A case independent approach on the impact of climate change effects on combined sewer system performance

A case independent approach on the impact of climate

change effects on combined sewer system performance

M. Kleidorfer, M. Moderl, R. Sitzenfrei, C. Urich and W. Rauch

ABSTRACT

M. Kleidorfer

M. Moderl

R. Sitzenfrei

C. Urich

W. Rauch

Unit of Environmental Engineering,

Faculty of Civil Engineering,

University of Innsbruck,

Technikerstrasse 13,

A6020 Innsbruck,

Austria

E-mail: [email protected];

[email protected];

[email protected];

[email protected];

[email protected]

Design and construction of urban drainage systems has to be done in a predictive way, as

the average lifespan of such investments is several decades. The design engineer has to

predict many influencing factors and scenarios for future development of a system (e.g.

change in land use, population, water consumption and infiltration measures). Furthermore,

climate change can cause increased rain intensities which leads to an additional impact on

drainage systems. In this paper we compare the behaviour of different performance indicators

of combined sewer systems when taking into account long-term environmental change effects

(change in rainfall characteristics, change in impervious area and change in dry weather flow).

By using 250 virtual case studies this approach is—in principle—a Monte Carlo Simulation in

which not only parameter values are varied but the entire system structure and layout is

changed in each run. Hence, results are more general and case-independent. For example

the consideration of an increase of rainfall intensities by 20% has the same effect as an

increase of impervious area of + 40%. Such an increase of rainfall intensities could be

compensated by infiltration measures in current systems which lead to a reduction of

impervious area by 30%.

Key words | climate change, combined sewer system, land use, modelling, performance

indicator, rainfall intensities, uncertainties, urban drainage

INTRODUCTION

The use of computer models is a state of the art instrument

in different fields of water resources and environmental

engineering (e.g. Chau 2006). Beside a sufficient model

calibration (e.g. Cheng et al. 2002; Kleidorfer et al. 2009) the

consideration of possible future development scenarios

becomes more and more important. Design and construc-

tion of urban drainage systems has to be done in a

prospective way, as the average lifespan of such investments

is several decades. The design engineer has to assess many

prospective influencing factors e.g. change in land use,

population, water consumption and infiltration. Further-

more climate change can cause increased rainfall intensities

which leads to an additional impact on drainage systems

(Rowell 2005; Arnbjerg-Nielsen & Fleischer 2009). Although

climate change models contain substantial uncertainties, the

consideration of climate change effects has become an

important issue to estimate the possible impact on existing

drainage systems (respectively costs of possible climate

change adaption measures) (Ashley et al. 2005). For example

Butler et al. (2007) evaluated the impact of climate change

on sewer storage tank performance using one case study.

Semadeni-Davies et al. (2008) took into account the impacts

of climate change, as well as the land use change due to

urbanisation.Mark et al. (2008) describes the climate change

adaption strategies of the Scandinavian countries and

presents three case studies, where climate change adaption

plans have been arranged. In one case study even future city

development due to urbanisation was taken into account.

doi: 10.2166/wst.2009.520

1555 Q IWA Publishing 2009 Water Science & Technology—WST | 60.6 | 2009

To evaluate climate change impact on combined sewer

systems, performance indicators have to be defined for the

drainage system, e.g. prediction of flooding or prediction of

discharge to receiving water bodies. Impact of climate

change can be analysed under various aspects (e.g. regard-

ing impact on urban flooding or impact on pollutant

discharge). Therefore a multitude of different performance

indicators can be evaluated (Berggren 2008). These per-

formance indicators respond differently on a variation of

boundary conditions, consequently they hardly can be

compared. Hence, when possible future rainfall conditions

are considered for evaluating specific system behaviour (e.g.

urban flooding) the impact on other performance indicators

(e.g. pollutant discharge) is not clear, but can only be

analysed for a specific case study.

In this work we compare the behaviour of different

performance indicators of combined sewer systems when

taking into account long-term environmental change effects

(climate change, land use and population change). Applying

the case study generator presented by Moderl et al. (2009)

250 virtual case studies are generated and subsequently

simulated. While the approach is—in principle—a Monte

Carlo Simulation, not only some parameter values are varied

but the entire system structure and layout is changed in each

run. Further, for each of these generated case studies

parameters are varied to predict environmental change

scenarios. Thus not only one specific case study is assessed

in terms of environmental change effects, but a whole range

of systems. The simulation results are evaluated statistically.

As different performance indicators and their interaction are

taken into account this study helps to understand the

correlation between possible future scenarios and combined

sewer system performance in a case-independent way.

Hence, this study focuses on two questions:

† Is there a correlation between different performance

indicators for evaluating the impact of climate change on

system behaviour? How do adaption strategies for

preventing urban flooding influence combined sewer

overflow (CSO) discharge?

† What is the impact of climate change—induced by

increase of rainfall intensities—as compared to the

impact of future development (urbanisation, land use-

change) and data uncertainties?

METHODS

Model

This work is based on hydrodynamic computer simulation

of combined sewer systems. The model used in this study is

the well known hydrodynamic model SWMM 5.0 (Rossman

2008). Hence the impact on flooding as well as on combined

sewer overflow (CSO) discharge to the receiving water

can be analysed based on a detailed assessment of the

hydraulics in the system.

Case studies

For a case independent conclusion 250 virtual case-studies

(VCSs) with varying system properties were generated using

the case study generator described by Moderl et al. (2009).

The virtual case studies were generated based on a

stochastic approach and designed according to state-of-

the art design guidelines and can be downloaded from the

Institute’s homepage. Of course the VCSs are simplifica-

tions of reality neglecting aspects of real world systems. For

example the VCSs are generated using a branching process

and therefore do not reflect loops in the system. Negative

slopes of conduits were not considered in the used set of

VCSs. Furthermore they are not associated with a natural

urban development which would lead to areas of different

degree of safety for flood protection (i.e. design of the sewer

system on higher return periods in areas of special interest

(e.g. industrial areas) than in rural areas). CSO structures

and detention volumes are located randomly distributed in

the system not only close to the receiving water and the

sewer layout is not interacting with other infrastructure

systems (e.g. water supply systems, roads). Nevertheless

Moderl et al. (2009) showed that the VCSs sufficiently

perform similar as real world systems for prediction of

surface flooding and CSOs (Moderl 2009). Additionally a

real world case study (RWCS) is used for comparison. Some

system characteristics of the real world case study are

presented in Table 1 and Figure 1 shows the system layout

of the real world case study compared to one of the 250

virtual case studies.

Figure 2 shows the 5%, the 50% (median) and the 95%

percentile of empirical cumulative distribution functions

1556 M. Kleidorfer et al. | Climate change effects on combined sewer system performance Water Science & Technology—WST | 60.6 | 2009

(CDFs) of the 250 VCSs (grey) and the CDF of the RWCS

(black) for impervious area per node (Aimp/node), for the

conduit diameters and for dry weather flow (DWF/node)

per junction. Hence each CDF illustrates the variance of

these parameters within the case studies. For example in the

RWCS about 50% of the nodes are connected to a conduit

diameter of about 1m or less (Figure 2, middle, black line).

A more detailed analysis and description of system

characteristics of those 250 VCSs is available in (Moderl

2009). This comparison of VCSs and the RWCS shows that

the VCSs are designed in similar way as the RWCS. Only

the distribution of DWF per junction is different as in the

RWCS there is no DWF at each node. Nevertheless, the

impact of this difference is expected to be insignificant.

Rainfall input for the VCSs as well as for the RWCS are

design storm events (Euler II) with different return periods

(RP) (RP ¼ 0.5; 1; 2; 3; 5 and 10 yr) evaluated from real

rainfall data. RP ¼ 10 yr. was chosen as upper limit

according to guiding rules (OWAV-RB 11 2009) as this is

maximum return period required for the proof of sufficient

hydraulic sewer capacity. The rain gauge used is located in

Innsbruck, Austria and the evaluation is described in the

description of the Austrian rainfall database (Rauch &

Kinzel 2007). Innsbruck is a city located in Tyrol, Austria.

The climate is alpine, so the region is characterised by cold

winters and summers with intense rainfall. The duration of

the design storm events is assumed with 120 minutes and

the time step used is 5 minutes. Figure 3 shows diagrams

of the design storm events exemplified for RP ¼ 1 yr and

RP ¼ 10 yr. The Euler II design storms are as well as the

temporal resolution and the length of the rainfall event

chosen according to guiding rules of Austria (OWAV-RB 11

2009) and Germany (DWA-A 118E 2006). Furthermore Lei

(1996) and Rauch et al. (1998) show that 5 minute timesteps

are sufficient for simulation of urban drainage systems.

Applications and a further description of the Euler II design

storm is also available from De Toffol (2009).

Performance indicators (PI)

Usually sewer system performance is expressed by means

of performance indicators PI which should provide key

information needed to assess the efficiency of the system.

A PI is always a qualitative index of a particular aspect.

(De Toffol 2009). In this study we evaluated 7 different

performance indicators according to Moderl (2009) for

combined sewer system behaviour to analyse different

aspects of system performance. Four of them are evaluated

in detail and presented below. Performance indicators PI1,

PI2 and PI3 are emission based and represent CSO

discharge to the receiving water. PI4 represents flooding

in the system during extreme events. All performance

indicators used range between 0 (total system failure) and

1 (perfect system behaviour). While legal and societal

requirements with respect to the performance indicators

have changed in Europe during the last years for PI1, PI2

and PI3 with introduction of the EU Water Framework

Table 1 | System characteristics of real world case study

Parameter Real world case study

Subcatchments 200

Junction nodes 246

Outfall nodes 34

Storage volume 5,100m3

Total area 2,500ha

Impervous area 774ha

Average fraction imperviousness 0.31

Inflow to wastewater treatment plant 2.2m3/s

Figure 1 | Virtual case study (left) and real world case study (right) represented by the Software SWMM 5.0.

1557 M. Kleidorfer et al. | Climate change effects on combined sewer system performance Water Science & Technology—WST | 60.6 | 2009

Directive (2000/60/EC 2000), requirements for PI4 expres-

sing a degree of safety regarding flood protection (CEN

1996, 1997) have been almost constant over time. Further

discussion regarding that topic is available for example

from De Toffol (2009) or Engelhard (2008).

Hydraulic CSO efficiency: PI1

The CSO efficiency is used in Austrian guidelines to

evaluate a combined sewer system’s performance over a

simulation period of at least 10 years with recorded rainfall

data. The indicator represents the percentage of surface

runoff which reaches the waste water treatment plant as an

average over the simulation period. Here Euler II design

storms (single events) are simulated to reduce computing

time and so PI1 does not concur with the methodology from

the guidelines. However it can be used to compare

behaviour of systems due to different boundary conditions.

Nevertheless for future studies a comparison with sewer

system performance using long-time rainfall series is

recommended. PI1 is calculated from total overflow volume

of the entire system (VCSO) and total surface runoff

generated (VR) after

PI1 ¼ 12VCSO

VR½0j1�:

Pollutant CSO efficiency: PI2 and PI3

The CSO efficiency for pollutants is a modification of PI1.

Therein the CSO discharge of the pollutant mass (MCSO) is

calculated and normalized by the sum of the pollutant mass

generated by surface runoff (MR) and by dry weather flow

(MDWF). PI2 is calculated for total suspended solids (TSS)

and PI3 for total nitrogen (TN) after

PI2;3 ¼ 12MCSO

MR þMDWF½0j1�:

The two water quality parameters, total suspended

solids (TSS) and total nitrogen (TN) were selected for

Figure 2 | CDFs of VCSs compared to RWCS.

Figure 3 | Design storm event Euler II for RP ¼ 1 yr (left) and RP ¼ 10 yr (right).

1558 M. Kleidorfer et al. | Climate change effects on combined sewer system performance Water Science & Technology—WST | 60.6 | 2009

the analysis since they exhibit very different behaviour; TSS

is particulate, while TN is mainly dissolved (Taylor et al.

2005). Although estimation of pollutant concentrations

contains substantial uncertainties (see e.g. Dotto et al.

2008 or Freni et al. 2008 or Kleidorfer et al. 2009 (in press))

the performance indicators can be used to compare

performances of different systems.

Flooding efficiency: PI4

When estimating urban floods due to sewer system over-

loads usually a highly spatial distributed evaluation is

required to identify possible weak points in the system.

Using 1D models often the flooded volume per manhole is

assessed. Here we use a system-wide performance indicator

which also follows the idea of calculating a normalized

efficiency in order to compare system behaviour. Hence,

PI4 is calculated from total ponded volume (VP) and surface

runoff (VR) after

PI4 ¼ 12VP

VR½0j1�:

Environmental change scenarios/data uncertainties

Possible scenarios for future developments of urban

drainage systems cannot be generalised due to completely

different conditions in different regions. For this study the

impact on three main parameters in urban drainage

modelling is considered: dry weather flow (DWF), paved

area (AP) and rainfall intensities (I). The parameter values

for the predictions are taken from literature. As a bandwidth

of possible future scenarios is analysed, a realistic set

of values rather than a perfect prediction is sufficient.

The parameter variation can also be interpreted as

uncertainties in data collection. Table 2 shows the impact

of possible future scenarios and data uncertainties on model

parameters.

Climate change

The change in precipitation is not calculated by means of

climate change models but rather straightforward as given

change of the rain intensity (in variable proportions).

Arnbjerg-Nielsen (2008) calculated climate factors for

consideration of climate change in design of urban drainage

systems for Denmark. Therein three different approaches

are used. He estimates the increase in design intensities as

climate change factor (CCF) by 10–50% depending on

duration, return period and anticipated technical lifetime

of sewer systems. Although this CCF regionally varies

(De Toffol et al. 2006) a bandwidth of increase in design

intensities due to climate change of 10–50% is assumed.

Consequently for each 5min timestep j the rainfall intensity

under consideration of climate change conditions ICCF,j is

calculated from original rainfall intensity Ij and the climate

chance factor CCF after

ICCF;j ¼ Ij·CCF:

Land use change and population change

Land use change due to urbanisation and expansion of

cities has an impact on pavement of urban areas which can

lead to an increase of the fraction imperviousness. But

on the other hand measures for on-site infiltration of

stormwater reduce surface runoff. Dry weather runoff is

influenced by a possible change in population and by a

Table 2 | Environmental change effects and data uncertainties in urban drainage models

Parameter Dry weather flow Paved area Rainfall intensity

Environmental change Change in population Urbanisation/change in land use Climate-change

Change in water consumption

Data uncertainties Measurement uncertainties Determination of fraction of area Spatial distribution

imperviousness Resolution

Measurement uncertainties

1559 M. Kleidorfer et al. | Climate change effects on combined sewer system performance Water Science & Technology—WST | 60.6 | 2009

change in water consumption. Hence both parameters—

effective impervious fraction (EIF) and dry weather flow

(DWF)—can increase or decrease in future development. In

this study a bandwidth from 2 60% to þ 60% for both

parameters is assumed. Consequently the parameter for

possible future conditions (effective impervious area EIFfA

or respectively dry weather flow DWFfDWF) is calculated

from current conditions and an area factor fA or a dry

weather flow factor fDWF after

EIFfA ¼ EIF·fA and DWFfDWF¼ DWF·fDWF:

For example a þ 60% increase of impervious area for

the RWCS leads to an increase of the effective impervious

fraction from 0.31 to 0.5.

Simulation and analysis

One real world case study and 250 virtual case studies

are simulated with design storm events (Euler II) of

different return periods. Deviation of impervious area

and dry weather flow is considered in steps

of 2 60%, 2 30%, ^ 0%, þ 30% and þ 60%. Additionally

climate change factors of 1.1, 1.2, 1.3, 1.4 and 1.5 are

analysed.

All together more than 30,000 simulation runs over a

24 hour period are evaluated.

RESULTS AND DISCUSSION

Correlation of performance indicators

Figure 4 shows distributions and correlations of the

performance indicators analysed for the return periods

(RP) 1 (year) and 10 (years) (RP ¼ 1 yr and RP ¼ 10 yr).

Here the dot plots illustrate the correlation between the 4

performance indicators analysed. For example in the plot in

the first column and the fourth row the correlation between

PI1 and PI4 is presented, whereas PI1 is plotted at the

abscissa and PI4 is plotted at the ordinate of the subplot.

The diagonal plots show the distributions of the perform-

ance indicators as histograms. The performance indicators

in all scatter plots range from 0 to 1.

The emission based performance indicators PI1, PI2

and PI3 are clearly correlated, for example high values

for PI1 come along with high values for PI2. On the

other side PI4— which represents flooding—decreases

with increasing emission based performance indicators.

This corresponds with findings by Butler et al. (2008) who

analysed the relationship between flood volume and

receiving water quality in an integrated urban wastewater

system. Increased conduit diameters improve the system

capacity leading to a reduction of PI4. On the other hand

more combined flow can be conveyed downstream

resulting in an increase of CSO discharge (i.e. a decrease

of PI1, PI2 and PI3).

Figure 4 | Distribution of performance indicators for return period 1 yr (left) and 10 yr (right).

1560 M. Kleidorfer et al. | Climate change effects on combined sewer system performance Water Science & Technology—WST | 60.6 | 2009

The comparison of different return periods in Figure 4

shows that an increase of the rainfall intensities leads to a

shift of the parameter distributions from high values to

lower values. This effect is highest for PI4. One can clearly

see that for RP ¼ 1 yr PI4 is 1 (or close to 1) for most virtual

case studies, which indicates that no flooding occurs. For

RP ¼ 10 yr the distribution of PI4 changes significantly

indicating that more flooding occurs.

Due to the good correlation between PI1 and both

pollutant performance indicators PI2 and PI3 we assume

that that emission based performance indicators behave

equally. Consequently for a clear arrangement of the

following analysis only PI1 and PI4 are presented.

Figure 5 presents box plots of PI1 and PI4 in order to

illustrate the variance of PI1 and PI4 of the different case

studies and to demonstrate how performance indicators

change with return periods. On each box, the central mark

is the median, the edges of the box are the 25th and 75th

percentiles and outliers of that range are plotted

individually.

Here again one can clearly see how PI1 and PI4 as

expected decrease with increasing rainfall intensities.

Additionally one can see that the evaluation of the

performance indicators of the VCSs cover a wide range of

possible simulation results.

Impact of environmental change scenarios

For a better estimation of impact of different environmental

change scenarios the effects are evaluated pairwise. Per-

formance indicators are (nonlinear) functions of model

structure Q (including all aspects of the model with respect

to the case study analysed as e.g. system layout, DWF,

catchment area …), rainfall event R, CCF, fA and fDWF. For

a specific case study i and a specific rainfall event n PI can

be written in generalised way as a function of CCF, fA and

fDWF for given Q and R

PI ¼ fðCCF; fA; fDEFjQi;RnÞ:

Consequently by equalising sewer system performance

respectively with two factors kept constant, one factor can

be expressed as function of another one. Hence impact of

specific environmental change effects can easily be com-

pared. For example by equalising

fðfA;CCF ¼ 1; fDWF ¼ 1Þ ¼ fðCCF; fA ¼ 1; fDWF ¼ 1Þ

the area factor fA can be expressed as a function of CCF

fA ¼ fðCCFÞ

and finally it is possible to evaluate which factors lead to the

same deviation of performance indicators. This evaluation

is done by means of linear interpolation between calculated

points.

Figure 6 presents the impact of CCF compared to the

impact of fA for the VCSs (median and boxplots) and for the

RWCS for PI1 on the left hand side and for PI4 on the right

hand side. Here is important to note that the figures don’t

present absolute values of PI, but PI is only used to

compare the impact of CCF with respect to the impact of fA.

The figure shows the evaluation for RP ¼ 5 yr but other

return periods behave very similar. Results show for

example that a CCF of 1.2 has the same effect on system

performance as an increase in impervious area of

about þ 40% when regarding PI1. This increase of rainfall

intensities could be compensated by a reduction of

impervious area (e.g. by infiltration measures) by about

30%. We expect that the effect that a reduction of

impervious area can have higher impact on system

performance than an increase of impervious area is true

due to nonlinear relations between system characteristics

and performance indicators (e.g. mobilisation of storage

volume due to backwater effects).

VCSs and the RWCS behave similar. Only for high

fAs . 1.4 the impact of an increase of paved area is

higher for the RWCS compared to VCSs when regarding

PI1. This indicates that impact of possible environmentalFigure 5 | Box plot of PI1 (left) and PI4 (right) for different return periods.

1561 M. Kleidorfer et al. | Climate change effects on combined sewer system performance Water Science & Technology—WST | 60.6 | 2009

change scenarios (i.e. change of performance indicators

compared to current conditions) is rather similar for

different systems although absolute value of PIs vary as

shown in Figures 4 and 5.

Impact of DWF is not presented here as analysis of

simulation results showed that a change of DWF has

marginal effects on PI1 and PI4. Furthermore a CCF

cannot be compensated by a reduction of DWF in a

reasonable range.

CONCLUSION

In this study the possible impact of climate change induced

increase of rainfall intensities on combined sewer system

performance is analysed in a case independent way by using

simulation results of 250 virtual case studies and one real

world case study. In addition the impact of possible change

in DWF (e.g. increase by population growth or decrease

water saving measures) and possible change in impervious

area (e.g. increase by further pavement in urban areas or

decrease by a boost of infiltration measures) is taken into

account.

The different performance indicators analysed represent

different system behaviours (e.g. PI1 = CSO discharge;

PI4 = flooding) but show a very similar pattern. The impact

of an increase of rainfall intensities has the highest impact

on system performance followed by the impact of a

variation in impervious area. Variation of DWF marginally

changes system behaviour especially when only minor

changes in future development are realistic.

VCSs and the RWCS behave very similar. This indicates

that impact of possible environmental change scenarios

(i.e. change of performance indicators compared to current

conditions) is rather similar for different systems although

absolute value of PIs vary as shown in Figures 4 and 5.

For example the consideration of a climate change

factor of 1.2 has the same effect as an increase of impervious

area of þ 40%. Such an increase of rainfall intensities by

1.2 could be compensated by infiltration measures in

current systems which lead to a reduction of impervious

area by 30%. Certainly this different behaviour of area

reduction and area increase needs further examination in

upcoming studies.

Concluding, in this paper some general coherences of

sewer system behaviour under future development scen-

arios and climate change scenarios are presented. This is a

contribution to enhanced system understanding taking into

account long-term environmental change effects.

As this study is limited by some (inevitable) short-

comings of the VCSs further research should focus on more

real world case studies with different regional character-

istics and on more realistic VCS. For example Sitzenfrei

et al. (2009) and Urich et al. (2009) develop a agent based

Figure 6 | Comparison of climate change factor and area factor.

1562 M. Kleidorfer et al. | Climate change effects on combined sewer system performance Water Science & Technology—WST | 60.6 | 2009

software for generating virtual case studies based on cellular

automata which is expected to represent interactions of

different infrastructure systems in a more realistic way.

Additionally using real rainfall time series instead of design

storm events may lead to an improved analysis especially for

estimating impact on performance indicators related to

CSO discharge.

REFERENCES

2000/60/EC 2000 Directive of the European Parliament and of the

Council establishing a framework for Community action in the

field of water policy.

Arnbjerg-Nielsen 2008 Quantification of climate change impacts on

extreme precipitation used for design of sewer systems. In:

Proceedings of the 11th International Conference on Urban

Drainage, Edinburgh, Scotland, UK, 2008.

Arnbjerg-Nielsen, K. & Fleischer, H. S. 2009 Feasible adaptation

strategies for increased risk of flooding in cities due to climate

change. Water Sci. Technol. 60(2), 273–281.

Ashley, R. M., Balmforth, A. J., Saul, A. J. & Blanskby, J. D.

2005 Flooding in the future—predicting climate change, risks

and responses in urban areas. Water Sci. Technol. 52(5),

265–273.

Berggren, K. 2008 Indicators for urban drainage system—assessment

of climate change impacts. In: Proceedings of the 11th

International Conference on Urban Drainage, Edinburgh,

Scotland, UK.

Butler, D., McEntee, B., Onof, C. & Hagger, A. 2007 Impact of

climate change on sewer storage tank performance. Water Sci.

Technol. 56(12), 29–35.

Butler, D., Fu, G. & Khu, S.-T. 2008 The relationship between

sewer flood volume and receiving water quality in an

integrated urban wastewater system. In: Proceedings of the

BHS 2008, Exeter, UK.

CEN 1996 EN 752—Drain and sewer systems outside buildings—

Part 2: Performance Requirements. European Commitee for

Standardization (CEN), Brussels, Belgium.

CEN 1997 EN 752—Drain and sewer systems outside buildings—

Part 4: Hydraulic design and environmental considerations.

European Commitee for Standardization (CEN), Brussels,

Belgium.

Chau, K. W. 2006 Particle swarm optimization training algorithm

for ANNs in stage prediction of Shing Mun River. J. Hydrol.

329(3–4), 363–367.

Cheng, C. T., Ou, C. P. & Chau, K. W. 2002 Combining a fuzzy

optimal model with a genetic algorithm to solve multi-

objective rainfall–runoff model calibration. J. Hydrol. 268,

72–86.

De Toffol, S. 2009 Sewer system performance assessment—an

indicator based methodology (PhD Thesis). Innsbruck

University Press, Innsbruck.

De Toffol, S., Burstrom, Y. D. S. & Rauch, W. 2006 On the

effect of spatial variances in historical rainfall time series to

CSO performance evaluation. Water Sci. Technol. 54(6–7),

25–31.

Dotto, C., Deletic, A. & Fletcher, T. D. 2008 Analysis of uncertainty

in flow and water quality from a stormwater model. In:

Proceedings of the 11th International Conference on Urban

Drainage, Edinburgh, Scotland, 2008.

DWA-A 118E 2006 Hydraulic Dimensioning and Verification of

Drain and Sewer Systems. Deutsche Vereinigung fur

Wasserwirtschaft, Abwasser und Abfall e.V. (DWA).

Engelhard, C. 2008 Assessing the Impact of Urban Draining

Measures with Regard to the Water Framework Directive (PhD

Thesis). Innsbruck University Press, Innsbruck.

Freni, G., Mannina, G. & Viviani, G. 2008 Uncertainty in urban

stormwater quality modelling: The effect of acceptability

threshold in the GLUE methodology. Water Res. 42(8–9),

2061–2072.

Kleidorfer, M., Moderl, M., Fach, S. & Rauch, W. 2009

Optimization of measurement campaigns for calibration of a

conceptual sewer model. Water Sci. Technol. 59(8),

1523–1530.

Kleidorfer, M., Deletic, A., Fletcher, T. D. & Rauch, W. 2009

Impact of input data uncertainties on urban stormwater model

parameters. Water Sci. Technol. 60(6), 1545–1554.

Lei, J. H. 1996 Uncertainty Analysis of Urban Rainfall–Runoff

Modelling. Norwegian University of Science and Technology,

Department of Hydraulic and Environmental Engineering,

Dissertation.

Mark, O., Svensson, G., Konig, A. & Linde, J. J. 2008 Analyses and

Adaptation of Climate Change Impacts on Urban Drainage

Systems. In: Proceedings of the 11th International Conference

on Urban Drainage, Edinburgh, Scotland, UK, 2008.

Moderl, M. 2009 Modelltechnische Analyse von Netzwerksystemen

der Siedlungswasserwirtschaft. University of Innsbruck, Unit of

Environmental Engineering, Dissertation.

Moderl, M., Butler, D. & Rauch, W. 2009 A stochastic approach for

automatic generation of urban drainage systems. Water Sci.

Technol. 59(6), 1137–1143.

OWAV-RB 11 2009 Richtlinie fur die abwassertechnische

Berechnung und Dimensionierung von Abwasserkanalen; 2.

vollstandig uberarbeitete Auflage. Osterreichisches

Normungsinstitut, Wien.

Rauch, W. & Kinzel, H. 2007 OKOSTRA—Eine Datenbank zur

Ausgabe von hoch aufgelasten Niederschlagsserien aus

Osterreich. Wiener Mitteilungen, pp. I1–I14.

Rauch, W., Thurner, N. & Harremoes, P. 1998 Required accuracy of

rainfall data for integrated urban drainage modeling. Water

Sci. Technol. 37(11), 81–89.

Rossman, L. A. 2008 Storm Water Management Model—User’s

Manual Version 5.0. National Risk Management Research

Laboratory—U.S. Environmental Protection Agency.

Rowell, D. P. 2005 A scenario of European climate change for the

late 21st century: seasonal means and interannual variability.

Clim. Dyn. 25, 837–843.

1563 M. Kleidorfer et al. | Climate change effects on combined sewer system performance Water Science & Technology—WST | 60.6 | 2009

Semadeni-Davies, A., Hernebring, C., Svensson, G. & Gustafsson,

L.-G. 2008 The impact of climate change and urbanisation on

drainage in Helsingborg, Sweden: combined sewer system.

J. Hydrol. 350(1–2), 100–113.

Sitzenfrei, R., Fach, S., Kinzel, H. & Rauch, W. 2009 A multi-layer

cellular automata approach for algorithmic generation of

virtual urban water systems—VIBe. In: Proceedings of the 8th

International Conference on Urban Drainage Modelling, 7–11

Sept. 2009.

Taylor, G. D., Fletcher, T. D., Wong, T. H. F. & Breen, P. F. 2005

Nitrogen composition in urban runoff—implications

for stormwater management. Water Res. 39(10),

1982–1989.

Urich, C., Sitzenfrei, R., Moderl, M. & Rauch, W. 2009 An agent

based approach for generating virtual sewer systems in the

software VIBe. In: Proceedings of the 8th International

Conference on Urban Drainage Modelling, 7–11

Sept. 2009.

1564 M. Kleidorfer et al. | Climate change effects on combined sewer system performance Water Science & Technology—WST | 60.6 | 2009