Rainfall data for the design of sewer detention basins

CHALMERS TEKNISKA HOGSKOLA
GEOHYDROLOGISKA FORSKNINGSGRUPPEN
Geologi
Geoteknik med grundlaggning
Vattenbyggnad
Vattenforsorjnings- och avloppsteknik
ISSN 0347 · 8165
RAINFALL DATA FOR
DESIGN OF
DETENTION
SEWER
BASINS
THE
Rainfall intensity
mm/h
40 --
I
0 120
1:
Ttot
180
Tm
Time
240 min
Viktor Arnell
Henriette Melin
Meddelande nr 76
Goteborg 1984

ISSN 0347 - 8165
mm/h
60
40
0
240
Time
Adress:
Geohydrologiska forskningsgruppen
Chalmers tekniska hogskola
412 96 GtHEBORG

PREFACE
This report includes the result of research on rainfall data for
design and analysis of detention basins. Earlier reports on rainfall
data deal with the evaluation of new Intensity-Duration-Frequency
curves for Goteborg, rainfall statistics on durations and
volumes for different rainfall intensity levels, and rainfall
data for design of sewer pipe systems with computer models. After
the present study the research continues with studies of rainfall
data for the design of detention basins aimed to reduce overflow
volumes.
The manuscript was typed by Ann-Marie Hellgren and the figures
were drawn by Alicja Janiszewska.
Programming, preparation of rainfall data and computer simulations
were made by Henriette Melin.
The work was financially supported by the Swedish Council for
Building Research (project No. 810284-7), Chalmers University of
Technology, the local Water and Sewage Works in Goteborg and VIAK
AB.
Linkoping June, 1984
Vi ktor Arne 11

i i
LI
OF CONTENTS
Page
PREFACE
LIST OF CONTENTS
SUMMARY
1 INTRODUCTION
i
i i
iii
1
2 REALIZATION OF THE DESIGN OF DETENTION BASINS
5
2 1 Definition of an Independent Flooding Event and
Design Criteria
5
2.2 Rainfall Data Used in the Investigation
5
2.3 Description of the Runoff Model Used
14
2.4 Description of the Detention Storage Model 16
2.5 Description of the t Catchments and Detention Basins 19
2.6 ign of Detention Basins for Various Types
of Rainfall Data 24
2.7 Traditional Design of the Detention Basins 26
3 COMPARISON OF DESIGN FOR VARIOUS TYPES OF
RAINFALL DATA
30
3.1 Factors Influencing the Comparison of Design for
Different Types of Rainfall Data 30
3.2 Result of Design Using Historical Storms 33
3 3 Result of Design Using Design Storms
3.4 Result of Design by the Traditional Method
34
39
4 CONCLUSIONS
APPENDIX
REFERENCES
42
44
70

i i i
SUMMARY
This report describes the on of rainfall data for design
of detention basins The use of ign storms (the uniform intensity
design storm, the Chicago design storm and the Sifalda design
storm) were compared with the use of historical storms where
the historical storms were assumed to give the most correct result.
Historical storms were selected from an 18-year rainfall record
and local coefficients for the design storms were estimated from
the same record.
The historical storms as well as the design storms were used for
design of detention basins located at the outlet in each of three
test catchments of the sizes 0 154 km 2 , 1.45 km 2 , and 0.185 km 2 .
The detention basins were emptied through outlets of nozzle-type
located at the bottom. Two permitted maximum water depths, namely
2.0 and 3.5 m, were tested and the permitted maximum outflows
were varied between 5 and 30 1/s ha counted for the contributing
surfaces only.
For the runoff simulations the CTH-Urban Runoff Model was used,
which includes the processes of infiltration, surface depression
storage, overland flow, gutter flow, and pipe flow. A special
sub-model simulated the flow through the detention basins and
designed the size of the horizon ta 1 a rea of the basins so the
maximum water depth was not exceeded. The outflow of each basin
was a function of the water depth
The result showed that it is important to simulate the inflow
hydrograph to the basins and that the design storm has a correct
total rainfall volume. The resulting basin volumes were not much
influenced by the maximum water depth. The uniform intensity design
storm underestimated the basin volumes by on average 18% and
should not be used for practical applications. The use of the
Chicago design storm resulted in underestimated basin volumes of
on average 9% and the Si lda design storm caused overestimated
volumes of on average 13%.

iv
A manual
ish Water and
Works soci ion (VAV P31) were also . The inflows the
ins were calcul by a linear t rea method and the time
of ion by an equation king into account the area of
the catchment, length and slope of the main pi , and the
rainfall i
ign rainfall intensities were taken
from i curves The i gn outflows
from the ins were obtai as the mean values of the outflows
ins were full and empty
The outflow is obtained was found important in the manual
method The method gave underestimated basin volumes of 5-20%.
The timations were larger in the large catchment than in
the two smaller catchments. Also, the underestimations were
larger the small outflows than for the larger outflows. To
compensate for the underestimations when using the manual method
the in volumes should be increased by 5-20% depending on the
total size contributing areas and the design outflow from
sin

1
1 INTRODUCTION
Detention basins in sewer systems are used to reduce the runoff
peak flows downstream of the basins or to control the amount of
overflow of untreated sewage water to the recipient, or to reduce
the amount of pollution in the sewage water
Rainfall data for design of the size of a basin must be selected
to meet the aim of the basin. As a genera 1 rule basins are designed
to store a specified volume of water while sewer pipes are
designed to carry peak flows. The same rainfall data may not be
appropriate to use for the design of basins as for the design of
pipes. Other factors influencing the selection of rainfall data
are if the basin is located off-line or on- ine. Off-line basins
are located separately from the sewer system and the water enters
the basin via, for example, a weir. On-line basins are a part of
the hydraulic flow system where the water is flowing through the
basin and is detained by controling the outflow from the basin
by, for example, a nozzle, a weir, or a pump. This report describes
the selection of rainfall data for design of on-1 ine
basins with an outlet of a nozzle-type located at the bottom. The
permitted maximum outflow and the permitted maximum water depth
in the basin are limited and flooding is not permitted more
quently than stated by the design return period used.
Two different kinds of rainfall data can be used for the design
of detention basins:
*
*
Design storms estimated from Intensity-Duration Frequency
(I-D-F) relationships or from historical rainfall data.
Historical storms or time series generated by statistical
methods.
A design storm is a rainfall which is developed for a certain
design return period, and the flow value which is calculated by
means of the storm is said to obtain the same return period as
the storm.

2
When historical storms are u a number of storms are run
through a runoff model and stati ical analysis is applied
to the simul flow values to find the flow value coupled with
the ign return period.
The conclusion is that when ign storms are u , the design is
based on rai 11 statistics and when historical storms are used,
the ign is on flow istics The latter must be more
since flow values are the ting ign varia
bles
The ign storms have the advantages of ing easy and inexpen
sive to use They are, however, evaluated for normal runoff areas
only, and can give di rent results for different types of
areas The historical storms give more correct results for all
types of areas, and non-linear effects in the runoff process that
influence the flow statistics can be taken into account. Historical
storms can be more expensive and complicated to use, but that
problem can minimized through well-developed manuals and computer
routines
A number of earlier investigations about detention basins are
reported in the literature but only a few deal with the selection
of rainfall data for design of the basins. Johansen (1979) found
that the use of historical rainfalls, combined with a unit-hydrograph
to the basin, gave approximately the same sizes of the basins
as by using the constant rainfall intensity obtained from
the I relationships as direct inflow to the basin. For outflows
less than 10 1/s hared the use of the I-D-F curves gave a
small underestimation and for outflows larger than 10 1/s hared
the use of the I-D-F curves gave overestimations. The outflow was
assumed to be constant as a function of time. The use of the his
torical rainfalls as di inflow to the basins gave overesti
mated volumes compared to volumes obtained by using the I F
curves. Thus, it was found important to use some sort of runoff
model to calculate the inflow to the basin.
Stahre (1981) found that the use of a model like ILLUDAS (see
Terstriep and Stall, 1974) gave smaller basins than the use of a
linear time-area diagram when using the same rainfall data in the

3
form of a constant rainfall intensity obtained from the I-D-F
curve. The use of a design storm presented by Sifalda (1973) and
shown in Fig. 2.5 gave much larger volumes of the basins than the
I-D-F rainfall data. This could be expected since the Sifalda
des i g n storm i n c 1 u des the I F i g n storm and w i t h r a i n fa 11
added prior to and after this storm. Stahre found it important to
specify the outflow correctly, for example if the given outflow
is the maxi mum permi peak outflow or the maximum permitted
average outflow.
Marsalek (1978) obtained larger outflows than the design outflows
when routing the Chicago design storm through basins designed by
using historical storms. He explained that by the Chicago design
storm caused larger runoff volumes than the historical storms
did.
Pecher (1978) found that the use of historical rainfalls and a
unit-hydrograph method combined with a detention basin model
(function not specified) resulted in smaller basin volumes than
conventional design of retention basins using the I F relationships.
Different studies on design of detention basins have also been
reported by Colyer and Wooldri (1979) Colyer (1980) Wooldridge
( 1981), and Donahue et a 1 ( 1981), but in none of these
last references have comparisons of design of detention basins
for different types of rainfall data been made
In the present report a definition independent flooding events
and the criteria for ign of the basins is rst given. The
precipitation data used in the form of an 18-year continuous historical
record and the ign storms deri from the record are
described. For the runoff simul ons CTH-Model (Arnell,
1980) are used. The detention basin model calcul the outflows
as a function of the water depths in the ins and determine the
areas of the basins given the maximum water depths and the maximum
outflows. Three test catchments are u and detention basins
are located at the outlet of basin The detention basins are
designed both by using historical storms and by using different
design storms. Traditional ign by using I F curves and

4
l i t a ca son
a
use
his CTH-Model givi most
correct ions are given ical ica
tions

2 REALIZATION OF IGN OF I BASINS
5
2.1
Criteria
Independent runoff events must ned to make possible a statistical
analysis of basin volumes calcul for the historical
storms. The basins are using the same flooding frequencies
as for the surrounding sewer pi Arnell (1982) found by
using autocorrelation analysis necessary time between independent
rainfall intensity values to 4 hours and he used that
as the necessary rain-free me period for separation of independent
rainfall events used for i gn of sewer pi pes The same
definition is used in the present investigation combined with the
condition that the detention basins must be empti between two
rainfall events otherwise the events mu be considered as one
event.
The detention basins are desi r the constrai of a permitted
maximum peak outflow varied between 5 30 l/s~hared
(hared = size of the areas contributing to runoff), and a permitted
maximum water depth in the basin vari between 2 and
3.5 m. A specifi baseflow f-cleaning, given in proportion
of the peak outflow, is allowed to pass the basins before
the storage start. The flooding frequencies are varied between
six months and 5 years.
2.2 infall Data U
In the investigation hi cal rainfall
been u from
Lund by, Goteborg, for the period 1 making an 18-year
(1922 excluded because of missing continuous rainfall series.
The time increments vary because only the break-points were
registered when the rainfall curves were formed to the computer.
For the runoff simulations by CTH-Model the data of
interest were transformed to hyetographs with a time of one
minute and divided into individual
using the result of
the earlier mentioned autocorrelation anal is
A rainfall event was
values where:
as a es rai l intensity

11
a co
screeni
in vo 1 ume
t umes
lcul
conas
sin
(2 1)
in
in
water is
times j and 1
(2 2)

7
The method used for screening of the cant i nuous ra i nfa 11 series
resulted in a reduced number of rainfalls which include all
events necessary to make a correct statistical analysis of the
basin volumes and it also includes a safety-margin since the va
lue of the constant outflow Q~ was chosen small enough. Johansen
{1979) used the method above for direct design of the basins and
found that it gave larger basin values than the use of historical
storms and a unit hydrograph method for calculation of the inflows.
The value of Q~ was varied between 1.0 1/s·hared and 30 1/s.hared·
For each outflow the rainfalls corresponding to the 100 largest
basin volumes were identified. Since many rainfalls were identi
cal for the different outflows a total number of 176 rainfalls
were selected from the tota 1 number of about 2300 events. Data
concerning the events are given in Table 2.1.
Designs of the detention basins have been done for three different
types of design storms:
(1) Average-Intensity-Duration {1-D-F) design storm
{2) Chicago design storm
{3) Sifalda design storm.
All these design storms are in one way or another connected with
the I F relationships. Arnell {1982) has presented I-0-F curves
{see Fig. 2.1) and the following mathematical expressions fitted
to the curves:
im = + c (2.3)
where
i
maximum average intensity during the duration T
m
a, b' c = constants listed in Fig. 2 .1.

8
RAINFALL
mm/h
110
im
100
i m =
a
T + b
+ c
90
Return period
year
a
Constants
b
c
80
1 I 3
445
7 2,5
1 I 2
535
7 2,5
725
8 2,5
60
2
965
9 2,0
50
5
10
1325
1700
10 1,5
11 0,5
40
30
F = Return period, years
10
0
0
20 40 60 80
100 120 140
160 180 200 220 240
DURATION, T) min
Figure 2 1
Intensity-Duration-Frequency curves for Lundby 3
Goteborg 1921-1939. After Arnell (1982).
Table 2 1
Data concerning selected rainfall events for design
of detention basins~ Lundby 1921-1939.
Average
value
Standard
deviation
Number per year 10 4
Volume per event (mm) 18.9 9 6
Duration per event (min) 582 426

9
is just the
constant rainfall intensity obtai ned from the I curves (see
Fig. 2.2). When a detention basin is designed, rainfalls with
different durations are tested to find the duration that gives
the maximum basin volume. One disadvantage with this simplified
design storm is that it represents only a part of the total rain
volume of the real rainfalls, as can be seen in Fig. 2.3
100
Rainfall
intensity
mm/h
80
I - 0- F curve for a return
period of one year
60
40
20
0
0
5
Duration
10 15 m1n
Figure 2.2
Example of the evaluation of an Average-Intensity­
Duration design storm from an Intensity-Duration­
Frequency curve. A constant intensity of 41.5 mm/h
during 10 minutes. Return period one year. After
Arne ll ( 19 8 2) .

c
100 ·-
90
80
70
60
50 ~
40
30
30
20

where
t = time
T = total duration chosen equal to 720 minutes
r relationship between the time prior to peak intensity
and the total duration T.
11
The value of r was evaluated from the local data for Lundby and
is given in Table 2.2. An example of the local Chicago design
storm is given in Fig. 2.4.
Table 2.2 Average values of the relationship_, r_, between the
time prior to peak intensity and the total duration
of 720 minutes. Lundby_, Goteborg_, 1921-1939.
F
r
1/5 ~ F ~ l/2
1 ~ F ~ 10
0.43
0.35
Rainfall
inte2nsity , i
mm/h
b )2
+ c (before the peak)
= __ _:;,__ __
2
+ c ( after the
(-......__+b)
a, b, c = c on stan obtai ned from Fig. 1
0
0
r · T
T
Time, t
Figure 2.4 Design rainfall_, suggested by Keifer and Chu
(1957)_, derived from the Intensity-Duration-Frequency
relationship for Lundby_, Goteborg_, 1921-
1939. Return period one year. After Arnell (1982).

maximum
12
, 1 is compounded
ipi ion
and
g. 2 5) The
of the Si
the
ll
t
mm/h
Figure 2 5
Example of a design storm of the Sifalda-type_,
estimated from data for Lundby_, Goteborg 1921-1939
Return period one year. After Arnell (1982).
storm compa to the I design storm is that the total volume
is more
ign storm for Lundby the total
v o 1 u me and rat i on and volumes and durations of the parts
before and a the main part were determined by regression analysis
of heaviest historical rainfalls for Lundby 1921 1939.
Local ign storms were estimated by the following equations
(see also Fig. 2 6):
Ttot u + V·ln T (2.6)
TIII
=
(e + f T) Ttot
1
(2.7)
TI = Ttot - TII TIII (2 8)
ptot
= g + h·ln T (2.9)
PIII
=
(k + 1 T) Ptot
1
(2.10)
PI Ptot PII PIII (2 11)

13
in which the definition of the symbols are given in Fig. 2.6 and
the values of the constants are listed in Table 2.3. The distribution
in time of the volumes was done with the added restriction
that the intensities at the beginning and at the end of the rainfall
are equal to 0.1 mm/h, which was the intensity value used
for separation of the independent rainfalls.
----Intensity obtained
from I- D F curves
0.1 mm/h~
0.1 mm/h
Figure 2.6
Definition of parameters of the local Sifalda design
storm.
Further details concerning all the local design storms are given
by Arnell (1982).

Table 2.3
Constants for estimation the local Sifalda design
storm. Lundby 3 Goteborg~ 1921-1939
Cons
u
v
e
f
g
h
k
1
r)
1/3 1 2 5
.09
111 17
31 70
11 -0 19
11 07 9 7 8.
2 5 04 5.
18.
12 -0 11 .07 03
2 3
inflow
ban Runoff
sins were calcul
CTH-Ur-
CTH-Model is a typical ign/anal is single-event model.
The s re
ich is shown in Fig 2.7, in
eludes the
overland
ow pi ow
sion storage,
model is
plied the runoff area is divi into a number of subare
catchments.
uniformly
ion input
rain
ion is calcula
given as over
values at cons
Horton's
area
time inand
the su
s ion s supply rate is calcul by an
exponential ationship ts ove a flow to s rt
ress ion s
Overland flow is cal
cul according to a ki ic wave theory combi with a relationship
the outflow and the tion storage on
the su Simulation of flow is only a summation of the
overland flow along the gu From the gu wa is fed
through inlets into pi tern pi ulic submodel
works accordi to a kinematic wave theory call a non linear
reservoir ca that allows a realistic attenuation to be simu-
tern
and c flow in a conve tree sewer

15
Infiltration
f
---,
I
~ ....J
Diameter
Flow
a
Figure 2.?
Structure of the CTH-Urban Runoff Model. After Arnell
(1980).
The CTH-Model was chosen for the rainfall study because it is a
detailed model which has proved to make "correct" runoff simulations
for measured rainfalls, and thus the model has not great
influence on the conclusions concerning the choice of design
storms. An example of another model that could have been used is

1
t
ion is
i
ion
The
was u one
ow
all d i
me only on rai
were
ll in
on a
ion
file
ins
2 4
The
ins
water is
sins are empti
pheric sure
flows are functions of
outflows
Given ba
detention
or
in this investi tion are on-line
ng through sins
es nozzle-type with atmos
si (see Fig 2.8) The outpth
s in sins the
their maximum values when sins are full .
are pass the real
start. This is to maintain sel leaning the
ins ion ins are assumed to cause no surcha ing
ins.
The s ou ows calcul wi the ions:
dM
Ql (2 12)
K Hm (2 13)
where
~~ volume r in ion basin
t
= time
Ql and inflow to outflow from the sin
H water in sin center of the
outlet le
K and m constants ich on the geometry of the
outl cons ion

17
the pipe
Figure 2.8
The outlet for a detention basin.
Eq. (2.12) is solved by the following finite difference scheme.
where
j and j+1
As
(2.14)
time j and j+1
= area of the water surface in the basin
(constant, independent of water depth)
= length of time step.
The outflow through the nozzle-type outl
the equation
hole is calculated by
(2 15)
where
size of the outlet hole
acceleration gravity
coefficient for calculation· of head loss at the outlet
hole.
The Eqs 2.14 and 2 15 are sol
ny Newton-Raphson iterations.

size
ion
a
ou
inc
routine
in
ign i
in
maximum
1
value
drau 1 i
ter
ou et
maximum
k.
1
re given (
e
value
calcul
k. is
1
to k r cl
not smaller n 1
2
diame-
2
In
to
ter level
inves ion
tart water level
e minimum ow r leani
was esti nsertion H
outlet ) in 2 15
(d
was se-
wa-
diame-
3
size
by a
zontal area
rou
mi
area until
or comes close to maximum
Usually 5 ls a necessa
sin
is
area
Data for di ins a given in ion
2 5
To k if i ion
the in
rai 11
necessa to
sin was calcul
wa
ion:
in
time
2
1
2
(2
time neces
at
K = Ao(
1
.))2
1
to
rai
outlet
sin
ll to
e
Hred

2 5
As catchments for tests of the different types of rainfall data
three areas were chosen, which had been used by Arnell (1980,
1982) in his validation of the CTH-Model, and in his study of
rainfall data for design of sewer pipes. The areas have different
sizes, topography, and types of buildings as shown in Table 2.4
and were chosen because input data was already prepared. Arnell
(1980) has des bed each catchment in more detail
Table 2. 4 Test catchment data and swnmary of runoff-simulation
input data
Catchment data
and runoff input data
Bergsjon
Linkoping 1
Linkoping 2
Size (
0 154
1.450
0.185
Impermeable part (%)
38
46
34
Land use
Apartment
complexes
Mixed housing
and commercial
buildings
Single family
detached
houses
Slopes
Number of pipes used
in the simulations
Number of inlets used
in the simulations
Sizes of 2
contributing
areas (m )
part of the total
a rea (%)
Surface depression
storage capacity (mm)
Length of the main
pipe (m)
Average slope of the
main pipe
Steep Flat
73 125
47 54
40 100 493 000
26 34
0.42 0 70
606 2824
0.045 0.0063
Medium
54
49
57 100
31
0 63
845
0.012

u
1 i
imu
in
u area
a linear
umes
ion anal
i
ll
umes
simulations
ni cant ow in
ini
(1 1)
rai
su
increase in
Bergsj sin
total a rea
u s
areas
can
tistical di
return Also if
rai lls
runoff
The influence
also
This is
from
reas was
in
i in
area mus
umes from an
runoff
rai ll
ing
no result
meable areas are clea
rounded by cu
a
sewer pi tern no di are u
basins
no
or no inl
as ion
ies t
most e
comno
ca
i
to
ll s
e
ity
sibilities
lly
ipitation was
no uence
measureate
volume
divi
ve
are surof

21
One detention basin is located at the outlet of each catchment
The maximum permitted water depth of each basin was varied between
2.0 m and 3.5 m above the center of the outlet hole which
combined with the maximum permitted outflows of 5 to 30 1/s. hared
gave the detention basin data listed in Tables 2.5 and 2 6. The
outflow given in 1/s.hared is counted for the contributing areas
only. The va 1 ue of the head 1 oss ( ki) at the outl was chosen
equal to 0.5
The diameters of the outlet holes given in Tables 2.5 and 2.6 are
in some cases sma 11 er than is recommended, due to the risk for
clogging (see Section 2 4), but the listed diameters are u
despite making the results comparable.
The detention of water starts when the inflow is larger than the
minimum flow for sel eaning which appears when the water
reaches the head of the outlet hole Thus, the available storage
depth is 2.0 m and 3.5 m respectively minus half of the diameter
of the outlet hole.

Table 2 5
Detention basin data. Maximum permitted water depth
2 0 m.
Li 1 Linkoping 2
r-Ja;

23
Table 2.6
Detention basin data. Maximum permitted water depth
3.5 m.
Basin data
Bergsjon
Linkoping 1 Linkoping 2
Maximum permitted water
depth (m)
3.5
3.5 3.5
Maximum permitted outflow
= 5.0 1/s·hared
Maximum outflow (m 3 /s) 0.0201
Diameter of the outlet
hole (m) 0.062
Minimum flo~ for sel
cleaning (m /s) 0.0019
0.2465 0.0286
0.215 0.073
0.0432 0.0029
Maximum permitted outflow
= 10 1/s· hared
Maximum outflow (m
3
/s)
Diameter of the outlet
hole (m)
Minimum flo~ for selfcleaning
(m /s)
0.0401
0.087
0.0045
0.4930 0.0571
0.305 0.104
0.1028 0.0069
Maximum permitted outflow
= 20 1/s·hared
Maximum outflow (m 3 /s) 0.0802 0.9860 0.1142
Diameter of the outlet
hole (m) 0.123 0.431 0.147
Minimum flo~ for sel
cleaning (m /s) 0.0106 0.2446 0.0165
Maximum permitted outflow
= 30 1/s·hared
Maximum outflow (m 3 /s) 0.1203 1.4790 0.1713
Diameter of the outlet
hole (m) 0.150 0.528 0.180
Minimum flo~ for sel
cleaning (m /s) 0.0176 0.4060 0.0274

2
i Hi cal
In t catchment the ins were igned for the 100 la
est his ca 1 storms for each outflow 1 eve 1 found through
screeni
1 rainfall es described in ion 2.2
Since many storms were the same for the different outflow
s total number of storms were reduced to 176 These 176
were rst run through the CTH-Model and the inflows to
sins for all storms were stored on a computer le That
fi 1 e was u for the design of the basins for different
outflow s Thus, the runoff for the different storms for
ba in needed to be calculated only once.
When doing the
ign it was controled that the basin was empti
ing rainfall.
rough i gn the necessary vo 1 ume of the basin for each
storm was determined to prevent the basin from flooding. A series
of basin volumes for each outflow level was obtained corresponding
to the 176 ra i nfa 11 s. These basin vo 1 umes were ranked in
cending order and the largest (corresponding to a return
period of six months) were plotted on a statistical probability
paper using the plotting formula
y.
1
=1,2 ... ,N (2.17)
yi plotting position for the basin volumes in increasing
(yi = lnF; where F is return period)
N = number of ted basin volumes (in this case 36)
The resul are shown in Appendix I.

25
ign Using ign Storms
The different basin vo 1 umes
following design storms:
have a 1 so been determined for the
(1) Average-Intensity-Duration design storm
(2) Chicago design storm
(3) Sifalda design storm.
The designs were carried out for the return periods of six
months, 1, 2, and 5 years. For each return period designs were
made for different durations of the I-0-F design storm and of
part I I of the S i fa 1 d a des i g n storm to f i n d the dura t i on that
gave the maximum basin volume.
For each design storm and each return period one maximum basin
volume was obtained for each outflow level. The result is plotted
on the graphs given in Appendix I.
In Table 2.7 are listed the durations that caused the maximum
basin volumes and for a return period of one year for the different
design storms and outflow levels.

2 ? The storm storm
II)
vo
the I-D-F design
SifaZda
traditional design. Return
one year.
Catchment
Maximum Design storm and maximum permitted
permitted water (m)
outflow 1-D-F Sifalda Traditional des.
1/s• 2.0 3.5 2.0 2.0
0 360 360 240 240 360 360
10.0 110 120 110 110 120 200
20.0 35 35 40 35 70 70
30.0 25 25 25 25 30 35
Linkoping 1
Li ng 2
5.0 360 360 240 240 360 360
1 0 120 120 110 140 120 120
20.0 60 60 50 40 50 60
30.0 35 35 30 30 25 35
.0 360 360 240 240 360 360
0 120 120 140 140 120 200
0 35 35 35 45 70 70
30 25 25 25 30
2
c
ionsh
i
i
s developed
time-area diagram
n ra i 11
outflow from
tion
ume
Works Association has publis
detention basins (
method, which in
estimation
from
sin. The
sin is
0.
, T
m
, T
m
(2

where
27
M = volume of the basin (m3/hared)
tc
= time of concentration (min)
i = m
rainfall intensity obtained from g. 2.1 or Eq. (2.3)
(1/s·hared)
T duration of the rainfall (min)
Qc = constant
2
outflow (l/s· hared)
The maximum volume of the basin is obtained when
dM = O
dT
(2.19)
The outflows were in this investigation approximated by the equation
(2.20)
where
Q
max
Q
min
= maximum outflow when the basin is full
flow for which the storage starts.
The values of Qmax and Qmin were taken from Tables 2.5 and 2.6
and the resulting values of the outflows Q~ are listed in Table
2.8.

Table 2 8
Outflows for use in design with the traditional
method. Q 0 (Q + Q . )/2. Values in l/s ha d"
2 max m~n re
t
Maximum permi
outflow (1/S· ha
0 0
Lin
Li
1
2
5 11 17
2 6 04 12
2 77 5 11 45 17.
--- --- ---
Li
Lin
i
1
2
2.84 5 79 11 88
3 12 6 48 13 20
2 87 5 87 12 05 18.41
me
on, was calcul by (2 21) given
t ( 1)
pL
Lh80
(2 21)
where
Lh 80
Sh
length of the main pipe plus 80 m (m)
= total sizes of contributing areas (ha)
= average slope of the main pipe.
Values , and Sh were taken from Table 2.4, and
values kl, pl, pi, pA and pS were taken from
e 2.9.
sin volumes were calcul by changing rainfall duration
stepwise nd one that maximum volumes. The
results are listed in Tables 3 5 and 3.6 and plotted in Appendix
I for two maximum water depths of 2.0 m and 3 5 m for each
basin.

29
Table 2.9
Values of constants for estimation of time of concentration
by Eq. (2.21). After Lyngfelt (1981).
Constants
k1
pl
pi
pA
Bergsjon
Linkoping 2
0.490
0.50
0.32
0.10
Catchment
Linkoping 1
0.079
0.71
0.32
0.05

3 I I
N
3 1
i
ion
result
* errors a u
u
real
ca
i 1 i s i mu 1 i
*
i
c
bi
inties
ection
errors
uncertainites in
rai 11
over-
in return
timations
umes
in volumes
to
off volumes
s
vali
This also incl
tion of in
ows thin a
sizes
t sins were
so same
inties
accu
to imulate run
error of about -
reas In
a
r
can
ec
The di
ign storms are
on-
ationshi
uncertainties the I F curves
He also shows that ch in
approx
cul
ly the same
volumes mu t
rel
to
in Fig
11
cul
in
In
ha
the
2 1 14%.
intensities cau
k ows
same
The in ion resul sin volumes for the di
rent rainfall ta should also on for
over- and timation sin volumes cost incl
investment costs and flooding costs Stahre (1981) has published
a res on inves cos s in Fig 3 1, but values
are va to rna e an estimation the increase

31
000
GJ
E
:J
0
>
2 000
~
w

M = in ume
F return
m = lower value di ion ion
z di ion ion
(3 1) imately lid return
d
on can
1/z F
. 1 (3 2)
corres ing to sin volumes M 1
The value i y 0 1 maximum ows
l/s· l/s·ha s'
an ume 10%
return od y
di uncertainties errors are of same size as
were by 1 ( in a similar mat ion
of k flow values same us ions can
The c in volumes for
storms in umes hi storms
should as small as ibl
* s i ons di basin
volumes should as small as all
inties i
runoff, uncerthout
inties of 10-1 can
la increases
in 1 uncertainties
i
tics of
in
phys i ca 1 and s is cal
ign storms a
on
ir opment bust
incl
s

3.2
The complete results of design using historical storms are given
in Appendix I. The basin volumes corresponding to the return
periods of ·six months, 1, 2, and 5 years are listed in Table 3.1.
The volumes in Table 3.1 were estimated by linear interpolation
between the plotted points in Appendix I.
The basin volumes obtained for the historical storms are assumed
to be the most correct. No theoretical distributions were fitted
to the data becasue the p 1 otted points themse 1 ves are the most
correct values available.
Table 3 1
Ba3in volumes calculated for the historical storms
(m /ha d).
re
Catchment
Bergsjon
Linkoping 1
Linkoping 2
Maximum
Maximum basin depth
permitted 2.0 m 3.5 m
outflow Return period Return period
1/s·hared
1
1 2 5
1
1 2 5
2 2
5 145 195 248 318 144 188 242 310
10 116 136 175 242 112 135 173 238
20 68 103 129 164 69 103 123 164
30 53 82 102 150 52 83 104 149
5 141 188 215 323 145 191 250 323
10 105 125 166 230 106 127 170 235
20 56 85 110 147 59 90 115 156
30 31 63 80 122 34 65 85 127
5 140 187 241 308 139 186 239 310
10 110 129 164 231 109 131 168 232
20 61 97 120 157 62 97 119 160
30 45 79 97 140 45 78 98 140

in
were
i
water
in
umes
his
3 1 s
di
i two in s 2 0 m
va ation in sin volumes
sma 11 .
cal storms a
in
1 is
umes
3 5 m i nd i
maximum rmi
For smallest ou ow 5 l/s·ha
vo 1 umes were i
of l/s· ha di t
for the sma 11 Bergsj sin and
were found explanation is t
volumes are the most important
both rain volumes and variation in
ly same sin
ou ow
sin sizes
la r L inkoping 1 sin
small ows rain
, while for larger outflows
inflow hydrographs to
detention stora are important Thus, it is important to use a
runoff model to calcula inflows to detention sins for
la catchments outflows which conclusion was also drawn
by Johansen (1 ).
screeni rainfall se es (see ction 2.2) us
the of ra i 11 events to us in the runoff
simul ions ve a sufficient number of rai lls to make poss
ible a correct statistical ign the sin volumes for return
peri of six months and longer
The 176 rainfalls u in
rainfalls u by Arnell (1
of peak-flows ign of
rai lls can u both
tion basins for rn
study include all the 110
in a similar study for estimation
conclusion is that the 1
ign sewer pi and
six months longer
3.3
The basin umes for the i gn storms for the di
design return periods were plotted in the same diagrams in Appen
dix I as where the basin volumes for historical storms were
plotted Over- and underestimations of basin volumes obtained for
the design storms were estimated by compa ng with the basin vol
umes obtained the historical storms. The resulting di
ences are lis in Tables 3 2 and 3.3 ble 3 4 gives the sum-

35
mary of the result as mean values and standard deviations of the
differences for the different design storms, runoff areas and
maximum water depths
Table 3.2
Deviations in percent between the basin volumes for'
design stor'ms and the basin volumes foY' historical
storms MV is the mean value and a the standar'd
deviation. Maximum basin depth 2.0 m.
Catchment
Maximum 1-D-F Chicago Sifalda Traditional design
permitted Design Design Design 1-D-F curves and
outflow Storm Storm Storm time-area method
Return period Return period Return period Return period
1/s ha red ~ 1 2 5 ~ 1 2 5 ~ 1 2 5 ~ 1 2 5
Bergsjon
Li nkopi ng 1
5 -16 -19 -24 -201 -11 -16 -23 -21 +32 +23 +11 -3 -10 -12 -17 -14
10 -27 -17 -18 -23! -20 -10 -13 -14 +4 +21 +13 -4 -17 -10 -13 -17
l
20 -18 -22 -17
-14 -9 -1 +6 +10 +12 +9 -7 -15 -12 -7
30 -17 -21 -11 ~~~I ~: -12 -5 -7 -6 +5 +15 +1 -8 -12 -4 -8
MV;o -18; 4.6 '-12; 6.4 +9; 10.3 -11; 4.0
i
5 -15 -16 -23 -21 l -10 -13 -22 -22 +37 +29 +15 -4 -14 -15 -22 -21
10 1-25 -14 -17 -221 -16 -6 -11 -13 +12 +29 +17 0 -22 -14 -17 -21
20 -23 -22 -16 -121 -2 -6 -3 +3 +9 +19 +20 +13 -21 -20 -15 -8
30 1 -19 -27 -13 -111 +23 -5 +8 +5 +19 +11 +25 +9 -16 -24 -9 -7
-6; 11.6
+16; 10.6 -17; 5.3
t
I
5 -14 -17 -23 -19! -10 -14 -22 -20 +35 +27 +13 -2 -9 -9 -15 -12
10 -25 -15 -15 -21 i -18 -7 -9 -12 +7 +24 +18 -2 -15 -7 -9 -15
MV;CJ ,-19; 4.9
Linkoping 2 20 -15 -22 -14 -10 +3 -11 -6 +1 +11 +12 +16 +10 -2 -12 -8 -4
30 -13 -23 -12 -11 +9 -13 -3 -4 +4 +4 +15 +4 +2 -14 -3 -4
~1V;o -17; 4.7 -9; 8.4 +12; 10.2 -9; 5.2
I
l
MV = -17.9; 0 = 4.7j MV -8.5; CJ = 9.2 MV = +12.6;0 = 10.6 MV = -12.2; 0 = 5.9

Table 3 3
Deviations in percent between the basin volumes for
design storms and the basin volumes for historical
storms MV is the mean value and a the standard
deviation. Maximum basin depth 3 5 m.
Catchment
Maximum
permitted
outflow
1/s hared
1-D-F Chicago Sifalda Traditional design
Design Design Design 1-D-F curves and
Storm Storm Storm time-area method
Return period Return od Return period Return period
J:z 1 2 5 J:z 1 2 5 J:z 1 2 5 J:z 1 2 5
5
10
Bergsjon 20
30
MV;a
5
10
Li nkopi ng 1 20
30
MV;a
5
10
Li nkopi ng 2 20
30
MV;a
-15 -16 -22 -18 -9 -12 -20 -19 +35 +29 +14 ±o -9 -9 -14 -12
-23 -16 -17 -21 -16 -9 -10 -13 +9 +21 +14 -2 -13 -8 -10 -15
-17 -21 -12 -10 -3 -12 -3 ±o +6 +11 +19 +10 -6 -13 -6 -5
5 -20 -13 -11 ±O -11 -4 -5 ±o +6 +14 +3 -2 -11 -4 -6
- 7; 4.0 6.2 +12; 10.4 -9; 3.9
7 -16 -24 -21 -10 -13 -22 -21 +35 +29 +12 -3 -15 -16 -21 -19
-15 -18 -22 -14 -5 -11 -12 +14 +29 +16
:~I
-19 -12 -16 -20
-24 -18 ~15 -2 -8 -3 ±o +10 +16 +17 -19 -20 -15 -12
-26 -15 -12 +18 -3 +5 +2 +15 +14 +24 +9 -12 -20 -8 -7
- 4.5 10.0 +15; 10.2 -16; 4.4
-14 -17 -23 -20 -8 -13 -21 -21 +36 +27 +14 -3 -6 -9 -14 -12
-24 -15 -15 -21 -16 -8 -10 -12 +9 +23 +15 -2 -12 -6 -8 -13
-15 -21 -13 -11 +3 -9 -3 +6 +13 +13 +18 +9 ±o -10 -5 -4
-11 -22 -12 -10 +11 -9 -2 -2 +7 +8 +16 +6 +4 -9 -1 -2
-17; 4 7 -7; 9.0 +13; 10.0 -7; 5.1
!
MV = -17.6; a = 4.6 MV = -7.5; a = 8.5 MV = +13.4;a 10.1jMV = -10.4; a = 5.9

37
Table 3 4
Deviations in percent between the basin volumes for
design storms and the basin volumes for historical
storms. The mean values (MV) and standard deviations
(a) are for an average of all return periods
and all maximum outflows.
Rainfall data
1-D-F
design storm
Chicago
design storm
Maximum Bergs jon Linkoping 1 Linkoping 2 All basins
water
depth (m) MV a MV a MV a MV CY
2.0 -18 4.6 -19 4.9 -17 4.7 -18 4.7
3.5 -17 4.0 -20 4.5 -17 4.7 -18 4.6
2 0 -12 6.4 - 6 11.6 - 9 8.4 - 9 9.2
3.5 - 9 6.2 6 10.0 - 7 9.0 - 8 8.5
Sifalda
design storm
2.0 + 9 10.3 +16 10.6 +12 10.2 +13 10.6
3.5 +12 10.4 +15 10.2 +13 10.0 +13 10.1
--------------------!=--------- -------------1-------------- -----------· -=--------
Traditional design
1-D-F curves and 2 0 -11 4.0 -17 5.3 9 5.2 -12 5.9
time-area method 3.5 9 3.9 -16 4.4 - 7 5.1 -10 5.9
The basin volumes corresponding to the I-D-F design storm are on
average 18% smaller than the volumes obtained for the historical
storms. The underestimations are slightly smaller for the larger
outflows than for the smaller outflows. The too small basin vol
umes for the I F design storm are mainly due to the total rainfa
11 vo 1 umes for the I F i gn storms which are sma 11 er than
the total volumes of the historical storms used for estimation of
the I -D- F curves. When I F curves are estimated the maximum
intensity parts only of the real storms are regarded. The rainfall
coming prior to and after the maximum parts are ignored and
thus giving the I-D-F ign storm a too small total volume, see
g 2.3 and Arnell (1982)
The standard deviations of the di for the I F ign
storms are small which means that there is a small variation in
the result only and that the conclusion is clear: The I
design storm gives an underestimation of the basin volumes.

use Chi i storm resul in ti
in volumes on a ow 5 1/s.
timations were 15-20%
1/s.ha
vo 1 umes were both over-
explanation
timations is mainly I F ign
storm, namely total volume of storm is
smaller than the total volumes
cal storms u
evaluation the ign storm it is slightly
than I F ign storm se 1 ion of
storm.
The use of the Chicago design storm resul in more basin
vo 1 umes for 1 a rger outflows which is 1 i ca 1 because the high
peak intensity of the storm has a r on the basin
volumes for large outflows than for small outflows.
When using the Si lda ign storm overestimated ion basin
volumes of on average 13% were
The overestimations were
larger for the Linkoping 1 basin than smaller basins. It
was a 1 so 1 a rger for the sma 11 outflow 5 1 Is. ha than for
outflow of 30 l/s ha
The overestimated basin volumes for the Si ign storm were
obtained mainly because the volumes of part I of the storms are
too 1 The explanation is found in estimation of the regression
equations (2. .11) storm see Arnell
(1982). It was assumed important 1 volumes of the
ign storm are close to volumes the historical storms
which is expressed by (2.9) volumes of part I and III
a regression equation for III was chosen because a higher
value than for part I of the linear correlation coefficient
between the volumes of part III and du ions of part II was
obtai ned S i nee the tot a 1 vo 1 ume vo 1 ume of pa I I I are
given, and the volume of part II taken from the I F curves is
smaller than the average value historical storms, volume
of part I will be too large This will result in an overestimation
of the detention s
An improvement of the Si ign storm could achieved if
the regression could based on part I ins of part III or if
both part I and part III were u

The overestimations for Sifal storm are less the return
period 5 years than for the other return periods ted, probably
because the volume of part II of the storm is larger compared to
the volumes of part I and III for the return period 5 years than
for the shorter return periods
3.4
The basin volumes calcul tradi onal method and
scribed in Section 2.7 are li in Tables 3.5 and 3.6. The volumes
are also plotted in Appendix I together with the resulting
basin volumes from design with historical storms and design
storms. Over- and underestimations of the basin volumes obtained
by the traditional method compared with the volumes obtained by
the historical storm are listed in Tables 3 2 and 3.3, and a summary
of the results are given in Table 3.4.
As can be seen in Tables 3 5 and 3.6 negligible differences in
basin volumes were obtained for the two water depths 2.0 m and
3.5 m.
Table 3.5
Volwnes of the detention basins obtained with the
traditional method expressed by Eq. (2.18) M~ximwn
water depth 2.0 m. The voZwnes are given in m /ha
re
a·
Return
period
years
Catchment
Maximum permitted outflow (1/s·hared)
5 0 10.0 20.0 30.0
Bergsjon 130 96 63 49
Linkoping 1 121 82 44 26
Linkoping 2 128 60 46
1
Bergsjon 171 1 88 72
Linkoping 1 160 108 48
Linkoping 2 170 120 68
Bergsjon 206 152 114 98
Linkoping 1 191 138 73
Linkoping 2 204 150 111 94
on 2 200 1 138
Linkoping 1 2 1 114
Linkoping 2 270 150

40
Table 3.6
Volwnes of the detention basins obtained with the
traditional method expressed by Eq. (2.18) M~imum
water depth 3.5 m. The volumes are given in m /ha
re
a·
Return
period
years
Catchment Maximum permi outflow (1/s ha d) re
5.0 10 0 20.0 30 0
Bergsjon 131 51
Linkoping 1 1 86 48 30
Linkoping 2 130 96 47
Bergsjon 171 124 90 74
Linkoping 1 161 112 72
Linkoping 2 169 123 87 71
Bergsjon 208 155 116 100
Linkoping 1 197 142 98 78
Linkoping 2 206 154 113 97
Bergsjon 274 203 156 140
Linkoping 1 262 188 138 118
Linkoping 2 272 201 1 137
The basin volumes were underestimated by, on average, 10-12%. The
underestimations are larger for the larger Linkoping 1 catchment
than for the smaller catchments probably because the routing of
the hydrographs are more important in a large catchment than in a
small one. The underestimations are also larger for the small
outflows of 5 1/s·hared and 10 l/s·hared than for the other outflows
Sjoberg and Martensson (1982) obtained underestimations of approximately
15-40% when comparing design of percolation basins by
h i s tori c a 1 storms w i t h des i g n by the t r ad i t i on a 1 method . I t i s
clear that the traditional method gives underestimations that are
larger for small outflows than for larger outflows. A summary of
the results obtained in the present study and the results obtained
by Martensson and Sjoberg are shown in g 3.2

41
LEGEND
30
XX
Results reported
by Sjoberg and
Martensson (1982)
X
Bergs jon
0 · Linkoping
X f). Linkoping 2
20
0
15
X
ox
X
0
-0
1 0.
0
5
0
0
5 10 30
sign outflows ( l
Figure 3.2
Size of underestimations of detention basin volumes
obtained by design with the traditional method.
Results of the present study and results reported
by Sjoberg and Martens son ( 1982). Return period
2 years.
To compensate for an expected underestimation the basin volumes
obtained by the traditional method must be increased by 5-20%
depending on the permitted maximum outflow.
The durations of the rainfall for which the maximum volumes were
obtained are listed in Table 2.6. The durations vary between 25
minutes for the outflow 30 1/s· hared and 360 minutes for the outflow
5 1/s· hared

4 CONCLUSI
storms
t ign of on
ins i
usi
his
cal
vJhen using a ign storm it was i
ign
storm a 1 ra i 11
is es
ially pronounced ins
are
la and 1 a r ows time va tion
of rainfall and the runoff calculation is also of significance
resulting
the maximum
for large
in volumes
depth but
m) and small depth
seem to
investi
m)
much i nfl
ons are necessa
The I F design storm and Chicago ign storm mated
the detention in volumes to small total volumes
pee i ally, the underestimations
I F si storm were
significant can not i
The use of the Sifalda storm resul
basin volumes The Si storm can i
ins of part III for esti ions
using both part I and part I I I
The use of t itional
estimations of the in volumes. The
compensated for by increase of rai 11
crease of the in volumes
in overestima
by using rt I
volumes, or
si ificant r-
timations can
-i
I ities, or in
For practical applications
signing detention basins:
followi
is recommended when
*
*
A ru model should u
to the basins.
Historical storms should be u
able
imul ion of inflow
if is such ava i 1

43
*
*
*
The Chicago design storm or the improved S i fa 1 da design
storm can be used if historical storms are not available.
The Intensity-Duration-Frequency design storm should not be
used.
When using the traditional method for design the resulting
basin volumes must inc by 5-20% depending on the
permitted outflow.
The investigation in this report is based on point precipitation
Therefore, the result is limited to small catchments. For larger
basins the use of point precipitation data will result in overestimation
of the basin volumes. Further research is necessary
concerning the spatial variation of rainfall.
The conclusions are valid for catchments with runoff from impermeable
areas only. For catchments with runoff from permeable
areas it is important that the total rainfall volumes and the
rainfall temporal patterns are correct, otherwise, the over- and
underestimations will be larger than for catchments with runoff
from impermeable areas only.

APPENDIX
STATI CAL DISTRIBUTIONS FOR CALCULATED IN VOLUMES

45
After calculation of the basin volumes with the CTH-Model for the
historical storms, the 36 largest basin volumes for each catchment
and each outflow were ranked in descending order and plotted
on exponential distribution papers with a logarithmic scale. The
following formula was used for the estimation of the plotting
positions
i
y. = l:
1 .
J
i 1,2, .. ,N . . . ( I 1)
where
yi = plotting positions for the calculated basin volumes in
increasing order.
N
number of treated basin volumes which are chosen as
equal to the number of treated time periods, in this
case 36 !-year periods.
yi is related to the return period through the relationship
Y; = ln F .... (I.2)
where
F
return period in number of treated time periods, in
this case t-year periods (in the figures, F is given in
years).
In order to make the plottings more clear, the 7 largest basin
volumes were plotted as individual points and the remaining basin
volumes were plotted as average values of 3 volumes and 3 y-values.
The basin volumes for the different design storms were plotted in
the diagrams for the return periods t~ 1, 2, and 5 years.

GSJON
x1mum basin 2.0 m
flow
5 l s ·no red
volumes
r Each point (x) corresponds Each point ( x) corresponds
value I to one historical storm
-
of historical storms
X
X
,,
X
X
.X
I I I
I
LEGEND
......
s1n
mes
X -
m
® =
/!:,.
=
~ =
-E-- =

BERGSJO
xtmum basin depth 2.0 m
f[ow 10 ( S · na.rod
mes
( x) cor responds
to the average value
of historical storms
Each
(x) corresponds
orical storm
X
1
X
X
X
X
X
X
X
L
Basin
END
X =
® =
8 =
~ =
-E-- =
R rt• D
mes
m
m
to
,.j:::..
-..]
R
1
2
3
5

48
X
X
tf)
())
E
:::J
w
II II II II II
E
0
'U
())
L.
tJ1
u
c
8_E
tJ1 0
Q)
L
-..
("")
E
0
0
0
N
0 0
0 l()

ERGSJO
axtmum stn
2.0 m
flow 30 l/s ·no red
s
2 0
Each point (x) corresponds
to the average value
of historical storms
Each point ( x) corresponds
to one historical storm
X
X
X
X
X
X
X
X
X
X
X
X
X
L END
X =
® =
& -
0 =
~=
umes
m
m
,J:;::.
1,0
2
2
3 5

mes
L K G
1mum s;n 2.0 m
flow 5 l s ·no red
cor responds point x) corresponds
onca storm
X
X
I X
I
I
X
X
A\''
X
LEGEND
srn volumes to Ul
0
- -
n n• n
'
® =
~ -
~ -
-E-- =
2 5 2

LIN
1mum
mum
G
s1n 2.0 m
flow 10 l s · red
vo
mes
red
Each
to
of
cor responds
value
storms
Each
(x) corresponds
orical storm
X
X
X
X
X
X
X
X
L
x
X
srn
END
® =
& =
0
-E-- =
umes
= Histprical storm
m
to
Ul
......),
2
2
3 4
5
0

5
LINK G 1
1mum basin 2.0 m
outf ow 20 lis·
point (x) corresponds
point ( x) corresponds
to the average value to one historical storm
of historical storms
I
I
X
X
X
X
X
"""'----
EGEND
srn umes
I I X -
m
. '
® =
b =
0 =
r"'!l'"\l""i""!I'"\C"' ,....,,.....11'"'\riiii'"'\A""''I If"\ N
U1
~=
5 1

s
LINKOP G
1mum s1n 2.0 m
outflow 30 l/s ·no red
Each point corresponds Each point ( x) c·orresponds
to the average value i to one historical storm
of historical storms
X
X
X
100
50 X
X
X
X
X
X
X
X
X
LEGEND
X
=
® =
& =
G =
-E-- =
umes cor
storm
m
0
U1
w
25
2 2 3 4 5
10

5
lf)
ClJ
E
w
w
II II II II II
El
0
N
E
'D
ClJ
'--
lf)
-....
l()
X
X
N
~
(.9 c
0
0
..0
0
---
--
---
E
__J :J
E
X
0
X
X
X

LINKOP G 2
x1mum
basin depth 2. 0 m
outflow 10 l s ·no red
Basin vo mes
m3 nared
300
Each point (x) corresponds
to the average value
historical storms
point (x) corresponds
to one historical storm
X
X
X
X
X
200
1
X X X
X
X
X
X
LEGE
Basin v
X
® =
b. =
1::1
mes correspond·
m
y-
m
0
Ul
Ul

56
X
X
lf)
(1J
E
::J
X
0
z
w c
(/)
_J
II
X
II
181
II

LIN
G 2
X
m
s1n
2.0 m
s1n
volumes
mum
low 30 l/s ·
red
m
2
red
Each
to
cor responds
value
storms
Each
( x) corresponds
orical storm
X
X
X
X
X
100
X
X
X
X
X
X
X
LEGEND
Basin
X
® =
umes cor
= Historical storm
. '
B = __ .... ---- -·- ---·:;}·. -- - ...
0 = Chicago m
U1
-......]
-E-- =
l
Re rn ri
1/2
1
2
3
4
5
0

X
X
X
2 0
X
EGEND
......
srn
X = H
® -
umes cor 0
Ul
00
&. -
~ -
~=

mes
RG
x1mum
mum
Each point (x) corresponds
to the average value
of historical storms
3.5 m
10 l/s · red
corresponds
orical storm
X
X
X
X
X
X
2
0
X X X
X
X
X
X
LEGEND
s1n
X =
®
&
=
G =
-E-- =
mes cor
m
y­
storm
m
to
U1
1..0
2
2
3
4
5
0

'D
OJ
L.
lf)
(Y)
U1

volumes
m 1mum s1n 3.5 m
2
flow 30 l/s ·no red
X
Each point (x) corresponds I Each point ( x) corresponds
X
to the average value I to one historical storm
of three historical storms
X
X
I
1 X
LEGE
5
X
X
X
X
X
X
X
X
Basin umes 1na to 0'\
--J.
X = HI

62
X
X
til
OJ
E
E
L()
u
OJ
L..
(j)
"D
c
0
Q. E
(j) L.
CIJ __. 0
L. (j)
L.
0
u
0
II If II II II
_j g,
(j)
E
L.
0
(j)
----
..c
u
co
w
lf)
OJ
E
:J
0
>
X
X

es
LINK
x1mum
1
sn 3.5 m
flow 10l/s·nored
point (xJ corresponds
to the average value
of three historical storms
Each point (x) corresponds
to one historical storm
X
X
200
X
X
X
X
X
X
X
2
2
3 4 5 0

s LINK G 1
1mum DO Sin 3.5 m
cor responds
average value
of historical s!
I
I' I ........ !
2
X X
red
X
L E 0
0'1
s umes
,.!::::.
.. =
® =
~
=
0 =

LINK
G
1mum
s1n
3.5 m
vo
mes
mum
point (xJ corresponds
the average value
storms
flow 30 l s ·no red
Each point ( x) corresponds
to one historical storm
X
X
X
X
1
X
X
X J'" I
I
LEGE
X
X
X
I
srn umes to Cl"\
Ul
X = storm
® =
X
b - ....,
0 - m
-
-E-- =
5

66
u
tf)

LINK G 2
ax1mum s1n 3. 5 m
outflow 10 l s ·no red
point (xJ corresponds
the average value
historical storms
Each
( x) corresponds
orical storm
X
X
X
X
X
X
2
X
L
END
0'1
-......!
X X X
X
X
X
Basin volumes
X = storm
® =
& - --- -- --- ".:::;J"-
0 =
m
-E-- =
3
5
2

LINK
x1mum
mum
G 2
3.5 m
flow 20 s · red
point xJ corresponds
to the average vatue
of
storms
Each
(x) corresponds
orical storm
X
X
X
1 0
X
I
I
I
I
X
X
X
,d,
I
X
EGE
s1n
I ® =
-
0
volumes
01
r nrr.oc nnnrl. nn 00
m
X
~ I
I
B =
0 =

2
vo
Ll K
1mum
(x) corresponds
to the averaae value
I
I
s1n
2
flow 30 l s ·
3.5 m
red
(x) corresponds
to onP hic,.t orir;::~ I storm
X
X
X
X
X
X
X
X
LEGE
- umes cor
X
® =
= Historical storm
..
0
0'\
1..0
X
X
X
X
ffi
0 =
-E--- =
= Si
,...,
1 2
3 5
2

Arnell V 1980: ription
Runoff Model Chalmers Universi
of Hydraulics Report es A.5.
ti on of
ogy
CTH-Urban
Department
Arnell V., 1982: Rainfall
Chalmers Universi
Hydraulics Report es A.8
Pi
rtment
Bergstrom, T , 1976: Detention Basins in
pal Design, Application and Sizing
Water Works Association, VAV 1.
tems Princi
ish and Waste
kholm. (In Swedish)
Colyer, P J , 1980: ion in rm Ora i
Systems. Progress in Water
Brighton.
ogy, Vo 1 ' pp 21
Colyer, P.J and Wooldri G
Systems Off-Line Tanks in
Hydraulics Research ion,
Oxon.
in Ora i
inage
No IT 188 Wallingford
Donahue, J.R., McCuen, R H. and id T R , 1981: Comparison
of Detention Basin Planning and Design Models. Journal of
the Water Resources Planning and Management Div., ASCE, Vol
107, No. WR2, proc.paper
Johansen, L , 1979: Design Rainfalls for tems Department
of Sanitary Engineering, Technical University of Denmark,
Report 79-2, Kopenhamn (In Danish)
Ke i fer , C . J . and C h u , H H , 1 · S y nth e t i c rm Pattern for
Drainage Design. Journal of the Hydraulics div ASCE, Vol.
83, No. HY4. Discussion by McPherson in Vol 84, No. HY1,
Feb. 1958.
Lyngfe 1 t, S. , 1981: Design of Storm
~1ethod. Cha 1 me rs Un i ve rs i ty
logy Research Group. Report
tems. The Ration a 1
no 1 ogy Urban Geohydro-
. (In sh)

71
Marsalek J., 1978: rch on the Design Storm Concept. ASCE
Urban Water Resources Research Program, Technical Memorandum
No New York
Pecher, R., 1978: Design of Storm-Water Retention Basins. Seminar
on Retention Basins, Marsta, November 7-8, 1978 Nordic Research
Project, Operation of Treatment Plants Report 1.
Stockholm.
Sifalda, V, 1973: Entwicklung eines Berechnungsregens fUr die
Bemessung von Kanalnetzen. gwf wasser/abwasser 114(1973)H9.
Sjoberg, A. and Martensson, N., 1982: The Rain-Envelope Method.
An Analysis of the Applicability for Design of a 2-years
Percolation Basin. Chalmers University of Technology, Urban
Geohydro 1 ogy Research Group Report No. 64. Goteborg. (In
Swedish)
Stahre, P., 1981: Flow Detention in Sewage Systems. Swedish
Council for Building Research, Report Tl3:1981. Stockholm.
(In Swedish)
Terstriep, M. and Stall, J.B., 1974: The Illinois Urban Drainage
Area Simulator. Illinois State Water Survey, Bulletin 58
Urbana, Ill.
Wooldridge, G., 1981: Performance of Storage Tanks in Storm
Drainage Systems. In B.C. Yen (Ed ): Proceedings of the
Second Internati on a 1 Conference on Urban Storm Dr a i ange,
Urbana Illinois, June 14-19, 1981. pp. 350-359. Urbana, Ill.

CHALMERS TEKNISKA HOGSKOLA
lnstitutionerna for
Ceologi
Ceoteknik med grundlaggning
Vattenbyggnad
Vattenforsorjnings- och avloppsteknik
nr 1
nr 2
nr 3
Leif Carlsson: Crundvattenavsankni
med hjalp av provpumpningsdata. 1
Leif Carlsson: Crundvattenavsankni
diffusivitet med hjalp av
nning och grundvattenbildning. Lagesrappor­
(Utgangen)
Del 1. Evaluering av akviferers geohydrologiska data
• 67 sidor.
Del 2. Evaluering av 1
ngsdata. 1 • 17 sidor.
a lagers hydrauliska
nr 4 Viktor Arnell: Nederbordsmatare En sammanstallning av nagra ol ika matartyper. 1973. 39
s i dor. ( Utgangen)
nr 5 Vi ktor Arne 11 : I ntensitets-va rakt i ghetskurvor for hafti ga regn i Cote borg under 45-a rsperioden
1926-1971. 1974. 68 sidor.
nr 6
nr 7
Urbaniseringsprocessens inverkan
ter (1973-03-01 - 1974-02-01). 1
Olov Holmstrand, Per 0 Wedel:
s i dor. ( Utgangen)
lngenjorsgeol
nning och grundvattenbildning. Lagesrapporska
ka rtor 1 i tteraturstudi er. 1974. 55
nr 8 Anders Sjoberg: Interim Report Mathematical Models for Gradually Varied Unsteady Free
Flow. Development and Discussion of Basic Equations. Preliminary Studies of Methods for
Flood Routing in Storm Drains 1974. 74 sidor. ( )
nr 9
nr 10
nr 11
nr 12
nr 13
nr 14
nr 15
Olov Holmstrand (red.) Seminarium om i
Vi ktor Arnell, Bar je Sjolander Matni
Stationsbeskrivning. 1974. 53 sidor. (
ska kartor. 1974. 38 sidor. (Utgangen)
nederbordsintensiteter
)
onen.
Per-Arne Malmquist, Gilbert Svensson: Dagvattnets beskaffenhet och egenskaper. Sammanstallning
av utforda dagvattenundersokningar Stockholm och 1969-1972. Engelsk
sammanfattning 1974. 46 sidor (Utgangen)
Viktor Arnell, Sven Lyngfelt: lnteri
flode inom bebyggda omraden Ceohydrol
Coteborg. 1975. 50 sidor.
. Berakningsmodell for simulering av dagvattenska
forskningsgruppen i samarbete med VA-verket i.
Viktor Arnell, Sven Lyngfelt: Nederbords-avrinningsmatningar i Bergsjon, Coteborg 1973-1974.
1975. 92 sidor.
Per-Arne Malmquist, Gilbert Svensson Delrapport. Dagvattnets sammansattning i Coteborg.
Engelsk sammanfattning. 1975. 73 sidor.
Dagvatten. Uppsatser presenterade vid konferens om urban hydrologi i Sarpsborg 1975 1976.
33 sidor. Foljande uppsatser ingar:
Arnell V. Berakningsmetod for analys av dagvattenflodet inom ett urbant omrade
Lyngfelt S. Nederbords-avrinningsstudier i Bergsjon, Coteborg.
Sjoberg A. CTH-ledningsnatmodell DACVL-A.
Svensson C. Dagvattnets sammansattning, inverkan av urbanisering (Utgangen)
nr 16
nr 17
nr 18
nr 19
nr 20
nr 21
Crundvatten. Uppsatser presenterade vi d konferens om urban hydro 1 ogi i Sarpsborg 1975.
1976. 43 sidor. Foljande uppsatser i
Andreasson L, Cederwall K. Rubbningar av grundvattenbalansen i urbana omraden.
Carlsson L. Djupinfiltration i slutna akviferer.
Torstensson B-A. Foljder av grundvattensankning inom leromraden.
Wedel P. Exempel pa dranering av jordlager grund av tunnelbyggande. (Utgangen)
Olov Holmstrand, Per Wedel: Markvattenundersokningar i ett urbant omrade
1976. 127 sidor.
Coran Ejdeling: Berakningsmodeller for prognos av grundvattenforhallanden. 1978. 130
sidor.
Viktor Arnell, Jan Falk, Per-Arne Malmquist Urban Storm Water Research in Sweden. 1977.
30 sidor.
Viktor Arnell: Studier av amerikansk dagvattenteknik. Resa
sidor.
december 1976. 1977 64
Leif Carlsson: Reserapport fran studieresa i USA samt del,.cn .. a••uc; 2nd International
Symposium on Land Subsidence in Anaheim, USA. 29 nov-17 dec 1
61 sidor.

73
nr 22
nr 23
nr 24
nr 25
nr 26
nr 27
nr 28
nr 29
nr 30
nr 31
nr 32
nr 33
nr 34
nr 35
nr 36
nr 37
nr 38
nr 39
nr 40
nr 41
nr 42
nr 43
nr 44
nr L~S
nr 46
nr 47
nr 48
nr 49
Per 0 Wedel Grundvattenbildning, samspelet jordlager och berggrund. Exemplifierat fran
ett forsoksomrade i Angered. 1978. 130 sidor.
Viktor Arnell: Nederbordsdata vid dimensionering av dagvattensystem med hjalp av detaljerade
berakningsmodeller. En inledande studie. 1977 29 sidor.
Leif Carlsson, Kl as Cederwa 11 : Urbani seri ngsprocessens i nverkan pa ytvattenavri nni ng och
grundvattenbildning. Geohydrologisk forskning vid CTH, Sektion V, under perioden 1972-75.
1977. 17 sidor.
Lars 0 Ericsson {red.): Lokalt omhandertagande av
verksamhetsaret 1976-02-01 1977-01-31. 1977. 120 sidor.
dagvatten.
Del rapport fran forsta
Ann-Carin Andersson, Jan Berntsson: Kontrollerad grundvattenbalans genom djupinfiltration.
En inventering av djupinfiltrationsprojekt. 1978. 273 sidor.
Anders Eriksson, Per Lindvall: Lokalt omhandertagande av dagvatten. Resultatredovisning av
enkat rorande drift och konstruktion av perkolationsanlaggningar. 1978. 126 sidor.
Olov Holmstrand (red.) Lokalt omhandertagande av dagvatten Delrapport nr 2 fran perioden
1977-02-01 - 1977-11-30. 1978. 69 sidor.
Leif Carlsson: Djupinfiltrationsstudier
Lars 0 Ericsson
och Utvardering
lnfiltrationsprocessen
1978. 45 sidor.
Angered. 1978. 70 sidor.
en dagvattenmodell. Teori, Undersokning, Matning
Lars 0 Ericsson, Permeabilitetsbestamning i falt vid perkolationsmagasin. Dimensionering.
1978. 15 sidor.
Lars 0 Ericsson, Stig Hard: lnfiltrationsundersokningar i stadsdelen Ryd, Linkoping. 1978.
145 sidor.
Jan Hallgren, Per-Arne Malmquist: Urban Hydrology Research in Sweden 1978. Swedish Coordinating
Committee for Urban Hydrology Research. 1978. 14 sidor.
Bo Lind, Gote Nordin: Geohydrologi och vegetation i Dalen S, Karlskoga. 1978. 63 sidor.
Ei vor Bucht, Bo Lind: Metodfragor vi d naturanpassad stadsp 1 aneri ng - erf arenheter fran
studie i Karlskoga. 1978. 65 sidor.
Anders Sjoberg, Jan Lundgren, Thomas Asp, Henriette Melin: Manual for ILLUDAS (version
52) Ett datorprogram for dimensionering och analys av dagvattensystem. 1979. 67 sidor.
Per-Arne Malmquist m fl: Papers on Urban Hydrologi 1977-78. 99 sidor.
Vi ktor Arne 11, Per-Arne Ma 1 mqui st
Dagvattenteknik. 1978. 30 sidor.
Bo-Goran Lindquist, Gi 1 bert Svensson: Uppsatser om
Bo Lind: Dagvatteni nfil trati on - forutsattni ngar i nom ett bergsomrade, Ostra Ga rdsten
Goteborg. 1979. 32 sidor.
Per-Arne Ma 1 mqui st (red ) : Geohydro 1 ogi ska forskni ngsgruppen 1972-78. Sammanstall ni ng av
uppnadda resultat. 1979. 96 sidor. Kostnadsfri.
Gilbert Svensson, Kjell 0ren: Planeringsmodeller for avloppssystem. NIVA-mode1len til1ampad
pa Torslanda avrinningsomrade 1979. 71 sidor.
Diskussioner och figurer fran CTH-semina­
Per-Arne Malmquist (red.): lnfiltrera dagvatten
rium 1979-04-20. 1979. 86 sidor.
Bo Lind: Dagvatteninfiltration - perkolationsanlaggning i Halmstad. 1979. 58 sidor
Viktor Arnell, Thomas Asp: Berakning av braddvattenmangder. Nederbordens varaktighet och
mangd vid Lundby i Goteborg 1921-1939. 1979. 80 sidor.
Stig Hard, Thomas Holm, Sven Jonassen Dagvatteninfiltration pa gronytor - Litteraturstudie,
kunskapssammanstallning och hypotes. 1979. 278 sidor
Per-Arne Malmquist, Per Lindvall: Draneringsrors igensattning- en jamforande laboratoriestudie.
1979. 44 sidor
Per-Arne Malmquist, Gunnar Lanner, Erland Hogberg, Per Lindvall: SODRA NASET - ett exempel
pa forenklad utformning av gator och dagvattensystem i ett upprustningsomrade. 1980.
Vi ktor Arne 11
stadsdelen
Hakan Strandner, Gi 1 bert Svensson: Dagvattnets mangd och beskaffenhet
i Linkoping, 1976-77. 1980.
Lars 0 Ericsson, Stig Hard: Termisk
Termovisionsforsok i klimatkammare. 1
strering en metod att kartera markvattenhalt
65 sidor.

nr 50
nr 51
nr 52
nr 53
nr 54
Viktor Arnell Dimensionering och analys av dagvattensystem. Val av berakningsmetod. 1980.
56 sidor 22 figurer.
Lars 0 Ericsson: Markvattenforhallanden i urbana omraden. Slutrapport Goteborg 1980. 115
sidor.
Olov Holmstrand (red.): lngenjorsgeologisk kartering. Seminarium 1980-04-17. 110 sidor.
Olov Holmstrand: Lokalt
nfiltration vid
Olov Holmstrand, Bo Lind Per Lindvall
rade. Lokalt omhandertagande av
av dagvatten Sammanfattni ng av forskni ng om
sidor.
Lars-Ove Sorman
i Bratthammar,
ett lerom-
nr 55
Gunnar Lanner:
Danmark Holland
aneri ng i bostadsomraden i utl andet. Nya pri nci per
and. 1981 10 sidor.
nr 56
nr 57
nr 58
nr 59
Sven Lyngfelt: Dimensionering av dagvattensystem. Rationella metoden. 1981
Erland Hogberg: Samband mellan gatustandard och trafiksakerhet
forstudie. 1981.
82 sidor.
bostadsomraden.
Jan A Berntsson: Portryckforandringar och markrorelser orsakade av tradvegetation. 1980.
121 sidor
Per-Arne Malmquist, Stig Hard Grundvattenpaverkan av dagvatteninfiltration. 1981.
En
nr 60
Annika Lindblad: lnfiltrationsmatni
1972-80. Sammanstallning och statisti
utforda vid Geol
bearbetning. 1981.
ska institutionen, CTH/GU,
sidor.
nr 61
nr 62
Lars 0 Ericsson, Stig Hard: Termisk registrering
Slutrapport. 1981. 18 sidor.
Jan Pettersson Elisabeth
alternativa upprustningsfors
en metod att kartera markvattenhalt.
rorande tva
nr 63
Olav Holmstrand: Praktisk tillampning av ingenjorsgeologisk kartering. 1981
114 sidor.
nr 64
nr 65
nr 66
nr 67
nr 68
nr 69
nr 70
nr 71
nr 72
nr 73
nr 74
nr 75
Anders Sjoberg, Nils Martensson: REGNENVELOPEMETODEN. En analys av metodens tillamplighet
for dimensionering av ett 2-ars ati n. 1982. 29 sidor.
Gosta Lindvall ENERGIFORLUSTER LEDNINGSBRUNNAR Litteraturstudie. 1982. 35 sidor.
Per-Arne Malmquist: Lathund for berakning av Dagvattnets fororeningar. 1982. 32 sidor.
Sven Nystrom: Kommuns skadestandsansvar mot VA-abonnent for oversvamningsskador. 1982 71
sidor.
Sven Lyngfelt, Gilbert Svensson: Dagvattenavrinning fran stora urbana omraden. Simuleringsmetodik
exemplifierat pa Goteborgsregionen. 1983. 118 sidor.
Hans Backman, Gi 1 bert Svensson: Fl odesmatni
Matnoggrannhet under kontro 11 erade forha 11
sidor.
med portabla utrustningar.
s betongledning. 1983. 51
Olav Holmstrand (red): Naturanpassad stadsplanering i Dalen S, Karlskoga. Erfarenheter av
planeringspracess ach teknik under och efter byggandet. 1983 114 sidor.
Olov Holmstrand (red): Reservvattentakter. Redovisning av diskussiansdag 1983-05-18. 1983.
115 sidar.
Gilbert Svensson, Hakan Strandner (overs. och bearb.)
for simulering av floden i av1oppsnat. 1983. 101 sidor.
Gi 1 bert Svensson (red): Byggande drift och
dri ftstorni ngarna omfattande? -Projekterar vi
1984 98 sidar.
NIVANETT manual. Ett datarprogram
se av kommunala va-ledningar. -Ar
satt? - Var 1 igger kastnaderna?
Hans Backman: Avloppsledningar i svenska tatorter i ett histariskt perspektiv. -Ett
sammandrag av tekniska forutsattningar, ideer och diskussioner under 1900-talets ledningsbyggande.
1984. 123 sidor.
Ann-Carin Andersson, Olav Halmstrand Erik Almling, Rolf Rosen, Kjell Soderstrom: Infiltration
ach alternativa atgarder vid grundvattensankning. Jamforande beskrivningar och val
av metoder. 1984 115 sidor.