-runoff data in existing sewer
systems and appropriate research in this field.
At present, the accuracy of input parameters seems to be of much greater

importance than the possible further improvement of computer models.
Numerical Simulation of an Automatically Controlled Flushing System

for the Deep Interceptors in Hamburg
The control of pollution, quality of dry-weather flow and the conveyance

of storm runoff in combined sewer systems gain more importance with an increasing
consciousness of environmental problems. To improve the historically adopted
combined sewer system and to treat sewage and stormwater biologically before it
is discharged into receiving waters, an additional system of deep interceptors
is being realized in Hamburg. To guarantee maximum usage for the future,
consideration of construction cost, intended performance and operation and
maintenance is of paramount importance. Economic aspects set limiting conditions.
This implies a review of conventional methods of cleaning sewers, and

the need for alternative solutions. One possible way of cleaning sewers is to
use the flushing action of artificial waves. To investigate the efficiency of
flushing, an extended version of SESIM (Stadt-EntwäBserungs-Simulationsroodell)
is used. SESIM was developed from the Storm Water Management Model (SWMM) for
application in Germany.
Procedures that take into account sediment diameters(17) transported

and deposited on the basis of shear stress were incorporated together with
automatic local control to operate sluice gates depending on local water levels
and spatial gradients of water levels. For calculating the entire sediment
quantity,(18,19) u s e ±B made of the expressions for an average volume rate of
sediment per unit width. Also adopted is a sediment diameter of 0.7 mm,
considered representative of sands in sewage.
(20,21)
The investigated sewer system includes about 16 km of deep interceptors
and 16 km of trunks of the old sewer system. For the simulations, the system was
divided into two subsystems. At first, the possible self-cleaning capacity under
average dry-weather flow was considered in order to determine the efficiency of
flushing in removing depositions from trunks and interceptors. The purpose of
the subsequent simulation runs was to study the effect of automatically controlled
regulator gates on the flushing operation. Therefore, the controlled gates are
installed in manholes in the regions which had depositions during dry-weather
flow. During this phase of investigation, all automatic regulator gates were
controlled by water levels and local water level gradients at gages upstream of
the gates. At the beginning of the simulations, the initial water levels are
the dry-weather flow levels. The control-algorithm directs the gates to close
and to store the dry-weather flow. The water level has to reach a fixed height
below the vertex to open the gate. The period of storing and releasing the water
may be called the "flushing-cycle". The highest velocity occurs at the top of
57
the flushing wave near the gate. An observed velocity of 2.79 m/s is lower
than the 3.0 m/s, which is considered dangerous for concrete conduits.(22)
This study results in knowledge of the places that are subject to sedimentation,
and in recommendations for the installation of remote sensing and control
facilities for operational purposes.
Mathematical River Basin Models for the Optimization of Water Quantity

and Quality Management in Urban Areas
On the initiative of the Working Group on "Mathematical River Basin

Models," which had been instituted in 1972 by the Federal Ministry of the
Interior to comprise scientists and representatives of the water management
administration and of the most important water management associations, details
of the mathematical models developed within the framework of water management
during the last two. decades at home and abroad were compiled in two study papers.
The first paper, completed in October, 1975, concerns the details of
quantitative and qualitative models, and the model components used.(23) ^he
paper describes pure water quantity and pure water quality models, as well as
combined water quantity and quality models. It shows that the" basic structure
of all domestic and foreign models examined can be described in six "macro-
components" (input, physical, chemical and biological system, system-analytical
formulation, economic input, algorithms, and output). Further subdivision into
smaller "components," the so-called "micro-components," showed that about 100
micro-components are used, which were then described in detail. This procedure
allows comparison of the various models and of the setups used, and of the
required data volume and the possible results.
Another paper, "Study of Economic Considerations,'^2^) published in

December, 1977, explains the economic model structures and describes and
evaluates critically the economic elements of selected river basin models. The
study tries to show the interdependence between the structures of hydrologie
engineering models and economic models, with special attention to model purpose,
given set-up conditions, and the problems involved in information procurement.
In addition, an assessment of the suitability of the economic fundamentals used
in river basin models is made. For this purpose, the paper shows which economic
model components are suited for the evaluation of the results of river basin
models in connection with specific problems, and which should be used in future
for similar investigations. In this context, the following points are discussed:
- design function and optimization range in consideration of the economically

relevant decision criteria;
- the question of the expansion of the optimization process to include
maximization of economic and non-economic benefits or net advantages;
- dimensions of benefit and cost, or of advantages and disadvantages (indicators
and monetary magnitudes);
- methods for covering economic and non-economic advantage and disadvantage
components ; and
- selection of evaluation standards and possible evaluation procedures.
Besides a large number of foreign models, the two papers describe in

detail the "Prognostic Model for the Neckar River" and the "Aller/Leine/Oker
River Model". The purpose of the studies within the framework of the "Prognostic
Model for the Neckar River"
(25)
was the development of an operational
instrumentation for the planning of measures for the improvement of water quality
in a river basin which is representative of our country and of the prevailing
58
socio-economic boundary conditions. The main task was to examine measures
taken in the river and in respect of the introduction of wastewater and
rainwater for their effect on water quality. At the request of the Federal
Ministry of Science and Technology, and within the framework of technical
coordination by the firm Dornier System of Friedrichshafen, an interdisciplinary
team of experts from five State and university institutes worked on this model.
To develop models usable not only for the solution of the specific
problems encountered in the Neckar basin, the planning goals of the study had to
be laid out on a very comprehensive pattern. Expressed summarily, the aims of
the wo
rk (26)
were as follows:
- development of instrumentation for the coverage of the influence of certain
measures on water quality (generally referred to as "descriptive and/or
prognostic water quality model");
- development of instrumentation for the forecasting of utilization data and
data on benefit and damage; and
- collection and processing of illustrative and confirmatory data for the
formulation and application of water quality models and for the use of
forecasting methods.
As a result of the investigation, a large number of improvement
measures could be judged under specific valuation criteria. However, as far as
reliable prognoses of future water use and water pollution are concerned, there
are still uncertainties due to the lack of forecasting methods and to the
insufficiency of available input data.
The Aller/Leine/Oker model,(23>24) developed by the Institute of Water
Management, Hydrology and Agricultural Hydraulic Engineering of Hanover Technical
University, alms at the economic optimization of flood protection measures in the
area under investigation. Toward this end, the most economic solution for
alternative planning of extension or construction of about 50 retention basins
and reservoirs, and of the related river development of about 1,000 km total
length in the entire investigation area, must be worked out. In the course of
this work, substance and extent of the individual construction measures, as well
as their development in time and the future system control of the various
planning variants, are optimized. The solution most recommendable from the
economic point of view is determined in an optimization model with the aid of a
cost/benefit analysis. The ultimate aim is the maximation of the net benefit
of water management investment measures, with fixed expenditure limits, in flood
protection. Proceeds from the use of permanently impounded flood retention basins
for recreation purposes are taken into account in the financial considerations.
For the solution of the complex problems involved in the optimization

of water management data of river basins, an international exchange of expertise
is of particular importance. For this reason, the Specialized Conference on
River Basin Management of the International Association on Water Pollution
Research (IAWPR) was held in September, 1977, in the Ruhr region, giving experts
from at home and abroad an opportunity to discuss in particular the various
problems, and solutions for water management problems, encountered in
agglomeration areas.(27)
59
Sensitivity Model for the Urbanized Lower Main Region
From 13 to 16 March, 1979, an expert panel on the subject of "The

Development and Application of Ecological Models in Urban and Regional Planning"
convened at Bad Homburg, near Frankfurt, FRG. The meeting was organized by the
German Working Group on Unesco-MAB Project No. 11, and by the "Regional Planning
Group for the Lower Main," Frankfurt. Within the framework of cooperation
between IHP Project No. 7 and MA.B Project No. 11, Dr. Massing participated in
some of the expert discussions.
At the meeting, Dr. Vester of Munich delivered a detailed paper on a

"Sensitivity Model for the Lower Main Region," which had been developed by his
study group "Biology and Environment". The "purpose of the study is the
development of an operational model allowing the planning staff to understand a
human living space as a biokybernetic system and to draw therefrom decision aids
which may help to improve the survival capacity of the observed system".(28)
The starting point of the study is the realization on the basis of system-
theoretical considerations that exact prognoses of the future development of an
agglomeration area cannot be obtained by way of pure trend analyses. The
behavior of complex dynamic systems can be covered only kybernetically. The
classical cause analysis models are being abandoned. The aim is system
interpretation on the basis of the interrelationships between the variables.
With the aid of the model, the total Lower Main region, as well as component
partial systems, such as, e.g., a river drainage area, a city, the area of the
Frankfurt air terminal, or a large factory, are investigated. Subject areas are:
national economy, population, land use, ecology of the human environment, natural
elements (water, air, soil), infrastructure, and social components. From these
subject areas, 600 data descriptors were selected and reduced to 180 groups
which cover the basic domains of the urban system. A further reduction to 25
key indicators, which contain many other indicators by implication, was attained
through the evaluation of questionnaires.
Development of the sensitivity model proceeds in the following

steps:(29)
1. Collection and check of the data

1.1 Variables (collection, systematization, quantification,
interrelationships, check)
1.2 Invariables (construction of the conditions grid, dynamicization,
effects of changes in the conditions)
2. Model structure
2.1 Definition of different coarse structure levels
2.2 Development of coarse structures
2.3 Development of the overall network
3. Computerization and dynamicization
4. Adaptation of the model to application requirements.
The model has been completed. The authors of the model point out
that its language is easy to understand, that it is oriented towards actual
application, and that it is versatile and flexible. It is intended to serve
as an aid for diversified user groups (politicians, scientists, planning staff,
enterprises, authorities, international organizations) in different
60
applications (decision-making, prophylaxis, therapy, prognosis, research,
emergency management).
Summary and Recommendations
The preceding sections of this paper provide only very brief

information on investigations, planning and projects. Due to the limited
space available, a survey such as given in, for example, the Technical
Memorandum,^) was not possible here. Even so, however, the various component
elements mentioned show the importance of urban hydrology. This applies to
industrialized as well as developing countries.
The FRG IHP Working Group "Effects of urbanization and industrialization

on the hydrological regime and on water quality" is at all times prepared to
cooperate with the national Working Groups of other countries.
It seems now important that Unesco should engage in systematic

activity within the framework of its IHP. The necessary impulses are there,
material, experience and information are available, and the participants,
rapporteurs and organizers of workshops have assembled here. Before the
background of critical developments on the international scene, e.g., in the
field of energy and raw materials supply, special importance attaches to the
protection of the quality of life in urban areas, and thus also to urban
hydrology. Unesco should consider this as one of the keypoints of its
activities.
References
lt Massing, H., 1976: Urban hydrology studies and mathematical modeling in

the Federal Republic of Germany. Technical Memorandum No. IHP-6,
September 1976, American Society of Civil Engineers (ASCE), New York
2. Wohlrab, B., 1978: Brache T Definition und Abgrenzung. In: Brache und
Wasserhaushalt, Heft 34,'Schriftenreihe des DVWK, Verlag Paul Parey,
Hamburg und Berlin
3. Baitsch, B., 1977: Kulturlandschaft und Wasserhaushalt. Lehrstuhl für

landwirtschaftlichen Wasserbau und Kulturtechnik, Universität Bonn
4. Massing, H., 1978: Beziehung zwischen wasserwirtschaftlichen Ansprüchen

urbaner Gebiete und den Wirkungen der Brache auf den Wasserhaushalt.
In: Brache und Wasserhaushalt, Heft 34, Schriftenreihe des DVWK, Verlag
Paul Parey, Hamburg und Berlin
5. KWK/DVWW-Arbeitsausschuß "Seen und Erdaufschlüsse", 1978: Richtlinie

für die Gestaltung und Nutzung von Baggerseen. KWK-DVWW Regeln zur
Wasserwirtschaft, Heft 108, Verlag Paul Parey, Hamburg und Berlin
6. Lübbe, E., 1977: Baggerseen. Schriftenreihe des Kuratoriums für Wasser

und Kulturbauwesen, Heft 29, Verlag Paul Parey, Hamburg und Berlin
7. Blasy, L., 1979: Versickerung von anfallendem Drainwasser zur Aufrecht-

erhaltung des natürlichen Grundwasserhaushaltes - Planungen und Versuchs-
ergebnisse am Beispiel des geplanten Großflughafens München II.
Vortragsmanuskript des Internationalen Symposiums "Künstliche Grundwasser-
anreicherung" der FH-DGG und des IHP-Nationalkomitees der Bundesrepublik
Deutschland am 14.-18.5.1979 in Dortmund
61
8. Oberste Baubehörde im Bayerischen Staatsministerium des Innern, 1970:
Überleitung von Altmühl- und Donauwasser in das Regnitz-Main-Gebiet.
Eigenverlag, München
9. Schraidtke, R., 1978: The interbasin water transfer system Danube - Main.
Vortragsmanuskript, USA
10. German Research Society, 1972: Organization and Functions. Deutsche

Forschungsgemeinschaft, 2nd edition, Franz Steiner Verlag GmbH¿ Wiesbaden
11. Wasser-Kalender 1979: Jahrbuch für das gesamte Wasserfach. H. Hübner,

13. Jahrgang, Erich Schmidt Verlag, 1979
12. Deutsche Forschungsgemeinschaft, 1978: Die Arbeit der Senatskommission

für Wasserforschung 1975-1977. Kommission für Wasserforschung,
Mitteilung 1, Harald Boldt Verlag, Boppard
13. Sonderforschungsbereich 81, 1977: Finanzierungsantrag 1978-1980.

Technische Universität München
14. Deutsche Forschungsgemeinschaft, 1978: Städtische Hydrologie, Ergebnisse

des Rundgespräches vom 26./27. Januar 1978 in München. Senatskommission
für Wasserforschung, Eigenverlag der DFG, Bonn-Bad Godesberg
15. KWK/ATV-Arbeitsausschuß 4.4, 1977: Vergleichende Untersuchungen von

elektronischen Kanalnetzberechnungen. Arbeitsausschuß 4.4 "EDV bei
Ingenieuraufgaben in der Wasserwirtschaft". Schriftenreihe des
Kuratoriums für Wasser und Kulturbauwesen, Heft 31, Verlag
Paul Parey, Hamburg und Berlin
16. Keser, J., 1978: Comparative investigation of computer methods.

International Conference on Urban Storm Drainage, Southampton
17. Sonnen, M.B., 1977: Abatement of Deposition and Scour in Sewers,

EPA 600/2-77-212
18. Sündermann, J., 1978: Mathematische Modellierung des Sedimehttransportes.

In: Anwendung mathematischer Modelle von Küstengewässern und Tideflüssen,
Sonderforschungsbereich TU Hannover, Germany
19. Yalin, M.S., 1972: Mechanics of Sediment Transport. Pergamon Press
20. Central Engineering Laboratories - FMC Corporation: A Flushing System

for Combined Sewer Cleaning. Water Pollution Control Research Series
11020 DNO 03/72
21. Führböter, A., Mittelstadt, M., E. Macke, 1975: Modellversuche zur

Unterführung der Norderelbe mit einem Schmutzwasserdüker. Leichtweiss-
Institut für Wasserbau der TU Braunschweig, Germany
22. Fair, G.M., Geyer, J.Ch., and D.A. Okun, 1966: Water and Wastewater
Engineering, Vol. 1. John Wiley & Sons, Inc., New York
62
23- Bundesministerium des Innern, Bonn, 1975: Studie über bestehende
Flußgebietsmodelle, Teil A: Quantitative Modelle; Teil B: Qualitative
Modelle
24. Bundesministerium des Innern, Bonn, 1977: Studie über Wirtschaftlichkeits-
berechnungen in Flußgebietsmodellen
25. Bundesministerium für Forschung und Technologie, Bonn, 1977:

Prognostisches Modell Neckar, Berichte 1 bis 25 (1974 bis 1977)
26. Hahn, Hermann H. und Arrien Tiemon, 1979: Prognostisches Modell Neckar.
Wasserwirtschaft 69, 1
27. International Association on Water-Pollution Research, 1977:

Specialized Conference: River Basin Management, Ruhrverband Essen
28. Vester, F., 1978: Kurzbeschreibung der Studie "Darstellung der Gesamtdynamik
und Entwicklung eines Sensitivitätsmodells" am Beispiel der Region
Untermain. In: Ökologie und Planung im Verdichtungsgebiet - die Arbeiten
zu MAB-Projekt 11 in der Region Untermain. MAB-Mitteilungen Nr. 2, Bonn
29. Vester, F., 1976: Ballungsgebiete in der Krise. Deutsche Verlagsanstalt,

Stuttgart
63
Section 6 Progress since 1976 in Sweden
by Jan Falk*
Introduction
Background
Urban hydrological research in Sweden is financed mainly by the Swedish

Council for Building Research (BFR), the Research Committee of the National
Swedish Environmental Protection Board (SNV-F) and the Swedish Natural Science
Research Council (NFR), Research within the International Hydrological Decade
and the International Hydrological Programme has been supported by funds from
the NFR. Difficulties have sometimes arisen on specific projects due to the
problems of satisfying the particular concerns of more than one of these three
research councils.(1)
In 1976, the need for a coordinating group within urban hydrological

research was recognized and an organizational plan for such a group was proposed. (2)
In that report the research needs were also discussed. The Swedish Coordinating
Committee for Urban Hydrology (SUH) was thus established in 1977. The main task
of the SUH is to initiate, plan and coordinate urban hydrological research. The
SUH works as an advisory body to the research councils which retain the formal
responsibility. SUH consists of members from BFR, NFR, SNV-F, the National
Swedish Environmental Protection Board (SNV), the Swedish Geotechnical Institute
(SGI), the Swedish Geological Survey (SGU), the Swedish Meteorological and
Hydrological Institute (SMHI) and the Swedish Water and Sewerage Works
Association (VAV).
The future direction of urban hydrological policy is dealt with by a

Storm Water Group consisting of members from SNV, BFR, VAV, the Swedish Association
of Municipalities and the Swedish Board of Urban Planning. The Storm Water Group
assists the authorities, mainly the SNV, in their work on legislation and
developing guidelines in the field of storm water management. The interrelations
between the different agencies and groups are given in Figure 1.
On the initiative of the SUH, several national working groups have been
established. Examples are "planning of storm water management" and "snow melt
runoff". The SUH also initiates and arranges seminars.
Nordic cooperation has been established through FAG 2, a group connected

with the IHP. The task of the group is to promote the exchange of information
between researchers in the Nordic countries and to arrange informal seminars.
Four such seminars are planned and the first, on "runoff from urban surfaces,"
was held in Lund, Sweden, in May 1979.
Research Plan
During the last decade, problems connected to the handling of storm

water have become more important as Swedish municipalities have become adequately
supplied with treatment plants for wastewater from urban communities.(4»5) xhe
Department of Water Resources Engineering, Lund Institute of Technology,

University of Lund, Fack, S-220 07 Lund, Sweden. (Now at the Swedish
Environment Protection Board, Box 1302, S-171 25, Solna, Sweden).
RESEARCH COUNCILS
BFR NFR SNV-F
(decision«taking )
AUTHORITIES ANO ORGANISATION! COORDINATING COMMITTEE

( SUH )
SNV BOARO OF ASSOCIATION OTHERS
URBAN OF MUNICI- ( advisory and coordinating )
PLANNING PALITIES
STORM WATER GROUP
REFERENCE GROUP REFERENCE GROUP REFERENCE GROUP
RESEARCHERS
FIGURE 1 - RESEARCH ADMINISTRATION^ 3 )
aims of storm water management can be summarized:
Society must be adequately protected from flooding.

. The receiving waters must be protected from pollution due to
storm water and combined sewer overflows.
. The local natural water budget should as far as possible remain
undisturbed.
Costs should be minimized.
The overall aim may be expressed as follows: storm water should be regarded
as a resource that can b e used and not as a drawback.
In 1977, the SUH put forward a proposal-for a research plan,(6> 3 )

which was revised in 1978.(7) The plan is based on a model containing the
three basic steps of: planning; design and construction; and operation. Such
a model makes it possible to give priority to projects within the first and
third steps, because these have been largely neglected in research until now.
The following three research fields are specifically emphasized by the SUH:(7)
. Methods for planning, considering hydrology, geology and ecology

when urbanizing, rehabilitating and renewing with the application
of new techniques.
. Analysis, design and operation of storage facilities within storm
water systems - that is, to use in a optimal way the ground,
surface and structural facilities for the detention, retention and
infiltration of storm water.
66
. Long-terra observations of full-scale Installations for infiltration,
detention, etc., for studies of the operation, maintenance and
environmental effects.
Research Cost
The actual research is carried out mainly at the technical universities

but also at some of the larger consulting firms, governmental authorities and
at some of the larger municipalities. The total amount of money put into urban
hydrology research by the three research councils for the budget year 1978/79
amounts to 4.1 million Swedish Crowns, out of which BFR contributes 3.3, SNV-F
0.5, and NFR 0.3 millions.(3) In addition, universities, city authorities and
private firms spend money of their own on research and development. The overall
figure of 4.1 million Crowns may be. compared to the total cost of urban water
management estimated at 2000-million Crowns for 1974.
Urban Catchment Research in Sweden
The first urban catchment investigated in Sweden was instrumented in

1972. Swedish activities in the field of urban hydrology may therefore be said
to be of rather recent date. Catchment research has previously been summarized.(1>8)
Space permits reporting only some of the latest activities.
Precipitation Studies
In Sweden, evaluated records of rainfall intensities exist for only a

few places. This fact has led to several projects concerning analysis of more
rainfall series.
(9,10)
At the Swedish Meteorological and Hydrological Institute
(SMHI) a comprehensive system of routines have been developed for transfer of
historical records to computer media.(10) A similar transformation system has
been used by the Chalmers University of Technology.(9)
At the SMHI, a climatological analysis led to the definition of a
regional parameter (z) which reflects the pattern of convective rainstorms in
Sweden.(H) The variation of z for southern Sweden is shown in Figure 2. A
formula based on z permits the derivation of intensity-duration-frequency
relationships standardized to the period 1931-60 for any place in the country.
The durations may be chosen for periods of 3 minutes to 96 hours and recurrence
intervals from 1 month to 10 years. Tests on independent data show small
deviations.
Arnell(l2) made a comparison of calculated storm water discharges
with two kinds of rainfall data, viz. design rainfalls developed from intensity-
duration-frequency curves and measured time series of rainfall. These two
approaches were tested on a 0.154 km^ catchment area. The conclusion was that
the use of design rainfalls gave a more uneven dimensioning of storm water
systems. Statistical analysis showed that real rainfall sequences produced
significantly better simulations of observed frequency distributions of peak
flow.
The National Defence Research Institute (FOA) has developed a system

for measuring rainfall intensity.(13) The objective of the study was to
investigate the influence of raindrops on radio wave propagation at frequencies
higher than a few GHz. Raingauges were placed at each km along a 15 km path
just south of Stockholm,. The gauge works on the drop-counting principle with
67
FIGURE 2.- REGIONAL DISTRIBUTION OF z VALUES IN SOUTHERN SWEDEN^11)
a mean of 250 drops per mm of rainfall. All the gauges are connected to a
control center via telephone lines.
A similar project is being undertaken in Lund.

(14)
A network of
twelve 0.035 mm tipping bucket gauges has been established to cover an area
of 20 km2 in the central parts of the city. The data collection system is
operated by means of leased telephone lines (Figure 3). The objective is to
study the reduction in average precipitation intensity with increasing area,
the size, shape and movement of raincells and the effect of the city of Lund on
the precipitation pattern.
CLOCK CONTROL
TELEPHONE LOGIC
TELEPHONE
PLUGS PLUGS
RECEIVER
V TRANSMITTER -©v TELEPHONE CABLES iî COUNTER
MP
U L
L E
TRANSMITTER -© ¿ssas F
\i/
Y RECEIVER COUNTER T X INTERFACE
I E
R
V TRANSMITTER -©- "••Pi

K¿J
RECEIVER COUNTER
'• ; T/PE
1
PUNCHER
12 RAIN
GAUGES
NORD 10/S
FIELD INSTALLATIONS OFFICE INSTALLATIONS COfPUTER
FIGURE 3 - DATA COLLECTION SYSTEM FOR THE RAINGAUGE NETWORK IN LUND (14)
68
Malmquist(15) has.studied the quality of atmospheric fallout in
Gothenburg. The fallout contributed 20% of the organic matter content in the
storm water, 25% of the phosphorus and 70% of the nitrogen. The heavy metal
contributions from fallout ranged from 7% to 40%. Urban snow has been found to
have significantly higher pollutant concentrations than average storm water
from the same area. (15)
Runoff Studies
Chalmers University of Technology. Field measurements of runoff have

been carried out by the CTH in ten areas in cooperation with local water and
sewage works.(**) ¿11 catchments have separate sewer systems. The listing below
gives details of these catchments:
1. Bergsjb'n (Mellbyleden), Gothenburg. Area: 0.15 km2. Land use: apartment

complexes. Studies of: rainfall and runoff (quantity and quality) and
pollutant sources.
2. Bergsjösvängen, Gothenburg. Area: 0.05 km2. Land use: single-family
houses. Studies of: rainfall and runoff (quantity and quality) and
pollutant sources.
3. Floda (30 km northeast of Gothenburg). Area: 0.18 km2. Land use: single-
family houses. Studies of: rainfall and runoff (quantity and quality) and
pollutant sources.
4. Vegagatan, Gothenburg. Area: 0.06 km 2 . Land use: apartment complexes and
offices. Studies of: rainfall and runoff (quantity and quality) and
pollutant sources.
5. Linkb*ping 1 (180 km south of Stockholm). Area: 2.0 km 2 . Land use: mixed
housing and commercial buildings. Studies of: rainfall and runoff (quantity
and quality).
6. Linköping 2 (located within LinkCiping 1). Area: 0.22 km 2 . Land use:
single-family houses. Studies of: rainfall and runoff (quantity and quality).
7. Linköping 3 (located within Linköping 1 ) . Area: 0.04 km 2 . Land use:
apartment complexes. Studies of: rainfall and runoff (quantity and quality).
8. Torslanda (20 km west of Gothenburg). Area: 3.64 km 2 . Land use: mixed
housing and industrial buildings. Studies of: water balance, pollutant mass
flows, performance of the urban water systems.
9 and 10. Bratthammar and Halmstad. Areas: 0.07 and 0.02 km 2 . Land use:
single-family houses. Studies of: storm water infiltration.
The first catchment, Bergsjb'n, is the most investigated catchment in

the country.(^>8) The urban runoff process has been studied.(16) During 1976 the
catchment was investigated with respect to microsubstances, viz. 17 heavy metals
and PCB, DDT with derivatives, HCB and PAH.i 1 ?) The concentration of PAH in
storm water was found to be relatively high. The concentrations of the other
components investigated were not remarkably high if compared to wastewater. For
As, Cd, Cr, Hg, Sb, V and Pb, atmospheric fallout could explain a great deal of
the metal content in the storm water. The remaining metals and the organic
compounds (with the exception of PCB) are produced within the catchment. The
main sources are exhausts from vehicles, the wearing out of tires and asphalt and
the corrosion of vehicles and building materials.
69
A study on the first four catchments indicated above has proven the
feasibility of predicting storm water quality by means of the pollutant sources
in the area.(18) The mass flows of zinc and copper in storm water can be
predicted by atmospheric fallout and corrosion. Corrosion explains more of the
copper mass flow than of the zinc mass flow. The mass flow of lead can be
explained by the traffic volume; and, in this context, frequency and effectiveness
of street-sweeping is important. Mass flow of phosphorus Was found to be
correlated with atmospheric fallout as well as with population density. Traffic
volume and population density probably explained the mass flows of SS and COD.
The quality of air and atmospheric fallout markedly improved with the distance
from the city center, and corrosion rates and storm water quality showed
corresponding improvements.
The average concentrations of SS and P have been found to be higher

than those in biologically and chemically treated -wastewater, whereas the
concentrations of COD were of the same order-of-magnitude.(19) The concentrations
of pollution were approximately twice as large during the first third of a runoff
event than during the last third, thus indicating a first flush effect.
Catchments 5, 6 and 7 in the above list were instrumented with financial

support from the municipality of Linköping.(20) The objectives of the study were
to obtain information which could lead to improved methods of design of storm
water systems and improved methods of predicting the emission of pollutants. The
results of the project facilitated the planning of an adjacent residential area.
The experiments in catchments 9 and 10 were established in order to

investigate different aspects of local infiltration of storm water.\LL> Research
on storm water infiltration started at the beginning of the 1970's by the
construction firm BPA and the consulting firm Orrje at the same time as the
construction of the first storm water percolation basins.(22) jh e performance of
percolation basins are also being studied elsewhere. The consequences to the
environment of storm water infiltration are being investigated by the Chalmers
University of Technology. The effects on the groundwater quality are also being
examined at a residential area, an industrial area and a highway.
A joint project (Chalmers University of Technology and the consulting

firm VBB, Stockholm) is aimed at the development of methods for the optimization
of storm water sewer systems and for the choice of design flooding frequency.
(3,23)
Methods used will consider the costs of construction, operation and maintenance.
Costs of flooding (including social aspects and legal limitations) will also be
covered.
Royal Institute of Technology. Combined sewer overflows are the subject
of an investigation in the central 7 km2 pf Stockholm.(23) Continuous measurements
are made of water levels and SS in the wastewater. Some existing simulation models
are being tried for the quantification of overflows in the complex sewer system.
A closely-connected project covers several aspects of storm water
detention, viz., local infiltration, on-site detention, in-line detention and
off-line detention (Figure 4).(24) The study is split up into: (i) a survey of
existing installations; (ii) operation studies; (iii) studies of controlled
transition of storm water to wastewater treatment plants ; and (iv) summary,
conclusions and guidelines. Up to now, 8 reports have been published in Swedish
describing, for instance, detention in open ponds, pipe packages, tunnels and
local infiltration.
70
Local infiltration Infiltration
Percolation
Detention near Ground surface

the source Roof
Ditch
Open pond
Pipe network
Flow
Detention in the pipe Detention outside Pipe packages
network (lagging) —®-—- the pipe network
(storing)
Concrete basin
Underground
storage
Tunnel
Detention at the Concrete basin

treatment plant Open pond
FIGURE 4 - SYSTEMIZING DETENTION STORAGE ACCORDING TO LOCATION IN

SEWAGE SYSTEM AND EXAMPLES OF TECHNICAL PERFORMANCE^)
University of Luíeâ. During a normal winter, most parts of Sweden

are covered by snow. In Luleá (in the very north of the country), some research
interest has been devoted to the problems associated with snowmelt. No runoff
has been found to take place under the snow cover until fully saturated conditions
are reached in the soil water storage, the supply to which has to infiltrate
through the permafrost.(25)
University of Lund. During the years 1975-78, a total of 13 small

catchments have been studied in Lund. The areas of these catchments vary from
78 to 641 m2. The catchments are completely paved and the runoff was measured
in the storm sewer inlets.(1,26,27) Those studies in conjunction with other
studies in the United Kingdom have led to general results for the estimation of
depression storage from catchment slope.(28) infiltration from these Swedish
asphalt surfaces was found to be only a few per cent of the rainfall volume after
deduction of depression storage. Data from 9 of these catchments were used in an
International workshop held at the U.K. Institute of Hydrology in April 1978. (29)
Further details of this workshop, which included Swedish participation, are given
by the U.K. report by Lowing,(30; Section 4 herein.
Water Budget Inventories
A water budget estimate with respect to both an outer (the "natural"

cycle occurring in urban areas) and an inner system (the conveyance and
distribution of water for uses within urban areas) was made in 197ó(1) and
revised in 1977.(4) The budget for an average hydrological year comprising the
71
INNER SYSTEM OUTER SYSTEM
SOUK»
F« WATER
HO
fTATKM
MATER-WORK*
0 EVAPOKATKM
2 sao
SS <ÄV
aao
12»
1T0 LE PERMEABLE SURFACES
MS
IM
nue um
~&
OMOUWMMTER
«7
LEAKAOE
244
681
W Ü T E WATE«
«•WEM STORM «ATER
TREATMENT PLANTS OVERFLOWS 555
RECEIVING WATER
25 494
AL
FIGURE 5 - GENERAL URBAN AREA WATER BUDGET FOR SWEDEN, INNER AND OUTER SYSTEMs(^)
(VOLUMES, IN MILLIONS OF m3/YEAR, SHOULD BE REGARDED AS ROUGH APPROXIMATIONS)
total urbanized area in Sweden (4024 km2 in 1970) is given in Figure 5. These
figures are of course not representative for all urban areas. In an attempt to
make a distinction between the hydrological response during summer and winter,
a water budget estimate was undertaken for a hypothetical urban area of 10 km2
with a population of 30,000 inhabitants.(31) The seasonal runoff (for summer and
winter, respectively) from the outer system for the impermeable area and the
permeable area were calculated for 81 combinations of precipitation, potential
évapotranspiration, soil moisture storage and imperviousness. The most
significant finding was the fact that the runoff from paved and pervious areas
were of the same order of magnitude during the winter, while the summer runoff
was totally dominated by the contribution from the paved areas. These findings
were supported by recorded data from a rural basin (Värpinge) in Lund when
compared with that from an adjacent urban (Klostergarden) catchment.'1»-*2)
During several winter months the runoff from the rural catchment exceeded that
from the urban. During the summer almost no runoff occurs from the rural basin
and the vast majority of that from the urban basin could be explained as being
a contribution from impervious areas.
During the spring of 1978, a large number of gauging stations were

established in the City of Lund with the objective of making a quantitative as
well as a .qualitative water budget based on direct measurements of all the
included components of one specific city.(33) A comparison can be made of the
effects on the receiving water of three different pollution sources: treated
wastewater; combined sewer overflows; and storm water. To fulfill the aims of
the project, there are 12 rainfall recorders,
(14) recordings of water use, two
stations for sampling atmospheric fallout and corrosion, four stations for
measuring soil moisture, 10 stations for measuring groundwater level, five
stations for recording levels in combined sewer overflows, six stations for
measuring storm water outflow (covering 90% of the area that has separate
72
storm 8ewers), two stations for biologically and chemically treated waste water,
and three gauging stations on the receiving river (upstream and downstream of
the city and upstream of the treatment plant). Lund has a total developed area
of 20 km^, of which the central part (comprising 1/3 of the area) has a combined
sewer system. The derived water budgets will include water flows, SS, BOD, P and
heavy metals. The project is expected to provide the basis for an integrated
approach to the management of urban water systems.
A water budget study was undertaken in a suburban catchment close to

Gothenburg (catchment 8 in the listing given earlier).(34) The scope of the
study was to investigate the functioning of the separate sewer system. Both
quantity and quality in the storm and wastewater systems were taken into
account. Á water budget was evaluated based on measurements of rainfall,
drinking water use and wastewater and storm water flows. A distinction was
made between drainage water (base flow), wastewater and storm water in both
systems. The drainage water flow was found to be 0.4 - 1.5 times the average
wastewater flow. During rainfall the wastewater sewer received up to 30% of
the total runoff volume and the maximum leakage flow was five times the average
wastewater flow. The mass flows of pollutants were found to be less than for a
combined sewer system. The pipes in this system are laid in a stratum of
crushed rocks, and these results cannot be considered typical.
Urban Hydrological Modeling in Sweden
Currently there are several computerized runoff models available and

in operation in Sweden. Since 1974 a considerable development has taken place
within this field. However, practitioners almost exclusively use the rational
method or a time-area method suitable for graphic solution using simplified
design storms and time-invariant runoff coefficients.
The first steps towards the introduction of mathematical models were

taken by the Chalmers University of Technology (CTH) when presenting the CTH-
model.'*' For a thorough investigation of the routing process in sewer networks,
two special models, DAGVL-A and DAGVL-B, were developed. (1) The Urban
Geohydrology Research Group at the CTH ran a course on "Computer Calculation of
Storm Sewer Networks" in 1978. The course concentrated oh practical application
of a Swedish version of the ILLUDAS model, which has been made available at
university computer centers.
(35)
Some of the more comprehensive modifications
that have been introduced are a base flow component and a routine for storm
overflows. The model may be run using two types of design storm as well as with
historical rainfall events. Five different types of time-area relationships for
a variety of subcatchments have been introduced and the calculations may be
carried out in Si-units.
Professor W. James, McMaster University, Hamilton, Ontario, Canada,
spent a sabbatical year in Sweden (1977/78) and developed a computer package of
SWMM called SWESWMM. The major feature of this model is that it is possible to
run parts of the Runoff Block from a computer terminal in a conversational mode
with SI-units.(36,37) The total package is now implemented at the technical
universities.
A study of above-ground runoff was mentioned above.(26,30) This
study has led to the development of generally-applicable relationships for the
determination of the model parameters in a lumped hydrological model of surface
routing. This model is called a "nonlinear reservoir with time lag" and the
73
model parameters may be determined from slope and length.(27) Similar
relationships for several other models have been derived at an International
Workshop in Wallingford, U.K. ,(29) which was attended by some Swedish
participants.
SjOberg(38) has produced an overview of the state-of-the-art with

respect to runoff models used in Sweden. In a special analysis of the kinematic
wave model for surface routing, it was found to perform well in simulating urban
surface runoff.
(39)
Models for planning including quality have been tested at the CTH.
The most comprehensive study was carried out using the NIVA model (the Norwegian
Institute of Water Research model).(40) Three different separate storm water
sewer systems, including an existing one, and two alternative combined sewer
systems were simulated. Annual loads of P and Pb could not be reduced very
much. However, during wet weather the loads were reduced by a half when
comparing an efficient system to a comparatively inefficient system. The
largest reduction.in annual Pb load was achieved by converting a separate
system into a combined system.
Application of Research Results
The Swedish Environmental Protection Board (SNV) was recommending

separate sewer systems as the only solution to storm water problems up to 1976.
Knowledge has now increased in a way that makes alternative solutions possible
which could be better suited to the governing problems in the area in question.
The SNV published guidelines (in Swedish) for the renewal of urban sewage networks
in 1978 based on research findings. In 1979 a publication (also in Swedish) on
local infiltration of storm water was published jointly by the BFR and the SNV.
This publication was aimed at practicing engineers and decision makers, and was
followed up by a well-attended course arranged at the CTH. A project (managed
by the CTH) with the objectives' of planning a new residential district in the
city of Karlskoga was successfully carried out in 1978. In the project, several
experts representing different disciplines combined to develop a final solution
which was well suited to the existing geohydrological and ecological situation.
A course on the planning of storm water systems was held at the Royal Institute
of Technology in 1978 jointly led by several active researchers.
References
1. Lindh, G., "Urban hydrological modelling and catchment research". Unesco.

Research on Urban Hydrology, Volume 2, in Sweden, Technical Papers in Hydrology
16, The Unesco Press, Paris, 1978.
2. Carlsson, L a n d Falk, J., "Urban Hydrology - review of research needs" (in

Swedish), Division of Water Supply and Sewerage, Chalmers University of
Technology, Publication B 76/6, and Department of Water Resources Engineering,
University of Lund, Bulletin No. 56, Series A, Gothenburg and Lund, 1976.
3. Hà'llgren, J. and Malmquist, P.-A., "Urban Hydrology Research in Sweden 1978",

Swedish Coordinating Committee for Urban Hydrology Research, Nordic Hydrological
Conference 1978, Helsinki, In Urban Geohydrology Research Group, Publication
No. 33, Gothenburg, 1978.
74
4. Carlsson, L. and Falk, J., "Urban Hydrology in Sweden - an Inventory of the
Problems and their Costs", In Effects of Urbanization and Industrialization
on the Hydrological Regime and on Water Quality (Proceedings of the Amsterdam
Symposium, October 1977), IAHS-AISH Pub!. No. 123, 1977.
5. National Report to the United Nations Water Conference 1977, "Water in Sweden",
Ministry of Agriculture, Stockholm, 1977.
6. Coordinating Committee for Urban Hydrology Research "Preliminary Research Plan

for Urban Hydrology" (In Swedish), Stockholm, 1977.
7. Cederwall, K., Hällgren, J. and Malmquist, P.-A., "Research on Storm Water

and Groundwater in Urban Areas - Research Plan for 1979/80" (In Swedish),
Stadsbyggnad No. 11-12, 1978.
8. Arnell, V., Falk, J. and Malmquist, P.-A., "Urban Storm Water Research in
Sweden", Engineering Foundation Conference on Instrumentation and Analysis
of Urban Storm Water Data - Quantity and Quality, Easton, Maryland, USA, 1976,
In Urban Geohydrology Research Group, Publication No. 19, Gothenburg, 1977.
9. Arnell, V., "Intensity - Duration - Frequency Relationships for Heavy Rainfalls
in Göteborg during the 45 Year Period 1926-1971" (In Swedish), Chalmers Univer-
sity of Technology, Urban Geohydrology Research Group, Göteborg, Report No. 5,
1974.
10. Dahlström, B., "A System for Analysis of Precipitation for Urban Sewer Design",
In Urban Storm Drainage (edited by P.R. Helliwell), Pentech Press Limited,
Plymouth, Devon, U.K., 1978.
11. Dahlström, B., "Regional Distribution of rainfall intensities - a climatologi-

cal analysis" (In Swedish), BFR Report R 18:1979.
12. Arnell, V., "Analysis of Rainfall Data for Use in Design of Storm Sewer
Systems", In Urban Storm Drainage (edited by P.R. Helliwell), Pentech Press
Limited, Plymouth, Devon, U.K., 1978.
13. Wickerts, S. and Granath, S., "Measurements of rainfall intensity - Instrumen-

tation and Data Processing", FOA report C 30157-E2, E1, 1979.
14. Falk, J., Jönsson, 0. and Niemczynowicz, J., "Measurements of Rainfall Inten-
sities in Lund", Department of Water Resources Engineering, University of Lund,
Report No, 3023, Lund, 1979.
15. Malmquist, P.-A., "Atmospheric Fallout and Street Cleaning - Effects on Urban
Storm Water and Snow", Paper presented at the Ninth IAWPR Conference in
Stockholm, 1978, In Chalmers University of Technology, Urban Geohydrology
Research Group, Göteborg, Report No. 37, 1978.
16. Arnell, V. and Lyngfelt, S., "Some results from Urban Runoff Studies in
Bergsjön, Göteborg, In Geohydrological research at the Chalmers University of
Technology, Göteborg, BFR Document D 17:1977.
75
17. Horkeby, B. and Malmquist, P.-A., "Microsubstances in Urban Storm Water",
In Effects of Urbanization and Industrialization on the Hydrological Regime
and on Water Quality (Proceedings of the Amsterdam Symposium, October 1977),
IAHS-AISH Publ. No. 123, 1977.
18. Malmquist, P.-A. and Svensson, G., "Urban Storm Water Pollutant Sources",
IAHS-AISH Publ No. 123, 1977.
19. Malmquist, P.-A., "Pollution in Storm Water" (In Swedish), Educational Paper,
Göteborg, 1976.
20. Arnell, V., Lindqvist, B.-G. and Svensson, G., "Calculation of Runoff in Storm
Water Systems by means of detailed Simulation Models" (In Swedish), Stadsbyggnad
No. 10, 1978.
21. Cederwall, K. and Holmstrand, 0., "Local infiltration of Storm Water, Geohydro-
logical Research at the Chalmers University of Technology, Göteborg, BFR
Document D17:1977.
22. Raus, K., Andersson, R. and Carlstedt, B., "Storm water discharge by detention
and perflation" (In Swedish), BFR Report R23:1974.
23. "Urban Research Projects by July 1 1978" (In Swedish), BFR, 1978.
24. Stahre, P., "Detention Basins in Sewage Systems" (In Swedish), Royal Institute
of Technology, Rational Sewage Systems Publ. No 1, 1977.
25. Westerström, G., "Report on activities at the University of Luleâ, small ex-
perimental research basin Bensbyn", Sweden. Workshop Symposium Northern Research
Basins, Fairbanks, 1977.
26. Falk, J. and Niemczynowicz, J,, "Characteristics of the above-ground Runoff in

Sewered Catchments", In Urban Storm Drainage (edited by p,R. Helliwell),
Pentech Press Limited, Plymouth, Devon, U.K., 1978,
27. Falk, J. and Niemczynowicz, J., "Modelling of Runoff from Impermeable Surfaces",
University of Lund, Department of Water Resources Engineering, Report No. 3024,
Lund, 1979.
28. Falk, J. and Kidd, C.H.R., "Depression Storage on Paved Surfaces" (In preparation)
1979.
29. Kidd. C.H.R., (editor), "Rainfall Runoff Processes over Urban Surfaces",
Proceedings of an International Workshop, Institute of Hydrology, Wallingford,
Oxon, Report No. 53, 1978.
30. Lowing, M., "Progress since 1976 in the U.K.", Paper to be presented at the AGU
spring meeting, 1979.
31. Carlsson, L. and Falk, J., "Water Budget for Urban Areas in Sweden - a Rough
approximation", University of Lund, Department of Water Resources Engineering,
Report No. 3022, 1979.
76
32. Falk, J., "Comparison of the water budget for a rural and an urban catchment"
(In Swedish), Internal paper, University of Lund, Department of Water Resources
Engineering, Lund, 1979.
33. Andersson, B. and Falk, J., "Water Budget for the city of Lund" (In Swedish
with a short English summary), University of Lund, Department of Water
Resources Engineering, Report No. VA 26, 1978.
34. Malmquist, P.-A. and Svensson, 6., "Water Budget for a Housing Area in Göteborg",
IAHS-AISH Publ. No. 123, 1977.
35. Sjöberg, A. and Lundgren, J., "ILLUDAS manual - Manual for a Swedish version
of ILLUDAS" (In Swedish), Presented at a course given by the Urban Geohydro-
logy Research Group, Chalmers University of Technology, Göteborg, 1978.
36. James, W. and Larsson, R., "SWESWMM - Users Guide", Report presented at the
SWMM Course, May 29 - June 2, 1978 at the Division of Water Resources
Engineering, University of Luleâ, 1978.
37. James, W., "A Pre-and Post-Processing Program Package for the Storm-Water
Management Model", Proceedings Stormwater Management Model (SWMM), Users
Group Meeting, November 13-14, 1978, EPA 600/9-79-003.
38. Sjöberg, A., "On models to be used in Sweden for detailed Design and Analysis
of Storm Drainages Systems", Paper presented at the Nordic Hydrological
Conference 1978, In Division of Water Engineering, Helsinki University of
Technology, Report 10, Otaniemi, Finland, 1978.
39. Lyngfelt, S., "An analysis of parameters in a kinematic wave model of overland
flow in urban areas", Department of Hydraulics, Chalmers University of
Technology, Report Series B:13, Göteborg 1978.
40. Svensson, G., "Planning Models for the Evaluation of Storm Water Management
Alternatives", In Chalmers University of Technology, Urban Geohydrology
Research Group, Göteborg, Report No. 37, 1978.
77
Section 7 Progress since 1976 in France
by M . Desbordes*
Introduction
French research on urban hydrology has intensified since 1976. The

number of researchers, experimental studies and financial investments have all
grown. Municipal engineers are more and more conscious of the problems
encountered in urban drainage, and today they seek solutions with modern tools
born of research.
The most interesting phase is probably an equipment installation

program for experimental watersheds, that started at the end of 1976, in cities
all over France, in order to increase knowledge un runoff quantity and quality.
Research on runoff and underground water qualities have been initiated, whereas
such studies were not very numerous before 1976. New mathematical models for
sewer design or planning have also been tested, and some other models, intended
for frequent practical usage, have been improved in order to make them attractive
to most engineers and technicians.
Now, in France, it is clear that urban hydrology is an interesting

research domain to serve urban planning and management purposes. Many working-
groups have been organized, composed of researchers and engineers from public
and private firms. Annual meetings and conferences (1-3)
summarize the main
results and information.
Urban Catchment Runoff Quantity Research
The main study is the experimental program of the French Ministries

of Equipment and Interior which started at the end of spring in 1976, as
mentioned in the original report for France on urban hydrological modeling and
catchment research. This program is managed by the Loriferne Commission, (4)
which is responsible for studies of urban drainage in France.
The characteristics of the experimental watersheds are given in

Table 1. Catchments 1 through 10 have only a few rainfall and runoff gaging
stations. This is because the 1976-1977 measurement period was used to test the
measuring devices, to determine measurement and maintenance problems, and to
select only those watersheds that were among the best for future research. As a
result of data analyses, in 1978 catchments 2, 6 and 7 were chosen to be more
extensively equipped, w ) Data from the 1978 period are not yet available.
Equipment installation for catchments 11, 12 and 13 was also completed in 1978.
The data from the raingage networks of catchments 12 and 13 will be used to
study the spatial distribution of rainstorms. Note in Table 1 that multiple
measurement points have been used for most watersheds, consistent with our
interest in basic processes.
The 1976-1977 measurement period revealed that the most important

problems were: synchronization of recordings without using electronic devices
Universite des Sciences et Techniques du Languedoc, Montpellier, France.
79
Level
ement
ces
U -H f» CM CO
<t CM CM rH I-I vO CM rv CT« m
M 3 >
<U co <U
•u cd o
£3
cu
B co
S cu -u
O M tí CM CS CO CO CM CM iH r-i vt CM va- vO co
1-4 3 -H
cd P M
Raing ages
CM CM CM CM CM rH tH •-i CM CM in O as
i-l
a) 00 CO m CO vt m m vt O vt \o rv CM
o s-s O O CM vi- i-l o i-l o CT* vt rH i-i o
1-1
en
usness
O '-\ rv vo •S+ rv CO vO iH v+
•i-I S^S
m
CM
o CM
m
CM
m <T>
rv.
-i vT vt m CM CO CO CM CM
> ^
M
 m o o O m CO CO i-i vO o VO O O
a> cd CM CT> vO i-i CO i-l CM
M Xi CM
oo
CO
o CM
o\ <t 00
m
CM
vt I-I.
co CO
o\ CT« CM CM
.-i CM
w w CM vt
r-- r-<
ocation
>
RDANS0N
PQ
o\ a\
UVELLE
2 rH T-i
UISSE
S
OMIERS
tí tí
ent
EAU
•H •H
w CO CO
O
Jrf W en 4J •U
CJ X) A
Ce H-l
Z
t
g #25
>
i
S3
Oí
%
CO
tí
Ol
tí
j-i tí
W Ü á <ii H z od i H u
3 3
E d 3 g m
co H g- d Pu co en
CHAMBE
cd
y S3
CO > i 1 H 1 cd cd
Z g a§ O
W >-" 1 PM
co
M 
a C0
U
•H
en
W
•i-i
h P-i
o 1-1 CM CO vt m vO iv. co a\ o i-l CM CO
Z i-i i-l iH 1-1
S
80
(which are quite fragile under the adverse conditions encountered in urban
hydrology); acquisition of good flow measurements with only water-level measuring
instruments; the maintenance costs when the gaging stations are numerous over an
important watershed; etc. For this first period, data were manually digitalized
and put on punched cards. Tape recorders are still expensive and have not been
designed to work under sewer measurement conditions.
Urban Catchment Runoff Quality Research
Research attention on water quality aspects of drainage is very recent

in France.
(4) But now, it is known that it is a.very important problem that
must be studied carefully.(8) A recent literature review analysis,(9) of about
one hundred papers of researchers all over the world, has shown that the results
were quite poor despite the efforts. It is clear that pollution measurements are
difficult and often are not significant in terms of what one wants to measure.
Sampling devices are not yet very efficient, their results are seldom comparable,
and so is the analysis of samples. Some studies on samplers have been made( 1 0 )
in order to choose the best one. The Loriferne Commission is also seeking
standardization of sample analysis methods.
Some experimental studies have been made(l>H>12) and some others are
under way. Since the beginning of 1978, the Loriferne Commission and the Urban
Equipment Division of the Ministry of Equipment have undertaken different studies
in order to define runoff quality research methods. They have achieved the basis
of an experimental measurement program, which will be conducted during the middle
of 1979, and will complete the program for runoff quantity. The experimental
catchments are not yet chosen, but may be among those which are equipped for
quantity measurements.
Runoff Quantity Modeling Research
Research has been pursued from different points of view. Some existing
models have been extended in order to use them to solve some specific problems,
for example routing problems in special sewer works or singularities like siphons,
by-passes, changing slopes, etc.(13) Some parts of existing models have also been
improved or extended, as data became available.(14) Most studies have been
concerned with hydraulic routings, limitations of computer times, or runoff
modeling. Because of the lack of data, no significant research has been done on
storm characteristics since 1976. That will be possible by the end of 1979 with
the development of experimental programs.
Some other models have been developed in order to solve problems which
have not been yet studied, for example surcharging effects on sewers
(15)
and
runoff modeling from semi-urban catchments.(1") Some private firms have also
developed their own models,(17) often adapted from others, for practical uses.
It can be said that there are now a lot of models available in France
as in other countries but, as everywhere, they often have not been tested on
experimental data. Most of them are design models, and few can also be used in
planning.
(18) There exists no real-time operational model and this is a subject
for future research.
81
Runoff Quality Modeling Research
Runoff quality modeling is just beginning in France. The lack of

good data, as noted earlier, is probably the best explanation of this situation.
However, some literature reviews have been made. One of the most interés
tingia)
has as its main conclusion that the state of the art on runoff quality modeling
is quite poor despite the importance of studies on this matter. As was the case
for runoff quantity aspects a few years ago, the models are too sophisticated in
comparison with the general validity of water quality data. It is hoped that
the experimental program for study of runoff quality, relying on world
experience, will provide good data for modeling in the next few years.
c
Some research has been made on underground water quality modeling.
That domain may lead to more theoretical studies than on runo
ff.(20-22) T h e
data may be obtained from physical models. These studies are very important
for practical applications in runoff drainage, because these applications may
vary with land use and runoff quality.
It can be said that there has been an important growth in the number
of studies on runoff quality modeling since 1976. 'This growth will probably
increase during the next years because it is now well known that runoff quality
is one of the most important problems in urban drainage and that the solutions
adopted in the future for urbanization or drainage may depend on satisfactory
management of the quality of runoff.
Practical Applications
Besides theoretical studies on runoff quantity and quality, many other

investigations have been made from the point of view of practical applications.
As a result of this research, official rules on urban drainage have been revised,
and new ones were published in 1977.(23) ^ book on runoff quantity and quality.
is now available.
(24)
Many studies have also been made of special devices or works in order
to solve runoff quantity or quality problems.(25) For example, studies on porous
roadways,(26,27) retention basins and pollution probleras,(28>29) special works
like swirl concentrators,(24>30,31) e t c . Practical methods for the use of runoff
models have been developed, for example automatic determination of land UBe or
imperviousness using aerial photographs and an electronic camera.
(32)
Some cities have used runoff models in order to study their sewer
systems.
(15,33)
Important cities like Marseille, Paris, Lyon,''J^v etc., are
doing the same. Some will probably have their own models. Cities like Bordeaux,
Nice, etc., are seeking automatic control of their sewer systems, as was done in
the Seine-Saint-Denis Department (near Paris) before 1976.(1)
As a consequence of runoff research, economical aspects of drainage,
impacts of runoff on urban equipment and urbanization, etc., have also been
studied.(1.35,36)
Conclusions
Important progress has been made in urban hydrology since 1976 in

France. This has led to many practical applications, and more and more engineers
and technicians are involved with urban runoff problems in urban planning.
82
However, a major research effort will still be necessary to solve some
aspects of urban runoff pollution. As research in this domain is quite expensive,
international cooperation seems greatly desirable.
Another research direction during the next years will be automatic

sewer control in some important cities which already have drainage problems.
Soon after the preceding paper was presented, a meeting was held in
Paris to define the objectives of an Urban Hydrology National Association,
which may be formed this year. Among its activities will be the provision of
a suitable liaison with organizations in other nations to facilitate future
international cooperation.
References
1. Ministère de l'Equipement, "Qualité des eaux superficielles - Epuration -

Hydrologie urbaine" Journées d'Information Nationale ("Surface waters quality-
Sewage-Urban Hydrology" National Information Meeting). Paris (24-26 mai 1977)
207 p. Ed. Laboratoire Régional de l'Ouest Parisien (LROP) B.P. 108 78190-Trappes,
2. A.G.H.T.M., "L'assainissement pluvial urbain" ("Urban runoff drainage")

Actes du 58eme congres de l'AGHTM - BORDEAUX (29raai-3juin 1978).
Techniques et Sciences Municipales - 73*erae année - n" 8 - 9, p. 417-431
(août-septembre 1978). Techniques et Sciences Municipales - 73^me année -
n* 10, p. 470-499 (octobre 1978).
3. Société Hydrotechnique de France, "L'hydrotechnique au service d'une

politique de l'eau" ("The part of hydrotechnic in water management").
XVemes Journées de l'Hydraulique - Toulouse (5-7 sept. 1978). Ed. S.H.F.
199, rue de Grenelle, 75007-Paris.
4. Desbordes, M., Normand, D., "Urban hydrological modeling and catchment

research in France". ASCE Urban Water Resources Research Program T.M.
n* IHP-8, 57 p. (november 1976).
5. Cheron, J., "Les bassins versants expérimentaux français" ("French

experimental watersheds"). Commission LORIFERNE - Dossiers L.R.O.P.
n* 6139 et 6781. (décembre 1977). Ministère de l'Intérieur, D.G.C.L.,
Services Techniques 4-12, rue d'Aguesseau 75800-Paris.
6. Desbordes, M., Bellostas, J. M., Ramperez, A., "Commission LORIFERNE.

Bassins versants expérimentaux. Exploitation de la campagne de mesures
1976" ("Loriferne Commission. Experimental catchments studies on 1976
data"). Ministère de l'Intérieur, D.G.C.L., Services Techniques. Rapport
Laboratoire d'Hydrologie Mathématique de Montpellier (L.H.M.).n* 15/77,
185 p. (novembre 1977).
7. Desbordes, M., Quesada, B., "Commission LORIFERNE. Bassins versants

expérimentaux. Exploitation de la campagne de mesures 1977" ("Loriferne
Commission. Experimental catchments. Studies on 1977 data"). Ministère
de l'Intérieur, D.G.C.L., Services Techniques. Rapport L.H.M. Montpellier
n* 17/78, 190 p. (mai 1978).
83
8. Blanic, R., "La pollution des eaux pluviales" (Runoff pollution). Ed.
Ecole Nationale des Ponts et Chaussles (E.N.P.C.) Formation Continue - 28,
rue des Saints-Peres 75007-Paris, 60 p. (mars 1977).
9. Desbordes, M., Ribstein, P., "Etude de la qualité du ruissellement urbain.

Synthese bibliographique" ("Study on urban runoff quality. Literature
review synthesis"). Laboratoire d'Hydrologie Mathématique (L.H.M.) -
U.S.T.L. Place E. Bataillon 34060-Montpellier-Cedex. Rapport LHM 45/78,
55 p. (octobre 1978).
10. Ranchet, J., Grange, D., "Etude d'appareils "a préTevement automatique d'eau"
("Study on automatic water samplers"). L.R.O.P., Rapport n" 1-67-03-4,
43 p. (décembre 1976).
11. L.R.O.P., "Pollution des eaux de ruissellement en zone urbaine. Bassin de

Velizy-Ouest" (Runoff pollution in urban area, West-Velizy catchment).
L.R.O.P., rapport n* 4527, 32 p. (octobre 1976).
12. Coyne et Bélier - Ingénieurs Conseils, "Pollution des eaux de ruissellement

pluvial. Etude d'un bassin versant unitaire" ("Runoff pollution. Study
on a combined sewer system"). Direction Departementale de l'Equipement,
99, av. du Général de Gaulle 92-ROSNY-SOUS-BOIS, 100 p. (mai 1977).
13. Bocquillon, C , Desbordes, M., Raous, P., "Programme RERAM. Transformation

des hydrogrammes a la traversée d'ouvrages spéciaux" ("RERAM model. Routing
of hydrographs through special works or singularities"). Ministère de
l'Equipement. Centre d'Etudes Techniques d'Aix-en-Provence - Les Milles
13100-AIX-EN-PROVENCE. Rapport L.H.M. Montpellier n* 17/77, 38 p.
(décembre 1977).
14. Desbordes, M., Ramperez, A., "Extension du modele L.H.M./L.N.H. aux bassins
versants urbains de taille moyenne" ("Extension of the L.H.M./L.N.H. model
to urban catchments of middle size"). Laboratoire National d'Hydraulique
(L.N.H.)-6, quai Watier 78400-CHATOU. Rapport LHM Montpellier n° 20/77 -
30 p. (décembre 1977).
15. Choçat, B., "Un modele de simulation des écoulements dans les réseaux
d'assainissement pluvial" ("A routing model for storm sewers"). Ttiese
d'Ingénieur-Docteur - Institut National des Sciences Appliquées de LYON,
avenue Albert Einstein 69621 - VILLEURBANNE - 304 p. (mai 1978).
16. Mourad-Agha, A., "Les modélisations du ruissellement sur une parcelle

semi-urbaine" ("Runoff modeling from a semi-urban area"). These
d'Ingénieur-Docteur. Université des Sciences. Place E. Bataillon
34060-MONTPELLIER-CEDEX, 195 p. (novembre 1978).
17. Codaccioni, F., Bos, M., "Exemple d'utilisation de l'informatique:

Etude générale du ruisseau d'Ars par le modele d'assainissement ("A
computer use: the general study of the Ars river by the mean of a
drainage model"). Techniques et Sciences Municipales. 73eme année
n" 10, p. 492-496 (octobre 1978).
84
18. Chocat, B., Seguin, D., Peyretti, G., "Essai de classification des modeles
utilisas en hydrologie urbaine" ("A classification attempt of models used
in urban hydrology"). Techniques et Sciences Municipales, 74eme année
n" 3, 10 p. (mars 1979).
19. Hemain, J. C , "Etude bibliographique des modeles de qualité des eaux

drainées par les réseaux d'assainissement urbain" ("A study on urban runoff
quality models"). Service Technique de l'Urbanisme (S.T.U.) 64, rue de
la Fédération 75017 - Paris. Rapport LHM Montpellier n° 8/79, 160 p.
(avril 1979).
20. Peaudecerf, P., Sauty, J. P., "Application of a mathematical model to the

characterization of dispersion effects on groundwater quality" IAWPR 9th
International Conference, p. 443-454, Stockholm (12-16 June 1978).
21. Brissaud, F., Pappalardo, A., Peaudecerf, P., "Fluid mechanics and scale
effects on the phenomena in porous media" IARH International Symposium,
p. 223-239, Tessaloniki (29 août-ler Sept. 1978).
22. Brissaud, F., Couchât, Ph., "Interactions liquide-solide des solutés en

milieu poreus" ("Liquid-solid interactions of solutes in porous media").
Bulletin B.R.G.M. Ill, n° 4 (1978).
23. Ministère de l'Intérieur, "Instruction Technique relative aux réseaux

d'assainissement des agglomérations" ("Technical Instruction for urban
drainage"). Ministère de l'Intérieur, D.G.C.L. Circulaire n* INT 77
284, 126 p. (juin 1977).
24. Fouquet, P., et al., "Evacuation des eaux pluviales urbaines" ("Urban
runoff drainage"). Ed. E.N.P.C. - Formation Continue - 166 p. (novembre 1978).
25. Deutsch, J. C , "Problématique du choix d'un réseau d'assainissement"

("Problems in the choice of a drainage system"). Techniques et Sciences
Municipales n° 1, 74eme année p. 23-32 (janvier 1979).
26. Griselin, J. F., Christory, J. P., "Chaussées poreuses: étude de

faisabilité" ("Porous roadways: feasibility study"). Service Technique
de l'Urbanisme (S.T.U.), Dossier LROP n° 7305, 82 p. (décembre 1977).
27. Deutsch, J. C , Christory, J. P., Cainiard, L., "Les chaussées poreuses,

technique nouvelle en assainissement urbain: etude préliminaire" ("Porous
roadways, a new technic for urban drainage: preliminary study").
Symposium OCDE sur le.drainage des routes, p. 149-156 BERNE (22-24 mai 1978).
28. Bergue, J. M., "Synthîese bibliographique sur les bassins de retenue"

("Literature review on retention basins"). S.T.U., 60 p. (novembre 1978).
29. Billecoq, M., Simon.H., "Les retenues d'eaux pluviales: conclusions d'une
étude comparative de 5 bassins" ("Retention basins: conclusions of a
comparative study on 5 basins"). Techniques et Sciences Municipales n° 1,
74*eme année p. 41.46 (janvier 1979).
85
Delmas, D., "Etude bibliographique de la séparation liquide-solide dans
le traitement des eaux pluviales" ("Literature review on liquid-solid
separation in the treatment of runoff"). Institut de Mécanique des
Fluides (I.M.F.), 2, rue Camichel 31000-Toulouse, rapport n° 360-1, 76 p.
(décembre 1978).
Domínguez, B., "Etude fondamentale d'un décanteur de particules solides:

détermination des critères de similitude" ("Theoretical study on a solid
particles separator: determination of similitude criterions"). Tiiese
d'Ingénieur-Docteur n° 7, 226 p., I.M.F. Toulouse (décembre 1977).
Nadeau, P., "Calcul automatique du coefficient d'imperméabilisation d'un

sol a partir de photographies aériennes" ("Automatic evaluation of
imperviousness coefficient using aerial photographs"). S.T.U., rapport
C.E.T.E. d'Aix-en-Provence, 50 p. (janvier 1979).
Desbordes M., et al., "Etude de l'assainissement de la ville nouvelle de

LILLE-EST. Rapport de synthîese" ("Study on drainage of the new town of
LILLE-EST. Synthesis report"). Etablissement Public d'Aménagement de
Lille-Est 59650-VILLENEUVE D'ASCQ. Rapport LHM Montpellier,n° 32/78, 34 p.
(septembre 1978).
Peyretti, G., "L'expérience du service assainissement de la Communauté

Urbaine de LYON" ("The experience of the sewage service of the city of
LYON"). Assises Nationales de l'Informatique Communale, 4 p. INSA
(juin 1978).
Seguin, D., "Equipements urbains d'assainissement pluvial: aides a la

prévision "a partir des documents d'urbanisme" ("Urban drainage equipments:
assistance to planning using urban documents"). Thfese d'Ingénieur-Docteur,
n" IDI 78011, INSA de LYON, 255 p. (octobre 1978).
Seguin, D., "Application de l'informatique a l'assainissement urbain"

("Computer applications in urban drainage"). Assises Nationales
de l'Informatique Communale, 9 p., INSA (juin 1978).
Section 8 Progress since 1976 in Norway
by Rolf Skretteberg*
Introduction
This presentation is to be regarded as a follow-on to a report presented

by N. R. Selthun in 1976.C1) As outlined in the 1976 report, most of the urban
hydrological work in Norway was initiated by the launching of a major Government-
backed research program for purification of wastewater in 1971. The program
consisted of some 40 projects, of which the following were closely connected to
urban hydrology:
- Project 4.1 Collection of rainfall intensity data.

- Project 4.2 The effect of urbanization on runoff from small catchments.
- Project 4.6 System analysis of urban water systems.(2)
- Project 4.7 Investigations of urban runoff composition. (3)
- Project 4.10 Preparation of storm data (for planning use).
(4)
In this report only Projects 4.1 and 4.2 will be dealt with as they were the only
projects that went beyond 1976. A summary of progress and the results thereof will
be given together with the present status of both projects.
Project 4.1. Collection of Rainfall Intensity Data. This project has

been administered by the Norwegian Meteorological Institute, Hydrometebrological
Division. The source of field data is an automatic rain recording system which
consists of a tipping bucket pluviograph that records increments of 0.2 mm of rain
on a magnetic tape at a time resolution of 1 minute. 35 signals a minute may be
recorded, which is equivalent to 7 mm of precipitation per minute, a value very
unlikely to be exceeded in Norway.
The Plumatic, as the Norwegian-developed recording instrument is called,

is powered by two small batteries and can operate unattended for up to six months.
The instrument, with its sturdy construction, has proven very reliable in the
field and the total number of discarded recordings has amounted to only some 57o
per season on the average.
In 1978 the station network consisted of 52 Plumatic pluviographs

nationwide. The data is stored on computer magnetic tape at the Norwegian
Meteorological Institute, Hydrometeorological Division, where processing methods
have been developed. The work until recently has mainly been data checking,
storing, tabulation and simple calculations. Some of the data series are, however,
now long enough for more advanced statistical work to be performed. Routine
calculations for the various stations are updated every year and are as follows:
Intensity - frequency - duration curves; Intensity - duration curves; and Intensity
- precipitation depth curves.
Norwegian Water Resources and Electricity Board, Oslo, 3, Norway.
87
Project 4.2. The Effects of Urbanization on Runoff from Small
Catchments. This project has been administered by the Norwegian Water Resources
and Electricity Board in co-operation with some local municipalities. The main
purposes of the project were:
- to study the effects of urbanization on mean, annual and monthly runoff;

- to study the effect of urbanization on the variability and frequency of
discharges, including flood flow frequencies and changes in low discharges in
urban areas ;
- to study the effects of urbanization on the seasonal distribution of runoff;
- to study the precipitation-runoff relationship and its dependence on
urbanization;
- to provide the practicing engineer with better design data for the design of
surface-water drainage systems; and
- to support model studies.
Field data were obtained from two types of small urban catchments, viz. 5 sets
of paired catchments and 7 single catchments.
Observations included in Project 4.2 were precipitation (1 minute time

resolution) and discharge (5 minute interval). In addition, observations of soil
moisture, ground water and evaporation were made at various stages of the work.
All registrations of precipitation and discharge were stored on magnetic tape in
a compact and standardized file structure. The files were operated through a
library of programs for data processing, retrieval, tabulating, plotting and
analysis.
The concluding stages of Project 4.2 were hampered by staffing as well

as financial difficulties. Due to slower development rates than anticipated in
the areas that were scheduled to be urbanized, and the long-term character of the
observations, financial backing for an extension of the project period beyond 1977
was applied for, but in vain. This brought the processing and analytical part of
the project work to a halt until September 1978, when new ways for completing the
scheduled work were found. More about this later.
Due to the reasons mentioned above, a full analysis of the results will
not be published until later this year. Graphs of daily mean values of precipitation
and discharge, tables with a 5 minute time resolution of signficant storm events and
seasonal variations of other variables for all the catchments up to date are to be
published shortly. However, a preliminary analysis based on the relatively short
data sets from the unpaired catchments may be made in the meanwhile.
These catchments are of moderate size (8-40 ha.) and with a relatively
high percentage of impervious area (18-97%). Summer rainstorms of the convective
type with high intensity and short duration give the highest flows, and as a
result only impervious areas contribute to the flow peaks. This applies to the
volume of flow as well. Rainfall/runoff correlations indicate that runoff
coefficients are on the order of 50 to 100% for the impervious areas directly
connected to the drainage system. Dominance by impervious surfaces is implied by
urban hydrological model studies, which show very small contributions to runoff
from pervious areas. This, however, does not seem to be the case for the larger
coastal catchments with a relatively low percentage of impervious areas. These
88
catchments seem to yield maximum flow peaks from autumn and winter frontal rains,
often concurrent with snow melt or rain on frozen ground. In these cases the
time of concentration for the catchment has to be extended to include the pervious
areas. If these precipitation events, occurring on saturated, frozen or snow-
covered ground, are critical in drainage system design, installation of retention
basins or other means of detaining runoff will be of little effect.
The preceding observations suggest that when the rational formula is

used, two cases of runoff formation are to be considered:
1) when only connected impervious areas contribute to runoff peaks, the

runoff coefficients depend on percentage of impervious area and the
time of concentration is short (a matter of minutes), and hence the
peak runoff is high; and
2) when saturation, snow or frost causes part of the pervious areas to

contribute to runoff, the time of concentration is higher (may be hours),
the precipitation intensity is relatively lower but runoff coefficients
are high (may exceed 1).
General Status of Projects 4.1 and 4.2. The two projects described are
to be carried on as before except for one major change. They have now been taken
over by their respective administrative bodies and are to be run on a semi-
permanent basis. The advantages of this are obvious, with continuation of the
observation series secured and a permanent staff that will preserve the continuity
of the work.
The analytical part of the Project 4.2, "the effect of urbanization on

runoff from small catchments," has been delayed for various reasons, but its work
and fulfillment of project aims are now part of the duties of a recently
established Urban Hydrological Section at the Norwegian Water Resources and
Electricity Board, Hydrological Division. One of the main objectives of this new
section is to set and co-ordinate1 common guidelines for future quantitative urban
hydrological work in Norway, and to provide storage for field data as well as
facilities for their analysis. This should ensure central steering of future urban
work in Norway and also ensure uniform data acquisition. An important asset of the
new Urban Hydrological Section is the continuation of the nationwide station network
established by Project 4.2 in 1973/1974.
In 1978 the Norwegian Institute of Water Research (NIVA) performed a

preliminary study of snow melt/precipitation/runoff relationships for the urban
areas of the coastal regions of Norway. Resolved by the study was a recommendation
from NIVA for public funds to be raised to support a thorough investigation of the
phenomena.
A steering committee is to be formed to organize a research project

similar to Project 4.2, on the effect of precipitation on snow-covered or frozen
ground and the effect on the peak rate of runoff in urban coastal areas.
Several background and update reference reports have been released.

(5-12)
89
Urban Hydrological Modeling in Norway
The Norwegian Institute for Water Research (NIVA) has developed four
computer programs for a total analysis and simulation of sewer systems. The
main objectives in these programs, written in FORTRAN, have been:
- to characterize the flow of wastewaters through the various components of

sewer systems;
- to minimize the leakage of wastewater from the systems; and
- to determine economically optimal systems.
Four specific models have been developed:
- a wastewater and stormwater runoff model for sewer networks;

- a wastewater treatment plant model;
- a sludge treatment plant model; and
- an overall model for the economics of a system (using an optimization technique).
The models may be linked together, but are mostly operated independently.
An up-to-date program description of the network model and the wastewater

treatment model were given in the 1976 report,(*•) but because a few new features
and improvements have been added to the models over the last few years, updated
descriptions of the models seem appropriate.
The Network Model. The main features of the network model are:
a) The rainfall intensity can vary with time, i.e. different intensities may be
given for each minute of rainfall.
b) The runoff coefficient can vary with time, i.e. different coefficients may be
given for each minute of rainfall.
c) Time of entry vs. contributing area functions can be given, permitting

consideration of different surface storage characteristics.
d) The storage capacity of each of the sewer lines is considered.
e) The velocity of flows is made a function of the water depth in the partially
filled sewer.
f) Storage tanks may be inserted at any point in the system. The necessary tank
volume for a given rainfall can be computed when a maximum outlet discharge
from the tank is specified.
g) Storm overflows can be considered at any point in the system. The total
bypassed and diverted volumes of water and pollution are computed.
h) The sum of industrial and domestic wastewater flow together with infiltration
water is considered as a constant discharge in time, and each sewer line may
have its own value for wastewater "production".
90
i) Pumping stations and other inflow hydrographs may be given at any point in
the system.
j) Transportation of pollutants per unit time will be computed at any specified

location as a function of time after the start of rainfall. The pollution
component is given as input data in the stormwater runoff, expressed as a
value for each minute after the start of a rainfall. The gram component
produced per person per day must also be given. Total amount of the amount
diverted from storm overflows is computed.
k) Besides computing water discharge (1/s) and transport of pollutants (g/s),

the model can find the smallest standard pipe dimensions for the desirable
sewer lines that will avoid backwater for the particular rainfall considered.
1) The model computes the capital costs for the total sewer network. Necessary
input data are per cent of rock in each cross-sectional trench area and
diameter of sewer when this is not to be computed by the program.
m) When the sewer network has too small a capacity, backwater may occur. The
backwater level may be computed for each point of the system and is presented
as a function of time after the start of a rainfall.
n) Tunnels, canals and pipes may be accommodated.
o) The computing step between each runoff situation may be varied.
The Wastewater Treatment Plant Model. The main objective in the use
of this model is to study the performance of a wastewater treatment plant
receiving both sewage and rainwater runoff over a period or a whole year.
The total program configuration, i.e. the combination of network model

and treatment plant model, is used to analyze how a combined or a separate sewer
system in a given area will perform. On the basis of rain intensity vs. duration,
total precipitation duration and frequencies for a year's rain activity, some
representative base rainfalls are chosen. These rainfalls are given as inputs to
the sewer network model for a given area. The results in terms of flow (1/s) and
BOD load (g/s), as functions of time, may further be used as inputs to the
wastewater treatment plant model.
The Sludge Treatment Plant Model. Essentially, this model is still at

the testing stage. At present the following units are considered: thickener,
anaerobic digestion, aerobic stabilization, sludge centrifuge, sieve band press,
sludge drying beds, vacuum filters and final disposal. Figure 1 shows how one
unit fits into the model. The combination of units to be included may be
designated by the user (process alternative). Each flow includes water discharge
(1/sec), suspended solids (rag/1), particulate BOD (mg/1), dissolved BOD (mg/l),
total phosphorus (mg/1) and nitrogen (mg/1).
Recent Model Research. Of the more recent model work performed in Norway,
two studies by the Norwegian Institute for Water Research (NIVA) deserve mention.(13)
One deals with an evaluation of the base rainfall method in urban drainage
simulations and the second makes a comparison of the SWMM runoff model'*^) with the
time-area method.
91
WATER RETURNED
TO PLANT
UNIT OPERATION
1
SLUDGE^ 1. FUNCTIONS OF SLUDGE
, 2 NEXT
REMOVAL EFFICIENCY
INPUT OUTPUT UNIT
2.FUNCTIONS OF COST
FIGURE 1 - UNIT OPERATIONS IN SLUDGE MODEL
Evaluation of the Base Rainfall Method in Urban Drainage Simulations.

When urban drainage models are used for planning purposes, there is often a need
to simulate runoff for long-term periods, e.g. one year or more. There are two
possibilities: to simulate the whole year's recorded rainfalls; or to distribute
the recorded rainfalls into a few (5-15) base rainfalls, and run the simulation
with these base rainfalls given the average rain intensity, the average rain
duration and the frequency of each base rainfall.
Simulations were made with both recorded rainfalls and base rainfalls.
In one simulation, a year of rain activity was represented by 89 rainfalls (minor
rainfalls with very little runoff were excluded). These rainfalls were converted
into 10 base rainfalls. In a simulation with a 112 ha combined sewer area with a
detention basin or an overflow at the outlet, the base rainfall method resulted in
about 10 per cent less overflow volume than the simulations with the actual
rainfalls. The computer costs were reduced by about 90 per cent when the base
rainfall method was used.
A Comparison of the SWMM Runoff Model with the Time Area Method. In the
sewer network model NIVANET, developed by the Institute for Water Research, there
are two possibilities for simulating the surface runoff from a catchment basin:
by using the time-area method; and by using the detailed runoff subroutine of the
SWMM model. The outflows from each subcatchment, calculated by either of the two
alternative methods, are then routed through the pipes. The routing technique in
NIVANET is based on the TRRL method.C15>
From a theoretical point of view, the simulation of overland flow

hydrographs by the SWMM runoff block is logically closer to reality than the time-
area method. Because of its more sophisticated description, however, the SWMM
runoff block requires more input data and more computer time. The purpose of the
study was to present a simple comparison of the two runoff models, and to give
some practical recommendations for the use of the two alternative methods.
Results of the Comparison. Simulated outflows from a subcatchment are

shown for two different block rains in Figures 2 and 3, which illustrate the
different ways the two models compute runoff. The differences in runoff volumes
in Figures 2 and 3 of 25 and 23 nw, respectively, are mostly the result of
ignoring depression storage in the time-area method (NIVA). Because of neglecting
depression storage, the time-area method also generates faster rises and
92
50 mn
FIGURE 2 - SIMULATED HYDROGRAPHS AT THE OUTLET FROM A SUBCATCHMENT

(Runoff Volumes: NIVA, 87 m 3 ; and SWMM, 62 m 3 )
0 70 (0 60 60 100 U0 160 m m
FIGURE 3 - SIMULATED HYDROGRAPHS AT THE OUTLET FROM A SUBCATCHMENT

(Runoff Volumes: NIVA, 138 m3; and SWMM, 115 nß) 93
recessions in computed hydrographs. Time of peak and peak flow may differ as a
function of intensity and duration of the block rain. When using synthetic
block rains, however, it may be realistic to neglect depression storage for SWMM
as well. This depends on the rules for constructing such rains. In most cases,
the small pre-rains that fill up the depression volume are excluded, and runoff
can be assumed to start immediately. The effect of detention storage can be
included in NIVA because it includes the possibility of giving the runoff
coefficient as a function of time.
Results from simulations using the hyetograph for the 10-year base
rainfall in Oslo are shown in Figures 4 and 5. Here, also, there is a difference
in runoff volume, due to depression storage. There is quite a good agreement,
however, for the time to peak, and the differences in peak flows are negligible
from a practical point of view.
Results of simulations using recorded rainfalls are shown in Figures

6 and 7. The differences between the hydrographs for the two methods are of no
practical significance.
All the results presented here, however, are functions of values

assumed for the model parameters. Calibration is necessary for practical use of
runoff models. Other studies have shown that calculated peaks in many cases are
earlier and larger than observed values, but that agreement between calculated
and observed hydrographs is improved with the magnitude of storm events.
When the SWMM runoff model is used, it is necessary to characterize the

catchment by way of 11 parameters. For the time-area method there are only 3
parameters. The cost of data acquisition and its preparation for the SWMM runoff
model will therefore be considerably larger than for NIVA. The costs of running
the two models also differ. Depending on the rain type and duration, the SWMM
requires 40-75 per cent more computer time than the time-area method. Computer
time requirements (CPU-time, UNIVAC 1110), are as follows:
Block Rain Hyetograph Actual Rainfall

Duration Duration Duration
Method 12 min. 30 min. 140 min.
Time-Area (NIVA) 1.45 sec. 2.53 sec. 3.21 sec.

SWMM Runoff 2.55 sec. 3.58 sec. 5.67 sec.
The larger time requirement of the SWMM runoff block reflects the difference in
the way the two models calculate runoff. SWMM requires an iterative calculation,
employing the Newton-Raphson technique, while the time-area method (NIVA) allows
a direct solution at each time step.
It must be kept in mind that there was no allowance for runoff from
pervious surfaces in these studies. If there is runoff from pervious surfaces,
the differences between NIVA and SWMM can be expected to be larger than for the
cases presented.
From a practical point of view, considerations of computer cost and

the cost of data acquisition and preparation have to be weighed against the
94
0-
Â s~l
100-
200
1
8000- 1 \ SWMM
Al NIVA
I 11^*
7000-
6000-
5000-
4000-
/|
3000- /l
/ I
/ 1
7000- / / 11
/ I
/ / \\
1000- / /
/ /
0-
G 10 20 30 40 5C
50 min
FIGURE 4 - SIMULATED HYDROGRAFHS FIGURE 5 - SIMULATED HYDROGRAPHS

AT THE OUTLET FROM A AT THE OUTLET FROM THE
SUBCATCHMENT WHOLE CATCHMENT
(Runoff Volumes: NIVA, (Runoff Volumes: NIVA,
265 m3; and SWMM, 237 m3) 7390 m3; an d SWMM, 6560 m3)
95
^
3 O IN
oä vo in
a
¡3
O
H
Ut
«t
ro
-s! e
> vO
H o%
55 ITl
^
in
en
3
iH TJ
O C!
> rt
>H ...
men
o E
G
3 en
oí r^
>—> m
W
OS
H
' 3a j
38JHHOST:Q
96
desired level of precision. The results presented in this paper show that
differences in calculated hydrographs are not direct functions of the
sophistication of the models. The costs, however, seem to be a function of the
degree of detailed description.
The time-area method is well known and a practicing engineer using this
method is able to work with well-known quantities. Lesser computer cost and data
acquisition and preparation favor the time-area method. As there do not seem to
be any significant differences in runoff, models based on the time-area method are
sufficient for most practical applications, as long as runoff from pervious areas
is not involved.
However, for detailed studies and research, more complex models such as
SWMM have obvious advantages. (16)
Abbreviations Used in This Paper
PRA: Research program for purification of wastewater, administered by the

Norwegian Ministry of Environmental Affairs, 1970-1976.
NVE: The Norwegian Water Resources and Electricity Board.
NIVA: The Norwegian Institute for Water Research.
Address List for Norwegian Institutions Referred to in This Paper
The Norwegian Institute for Water Research,

P.O. Box 233, Blindem, Oslo 3.
The Norwegian Meteorological Institute,

P.O. Box 320, Blindem, Oslo 3.
The Norwegian River and Harbour Laboratory,

Klebuvn. 153, N-7000, Trondheim.
The Norwegian Water Resources and Electricity Board,

P.O. Box 5091, Majorstua, Oslo 3.
University of Trondheim,
The Norwegian Institute of Technology, N-7000 Trondheim.
References
1. Selthun, Nils Roar, "Urban Hydrological Modeling and Catchment Research in

Norway," ASCE UWRR Program Technical Memorandum No. IHP-9, ASCE, New York,
N.Y., 18 pp., December, 1976. (NTIS No. PB 267 587).
2. Lindholm, 0.: "Systemanalyse av avl^psanlegg". PRA, users report no. 1,

ISBN 82-90130-00-4, 57 pp., Oslo, 1975.
3. Lindholm, 0.: "Forurensing i overvann", NIVA report 0-57/74, 55 pp., Oslo,

1976.
4. Lindholm, 0.: "Valg av modellregn", PRA users report no. 6, ISBN

82-90180-05-5, 16 pp., Oslo, 1975.
97
PRA project 4.2: "Introduksjon av maleprogram og malefelter." Project
report no. 1, NVE, Oslo, 1974. 140 pp., English summary.
Tveit, J.: "Insulation against ice at measuring weirs". Nordic Hydrology,

5, pp. 32-49, 1974.
Hetager, S. E. : "Presentasjon av forskningsprosjekt PRA 4.2: Urbaniseringens

innvirkning pa avl^psforholdene i sraâ nedbtfrfelt". (Presentation of a
Norwegian research project: The effect of urbanization on runoff from small
catchments). In PRA/IHD: "Kvantitativ urban hydrologi," a Nordic symposium
on urban hydrology, pp. 65-77, Oslo, 1975.
PRA project 4.2: "Dataoversikt 1972 - 1974". Project report no. 2, NVE,
Oslo, 1975. 197 pp., English summary.
Roald, L. A., and S. E. Hetager: "Mulige effekter av urbaniseringen pá

avrenningsforljípet i Moss-Oslo omrâdet". ("Effects of urbanization on
runoff from small catchments in the Oslo-Moss region"). In PRA/IHD:
"Kvantitativ urban hydrologi," a Nordic symposium on urban hydrology,
pp. 132-145, Oslo, 1975.
PRA project 4.7: "Forurensning i overvann". Report 0-57/74. NIVA, Oslo,

1976. 55 pp. + data appendix.
Balraer, P.: "Sn^smeltingens betydning for avl^psanlegg". (The effects of

snowmelt on wastewater plant). Report (4-23, NIVA, Oslo, 1978. 14 pp.
Selthun, N. R. : "Rainfall/runoff relationship in Norwegian urban catchments".

Presented in Nordic Hydrological Conference 1978, pp. 207-213, Finland 1978.
Norsk institutt for vannforskning/Computas A/S, 1978: Diraensjonering og

planlegging av avl^pssystem, Oslo.
Huber, W. C. et al., 1975:- Storm Water Management Model. User's Manual

version II. EPA-670/2. 75-017. USA.
Watkins, L. H., 1962: The design of urban sewer system - Road Research
Technical Paper No. 55. Dept. of Scientific and Industrial Research, London.
Waller, D. H. et al., 1976: Urban drainage model comparison for a catchment

in Halifax, Nova Scotia. Research Report No. 43. Environment Canada,
Ministry of the Environment, Ontario, Canada.
Section 9 Progress since 1977 in the Netherlands
by F. C. Zuidema*
Introduction
The national report on urban hydrology for the Netherlands was

completed early in 1977.'*' Contributions to that report were acquired in the
form of reactions to a letter of request from the Netherlands National Committee
for the I.H.P. to a number of experts and organizations. The same procedure has
been used to obtain information on progress since early 1977 for the present
paper, and the writer hereby acknowledges these vital contributions.
Various official bodies in the Netherlands are entrusted with

responsibility for control of the discharge and quality of surface water.
Municipalities have control of the sewer systems, and the open water of the
rural areas is controlled by the Provinces (delegated to "polder-boards" and/or
purification-boards) or by the Government (delegated to the Ministry of Public
Works). When the Pollution Surface Water Act came into operation in 1972, these
administrative aspects were regulated. For every discharge of water from an
urban area, dispensation must be granted by the proper authorities. Moreover,
absolute limits are set to the annual maximum of overflow frequency and the
maximum quantity of overflow water. The decision about the maximum allowable
emission of pollution is taken with the aid of models that have been verified
insufficiently in practice until now.
Lelystad Catchments
Details on the instrumented catchments in Lelystad were presented in

the 1977 report.(1) These catchments include a residential area, a parking lot,
a shopping and office center and two flat roofs, where investigations have been
in progress since 1969. During 1977 and 1978 the research emphasis was mainly
on models for the investigation of the rainfall-runoff process, and on the
quality of water in an urban area. Information was provided by C. H. van Dam
and G. A. Ven, IJsselmeerpolders Development Authority, Lelystad.
Rainfall-Runoff Modeling. In cooperation with the Technical University

of Delft, various routing models have been developed for transferring rainfall
into sewer inflow. These models have been optimized for ten events, and were
compared with each other. The routing models used were: the Nash cascade
model; a linear model based on Laguerre functions; a unit hydrograph model; and
a non-linear model based on a Volterra-series. A variable proportional-loss
was used, by which the losses at the outset of a storm are high and diminish
proportionately with-more rainfall. The more sophisticated models performed
better for the individual events. If, however, the reckoning was done with
mean parameter values, a distinct preference for one of the models was not
evident. The conclusion has to be that a model as simple as possible, such as
the linear reservoir or the Nash cascade model, is preferred for the description
of the rainfall-runoff process on impervious areas.
*: National Council for Agricultural Research, The Hague 2076, Netherlands.
99
In April 1978 in an International Workshop at the Institute of
Hydrology in Wallingford, U.K., data from three sources were available in a
data base, consisting of 188 rainfall-runoff events on 16 urban subcatchments.
Using those data a number of loss models and surface routing models were
investigated. The relationship between model parameters and catchment
characteristics was examined. One of the conclusions was that the choice of
a surface routing model is less critical than the way it is used.
(2)
Planned Extension. The practical utility of the hydrological
research will increase considerably if the research results can be made
applicable for other areas. To correlate the prevailing parameters as well as •
possible with the physical characteristics of the catchment area, it is
essential to have available a number of areas under investigation with different
properties (extent, impervious area, type of buildings). So it was decided to
add another catchment area to the already existing ones, a fairly densely
built-up residential area of 3.5 ha, situated in the immediate vicinity of the
other catchment areas. Contrary to the houses in the existing areas, the houses
in the new area have slanting roofs.
As in the other measurement areas, the following recordings will be

made in the added area: rainfall; subsurface drain runoff; stormwater runoff";
and groundwater level. Besides these quantity measurements, samples will be
taken of a number of water quality parameters. By means of intensive recording
and by combining water quality and water quantity research, it will be possible
to gain a better insight into the amount of pollution of the receiving water
caused by an urban area.
Water Quality Model. A sewer system in which wastewater and stormwater

are discharged jointly is called a combined system. In the Netherlands, the
wastewater of the combined system is pumped to a wastewater treatment plant,
which discharges into the water outside the urban area. During heavy rainstorms,
stormwater mixed with wastewater flows via weirs directly into the open water-
courses of the urban area. The rate at which this occurs depends; on the pumping
capacity of the wastewater treatment plant and the available storage capacity of
the sewer system. In a separate system, the wastewater is pumped to the
wastewater treatment plant while the stormwater flows either directly or via the
subsurface drainage system into the open urban watercourses. If there are
improper connections with the wastewater sewer, this stormwater contains also a
part of the wastewater. The choice of the type of sewer system in the new towns
in the IJsselmeerpolders is of importance in view of the water quality in the
urban area. A separate system will give a more or less permanent waste load on
the surface water, while a combined system deals with high peak waste loads. To
get a better idea of the relevant processes, water quality models have been
developed to simulate the two sewer systems. In those models that have been
evaluated up to now, the water quality of the surface water is described by only
one parameter, the chemical oxygen demand. Presently an extension is being
carried out, in which more water quality parameters are being modeled concurrently.
A schematic respresentation of the two types of systems modeled is given in
Figure 1.
On the basis of the results obtained from the models, in the combined
system the height of the specific peak loading appears to be highly decisive for
the water quality of the canals, while there is little correlation between the
amount of overspilled water and the pollution discharged at the same time.
Therefore, the quality of the water in the canals and also the annual pollution
100
coMbiMd syttM
TÇ5&¿d
¿«•31
^JP^^^—S^
J L
T r
1 mói«)! rmnoir
2 waiH Wittr u w t r
3 riidwittr m u e
iiptrat« «yttM
4 u(fem CMMtt
5 trHtnmt plant
6 fall« conntction
FIGURE 1 - WATER QUALITY SYSTEMS

MODELED, LELYSTAD
101
load on the canals is largely determined by the height of the specific peak
discharge and to a lesser extent by the number of overflows. So it should be
kept in mind that limitation of the number of overflows will not mean a
proportional improvement of the quality of the open water.
In the separate sewer system, the large quantities of water from the
stormwater sewer yield, in spite of the sometimes considerable waste load, a
fairly constant water quality in the open watercourses. Large specific peak
discharges, adversely affecting the water quality for a long time, practically
do not occur.
Comparison of the total annual pollution discharge of both systems

leads to a slight preference for the combined system. However, the total annual
pollution discharge, because it affects the quality of the open water, is less
important than the pollution concentration which occurs in the open water.
Comparison at last of the number and the height of the specific peak loadings on
the open watercourses from the separate and the combined systems leads to the
conclusion that in almost all investigated cases the separate system might be
better.
Neede Catchment.
The Neede catchment was described in the 1977 report.(*' Suffice it

to say that the combined sewer system being gaged has a service population of
about 9,000 persons. As many measuring points as possible have been connected
to a data logger. New information has been provided by F. Slijkoord. (3)
On the basis of the work of many other investigators, it is suggested

that the sewer system will react as a non-linear system of the first order under
the influence of rain events.
(4)
In estimating the parameters of the runoff
process of the sewer system, a quasi-linear approach will be followed, with the
time-constant and the runoff coefficient depending upon the morphology of the
rain considered. The aim of the investigations in the Village of Neede is to
verify and complete the current models.
In order to achieve correct information about the runoff from
precipitation on the catchment, information about the wastewater flow is
essential. From examination of the registration of the venturi meters on dry
days, it turned out that the discharge of wastewater to the treatment plant is
time dependent. Three different periodicities have been observed: daily, weekly
and seasonal. Unmistakably, the first two cycles are due to the behavior of the
inhabitants of the village. The seasonal effect is a long-term effect, and its
maximum contribution is of the same magnitude as the daily amount of wastewater
in the summertime. This effect might be caused by infiltration of groundwater
and by possible runoff of water from the rural surroundings into the sewer
system. In the system this effect will be considered as a base-flow, which
could be estimated from the recession of the wastewater discharge to the
treatment plant during, the night. Attempts have been made to derive the total
run-off to the treatment-plant from registrations of the water-table. It
turned out, however, that the calculated discharge is strongly dependent on
the position of the point of registration in the field. Under development is
a method to estimate the parameters of the runoff process by making a regression
of the precipitation direct to the water-table. The first attempts of this
exercise show promise. It appears that the sewer system reacts as two fast-
reacting reservoirs, each with its own time-constant. This phenomenon can be
102
interpreted on the basis of the construction of the network of pipes.
Larger Sector Modeling
Described in this subsection are some results of studies dealing with

the influence of urbanization on the water management of the IJsselmeerpolders
and on wastewater planning in Rotterdam.
Influence of Urban Runoff. Information on this subject has been

provided by J. E. G. Bouman of the IJsselmeerpolders Development Authority,
Lelystad. The new towns of Lelystad and Almere are being built in Flevoland,
one of the new polders in the lake IJssel. Lelystad is meant for 100,000
inhabitants and is situated in the northeastern part of Eastern Flevoland.
Almere is meant for 125,000 to 250,000 inhabitants and is situated in the
western part of Southern Flevoland. The Eastern and Southern Flevoland polders
are one unit. In terms of water management there are two sections, namely the
High Section with a water level of 5.20 m below sea-level and the Low Section
with a water level of 6.20 m below sea-level. The main canals and the pumping
stations are designed to accommodate a total discharge of 11 mm/day. In
Lelystad the rainwater discharges on the Low Section only, whereas in Almere the
rainwater discharges on both Sections. In order to design watercourses, culverts,
weirs, etc., within the urban area, and on the other hand to evaluate the
discharge from the urban area to the rural areas, a computer-model has been
developed to calculate flow and levels in open watercourses. The relevant data
for the model have been obtained from the "Urban catchment research project
Lelystad". In this research project rainwater discharges from areas with
different types of cover are investigated. The discharge from rural areas can
be obtained from a current research project on this matter in Flevoland (arable
land, pastures and forests).
In Lelystad and in Almere a separate sewer system is in use. One

system is used as a stormwater sewer and the other system is used to transport
only wastewater from households and industry. For water management, only the
stormwater sewer is important. Research showed that the time-lag between
rainfall and sewer discharge to the city canals is very small and can be
neglected in this type of calculation. The runoff coefficient of the paved
area (0.8) has to be taken into account. The surplus rainwater on the unpaved
areas within the urban areas is discharged through plastic drainpipes into the
city canals. The discharge of these drainpipes during design situations amounts
to about 17o of the discharge from the stormwater sewer. Research showed that a
constant discharge of 5 mm/day of the drainage system would describe the actual
situation for wet periods accurately enough. The amount of open water varies
from 3 to 7 per cent, with an average of 3.5 per cent.
The discharge from rural areas has been estimated by means of two
different simulation models (Nash cascade model and Neumann-Feddes groundwaterflow
model). These models have been calibrated with the results of the research on
discharge measurements in rural areas. In the calculations an additional
boundary condition has been introduced on the drain discharge. The maximum drain
discharge has been limited to 20 mm/day.
The existing water management system of Flevoland has been based on

discharge criteria which apply for rural areas. In order to establish a water
management system for future use it is necessary to evaluate the influence of
the new towns on the existing water management system. The evaluation was done
103
for such periods as are critical for the urban areas. For different parts of
the cities different types of periods are critical. In this case a wet period
of 10 hours was chosen with a total rainfall of 53 ran. It could be expected
that this type of rainfall would yield the highest rises in levels in the system
as a whole. The computer model calculates unsteady flows and levels. In order
to run the model, a scheme of "branches" and "nodes" has to be drawn. A branch
represents the flow capacity of a watercourse. A node represents the storage
capacity of a watercourse. Two alternatives have been taken into account with
regard to the land use In the polder: rural land use only; and mixed urban and
rural land use. In this way the influence of the urban areas on the watercourses
in the rural areas can be simulated. Figure 2 shows the highest calculated
water level rise in the High Section (Hoge Afdeling) and at the connection of
Almere-Stad.
The results of the modeling show that for a mixed land use the
maximum rise in level will be less than the maximum rise in level for rural
land use. This is in spite of the quick response of the discharge from the
urban areas. There are reasons for this behavior. Especially in the case of
Alraere, large water loads are incorporated in the water management system. This
leads to an increase of the storage capacity of the system and yields a smooth
discharger pattern. Furthermore, about half of the urban area is paved and the
runoff from these areas starts very quickly. The remaining part of the urban area
is also covered with a one-meter sandfill. This leads to a very slow discharge
process. And last but not least, a temporary storage can occur in the urban areas
in favor of the rural areas.
A related study was on the influence of the building of the New Town of
Alraere on the geohydrological situation of Flevoland. Space limitations prevent
a discussion of findings, but it is important to note that A. Overwater of the
IJsselmeerpolders Development Authority has found the groundwater model used to
be a useful device to describe the geohydrological situation in the area. The
results of the calculations have given a good insight into the changes in seepage
and piezometric heads within the area.
Wastewater Planning and Research in Rotterdam. According to Gerard

van der Kroon, Rotterdam Municipal Works Department, it is expected that
sophisticated modeling and measuring techniques together with comprehensive master
planning will become important tools for determining the future needs of the
Rotterdam wastewater collection and disposal system. It is emphasized that
collaboration and participation in research of experts of various disciplines
are prerequisites for reaching the goal of a rational wastewater management
system. Since the Dutch Clean Water Act became enacted (1970), cities, private
enterprises and communities which had not yet built wastewater treatment plants
are being required to develop plans for the treatment of their wastewaters (1970-
1985). As a consequence, cities like Rotterdam reorganized their technical and
administrative staffs to manage their future wastewater problems. Simultaneously,
Rotterdam has been developing plans to upgrade its existing sewer systems and
treatment plants. Of particular relevance Is the reorganization now under way of
the Rotterdam Wastewater Management Department into two main sections: operations
and maintenance; and planning and research. Tasks of the planning and research
section are the following: data analysis and operational research; standards
evaluation, and development of new facilities that accrue from newly developed
standards; planning of facilities improvements and normalizations; development
of design criteria and the planning and implementation of new facilities in town
development areas by alternative schemes; and planning and implementation of
large maintenance projects.
104
wattriaval M m + » i Itvol
ion
ring iN the dettinitioni
triturai m
400
470
Kighttt riM ¡n th*
Hogt Afdaling
- - : ^
44(1 ¿ > ^
480 * ^ . <
^^y. * th« connection of Almara-Stad
y>
S*
wrt* tha Hogt AhMing
4*0 /y
' • ^
SOO
S 70 J^K:
540
24 4« 60 M
FIGURE 2 - CALCULATED WATER LEVEL RISES
105
A raingage network is being operated and surface water quality data
is being collected from about 50 surface water sampling stations extending all
over the city, with a time interval of one month between samplings. Phreatic
groundwater stages are being measured periodically. There are in total 1,600
observation wells that are uniformly spread over the city and some have records
that date back to as early as 1920 or thereabouts. There is historical reason
for the presence of so many observation wells since at the beginning of this
century several buildings in the town centre had been severely damaged by
artificial, lowering of the groundwater level (1919). Data analysis and
verification of groundwater stages in Rotterdam is nowadays poorly accomplished.
This has very probably been caused by the vast amount of groundwater data that
requires the use of computer analysis and such special studies are now
undertaken by the land surveying section of Rotterdam Municipal Works to resolve
the groundwater stages in a satisfactory way.
Drainage catchments have minimum-maximum sizes of 10-250 ha of impervious

area inside the basins and this requires very variable design criteria for the
collection sub-systems. There is also some variety in land uses. Most catchments
are mutually connected, which is of advantage in case of an accident (i.e. a break
of a force main). The oldest catchments are characterized by a lack of sufficient
surface water storage (widely differing from 0 to 1% of the total surface). In
park districts, a minimum of 27» to 4% of the total surface is being occupied by
surface water; i.e. partly constructed for recreational purposes and partly for
storage of overflow water storage from the sewer system. All wastewater collection
sub-systems are of the combined type. The storage capacity of the present
collection system is widely variable and ranges from very large (i.e. 10 mm) to
very small (i.e. 2.6 mm). Rotterdam is fortunate to possess a well maintained
sewer system. The mean life time of its sewers is much smaller than 40 years.
Generally, the pumping of the collection system and of the watercourses are
intertwined.
In a future arrangement, pumping of the wastewater collection system will

be separated from that of the surface water drainage system because collected
surface waters are biologically, clean and do not need purification in a treatment
plant. At present, the wastewater collection system is so complex that present
knowledge of the wastewater control system is inadequate. The integrated master
plan calls for the concept of an automated system approach and very probably the
use of computers for system analysis. In this respect, it is expedient to make a
basic study of the capabilities of computer implementation starting with the
present information system. The first step must be an inventory of the potential
data sources and system facilities and the establishment of systematic wastewater
planning.
Conceptual models and their analysis were used as a means for efficient
and dependable facility planning and management, for resolution of design criteria,
and for use in connection with development of operational control strategies.
Generated models accommodate a variety of water balance inventories of many
catchment basins. Simple subroutines have been developed for the analysis of
dry-weather flows, general runoff characteristics, stormwater characteristics of
special rain events (1973-1979), sewer infiltration, stormwater overflow frequency,
control of stormwater overflow pumps, and last but not least for the evaluation of
impervious area. Additionally, operational control strategies have been developed
which are capable of handling wide diversities in management practice. It has
still to be considered whether hydraulic models in the future must also accommodate
conduit transport hydraulics. The need for hydraulic models will be lessened
106
because of the emphasis on the building of in-storage capacity. Field data
have been used in model tests. A feature of the Rotterdam model subroutines
is the capability of a continuous analysis of operational processes. Models
are being utilized in the analysis of existing sewer systems and in research
which has been executed in urban catchments on a full technical scale. Dynamic
groundwater flow modeling has been used in combination with estimates of in-sewer
infiltration rates and indications of unwanted discharges in the catchment basins
of Hillegersberg and Schiebroek (1977).
The stormwater overflow pump concept has been used to effectuate

continuous automatic control of combined wastewater and stormwater pumping to the
wastewater treatment plants and peak stormwater flow directly into the river in
the catchments of Hoek van Holland, IJsselmonde and Schiebroek (1973 - 1979). So
far, the concept of stormwater overflow pumping has been a great practical success
because the pump rate to the plant as well as the pollution of the surface waters
(inland water and river water) have been very much reduced. Future research
activities are needed to find the way for coupling the various model subroutines
and for improving the control methods to make research findings better
implementable and thus achieve better operational control. It is expected that
sophisticated modeling and measuring techniques together with comprehensive
master planning will become more and more important for determining the future
needs of the Rotterdam wastewater collection system.
National and Provincial Modeling
The development of a national balance model, the purpose of which is

to assist in the determination of water management policy, has been started some
years ago. From water balance studies it has been concluded that the total
quantity of water available in a dry June in the year 2000 would be sufficient,
but that additional infrastructural facilities would be necessary to safeguard
supplies at all times and places. The model is used for 10-day periods and for
33 water distribution points in the country. Recently the study has been
extended to a Policy Analysis for the Water Management in the Netherlands (PAWN),
which includes a strategy design after screening of alternatives, the assessment
of impact and a sensitivity analysis which will enable decision makers to choose
a preferred strategy.
Integrated water resources management has also been sought at the

Provincial level.
Province of Gelderland. Since 1971 a committee in the province of

Gelderland has been developing a scientific base for optimal planning and
management of available surface water and groundwater, from the standpoint of
both quantity and quality. (1) In this approach, the water resources management
system is divided into three types of elements, viz., social, natural and
artificial elements, and into a number of levels (sub-systems). These levels and
elements are being investigated at varying degrees of detail with the aid of
mathematical models, mostly by interdisciplinary teams. In this way, the results
of measures to be taken can be predicted, so that alternative plans can be
evaluated and mutually compared. This procedure includes the possibility of
dividing the complex decision process over various levels, so that it can be
better solved. The application of a planning procedure within the decision
process enables attainment of an integral water management plan. As a selection
procedure, multicriteria decision models as well as models with only one
criterion (utility) can be applied. Two recent multi-criteria decision models
107
are being applied in Gelderland, namely the method of Saaty (U.S.A.) and the
method of Paelinck (Netherlands). Both methods are very similar, but there are
some technical differences. (5)
The method of Saaty and the further developed
method of Paelinck may be very useful in the management and planning of water
resources systems. An attempt to "optimize" a complex water resource problem
(more factors besides the location) will be done this year by the Netherlands
Economic Institute.
Province of Drenthe. The province of Drenthe includes an area of
268,500 ha. The central part, with an altitude of 15 to 25 ra above sea-level,
inclines to the borders with heights ranging between 0 - 5 m above sea-level.
The territory of the province does not border on the sea and is not cut by a
river. During the growing season, there is mostly a shortage of water, with the
fields completely dependent on precipitation. Only a limited water replenishment
is possible by pumping water out of the Meppelerdiep canal, which has an open
connection with the IJsselmeer. As a result of many developments, the demand for
water is continuously increasing. The intensification of agriculture makes more
exact water management especially necessary. The water discharge system has been
completely modernized during.the last decades. However, the capacities of
possible water supply extensions are too small to cope with the increasing demand.
This was evident especially during the very dry summer of 1976. In order to get
a view on the extent of water shortages,, a study has recently been started to
obtain answers to a number of questions, such as the purposes of water supply,
the quantities needed for many years on an average, the sources of water and the
alternatives in case of different degrees of dryness. For a good comparison of
costs and benefits of the alternative plans and the further material and
immaterial advantages and disadvantages, data has been collected on dry-class
years, namely the year 1976, a 10%, a 20% and a 50% dry year, and on water
requirements for various kinds of land use.
Generally speaking, it is possible in the Province of Drenthe to
withdraw a part of the water required in the growing season from the quantity
which has been preserved in spring. Preserving can be especially effected by
raising the weirs in time after winter, by which means higher levels are obtained
in surface water and groundwater. During the remaining time of the growing
season, water has to be supplied from outside the Province, namely from the big
lake IJsselmeer.
In the determination of cost/benefit considerations a comparison has

been made between the situation with and the situation without a water supply
project. In the latter situation, the zero-situation, it is assumed that the
existing supply system has not been used at all. The plan, in which the
difference between the total benefits and the total costs of the water supply
system, the net plan-effect, is maximal, is the most attractive.
Calculations of sensitiveness prove that the order in the net plan-

effect does not change if the value of the complementary water supply is changed
for the benefit of various pressure groups. Attaching importance to the
different values of the various pressure groups may influence the choice. The
importance of the values will be determined in view of political considerations,
which will influence the order of plans.
Information on the Drenthe water supply study was provided by M.

Tamrainga, Drenthe Provincial Waterboard.
108
Future Elaboration
An extensive version of this Section 9 will be published by the

Netherlands National Committee for the IHP.
References.
1. Zuidema, F. C , "Urban Hydrological Modeling and Catchment Research in the

Netherlands," ASCE UWRR Program Technical Memorandum No. IHP-10, ASCE, New
York, N.Y., 42 pp., January, 1977. (NT1S No. PB 267 587).
2. Kidd, C. N. R., Rainfall-runoff processes over urban surfaces, Proceedings

of an International Workshop, April, 1978, Institute of Hydrology,
Wallingford, U.K., Report No. 53.
3. Slijkoord, F., Consequences of urbanization in relation to the quantitative

aspects of drainage, Proceedings and Information No. 24 Committee for
Hydrological Research TNO, "Urbanization and water management," The Hague,
1978, Netherlands.
4. Zondervan, J. G., Modeling urban runoff, a quasi-linear approach, Thesis,

Agricultural University, Wageningen, Netherlands (1978), Publ. PUDOC.
5. van de Nés, Th.J., Applicability of multi-criteria decision models with

multiple decisionmakers in water resources systems, Symposium on modeling
of water resources planning, Wageningen, 1979, (in press).
Section 10 Progress since 1976 in Poland
by Pawel Blaszczyk*
Introduction
This paper is an extension of the report "Urban Runoff Research in

Poland,"i 1 » 2 ) in which the state of research in Poland in April 1976 was
presented. Presented are the progress achieved and the results obtained during
the period April 1976 - May 1979.
During 1976-79 research was continued in the Warsaw sewer catchment

area characterized in the previous report.(1>2) As a result of research the
following was elaborated:
- formulas and calculation algorithms were changed for the rational method,
which is widely used in Poland to design rainwater sewers;
- method of forecasting runoff hydrographs in the rainwater sewer scheme based

on the "black box" model;
- method of simulating pollutographs of urban runoff; and
- principles of forecasting runoff pollution to meet requirements for planning

new drainage network schemes.
Supplements and Changes in the Formulas and Calculation Algorithms Used in Poland
in the Rational Method
The rational method is widely used in Poland to determine the transverse

dimensions of rainwater sewers. The basic formulas of this method, applied
throughout almost the whole of the country, relate the flow rate in the sewer to
the reliable rainfall intensity, catchment area, and runoff coefficient:
Q = 0.001 +Fq ,
where Q is the flow rate in the sewer in m3 / 8 ,

t is the average runoff coefficient for the catchment,
F is the catchment area in ha, and
q is the mean rainfall intensity in dwß/a ha at the concentration
time t.
For the requirements of rainfall sewer design, the following formulae) is used
throughout almost the whole of Poland:
47o^nr
q =
t 0 - 67
where C is the rainfall frequency and

t is the concentration time of reliable rainfall in minutes, where
Research Institute on Environmental Development» 02-078, Warsaw, Poland.

Ill
t - 1.2 Stp + 2 - 1 2 min./tp , the flow time of runoff through
n
segments p of the sewer, Etv> from the beginning of the sewer to
1 i
the cross-section under consideration.
The value of the runoff coefficient is assumed to be constant during the

concentration time of rainfall and determined on the basis of the relationship:^'
* = 0.75(1.2 - z) ,
where z denotes the ratio of the pervious areas (e.g., green areas and
undeveloped areas) to the catchment area F.
Studies over 1962-1977 of the relations between rainfall and runoff

for the urban sewer network in the area of Warsaw made it possible to draw
several basic conclusions and to introduce changes in the calculation method:
(5)
1. It was found that the flow rates through the rainwater and general runoff
collectors are considerably lower, often four or more times lower, than would
follow from use of the above-mentioned formulas. This effect was observed
particularly in flat catchments. It is not accompanied by the flooding of
low-lying streets and buildings.
2. The cause of this effect is the very high storage capacity of the rainfall
sewer network in areas of small slope, which is underestimated during
determination of the size of collectors.
3. For sewers and collectors with slopes equal to or less than 0.005 which
drain catchment areas equal to or greater than 50 ha, it was proposed that
a corrected formula be used to determine the concentration time of reliable
rainfall:
t = 1.2 SfPjL + f^ S l i f i . ^ ¥ t f l + 2-^12 min. ,
where 1. is the length of sewer segment in m,

2
f. is the cross-sectional area of sewer in m ,
3 3
V. is the storage volume of the network in m /ha - 20 -7 80 m /ha,
F. is the catchment of the sewer segment,
i 1
3
Q. is the flow rate in cross-section i in m /s, and
a is the storage utilization coefficient, ranging in value from

0.1 to 1, depending upon the catchment inclination.
4. The proposed changes were tested on the basis of research on rainfall and
general runoff sewer network schemes.
112
Method of Simulating Hydrographs of Runoff from Rainwater Sewer Schemes Based
on the "BlacK Box" Model
Studies conducted in the years 1972-76Í1) in catchments of 23 to 280 ha

led to the development of a method for forecasting runoff hydrographs on the basis
of observations of rainfall phenomena^). The relation between rainfall and
runoff was determined by the following runoff parameters:
1. Total runoff coefficient,

F t
r dt dF
/ /
if 3
where r is the runoff intensity in dm /s ha.
q dt dF
2. Maximum runoff coefficient,
. max
f max = ~-—
3. Inertia coefficient,
T
T -
*I
where T is the runoff concentration time and t is the rainfall concentration
time.
4. ff\ , a shift of the beginning of runoff from the beginning of rainfall.
The values of the runoff parameters should be determined on the basis

of a long period of observation spanning several years. In the case of the
catchment studied, they were determined utilizing 121 observations chosen from a
four-year sequence of studies. An example of the application of the above-
mentioned parameters for the forecasting of runoff hydrograms is presented in
Figure 1. The method for forecasting runoff hydrographs has found application
in the design and development of sewer schemes and evaluation of their influence
on receiving waters.
Method of Simulating Pollutographs of Urban Runoff
In view of the planning requirements for development of sewer schemes

in urban areas, it is necessary to know, apart from hydrographs of runoff from
rainwater sewer schemes, the runoff pollutographs at least in terms of basic
pollution indices: suspensions, COD, BOD^, dry residue and chlorides (7).
Comparison of the methods used for this purpose led to the choice of a statistical
method that satisfied the condition of practical applicability in the country
while taking into consideration the essential factors which determine the quality
of rainfall effluents within a limited scope of observation.
113
8-
•i ti
o
/t
ßAJHfML
ruiníoiL T = 65 min
V - 45666 m 1
Jod
Ato
4o
I V
'mx
r5490 m*
Mfl^ho
Ye «0,25
fur« 0.49
4D&J0.
T s \~Jomn
feo
T r»¿» 5o.mín.
o< = 4o míi|
too!û> ¿CO
WMt».
«o
m Ye
fit
-f --2.9
* °.ft>
- O.QO
V
T
- £738 m '
» 468 mí«
XíííL -0,<5
O» * 35 mi«
Ao-otí í etmutoted ru;nOff rujdroqrgpfi

«o raíi-ai:- 1, observed runoff fiudroqiuph
Q»?3 •
.i
^í
-I 1 r
_¿^l 1 ' •—TT"" ' — ' '—i—1—i—'—|*?r^l ' ""
J
46«° 17* Vfl" *• 4?00 *° * ~ h
F I G U R E 1 - HYDROGRAPH SIMULATION B A S E D
ON T H E "BLACK B O X " M 0 D E L ( 6 >
114
Às a result of analysis of available research findings, with
particular consideration given :to the studies conducted in a catchment of
280 ha, it was found that the value of instantaneous pollution load can be
simulated as a function of the flow rate, time from the beginning of runoff
and the breaks between precipitation. This dependence does not apply to thaw
effluents, for which the instantaneous pollution load depends only on the flow
rate.
The general form of the regression function, with which the observation
results can be approximated, can be written as follows: (7)
î = A QJ
where Q and -tí are the flow rate and pollution load in the i-th ten-minute
interval of time in dm^/s and in g/s, respectively. A and B are parameters that
depend on the kind of pollution index involved, with breaks between precipitation
and time of the beginning of runoff. The values of these coefficients determined
on the basis of two years of observation conducted in a housing-industrial
catchment of 280 ha are presented in Table 1(7). An example of a pollutograph
simulation is given in Figure 2.
TABLE 1 - PARAMETERS FOR THE FORMULA X. = A 0? ,
POLLUTION INDICES AND GENERAL SUSPENSIONS
Break Between Precipitation
T á 24 h T ^ 24 h
Parameters
Time from the Beginning of Runoff, T
T^lh l<r<2h T>2h TÉlh l<3r^2h T>2h

Coefficient A 0.0152 0.0142 0.0178 0.0143 0.0180 0.0093
Coefficient B 1,468 1.4034 1.2781 1.5995 1.4751 1.5033
Correlation
Coefficient 0.97 0.96 0.98 0.94 0.96 0.99
Principles of Forecasting Runoff Pollution to Meet Requirements for Planning New

Drainage Schemes
In the formulated "Principles of planning sewer schemes in urban-

industrial agglomerations,"(8) which summed up, among others, the results of
several research works in the field of rainwater runoff from urban areas, it
was recommended that, for the purposes of forecasting hydrographs and
pollutographs of runoff, the results be utilized of studies conducted in chosen
rainwater and general runoff network schemes within these systems. If such
studies are not available, it was recommended that a triangular shape be assigned
to runoff hydrographs and pollutographs and that average values of pollutants
obtained from observations elsewhere be used, e.g. general suspensions with an
115
tbo
Í600
<5-*25fc lüoo
4400
observed {lov hiydroqraph Atoo •o
_0 cow puled pollul oqraph 1ooo

"•aS
observed ponograph 9oo c
breafe between precipitûtion n.
-105 h 800
•Jbo
600
BCD
•400
\2co
Too
too
V
lo Ào 60 so loo <fo <4o <6o /go too'Mo "tune iC*"'"]
FIGURE 2 - POLLUTOGßAPH SIMULATION (7)
116
average of 300-500 and a maximum of 1000-3000 mg/1 and a BOD,, with an average of
15-50 and a maximum of 100-300 mg/1.
Research on the Influence of Urbanization on Changes in Catchment Runoff

Parameters
Observations carried out in the above-mentioned catchments were used

also to evaluate the influence of the urbanization of land on changes in the
parameters of runoff from the catchment. Preliminary analysis of observations
during the period 1972-78 has confirmed the research hypothesis that land
urbanization significantly affected runoff parameters, shortening its time and
increasing its maximum values(9)4 Quantification of this phenomenon for
selected catchments should be expected during the forthcoming period.
Models of Runoff from Urbanized Catchments
Attempts were also made to construct mathematical models of runoff

from urbanized catchmentsC10»11). The principle of construction of the models
is based on the mutual relationship of three submodels: surface runoff, inflow
by drains to the sewers, and sewer flow. While the elaborated models take into
consideration the unsteady flow conditions over the surface to the sewer, they
do not include backwater effects.
Models of Pollutant Propagation in a River
The need to evaluate the influence of surface runoff from urban areas
on the quality of receiving waters has led to the development of mathematical
models of the unsteady processes of pollutant propagation in rivers,
(12-14) l n
the form of one-dimensional and two-dimensional models.
In the scope of one-dimensional models, the models of changes of BOD5
and oxygen concentration in the receiving water were worked out for steady-state
conditions both in the receiving water and in the effluent discharging to the
receiving water and also for steady-state conditions of the receiving waters and
unsteady conditions of effluent discharge. The differential equations worked out
for these schemes describe the changes of concentration of organic pollutants and
oxygen in the river on the basis of the kinetics of the particle processes of
self-purification: decay, re-aeration, sedimentation, adsorption, respiration
and oxygen consumption by bottom sediments, and the processes of longitudinal
dispersion and convection.
The computer realization of algorithms enables evaluation of the

influence of effluent discharges of. steady and unsteady character on the
established states of water quality. In both algorithms for both states the
possibility of simulating anaerobic conditions in the river were taken into
consideration. The worked out models, algorithms and forecasts, taking into
consideration also the dimensionless parameters of the river, enable multi-
variant forecasting of the influence of pollutant discharges with time-variable
concentrations on the quality of the river waters along its full length,
Figures 3 and 4.
Two-dimensional models include the processes of mixing of the effluent

discharge with the receiving waters, for the cases of both steady and unsteady
discharges.
117
topado» ;/dm a
. Xlfloi °8/n9/( i w ^ - - - .
JC-IBiM
x>2fa
»flw (ZfiT* -^
X'2ftM
FIGURE 3 - TEMPORAL AND SPATIAL DISTRIBUTION OF BOD5
"Zo "S« Qî #S fà "#£ <í? i» Ti iü l* ?» Un»
FIGURE 4 - DISSOLVED OXYGEN IN THE RIVER

AS AFFECTED BY STORMWATER
DISCHARGE^ 2 )
118
References
1. Biaszczyk, P., Urban runoff research In Poland, ASCE UWRR Program

Technical Memorandum No. IHP-11, ASCE, New York, N.Y., Feb., 1977.
2. Biaszczyk, P., Urban runoff research in Poland, UNESCO, Research on Urban

Hydrology, Vol. 2, Technical Papers in Hydrology 16, Paris, 1978.
3. Biaszczyk, P., Determination of rainwater intensities to calculate drainage

networks on the basis of observations of the mean rainwater intensities
in Warsaw, Scientific Publications, Warsaw, 1967, Building Series No. 38.
4. Nowakowská-Biaszczyk, A., Blraszczyk, P., Waterworks and sewers in physical

planning, Warsaw, 1974, Arkady.
5. Blraszczyk, P., Improved methods of dimensioning rainwater and general

runoff sewers. Materials for PZiTS Conference, Wrociaw, Oct., 1977.
6. Fidala-Szope, N., Quantitative characteristic of rainfall effluent runoff

in an housing-industrial catchment on the basis of studies conducted in
Warsaw. Materials for PZiTS Conference, *<Sd¿, Nov., 1977.
7. Kalinowski, M., Research results and simulations of pollutographs of urban

rainfall effluents and their consideration in drainage calculations of
storage reservoirs. Report to the Conference of the Civil Engineers Section
of the Polish Academy of Sciences, May, 1979.
8. Biaszczyk, P., Principles of planning drainage systems in urban-industrial

agglomerations. Report to the Conference of the Civil Engineers Section
of the Polish Academy of Sciences, May, 1979.
9. Bielaitski, Zb., Influence of urbanization on rainwater effluent runoff.

Materials of the Water IV Seminar, Institute for Environmental Development,
Wrociaw, April, 1979.
10. WoioBzyn, E., Model of drainage network flow. Dissertation, Institute of

Hydrotechnics, Gdansk Technical University, 1977.
11. Blechert, B., Janikowski, R., Kreduszyñski, B., Mill, W., Mathematical
model of rainfall-runoff phenomenon from an urbanized catchment. Materials
for Water IV Seminar, Institute for Environmental Development, Wroclaw,
April, 1979:
12. Osmulska-Mróz, B., Mathematical model of aerobic river conditions as an

instrument for studying the effects of rainfall effluent discharge. Man
and Environment, No.. 2, 1977, Institute for Environmental Development.
13. Osmulska-MrcÍz, B., Analysis of the possibility of removing rainfall

effluents from urban areas to receivers. Dissertation, Warsaw Technical
University, 1978.
14. Osmulska-Mroz, B., Mathematical model of unsteady processes of pollutant

propagation in a river. Matherials for Water IV Seminar, Institute for
Environmental Development, Wroclaw, April, 1979.
119
Section 11 Progress since 1977 in India
by S. Ramaseshan*
Editor's Note
The national report for India was prepared in May, 1977, by S.

Ramaseshan and P. B. S. Sarma,^) and was the last issued of twelve such
national reports. While the senior author had fully expected to present a
paper on progress since then in India, he had to decline at the last minute
mainly because he had only recently been appointed to his new position and
because he had not found much new work to report on the topic. What follows
was presented by the editor of this report at the International Symposium.
The next subsection is from a letter to the editor sent by Dr. Ramaseshan on
April 28, 1979. This is followed by a recapitulation of a few of the points
made in the 1977 reportC1) and a description of India's new Institute of
Hydrology.
Some New Developments
The India Meteorological Department has filled a double exponential

distribution to the partial duration frequency of rainfall above a threshold
value. Also, the Department has derived point to areal rainfall values for
durations of 15 minutes to one hour from data for six rainstorms.
P. V. Rao of Punjab Engineering College has engaged in a laboratory

simulation of watersheds. He has studied the effect of size and shape of
watersheds on hydrograph peaks, and the shapes and peaks of hydrographe with
scale models of parts of Chandigarh.
A Master of Technology thesis at Indian Institute of Technology,

Kanpur, has indicated the importance of correct identification of impervious
and pervious areas as well as soil types in the hydrologie simulation of urban
watersheds using ILLUDAS.
While more field data have been collected since the 1977 national
report, no worthwhile results have since been derived for urban watersheds.
There have been difficulties in fitting observed data to model results,
indicating possible data and other errors.
The Central Board for Prevention and Control of Water Pollution has
started collecting water quality, data around Delhi. Some preliminary results
have been published, but more for the purpose of identifying problems than for
their solution.
A Developing Nation Perspective
This subsection was adapted from the 1977 report(^) in a summary of

the twelve 1975-1977 national reports.(2)
*: Director, National Institute of Hydrology, Roorkee-247672, U.P. India.
121
According to the 1971 census, 13% of the total national population of
547 million people in India live in urban areas. Of the urban population of
about 110 million, thrae-fourths are served by water supply facilities and two-
fifths are served by wastewater sewerage. Urban areas in India "have a small
proportion of built-up area to total area and hence urban drainage problems in
India as in most other parts of the developing world are quite different from
the 'concrete jungles' of the developed countries."
"In a developing country, the priorities for economic development and

investment are for food, shelter, clothes, health and education. Urban drainage
is generally not taken into consideration except when it affects significantly
any of the above factors, particularly as a part of the more general problem of
flooding of urban areas. As almost all important cities of India are on the
banks of rivers and are subject to flooding, drainage of urban areas and riverine
flood control are generally interlinked. Because of financial limitations
and because urban drainage problems constitute 'negative goods', very little
attention has been paid in India to urban drainage "
Low-lying areas prone to frequent flooding are "often encroached upon

by the poorest section of the population, and are covered with sprawling slum
areas with a high density of population and meager civil amenities. Failure to
provide an adequate urban drainage system seriously affects the life of these
people and exposes them to potential health hazards. Thus, urban drainage systems
are also linked with the overall problem of slum abolition, resettlement and urban
redevelopment."
"Urban hydrologie problems of India, as in other developing countries,

differ from those of developed nations in several important respects. They
include:
lateral rather than vertical development;

limited amounts of paved areas;
intimate interaction between urban drainage and flood control;
preference for open drains over closed ones;
limited availability of continuous records of precipitation, streamflow and
water quality;
low fiscal priority for drainage investment;
limited numbers of sewer connections and hence silting of combined sewers;
high cost for construction and modification of combined sewer systems; and
limited capacity for financial investment."
As in most countries, all three levels of government are parties to
urban drainage research: national,.territorial (provincial) and local. However,
involvement by universities and consultants appears to be smaller.
The Institute of Hydrology
The National Institute of Hydrology is an autonomous Society under

the Department of Irrigation, Government of India, located at Roorkee, U.P.,
India. It will conduct systematic and scientific studies in theoretical and
applied hydrology and water resources towards optimal integrated utilization
of surface and groundwater resources of India through development and/or
validation of methodologies, procedures, tools, etc., in hydrologie analysis
and design.
122
Activities of the Institute will include:
i) development of instrumentation, coordination and design of data collection

systems and networks, data compilation and preliminary analysis including
data storage and retrieval systems, remote sensing and telemetry;
ii) hydrologie analysis including precipitation, évapotranspiration,

consumptive use, field efficiencies, soil moisture, groundwater,
streamflow, rainfall-runoff relationships, flood routing, soil erosion,
sedimentation and water quality;
ii'i) hydrologie design including water budgeting, watershed simulation,

conservation and storage, groundwater simulation, flood forecasting,
estimation and control, drought estimation, agricultural planning models,
water quality models and urban systems;
iv) water resources planning including economic analysis and optimal design,
development and management of integrated surface and/or groundwater systems
using programming and simulation models;
v) data library, program library, and technical library; and
vi) publication, training and laboratory facilities.
The UNDP is assisting in the setting up of the National Institute of

Hydrology through a 5-year project.
References
1. Ramaseshan, S., and P. B. S. Sarma, "Urban Hydrological Modeling and Catchment

Research in India," ASCE UWRR Program Technical Memorandum No. IHP-12, ASCE,
New York, N.Y., 21 pp., May, 1977. (NTIS No. PB 271 300).

catchment research: International summary, Technical Papers in Hydrology 18,
Unesco, 48 pp., 1978.
123
Section 12 Progress since 1975 in the
United States of America
by M. B. McPherson*
Introduction
This presentation is an updating of a November, 1975, national state-

of-the-art report on urban hydrological modeling and catchment research.(1) As
in that report, sewered catchments will be featured in discussing;new developments
in quantity and quality aspects of urban runoff. However, one of the most
important new trends over the past few years has been a growing acceptance of a
perception of water in urban areas as a totality, as a resource.(2,3) This
integrated view is exemplified in a prognostication of needs for urban water
models in a 1976 report.
(4)
Areawide Planning
Under Section 208 of the Federal Water Pollution Control Act of 1972
(Public Law 92-500), areawide planning for water pollution abatement management
has been undertaken in nearly all metropolitan areas. Urban runoff considerations
are an issue in most of these areas. The initial thrust of PL 92-500 is universal
attainment of secondary treatment of municipal wastewaters and the equivalent of
secondary treatment of industrial wastewaters, collectively termed "point sources".
The next thrust is stated to be still further reduction in point-source pollution
and abatement of pollution from "nonpoint sources". Urban nonpoint sources are
the runoff discharged as combined sewer overflows and urban runoff from other
than combined sewers. While the runoff pollution of a metropolis is commonly
greater than its treatment plants' effluent pollution during storms under
current conditions, as the U.S.A. moves towards moré stringent point source
control the impact of nonpoint sources will increasingly predominate.
Some rather sweeping -amendments to PL 92-500 were made in the Clean

Water Act of 1977 (PL 95-217). Recognized was that the value of Section 208
studies was limited by a general inability to quantify the pollution from nonpoint
sources. Further, these studies imply a collective need for a very high level of
federal funding for construction. Funding of construction through grants for
stormwater projects was suspended until the nature of nonpoint pollution can be
better understood and interpreted in terms of appropriate federal funding. The
U.S. Environmental Protection Agency (EPA) has developed a Nationwide Urban
Runoff Program for 1978-1983 which will assess the nature, causes and severity
of urban runoff problems and means for controlling them.(5) Findings will be
reported to Congress in 1983. The strategy is to concentrate on about thirty
intensive projects nationwide, with the objective of demonstrating transferability
of the results.
EPA issued an areawide assessment procedures manual to assist Section

208 planning agencies.(6) Technical planning issues were widely discussed.('"")
EPA specialists annually review about 150 references on urban runoff and combined
sewer overflow. (13-15) '
Urban stormwater management and technology capabilities
have been summarized.
(16)
A nationwide assessment has been made of the extent
and costs for abatement of combined sewer overflows and urban stormwater
discharges.
(17)
*:
abatement
Director,
haveASCE
been
Still
Urban
implied.(1°)
higher
Watercosts
Resources
The
for
methodology
combined
Researchsewer
Program,
used overflow
to estimate
Marblehead,
pollution
costs
Mass.
in the
,125
U.S.A.
nationwide assessment have been outlined.(*'). Over 300 selected references on
urban runoff control planning have been cited.(20)
Catchment Research
In the U.S. report on urban hydrological modeling and catchment

research, (D 64 urban catchments were identified from which data had been used
to test some 16 urban runoff models. Citations were restricted to cases that
were publicly documented. Since 1975, the number of such catchments may well
have doubled, mostly as a result of Section 208 planning and jurisdictional
master planning activities, but it will be some time before the new ones can
be collectively documented because a number are components of on-going planning.
In a continuing project for the EPA,(21) urban catchment rainfall, runoff and
quality data are being placed on magnetic tapes in a common format and are
being entered in the EPA STÖRET data retrieval system.
Noted in the 1975 reference report(*^ was that a means for measuring
flow in circular combined or storm sewers had been developed by the U.S.
Geological Survey (USGS), as part of an integrated catchment data collection
system, and that the first installation of this system was in two adjoining
subcatchments in Philadelphia, Pennsylvania. Similar installations have been
made in. Porapano, Florida, and the data summaries for one of the four catchments
have been released.(22,23) Reports have been issued on data from catchments in
Denver, Colorado,(24,25) an(j Portland, Oregon.(26-28) A national guide for the
collection, analysis and use of urban stormwater data is available.(29)
Under a cooperative effort with the EPA, the USGS in 1978 embarked on
an Urban Studies Program. The initial formulation of the program(30) encompasses
data collection, its interaction with model refinement and calibration needed to
enhance understanding of processes, and the testing of alternative management
techniques, under the framework outlined in Figure 1.(30) A typical installation
of the monitoring system contemplated is shown schematically in Figure 2.
(30)
Catchment Modeling
Most of the urban runoff models now commonly used were developed in/
the 1967-1975 period. As noted in the subsection above, 16 models were
identified in 1975 which had been tested against actual field data and for which
the tests were publicly documented. Use of such models has increased
substantially since, particularly in conjunction with Section 208 studies,
only two of which are cited as examples.(31,32)
A survey of model usage in urban water management for problems in

hydrology, hydraulics, water quality and economics, substantiates this change,
not only for urban runoff modeling but for almost all other urban water manage-
ment aspects.
(33)
However, most frequent usage reported was for water quality
studies.
Numerous comparisons continue to be made of the features or
performance of urban runoff models.(34-37) The most widely used model for
metropolitan planning studies (Storage, Treatment, Overflow, Runoff Model)
is continually upgraded.(37,38) Similarly, the most widely used system analysis
model (Storm Water Management Model) is continually upgraded and users of this
model meet twice per year to review experiences with all types of urban runoff
models.(39,40) Refinements continue to be made to the best known system design
126
URBAN i> MODEL
REFINEMENT O BETTER
STORM-WATER UNDERSTANDING
AND
DATA CALIBRATION OF
COLLECTION
O C PROCESSES
A
V
TEST
ALTERNATIVE
MANAGEMENT
TECHNIQUES
Ä
iz ASSESS
IMPACT ON
RECEIVING
WATERS
(30)
FIGURE 1-FRAMEWORK FOR DATA COLLECTION.
0 *
ATMOSPHERIC SAMPLING
RAIN GAGES
DIGITAL
RECORDER
1
2 SYSTEM
3 CONTROL
4 UNIT
5
SAMPLE
6 COOLER
7
8
OPTIONAL SENSORS:
RAINFALL, STAGE,
TEMPERATURE,
CONDUCTIVITY,
pH, ETC.
FIGURE 2-TYPICAL INSTALLATION OF THE URBAN HYDROLOGY MONITORING

SYSTEM. (30)
128
model (Illinois Hrban Drainage Area Simulator) by the State agency that
developed it.
(41) Various receiving water models have been reviewed in terms
of planning applications.(42) Certain urban runoff model features have been
elaborated upon.(43)
An annual forum where modeling and related experiences are exchanged
is conducted by the University of Kentucky, and a total of 177 papers have been
presented in the five conferences held so far.(44-48) A. number of status
reports on urban runoff, wastewater system and water supply modeling were
presented at a national conference on environmental modeling and simulation.
(49)
Abatement of pollution from combined sewer system overflows has a
higher national priority than urban runoff pollution from separate storm sewer
systems and areas not served by underground stormwater conveyance systems.
Research and development has therefore been greater for combined systems. In
this connection, advances continue to be made on the development of capabilities
for automatic control of total jurisdiction systems.(50,51) Complete control
will require a capability for tracking storms as they approach and pass over a
metropolitan area.
(52)
Synthetic storms have been used for planning and design for decades.
In a challenge to this tradition, runoff simulation results have been compared
for actual and synthetic rainfall events.
(20,53)
Urbanization Effects
Few generalizations on the hydrological effects of urbanization that

were based on adequate documentation could be made in 1975. Recognized were that
urbanization increases the volumes and peak rates and decreases the lag times of
runoff from urban lands. There is some evidence that the relative amplification
of peak rates may diminish for rarer events.
(54)
Resolution of behavioral generalizations is handicapped by the limited
field data base alluded to earlier. The only interpretive tool for runoff that
incorporates regionally specific parameters on a national scale has been
developed by the Soil Conservation Service.(55) Synthetic unit hydrograph
parameters derived from a number of field measurements in several States
(20,56)
offer a means for extrapolating findings from local field observations to local
ungaged catchments in a given metropolitan area. These tools deal only with
runoff quantity, and it will probably be several years before results from the
field investigations noted earlier supplant these very rudimentary indicators
and add comparable capabilities for characterizing runoff quality in regional
terms on a national scale.
Unit hydrographs continue to be used in interpretive studies.
(57) A
methodology for analyzing urbanization effects has been advanced.(58) Effects
on runoff have received considerable attention.(59-64) Water quality aspects
have been emphasized in another study. (65)
An attempt has been made to classify urban catchments in terms of
stream ordering. (6è) Urbanization effects on channel stability have been
studied(67) and an overview on erosion effects has been provided.
(68)
Pollution control advances have been summarized.
(69)
Relationships
between atmospheric pollution and urban runoff quality have been investigated.
(70)
129
Special problems associated with urban lakes have been described.(71)
Effects of urbanization on such lakes have been investigated.(72-74)
Inadvertent weather modification of urban microclimates has received

new attention. Field research progress has been impressive. (75-77)
In an effort to minimize future adverse effects, preferred practices

for land development have been promulgated with regard to residential stormwater
management(78) ancj erosion and sediment control. (75)
Conclusions
While the term urban hydrology gained currency less than two decades
ago, the peak period for research attention was from about 1967 until 1975, when
numerous tools of analysis were developed and tested.(80,81) The period since
has been characterized by an explosion in the application of these tools, but
the level of supporting phenomenological research did not keep pace and has
declined steadily since about 1974. Much of this recent activity was in
conjunction with or as a result of areawide planning for water pollution
abatement management in most metropolitan areas.(82) With the first phase of
areawide planning having been completed in 1978, there will be a transition
period of perhaps two years for revaluation of the effectiveness of these
initial efforts and the initiation of the next phase.
Because the EPA's Nationwide Urban Runoff Program is geared to

culminate in a report to Congress in 1983, considerable advances in knowledge
should be made in the interim. Recognized everywhere is an acute need for
parameter characterization, at least in terms of major climatic regions, for
a variety of mathematical models. The gap between need and capability is much
greater for water quality parameters than for flow parameters. It is sincerely
hoped that the EPA Program coupled with the USGS Urban Studies Program will lead
to a much more reliable understanding and mastery of urban runoff processes.
It is anticipated that 1979-1982 will be a period of reappraisal,

refinement and significant extension of what is presently known. While past
research has been loosely coordinated, fractionalized, and with uneven
emphasis, impressive advances have nevertheless taken place since 1975.
If the period 1967-1974 can be called the period of tool making, then
the period since can truly be described as the period of tool wielding. As in
every scientific and technical field, the concomitant tremendous advances in
electronic computation capabilities were undoubtedly a factor.
Perhaps the most important development in a long-term sense was a-

much greater appreciation of the complex interconnections, interactions and
interrelations among the numerous facets of water in the urban environment.
It might be fairly stated that the concept of the hydrologie cycle on a
metropolitan scale has been rediscovered by some and is gaining acceptance by
many.
References
1. McPherson, M. B., Urban hydrological modeling and catchment research in

the U.S.A., ASCE UWRR Program Tech. Memo. No. IHP-1, Amer. Soc. Civil
Engrs., 49 pp., NTIS : PB 260 685, November, 1975.
130
2. Kohlhaas, C. A., Focus —Comprehensive water resources management, Water
& Sewage Works, 122(6), pp. 70-75, 1975.
3. Dendrou, S. A., J. W. Delleur and J. J. Talavage, Urban growth in water

resources planning: multilevel approach to urban water resources systems
analysis —Application to medium size communities, Purdue Univ. Wat. Res.
Center Tech. Report 100. 116 pp., NTIS: PB 282 388, 1978.
4. Sonnen, M. B., L. A. Roesner and R. P. Shubinski, Future direction of urban

water models, U.S. Envir. Prot. Agency BPA-600/2-76-058. 86 pp., February,
1976.
5. Nonpoint Sources Branch, Summary Document, Work Plan, Urban Runoff Program.
1978-1983, U.S. Envir. Prot. Agency, 22 pp., September, 1978.
6. Office of Research and Development, Areawide assessment procedures manual,

U.S. Envir. Prot. Agency EPA-600/9-76-014, in three looseleaf volumes,
July, 1976, et seq.
7. Pisano, M. A., Nonpoint sources of pollution: a federal perspective,

J. Envir. Engrg. Div., Amer. Soc. Civil Engrs., 102(EE3), pp. 555-565, 1976.
8. Pisano, M. A., 208: A process for water quality management, Civil

Engineering, 46(11), pp. 55-57, 1976.
9. Lienesch, W. C , and G. A. Emison, Planning for implementation under

Section 208, J. Wat. Res. Plan. & Mgt. Div.. Amer. Soc. Civil Engrs.,
102(WR2), pp. 283-295, 1976.
10. Billings, L. G., The evolution of 208 water-quality planning, Civil

Engineering, 46(11), pp. 54-55, 1976.
11. Shubinski, R. P., and W. N. Fitch, Appraisal of areawide wastewater planning,

J. Wat. Res. Plan. & Mgt. Div., Amer. Soc. Civil Engrs., 103(WR1), pp. 63-
72, 1977.
12. Westman, W. E., Problems in implementing U.S. water quality goals, American
Scientist, 65(2), pp. 197-203, 1977.
13. Field, R., J. Curtis and R. Bowden, Literature review — urban runoff and
combined sewer overflow, J. Wat. Poll. Control Fed., 48(6), pp. 1191-1206,
1976.
14. Field, R., R. Bowden and K. Rozgonyi, Literature review — urban runoff and
combined sewer overflow, J. Wat. Poll. Control Fed., 49(6), pp. 1095-1104,
1977.
15. Field, R., and B. B. Gardner, Literature review — urban runoff and combined
sewer overflow, J. Wat. Poll. Control Fed., 50(6), pp. 1170-1185, 1978.
16. Lager, J. A., et al., Urban stormwater management and technology: update
and users' guide, U.S. Envir. Prot. Agency Report EPA-600/8-77-014, 313 pp.,
September, 1977.
131
17. Sullivan, R. H., et al., Nationwide evaluation of combined sewer overflows
and urban stormwater discharges,- Vol. I: Executive summary, U.S. Envir.
Prot. Agency EPA-600/2-77-064a, 95 pp., September, 1977.
18. Giggey, M. D., and W. G. Smith, National needs for combined sewer overflow
control, J. Envir. Engrg. Div., Amer. Soc. Civil Engrs., 104(EE2), pp.
351-366, 1978.
19. Heaney, J. P., S. J. Nix and M. P. Murphy, Storage-treatment mixes for

stormwater control, J. Envir. Engrg. Div., Amer. Soc. Civil Engrs.,
104(EE4), pp. 581-592, 1978.
20. McPherson, M. B., Urban runoff control planning, U.S. Envir. Prot. Agency
EPA-600/9-78-035. 187 pp., October, 1978.
21. Huber, W. C , and J. P. Heaney, Urban rainfall-runoff-quality data base,

U.S. Envir. Prot. Agency EPA-600/8-77-009, 167 pp., July, 1977, et seq.
22. Mattraw, H. C. Jr., and C. B. Sherwood, Quality of stormwater runoff from

a residential area, Broward County, Florida, J. Res. U.S. Geol. Survey,
5(6), pp. 823-834, 1977.
23. Mattraw, H. C. Jr., J. Hardee and R. A. Miller, Urban stormwater runoff

data for a residential area, Porapano Beach, Florida, U.S. Geol. Survey
Open-File Report 78-324, 108 pp., 1978.
24. Ellis, S. R., Hydrologie data for urban storm runoff from three localities
in the Denver metropolitan area,. Colorado, U.S. Geol. Survey Open-File
Report 78-410, 135 pp., May, 1978.
25. Dawdy, D. R., J. C. Schaake, Jr., and W. M. Alley, Users guide for
distributed routing rainfall-runoff model, U.S. Geol. Survey Water Resources
Investigations 75-90, 146 pp., September, 1978.
26. McKenzie, S. W., and T. L. Miller, Basic data on urban storm-water-quality,

Portland, Oregon, U.S. Geol. Survey Open-File Report 76-594, 71 pp., 1976.
27. Miller, T. L., Urban storm-water-quality data, Portland, Oregon, and

vicinity, U.S. Geol. Survey Open-File Report 78-851, 23 pp., 1978.
28. Miller, T. L., and S. W. McKenzie, Analysis of urban storm-water-quality

from seven basins near Portland, Oregon, U.S. Geol. Survey Open-File
Report 78-662, 47 pp., 1978.
29. Alley, W. M., Guide for Collection, Analysis, and Use of Urban Stormwater
Data, Amer. Soc. Civil Engrs., 115 pp., 1976.
30. Lystrom, D. J., Data gathering for stormwater runoff and surveillance, in
Water Problems of Urbanizing Areas, edited by W. Whipple, Amer. Soc.
Civil Engrs., pp. 90-97, 1978.
31. Rimer, A. E., J. A. Nissen and D. E. Reynolds, Characterization and

impact of stormwater runoff from various land cover types, J. Wat. Poll.
Control Fed., 50(2), pp. 252-264, 1978.
132
32. Schultz, N. U., and A. Wilmarth, Water quality simulation and Public
Law 92-500, Water Resources Bull., 14(2), pp. 275-287, 1978.
33. Donigian, A. S., and R. K. Linsley, Planning and Modeling in Urban Water
Management, Hydrocomp, Inc., Palo Alto, Cal., 158 pp., October, 1978.
34. Huber, W. C , Modeling for storm water strategies, APWA Reporter. 42(5),
pp. 10-14, 1975.
35. Marsalek, J., et al., Comparative evaluation of three urban runoff models,
Water Resources Bull., 11(2). PP. 306-328, 1975.
36. Brandstetter, A., Assessment of mathematical models for storm and combined
sewer management, U.S. Envir. Prot. Agency EPA-600/2-76-175a, 510 pp.,
August, 1976,
37. Abbott, J., Testing of several runoff models on an urban watershed, Corps
of Engineers, The Hydrologie Engrg. Center Tech. Paper No. 59, 43 pp.,
October, 1978.
38. Hydrologie Engineering Center, Corps of Engineers, Guidelines for

calibration and application of STORM, Training Document No. 8, 48 pp.,
December, 1977.
39. Torno, H. C , editor, Proceedings Stormwater Management Model (SWMM) Users

Group Meeting May 4-5, 1978, U.S. Envir. Prot. Agency EPA-600/9-78-019,
244 pp., July, 1978-.
40. Torno, H. C , editor, Proceedings Stormwater Management Model (SWMM) Users

Group Meeting November 13-14, 1978, U.S. Envir. Prot. Agency EPA-600/9-79-003,
41. Terstriep, M. L., G. M. Bender and D. J. Benoit, Buildup, strength, and

washoff of urban pollutants, Preprint 3439, 29 pp., a paper presented at the
Amer. Soc. Civil Engineers convention, Chicago, October 16-20, 1978.
42. Grirasrud, G. P., E. J. Finnemore and H. J. Owen, Evaluation of water quality

models: A management guide for planners, U.S. Envir. Prot. Agency
EPA-600/5r76-004, 176 pp., July, 1976.
43. Overton, D. E., and M. E. Meadows, Stormwater Modeling, Academic Press,

358 pp., 1976.
44. Kao, D. T., editor, Proceedings National Symposium on Urban Rainfall and
Runoff and Sediment Control, Univ. of Kentucky Report UKY BU106, 246 pp.,
October, 1974.
45. Haan, C. T., editor, Proceedings National Symposium on Urban Hydrology and
Sediment Control, Univ. of Kentucky Report UKY BU109, 314 pp., November,
1975.
46. Barfield, B. J., editor, Proceedings National Symposium on Urban Hydrology,

Hydraulics and Sediment Control, Univ. of Kentucky Report UKY BU111, 386
pp., December, 1976.
133
47. Kao, D. T., editor, Proceedings International Symposium on Urban
Hydrology, Hydraulics and Sediment Control, Univ. of Kentucky Report
UKY BUI14, 382 pp., December, 1977.
48. Haan, C. T., editor, Proceedings International Symposium on Urban Storm

Water Management, Univ. of Kentucky Report UKY BU116, 348 pp., July, 1978.
49. Otti, W. R., editor, Proceedings of the EPA Conference on Environmental

Modeling and Simulation, U.S. Ênvir. Prot. Agency EPA-600/9-76-016, 847 pp.,
July, 1976.
50. Wenzel, H. G. Jr., J. W. Labadie and N. S. Grigg, Detention storage control

strategy development, J. Wat. Resour. Plan. & Mgt. Div., Amer. Soc. Civil
Engrs., 102(WR1), pp. 117-135, 1976.
51. Trotta, P. D., J. W. Labadie and N. S. Grigg, Automatic control strategies

for urban stormwater, J. Hydraul. Div., Amer. Soc. Civil Engrs., 103(HY12),
pp. 1443-1459, 1977.
52. Nguyen, V.-T.-V., M. B. McPherson and J. Rousselle, Feasibility of storm

tracking for automatic control of combined sewer systems, ASCE UWRR
Program Tech. Memo. No. 35, Amer. Soc. Civil Engrs., 29 pp., November, 1978.
53. Marsalek, J., Research on the design storm concept, ASCE UWRR Program Tech.
Memo. No. 33, Amer. Soc. Civil Engrs., 28 pp., September, 1978. (Included in 20.)
54. Hollis, G. E., The effect of urbanization on floods of different recurrence

interval, Water Resources Res., 11(3), pp. 431-435, 1975.
55. Soil Conservation Service, Urban Hydrology for Small Watersheds, U.S. Dept.
of Agriculture Tech. Release No. 55, 91 pp., January, 1975.
56. Espey, W. H. Jr., D. G. Altman and C. B. Graves, Jr., Nomographs for ten-
minute unit hydrographs for small urban watersheds, ASCE UWRR Program
Tech. Memo. No. 32, Amer. Soc. Civil Engrs., 22 pp., NTIS: PB 282 158,
December, 1977. (Included in 20.)
57. Gundlach, D. L., Unit hydrograph parameters versus urbanization, J. Irrig. &
Drain. Div., Amer. Soc. Civil Engrs., 102(IR3), pp. 388-394, 1976.
58. Mays, L. W., and Y.-K. Tung, Methodology for analyzing effects of
urbanization on water resource systems, The Univ. of Texas at Austin
CRWR 156, 158 pp., NTIS: PB 279 663, March, 1978.
59. Cech, I., and K. Assaf, Quantitative assessment of changes in urban runoff,
J. Irrig. & Drain. Div.. Amer. Soc. Civil Engrs., 102(IR1), pp. 119-125,
1976.
60. McCuen, R. H., Flood runoff from urban areas, Univ. of Maryland Wat. Resour.
Res. Center Tech. Report 33, 70 pp., NTIS: PB 244 504, June, 1975.
61. McCuen, R. H., and H. W. Piper, Hydrologie impact of planned unit

developments, J. Urban Plan. & Devel. Div., Amer. Soc. Civil Engrs.,
lOl(UPl), pp. 93-102, 1975.
134
62. Riley, J. P., E. K. Israelsen and 6. B. Shih, Regional analyses of
runoff characteristics for small urban watersheds, Utah Water Res. Lab.
PRWG87-1. 69 pp., NTIS: PB 243 517, April, 1975.
63. Shanks, R. W., and A. R. Rao, The efforts of urbanization on low flows
and total runoff, Purdue Univ. Wat. Resour. Res. Center Tech. Report 94,
101 pp., NTIS: PB 268 046, May, 1977.
64. Walesh, S. G., and R. M. Videkovich, Urbanization: hydrologic-hydraulic-

damage effects, J. Hydraul. Div., Amer. Soc. Civil Engrs., 104(HY2),
pp. 141-155, 1978.
65. Hydrocomp, Inc., Evaluation of the effects of urbanization on aquatic

ecology and hydrological regimes, Palo Alto, Cal., 159 pp., NTIS:
PB 247 095, July, 1975.
66. Grigg, N. S., and T. Sriburi, Order classification of urban catchments,

Water Resources Bull.. 14(1), pp. 63-71, 1978.
67. Graf, W. L., Impact of suburbanization on fluvial geomorphology, Water

Resources Res.. 11(5), pp. 690-692, 1975.
68. Wolman, M. G., Erosion in the urban environment, Hydrologie Sciences Bull..,
20(1), pp. 117-125, 1975.
69. Field, R., A. N. Tafuri and H. E. Masters, Urban runoff pollution control
technology overview, U.S. Envir. Prot. Agency EPA-600/2-77-047, 91 pp.,
March, 1977.
70. Betson, R. P., Bulk precipitation and streamflow quality relationships in

an urban area, Water Resources Res., 14(6), pp. 1165-1169, 1978.
71. Britton, L. J., R. C. Averett and R. F. Ferreira, An introduction to the

processes, problems and management of urban lakes, U.S. Geol. Survey
Circular 601-K. 22 pp., 1975.
72. Cherkauer, D. S., Effects of urban lakes on surface runoff and water
quality, Water Resources Bull., 13(5), pp. 1057-1067, 1977.
73. Cherkauer, D. S., Effects of urban lakes on quantity and quality of

baseflow, Water Resources Bull., 13(6), pp. 1119-1130, 1977.
74. Gunner, H. B., and J. Rho, The relationship of lake quality to specific
urbanization stresses, Univ. Massachusetts Wat. Res. Institute Pub. 88,
44 pp., NTIS: PB 281 579, June, 1977.
75. Huff, F. A., Urban effects on the distribution of heavy convective rainfall,
Water Resources Res., 11(6), pp. 889-896, 1975.
76. Huff, F. A., Effects of the urban environment on heavy rainfall distribution,
Water Resources Bull.. 13(4), pp. 807-816, 1977.
77. Huff, F. A., and S. A. Changnon, A hydrometeorlogical research program,

Water Resources Bull., 13(3), pp. 573-581, 1977.
135
78. ULI, ASCE and NAHB, Residential Storm Water Management, Amer. Soc. Civil
Engrs., 64 pp., 1975.
79. ULI, ASCE and NAHB, Residential Erosion and Sediment Control, Amer. Soc.
Civil Engrs., 63 pp., 1978.
80. McPherson, M. B., Special characteristics of urban hydrology, in

Prediction in Catchment Hydrology, Australian Academy of Science,
Canberra, ACT, pp. 239-255, 1975.
81. McPherson, M. B., editor, Utility of urban runoff modeling, ASCE UWRR
Program Tech. Memo. No. 31, Amer. Soc. Civil Engrs;, 126 pp., NTIS:
PB 261 460, July, 1976.
82. Torno, Harry C., "Model Application in EPA Planning Programs," paper
presented at ASCE Convention and Exposition, Boston, Mass., April 6, 1979,
ASCE Preprint 3526, 12 pp.
136
Section 13 Summary of symposium discussion
Some parenthetical remarks made during the floor discussion were too
faint to be recorded. The editor has paraphrased, condensed and moderately
rearranged some discussion comments to preserve clarity, continuity or content.
Discussants are identified by last name and country or organization, only.
Full names and affiliations are included in other sections of this report,
except for the following: Professor Miguel Medina, Duke University, Durham,
N.C.; Mr. Michael L. Terstriep and Dr. Krishan P. Singh of the Illinois State
Water Survey, Urbana, Illinois; Mr. Marshall Jennings, USGS, Bay St. Louis,
Mississippi; Mr. William H. Sammons, Soil Conservation Service, Lanham,
Maryland; Mr. Harry C. Torno, U.S. EPA, Washington, D.C.; and Professor
Jacques W. Delleur, Purdue University, West Lafayette, Indiana.
Laurenson (Australia). Two of the speakers referred to the very

small areas (subareas) that they used in their models, one mentioned inlet
areas as small as 100 m 2 and both referred to the small size of areas that
drain to any given gully pit. Some urban catchments are quite large, and I
wonder whether the amount of detail in defining the characteristics of such
minute subareas, starting from 100 m 2 , does not become a bit burdensome when you
are trying, say, to estimate the flood from an area of several square kilometers.
Does that same amount of detail have to be included all through larger catchments?
Lowing (U.K.). The smallest of the areas that we were gaging was about
100 m2: in size, but this is still bigger than some of the areas that we used for
calibrating the relationship between the model parameters and catchment
characteristics because we used laboratory catchment data also to pin down the
small end of the range. We were interested to explain area-dependent variation
in the values of parameters in a model of runoff from very small inlet areas.
I believe that the largest size we would have draining to a gully would be
around 500 m 2 or 600 ra2. In ray presentation I showed a set of nine standard
hydrographs, and in design use you would just specify each subárea as being one
of these several appropriate drainage densities. A certain group of gullies
draining to a manhole would have an average density of perhaps 100 m 2 per gully
or 200 m 2 or 400 m 2 , which, I believe, were the three size ranges that we used
(see Figure 1, page 4-10). So you would perhaps describe a whole area as having
this approximate drainage density. It is not necessary in design application of
the model to measure individual areas drained to a gully.
Falk (Sweden). Only some ten days ago we had a seminar at Lund, Sweden,
on runoff from urban surfaces. At the seminar the question was raised about the
use of knowing the runoff from such a small area. There must be an upper limit
in catchment size when attenuation of the hydrograph is not very important any
more. I think that for catchments at least up to a size of something like 2 km 2 ,
it may be of some importance to know what is happening in detail. We have found
in a residential area in Lund built around 1963, that the catchments draining
each gully range in size between, say, 80 m 2 and 700 m , and that must be
considerably smaller than in North America. For example, last week I was watching
a very heavy rainfall in Montreal and there were reaches of several hundred
meters before the water entered a gully.
137
Marsalek (Canada). I would like to comment on the measuring of
inflows into sewer inlets. I would be quite concerned about doing so in North
America because many inlets here are designed to intercept only part of their
tributary inflow. Runoff entering the inlet does not necessarily equal the
contribution of the drainage area. This is done quite intentionally to
exploit the maximum diversion capacity of inlets. We recently concluded a
study for the Ministry of Transportation and Communications dealing with highway
drainage. In some cases, the flow by-passing an inlet could be as much as 90%
of the gutter flow. Obviously, the European design of inlets differs.
Lowing (U.K.). I have been asked if we have had any problems with
the measurement of flows into gullies. We had to choose these installations
with care. In some cases at Southampton, we had to make miniature berms on
the road with tarmac in order to direct by-passing water into the gullies, but
we also tended to monitor them in nested groups so that we would have two or
three gullies monitored collectively. Hopefully, the sum total was right even
if there might have been some by-passing of individual inlets. However, the
main thrust of that data collection program was directed towards the calibration
of the routing part of the model, the storage routing constant.
Falk (Sweden). On three of the catchments in Lund we had devices like

that to avoid by-passing.
Lowing (U.K.). While we are on the subject of gullies, I would like

to raise the question of spacing. As we both said, we have in our countries
very small areas draining to gullies. We are wondering now whether this is so
clever because there is more storage available above the ground than had been
allowed for in previous methods. Previously used times of entry (2 or 3 minutes)
have now been shown to be rather large under-estimates of what the real delay
times are. If there is more storage above ground, why do we not use even more
of it by having fewer inlets, thereby having fewer lengths of first-order pipes
and lesser maintenance problems? As somebody pointed out in a recent committee
meeting, what if the water is flowing down the street for ten or fifteen
minutes after the storm? Nobody is going to be out on the street anyway, so
you might as well use the roadway as a primary drainage system. Only recently
has it been realized at home that the gully pots themselves act as concentrators
of pollution during dry periods, and the more gully pots you have the more oil
and rubber you have sitting there waiting to be flushed out the first.time it
rains. This has been shown to be quite a significant factor in affecting the
quality of the first flush runoff. So there are one or two arguments suggesting
that we might in future try to have fewer inlets.
Medina (Duke University). I would like to ask Mr. Skretteberg about

the assumption of no surface runoff from pervious areas. In some urban catchments
in the United States, there can be significant surface runoff from pervious areas
such as park land. Would you please comment on the NIVA time-area method versus
SWMM in this regard?
Skretteberg (Norway). I mentioned that there is a problem especially

along the coast with runoff from permeable areas, particularly in the winter.
This problem is going to be investigated under a new research project.
Zuidema (Netherlands). It is not a disadvantage that a catchment area

is as small as 700 m^ to determine the physical characteristics of a catchment
and to be assured that the results can be applicable for other areas in common
and will be reliable.
138
Falk (Sweden). During the summer, it can take quite a heavy rainfall
to create any runoff at all from permeable areas. But there are certain places
in Sweden where the ground surface is very hard, and you can have a considerable
contribution from such surfaces even during low intensity rainfalls.
Lowing (U.K.). In the Road Research Laboratory method, they

certainly made the assumption of no runoff from pervious areas, and I know that
Mr. Terstriep of Illinois found it did not work too well in the U.S. and
introduced a model component for the pervious area runoff. The Road Research
Laboratory itself introduced a model for pervious area runoff when it came to
apply the method in tropical Africa but it was not so happy when we wanted to
do the same thing in the U.K. because, looking at typical summer events, it
does not appear as though you get any runoff from pervious areas. Yet when you
do regressions of runoff volume on soil properties and antecedent conditions
(which you think would have more to do with the wetness of the soil rather than
the wetness of paved areas), these variables are significant (in the U.K.). So
one does not really know, but if they are significant variables in a regression
study they should be included and so we allow pervious area runoff to contribute
in theory. But having said that, the average runoff that we observe is certainly
less than 100% from the impervious areas only. So there might have been two
compensating errors in the original assumptions, but the one that assumed 1007.
from the impervious areas was the dominant error.
Terstriep (Illinois State Water Survey). Mention has been made of

the fact that we added a pervious area component to the RRL. The primary reason
was the difference in rainfall regime between most of the U.K. and at least
Illinois. I do not remember the numbers exactly, but for the 5-year, 1-hour
storm, for example, I believe we have nearly three times the expected rainfall
in Illinois than you see in most of the U.K. Also, in relating this to quality
of runoff of urban areas, in the quality modeling that we have done we have gone
back to considering the loadings from paved areas only. We did not consider any
loadings from pervious areas. Now we feel that we are probably ignoring certain
nutrients in doing this, but also we are dealing with a much smaller rainfall.
For the most part, a half-inch-of rain in an hour is about the maximum rainfall
that we have dealt with in looking for high concentrations of pollutants in
urban runoff. In the range of a half-inch or less the contribution from
pervious areas tends to be negligible.
Singh (Illinois State Water Survey). There may be considerably more

discharge from non-paved areas on steep slopes than from paved areas on flat
slopes. As far as pollution from a pervious area is concerned, it may not show
up as a peak but may be sustained for some time, and might have a serious
impact on the quality of receiving waters.
Marsalek (Canada). I would like to return to the question of directly

connected impervious areas. Because certain properties may have roof leaders
that are connected to sewer systems while others may not, it becomes practically
impossible to determine the directly connected impervious area just from plans.
I would like to refer to experiences we have had in Burlington, Ontario, where
we have been collecting data for some time. If we use the total impervious
area, which is something like 110% of the connected impervious area, we simply
139
have too little runoff or we cannot explain the variation from one event to
another. Once we use only the directly connected impervious areas, and we
introduce surface storage more or less as a constant value, we have extremely
good explanations of runoff volumes from events of various magnitudes. That
is, if one takes an incorrect value for the directly connected impervious area
then another source has to be found to explain where any residual measurement
error is coming from. I presume if I were to do a regression analysis, the
answer would be that it is coming from the pervious areas. So I think this is
one of the dangers. Staying on the same point, when looking at the regression
equation showing the possible contributions from the pervious areas via soil
moisture conditions, etc., Dr. Lowing mentioned that the magnitude of the
storm or the depths of precipitation did not matter at all. I would say that
this would probably contradict the assumption of a contribution from pervious
areas, because I would expect that for minor storms there would be no
contribution, but for rather severe storms there would be a sizeable
contribution.
Lowing (U.K.). Yes, I share that intuitive belief. I suppose we

should not start into this because we can argue intuition all afternoon. We
have tried to stick with analysis of actual measured data which is why we are
unhappy about talking about runoff from pervious areas because we cannot
separate it by measurement. But I do not want to be misunderstood on the other
issue. We certainly were considering only directly connected impervious areas
and we always do know where the roof leaders discharge. That was not the
problem so much as accidental mistakes in individual cases. It-has been suggested
that those should be investigated in detailed surveys and that it made a great
deal of difference whether the surface was old or recently resurfaced, and so
on, and one should make allowance for this, but we thought that was impractical.
Falk (Sweden). For one of 13 small asphalt catchments in Lund, we

found that something like 8% of the rainfall volume minus depression storage
was a continuous loss and we explained that as infiltration. But that was the
worst maintained surface of all, with a large number of cracks. For the rest
of the catchments that loss was. only something like 2% on the average.
McPherson (U.S.A.). I wonder if micro-catchment research might not

be useful that would include not just a single inlet area but maybe multiple
inlet areas in series, and such things as house leaders, not for quantity so
much as for quality, to try to track the accumulation and transport of
pollutants through urban areas. My feeling is that until we start going into
this kind of detail we end up heavily in conjecture. Any opinion on that point?
Terstriep (Illinois State Water Survey). I would like to support

that point. We propose one microbasin in our upcoming project in the National
Urban Runoff Program where we will measure quantity and quality at an inlet.
I had to argue strenously to keep this in our proposal. In fact,- I had two in
there originally and we ended up with one. I feel, for example, that the
washoff function that is used in quality models now is one of the weakest
portions of these models and there is virtually no data to test the washoff,
particularly on a particle size analysis, and I think this is a severe data
need right now.
140
Falk (Sweden). The inlet runoff gage used In Lund Is now at the
Chalmers University of Technology and the purpose there is to look at washoff.
They will make measurements on catchments of the same size but will especially
be looking at pollution.
Jennings (U.S. Geological Survey). I wonder if the question of micro-

catchment gaging perhaps bears on the original question that Professor Laurenson
asked. When the ultimate objective is to study a 50 mi 2 watershed, we need
information from some small catchments on basic processes.
McPherson (U.S.A.). At the design storm seminar at Ecole Polytechnique

in Montreal last week, we discovered that we were arguing with each other about
different kinds of problems. Someone was talking about metropolitan-wide planning
and someone else was talking about designing a particular neighborhood collection
system, and someone else was talking about automatic control of a system. Still
another was talking about research where they were trying to learn more about
basic fundamentals and processes. Therefore, I think that today we must define
what we are talking about. I was really talking about fundamental processes,
because that was supposed to be one of the objectives of the National Urban Runoff
Program in the U.S. It depends on what we are looking for. If you want a
demonstration, that is something else again, where you are talking about an area
of a square mile or something like that. I guess it all boils down to: what is
a representative sample?
Jennings (U.S. Geological Survey). We may have a possible cooperative

study in Rochester, New York, on an area of 100 mi 2 . All the point sources have
been taken care of yet water quality problems remain, so urban runoff is thought
to be the culprit. The question is: do we invest in very small catchments
studies or do we choose a relatively large catchment that becomes a subárea in
the total watershed model and study the processes on, say, 100 acres using very
detailed gaging? A 50 mi 2 to 75 mi 2 basin will have several hundred subareas in
any distributed rainfall-runoff-quality model. We all know what happens when you
put that many subareas into a model. Things kind of blend together in unusual
ways, and we found ourselves getting reduced accuracy of prediction when we had
more elements than if we had simplified to a smaller number of subareas. I
agree that with the micro-catchment we can really learn about processes, but
where does that leave us when we have to do management-type studies on practical
large basins?
McPherson (U.S.A.). Another aspect of this seems to be the struggle

between art and science. One side is the science of the processes and the
other is the art of engineering. Right now, the emphasis is on the art.
Lowing (U.K.). I am sure there are many different ways that you can
look at this question, but the one that occurs to me straight-away on the size
of the catchment you should be gaging is that we were looking at small source
areas because that is where it all happens. The runoff gets into the drainage
system by sitting around for awhile above the ground, and by measuring the very
first point you can, you learn more about the proportions of storage that
operate above and below ground. As I was saying earlier on, this is beginning
to affect our thinking about the number of gullies you should have and this has
implications on the costs of construction and maintenance of the smallest pipe
141
size and connections, and street-sweeping effectiveness,, and whether people are
happy seeing water in the streets or not. I think it is important to do both;
and you have to investigate the small end as well as the large end.
Sammons (Soil Conservation Service). What seems to be missing

completely is whether or not the source areas you are working with have any
transfer value. In the wide range of climatology that we have here in the U.S.,
we can have adjacent watersheds where you cannot transfer the information from
one to the other although you are gaging both of them, because one may have a
thermals,pring on it or something else that is really characteristic of the
geomorphology. Everybody wants to know rationally what goes on, but as Mr.
Jennings brought out, we are forever putting these things together in a model
and yet we do not know how to take the subareas into account or even how to
take two areas and bring them together. We do not really know what happens
downstream from two areas coming together unless we have some gaging. I have
seen estimates for areas of reasonable size being 5 and 10 feet off using
different methods of routing. We have all those kinds of problems, as well as
your carry-over effect, etc. And then here in the U.S., and especially in urban
areas, they believe in on-site storage. In other words, so that you do not harm
your neighbor, you have to store runoff on-site. You get into all kinds of
events, in two categories, what some people refer to as design concepts and
actual events. One of them is the real world and the other is merely a concept
and there is no realistic relationship between them.
McPherson (U.S.A.). I believe that this discussion points out the

frustrations engineers have about the state-of-the-art versus the scientific
content of urban hydrology, while at the same time urban America continues to
grow and to be rebuilt and the work must go on. We are faced with this
unfortunate dichotomy. On the other hand, a sarcastic view holds that lack of
data has never particuarly bothered us before, so why reform suddenly?
Laurenson (Australia). The problem of the size of areal discretization

that we have been discussing is quite analogous to the problem of time
discretization that all of us have thought about a long time. I think that in
the time domain it has come to be realized that processes have to be studied on
a time scale of as small as 1-minute or 5-minutes or so if one is to gain a
complete understanding of the processes that have been taking place, but if, on
the other hand, you are dealing with a design problem in which the relevant
storms have durations of 6, or 12, or 24, or 36 hours, then there is no way in
the world that in the design problem you are going to use a time increment of
1-minute or 5-minutes. You average things out over a much longer period and I
suppose the same thing applies also in the areal discretization problem. If we
are to research and understand the processes that are really taking place on the
catchment, then we need to work in terms of areas as small as 100 m2 if this is
the inlet area, but that in design problems for large areas of 10 km^, then
clearly one has to lump these things together to a scale of subárea that is
just adequate to represent the areal variability of the important parameters in
the particular design problem which is of concern. That is, I think there is a .
very strong analogy between the areal discretization problem and the time
discretization problem. To change the subject, I would like to pose a question
about the frequency of street-sweeping. This has come up in some of the comments
that have just been made and I would like information on just what is the order
of frequency of street-sweeping that people are thinking about when they talk
about this subject. Where I live, the streets outside my house get swept about
once a week, and in the city areas they get swept about once a day. Yet, last
142
week in Canada I heard someone comment about a frequency of sweeping of once a
year to remove the gravel and other larger particles that build up during the
winter. Just what frequency are we talking about in North America when we talk
about frequency of street-sweeping?
Marsalek (Canada). Street-sweeping practices vary, not only in Canada

but I guess all over the world. In fact, we had a project in which some of these
practices were surveyed in a number of countries. For example, in downtown
Zurich, Switzerland, people sweep about five times a day. In Canada, the
situation with which I am familiar is the city of Hamilton, which has about one-
half a million people if its suburbs are included, and Burlington, which has
about one hundred thousand people. In Burlington, the commercial or downtown
districts are swept daily. However, in outlying residential areas this frequency
is reduced dramatically, and in many areas streets are swept only once a year,
typically in the spring, to remove the sand and grit which remain on the catchment
after the winter. Having monitored water quality on one of these catchments which
is swept once a year, with removal of leaves in the fall, that is about all there
is to it. Water quality is not as bad as one would get by multiplying standard
daily loading rates by 365 days. Referring to the situation in downtown Hamilton,
commercial areas are swept daily, whereas residential areas are swept perhaps
once every two weeks to once every four weeks.
McPherson (U.S.A.). Some U.S. communities have only an annual spring

cleanup. At the other extreme is a city in the west that gets only 7 inches of
rainfall a year that both sweeps and washes its downtown streets every day
because its economy is based on gambling casinos.
Desbordes (France). Almost two weeks ago I asked the Director of Sewage
Service of the city of Montpellier about street cleaning and he said that with
6,000 street catchbasins and with the equipment and personnel available, each
catchbasin is going to be cleaned each six years! Street-sweeping is not the
only practice that affects urban runoff pollution control.
Torno (Ü.S. EPA). I am not sure that frequency is the entire question.
I live in the Washington, D.C., area and they run trucks down the middle of the
street, fire a little water off to the side, and I suspect they are probably
running at an efficiency rate of only a few per cent. I do think that frequency
and efficiency are related. One of the objectives of the National Urban Runoff
Program is to test the efficiency of what we choose to call in this country
"Best Management Practices," which is interpreted to mean remedial measures that
do not involve substantial capital construction works, and hopefully, we will be
able to shed more light on that question. I think the answer to your question is
that practices are highly variable as you have heard here.
Lowing (U.K.). I was in southern Brazil a few months ago, and I doubt
that they even sweep their streets at all in the town I was in, because I was
looking at the road gullies and they were totally choked up. There was no way, I
thought, that any water could get in them because they were totally choked with
rubbish. Then I saw how the street-cleaning was achieved because it rained. It
had all gone - the storm just cleared the streets and the gullies were emptied
by the pressure of water. Maybe this is the best way of doing it.
Marsalek (Canada). I personally think that if we are interested in

street-sweeping from the pollution control point of view, that not only is the
frequency of street-sweeping important but also the frequency of rain. If it
143
rains every third day and we sweep once every two weeks or once every month,
the catchment has been cleaned by rain before the sweeping machinery is put on
the streets.
McPherson (U.S.A.)* What goes up eventually comes down, and what we

have been overlooking in many cases is taking measurements of precipitation
quality especially in some of our larger urban centers. It is amazing that this
is often overlooked.
Falk (Sweden). In four of the catchments situated in central parts of

Gothenburg and towards the outskirts of that city, it was possible to relate much
of the runoff pollution to the atmospheric fallout that tended to decrease with
the distance from the towns. Also, they could explain some pollution from
traffic volume, especially lead.
Delleur (Purdue University). In making correlations of some models

in the U.S., I have found that the slope of the urban catchment is usually not
a variable, which I have attributed to the fact that the sewers in this country
are usually designed with limitations in velocity, usually on the order of about
2% feet per second for the minimum and about 10 feet per second for the maximum.
However, this is not the case for many European catchments where the slope is a
significant variable, such as in the U.K. I would like to ask what the
respective experience has been and if there is a physical reason why slope may
happen to be a statistically more important variable in some European places
than it seems to be in the U.S.
Lowing (U.K.). The statistical dependence of a response-time parameter

on slope was indeed observed in the U.K. but, in the urban context, we have
demonstrated it only on the above-ground sub-area draining to road gullies rather
than in pipes. I am,.however, confident that response times of complete pipe
networks in the U.K. would also show a marked dependence on slope even though
our range of permitted velocities is broadly similar to yours. I think the
difference can be explained in terms of the size of catchment being studied,
the extent of within-catchment-variation of pipe slopes, and the particular
method of quantifying the catchment slope.
Laurenson (Australia). In a recent study of the time-lags on some

urban catchments in Australia, I looked at two catchments, one of which had an
area of 0.7 km? and the other 5 km?. The lag time for the smaller catchment was
one hour and the lag time for the larger one was half an hour. The relevant
slopes of the main stream were something like approximately 0.5% for the smaller
one, and 2% and 3% for the larger one, and it seemed very clear that the reason
for the longer lag time of this much smaller catchment was the very much
greater, slope of the larger catchment. As for the reason, I think that
Australian engineers tend not to worry about extremely high velocities in the
pipes, and they prepare to lay concrete pipes on any slope whatsoever without
worrying about the velocity in them.
144 (I) SC.81/XXI-21-3/A

ISBN 92-3-101984-8

Recherche Desbord

Uploaded by

Document Information

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

Recherche Desbord

Uploaded by

Copyright:

Available Formats

Technical papers in hydrology 15

General editor : M . B . McPherson

Published in 1977 by the

Urban hydrological modelling and catchment research in U . S . A . 11

Urban hydrological modelling and catchment research in Australia 65

Urban hydrological modelling and catchment research in Canada 89

Methods for calculating m a x i m u m flood discharges for natural watercourses and

Urban hydrological modelling and catchment research in the United K i n g d o m 161

Scope of This Report

Figure l' 5 ' is a simplified description of the stormwater and wastewater

3. Sonnen, M. B., Larry A. Roesner and Robert P. Shubinski, "Urban Water

Shown schematically in Figure l' 1 ' is a simplified description of the

There is widespread interest in multi-purpose drainage facilities that

Anticipating the possible future availability of new field data, an ASCE

Space does not permit presentation of a more comprehensive background,

Mathematical models used for the simulation of urban rainfall-runoff or

Planning Models are used in macro-scale applications, such as for

Design/Analysis Models are hydraulically more sophisticated and thus are

Operations Models are likely to be more application-specific than planning

Sewered Catchment Models

In keeping with the unanimous recommendation of the IHD/Unesco Subgroup

Explanation of Table 1 Entries. Instances where more than one model

Comments on Table 1 Entries. Of considerable importance are the facts

Availability of Computer Programs

Illinois Urban Drainage Area Simulator. Entered in Table 1 is the

Environmental Protection Agency SWMM and Variants. Subroutines for

*: References indicated by letters appear in Table 1A.

ERL INFIL TRAT ION

RECEIV ING WAT

University of Cincinnati. Components of the University of Cincinnati

Hydrocomp. The Hydrocomp Simulation Program System, outlined in

University of Illinois Models. A user's manual with part of the

University of Nebraska. A model description and the computer program

Dorsch Consult. Analogous in complexity and capabilities with SWMM-type

Georgia Institute of Technology. A model description and the computer

Battelle Northwest. A general description of the model has been

FIGURE 4. THE HYDROCOMP SIMULATION PROGRAM SYSTEM 1 2 6 *

RAINFALL INFILTRATION AND

STREAMFLOW GUTTER PIPE

FIGURE 5. GENERAL STRUCTURE OF MIT CATCHMENT MODELth)

Chicago Hydrograph Method. Based on the first sewered catchment

Corps of Engineers (SSARR). A program description with a user's manual

Corps of Engineers (STORM). Model documentation,^' a user's manual,(43)

An extensive analysis of various models^"' compared the characteristics

Computer requirements and programming languages for almost all of the

svaav snoiAH3d woaa aaowia • •

Unit Hydrograph Tests

The report on the Georgia Institute of Technology m o d e l e also contains

An excellent manual on unit hydrograph analysis is available,(53) together

Unit hydrographs have been employed in recent studies investigating

Sewerage Operations Testing

Cited in Table 1 is a 29-ha portion of the Vicente Street catchment in

A compendium,(^^) two companion reports,(78>79) a North American summary( 8 ^)

Some Other Kinds of Models

A set of conceptual models tracing pesticide movement and behavior in the

Of course, modeling of groundwater quantity('2) has been growing all the

A model for metropolitan application has been developed which considers

A questionnaire circulated a decade ago(°7) revealed that all but a small

We are reminded that until each internal module of an overall catchment

It might be concluded that model development has already greatly out-

Epilogue. The presentation in this Section was completed in November,

A model developed by the University of Texas(107) has been tested on

As part of a project at Texas Tech University, the British Road Research

A series of papers dealt with development of unit hydrographs for gaged