Content uploaded by Violeta A. Montoya-Coronado
Author content
All content in this area was uploaded by Violeta A. Montoya-Coronado on Jun 24, 2024
Content may be subject to copyright.
Journal of Environmental Management 362 (2024) 121073
Available online 4 June 2024
0301-4797/© 2024 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/by-
nc/4.0/).
Review
Combined sewer overow mitigation through SUDS - A review on
modelling practices, scenario elaboration, and related performances
V.A. Montoya-Coronado
a
,
*
, D. Tedoldi
a
, E. Lenormand
a
, H. Castebrunet
c
, P. Molle
b
, G. Lipeme
Kouyi
a
a
INSA Lyon, DEEP, UR7429, 69621, Villeurbanne, France
b
INRAE, Research Unit REVERSAAL, Lyon, France
c
Universit´
e Grenoble Alpes, IGE, F-38000, Grenoble, France
ARTICLE INFO
Keywords:
Hydraulic disconnection
CSO
Large scale modelling
SUDS modelling
Urban drainage modelling
Urban planning
ABSTRACT
Hydrologic-hydraulic modelling of urban catchment is an asset for land managers to simulate Sustainable Urban
Drainage Systems (SUDS) implementation to full combined sewer overow (CSO) regulations. This review aims
to assess the current practices in modelling SUDS scenarios at large scale for CSO mitigation encompassing every
stage of the modelling process from the choice of the equation to the validation of the initial state of the urban
system, right through to the elaboration, modelling, and selection of SUDS scenarios to evaluate their perfor-
mance on CSO. Through a quantitative and qualitative analysis of 50 published studies, we found a diversity of
choices when modelling the status quo of the urban system. Authors generally do not explain the modelling
processes of slow components (deep inltration, groundwater inltration) and interconnexion between SUDS
and the sewer system. In addition, only a few authors explain how CSO structures are modelled. Furthermore, the
modelling of SUDS implementation at catchment scale is highlighted in the 50 studies retrieved with three
different approaches going from simplied to detailed. SUDS modelling choices seem to be consistent with the
objectives: studies focusing on dealing with several objectives at the time typically opt for a complex system
conguration that includes the surface processes, network, CSO, SUDS, and often the soil and/or groundwater
components. Conversely, authors who have selected a basic conguration generally aim to address a single,
straightforward question (e.g., which type of SUDS). However, elaboration and selection of scenarios for CSO
mitigation is mainly based on local constraints, which does not allow hydrological performance to be directly
optimised. In conclusion, to improve current practices in modelling SUDS scenarios at large scale for CSO
mitigation, authors suggest to: (i) improve clear practices of CSO modelling, calibration and validation at the
urban catchment scale, (ii) develop methods to optimize the performance of scenarios for CSO mitigation using
hydrological drivers, and (iii) improve parsimonious and user-friendly models to simulate SUDS scenarios in a
context of data scarcity.
1. Introduction
Surface water resources are increasingly under threat. An important
factor contributing to its degradation is Combined Sewer Overows
(CSO) (Bertels et al., 2023; Even et al., 2007). CSO occur when untreated
wastewater and stormwater are directly discharged into water bodies,
resulting from the sewer systems’ inability to handle large ows. These
overows introduce a surge of contaminants into the receiving waters,
including organic matter and nutrients (Seidl et al., 1998; Even et al.,
2007), microbial pathogens (Al Aukidy and Verlicchi, 2017),
pharmaceuticals (Ellis, 2006; Kay et al., 2017), biocides (Paijens et al.,
2021), microplastics (Di Nunno et al., 2021) and heavy metals (Xu et al.,
2018) in receiving waters. Such contamination poses signicant risks to
aquatic ecosystems and public health. As urban areas expand, the
traditional centralized sewerage system has proved ineffective due to
limited capacity. Moreover, the underground nature of these conven-
tional sewerage systems presents multiple challenges to address, espe-
cially with infrastructure ageing. These challenges include pipe
deterioration, sediment accumulation and ow rate increase issues, and
CSO discharge (Bertels et al., 2023; Chocat et al., 2001; Fenner, 2000).
* Corresponding author.
E-mail address: violeta-alexandra.montoya-coronado@insa-lyon.fr (V.A. Montoya-Coronado).
Contents lists available at ScienceDirect
Journal of Environmental Management
journal homepage: www.elsevier.com/locate/jenvman
https://doi.org/10.1016/j.jenvman.2024.121073
Received 31 January 2024; Received in revised form 1 April 2024; Accepted 30 April 2024
Journal of Environmental Management 362 (2024) 121073
2
Furthermore, the increased frequency of extreme weather events asso-
ciated with climate change poses a signicant concern, potentially
leading to increased CSO volumes during certain seasons and in specic
regions (Bonneau et al., 2023; Gogien et al., 2023; Zhou, 2014).
Considering these challenges, stormwater management has wit-
nessed a shift of paradigm towards more sustainable solutions. Sus-
tainable Urban Drainage Systems (SUDS) are increasingly being
implemented in urban areas. Field studies have shown that SUDS
effectively reduce peak ows, decrease runoff volume, and attenuate
pollutant loads towards surface waters (Al-Rubaei et al., 2017; Golden
and Hoghooghi, 2018; Ventura et al., 2021; Walaszek et al., 2018).
Consequently, SUDS help to mitigate the adverse impacts of traditional
sewer systems. As Europeans cities are under pressure to reduce CSO
(Botturi et al., 2021), researchers and policy-makers develop and/or use
urban drainage models to evaluate the potential of SUDS to address this
challenge at large scale (Bertels et al., 2023; Golden and Hoghooghi,
2018). Those are used to test and analyze a variety of scenarios, for
which the modeler tries to identify the "optimal" one in terms of per-
formance (according to one or more criteria). The optimal scenario
typically refers to the conguration of SUDS measures and their spatial
distribution within the catchment that maximizes CSO reduction while
considering additional criteria such as cost-effectiveness, ood mitiga-
tion, pollutant removal and environmental sustainability.
In the last 20 years, several review articles have emerged, aiming to
summarize the current state of SUDS modelling tools and approaches.
Elliott and Trowsdale (2007) proposed a review comparing ten model-
ling tools used to simulate SUDS. Some years later, Ahiablame et al.
(2012) reviewed the SUDS representation and focused on two modelling
approaches: a process-based modelling (inltration, sedimentation) and
a practice-based modelling (aggregation method). Jayasooriya and Ng
(2014) summarized the modelling tools data requirement, and sug-
gested the need to achieve best practices such as introducing transparent
modelling procedures. Furthermore, Kaykhosravi et al. (2018)
completed the last reviews by comparing and evaluating the SUDS
modelling tool approaches, giving more specications and gaps about
hydrological-hydraulic modelling methods and how SUDS compartment
can be integrated in the urban catchment to facilitate SUDS scenario
elaboration. Despite the increasing trend about SUDS modelling, Elliott
and Trowsdale (2007), Ahiablame et al. (2012), Jayasooriya and Ng
(2014) and Kaykhosravi et al. (2018) identied difculties on upscaling
modelling SUDS approaches at large scale and the optimization dif-
culty due to the multiple possible congurations.
In response to the SUDS conguration complexity problem, a set of
review articles focused on decision support tools to facilitate SUDS
scenario elaboration and selection. Those decision support tools mainly
consist on an urban drainage model (to simulate the production, losses
and transport of urban water within urban systems) coupled with an
optimization method (to adjust the system’s conguration iteratively
until the performance indicators reach a level considered optimal or
satisfactory) (Mikovits et al., 2017). These decision support tools are
developed to identify the optimal congurations of SUDS scenarios for
the specic and local objectives, with the aim of systematizing the
decision-making criteria and reducing the expert judgment. Lerer et al.
(2015), Torres et al. (2016) and Ferrans et al. (2022) classied three
types of decision support tools according to the three variables of de-
cision that constitute a scenario, including the type of SUDS, its place-
ment and design. Zhang and Chui (2018) reviewed the decision support
tools for spatial allocation. In addition, a focus was made on good
practices for developing reliable performance scenarios. More specif-
ically, Jayasooriya et al. (2020) reviewed the challenges and opportu-
nities for SUDS practices in industrial areas. The review highlighted the
importance of a systematic methodology for an optimum application of
SUDS to manage stormwater. Interestingly, a common thread among
these studies is the observed lack of generic use of these tools. Many
were developed using a specic case study, leading to numerous existing
tools, but with a marked limitation in their repeatability.
A recent work from Muttil et al. (2023) focused on SUDS capability
for CSO mitigation. The review highlighted that despite the increasing
number of case studies implementing SUDS strategies for CSO mitiga-
tion the past years, there are still few articles addressing the subject. This
might be due to the complexity and challenges related to modelling at
large scale. In addition, Muttil et al. (2023) pointed out that in existing
reviews about urban drainage models and decision support tools for
SUDS widespread planning, none address the modelling practices
throughout the whole process from the urban drainage system concep-
tualization to the evaluation of scenario planning performances.
Existing reviews on SUDS modelling tools and approaches for CSO
reduction mainly have three objectives: i) summarize the current state of
SUDS modelling tools and approaches; ii) expose decision support tools
to facilitate SUDS scenarios elaboration and SUDS scenarios selection for
CSO mitigation; and iii) summarize the present gaps on the process of
modelling SUDS scenarios for CSO mitigation.
The present work aims to go further by proposing a comprehensive
review encompassing every stage of the modelling process (Fig. 1). This
includes the construction, calibration and validation of the baseline
conguration of the urban system, as well as the elaboration, modelling,
and selection of SUDS scenarios to evaluate their performance on CSO
reduction.
The objective of the present review is to provide a comprehensive
analysis of the modelling approaches used at the urban catchment scale.
This includes:
1. Drawing up an inventory of the different temporal and spatial scales
used.
2. Identifying the various compartments and modelling approaches
considered, while also addressing any shortcomings in the concep-
tualization of the urban catchment system.
3. Identifying the different methods used to elaborate and implement
SUDS scenarios at the urban catchment scale.
4. Highlighting the performances of these methods in terms of CSO
mitigation at the catchment scale.
2. Methodology: selection of the sample of studies
A systematic search for relevant studies was made using a combi-
nation of four groups of keywords (Table 1: CSO keywords AND network
keywords AND disconnection keywords AND modelling keywords) in 3
databases (Science Direct, Scopus, and Web of Science). 1065 studies
were collected, however, only the papers meeting the following criteria
have been selected for the present review: (i) at least one of the article
objectives must be linked to modelling CSO mitigation strategies, (ii) the
scale of the study area must be an urban catchment, and (iii) papers
pointing out the simulation of the effect of SUDS distribution through
the urban catchment.
Then, the list was completed by additional papers identied through
the reference list and citations of read ones. A sample of 50 recent
studies over 1999–2023 implementing SUDS practices for CSO mitiga-
tion at urban scale was used for this overview. The sample of studies
were published before August 2023, which was the end date of the re-
view process. Percentages are used as units, with each 2% representing
one article.
3. General observations about the corpus
3.1. Geographical origin of case studies
North America and Europe, account for a substantial fraction (39%
and 41%, respectively) of the 50 retreived papers. This dominance re-
ects the strong research infrastructures and resources in these regions.
On the other hand, Asia contributes up to 16% of the retrieved papers.
Latin America and Caribbean, represents only 4% of the research while
no paper from Africa nor Oceania was present in the corpus.
V.A. Montoya-Coronado et al.
Journal of Environmental Management 362 (2024) 121073
3
3.2. Study drivers and CSO evaluation criteria
The 50 studies are included in the scope of decentralized urban water
management practices. The main objective of all the studies is to reduce
spill volumes through CSO. Besides, 18% of the papers have also
considered pollutants conveyed through CSO and impact on the
receiving body (e.g., Fan et al., 2022; Riechel et al., 2016; Sarminingsih
et al., 2019). Almost the same quantity (20%) of the studies have
considered the question of CSO mitigation in a climate change context
(e.g., Hou et al., 2020; Roseboro et al., 2021; Tavakol-Davani et al.,
2016). Then, 16% considered the quantication and comparison of the
construction cost to the CSO mitigation efforts (e.g., Joshi et al., 2021;
Montalto et al., 2007; Radinja et al., 2019). Furthermore, considering
the multiple co-benets of SUDS, 12% of the studies have also consid-
ered the assessment of ood risk (e.g., D’Ambrosio et al., 2022; Stovin
et al., 2013; Villarreal et al., 2004) and 4% considered the non-point
source pollution (e.g., Hou et al., 2020, 2021).
3.3. Modelling tools
The distribution of modeling software across the ensemble of studies
provides valuable insights into prevailing trends. These software choices
underscore the diverse strategies employed by researchers to address the
complexity of the studied systems. Particularly noteworthy among these
trends is the substantial 48% adoption rate of the Storm Water Man-
agement Model (SWMM). In contrast, 16% of the studies opt for Info-
works. Another 6% of the studies choose to utilize Mike Urban, while 4%
turn to the L-THIA LID model and 4% use coupled models (such as
SWMM and MODFLOW) to enhance their understanding of coupled
groundwater and surface water interactions. Two smaller subsets,
comprising 2% each, employ City Drain and MOUSE. On the other hand,
18% of the studies develop and apply their own customized models. This
diverse utilization of softwares reects researchers’ adaptability to
emerging questions, tailoring tools to address the unique challenges
presented by urban drainage systems.
4. Spatio-temporal model resolution
4.1. Spatial discretization
This section focuses on reviewing the spatial discretization
commonly considered for CSO mitigation objectives through SUDS
planning management.
4.1.1. How are catchments and sub-catchments delimited?
Catchments and sub-catchments can be delineated by different
criteria such as soil type, vegetation type, land use and topographical
features (Park et al., 2008). The choice of these criteria and delineation
method affects the degree of discretization which is important for
modelling (i.e., heterogeneity of land use will inuence the water vol-
ume going through the sewer network) and data analysis (i.e., deter-
mining suitable places for SUDS). Only 26% of the studies have
mentioned the methodology used to delineate the sub-catchment areas
using sewer system maps (Chen et al., 2019), digital terrain models
(Alves et al., 2016) and aerial photographs (Montalto et al., 2007).
Otherwise, Simperler et al. (2020) used block partitioning resulting from
the superposition of the land occupation types (e.g., residential, com-
mercial, mixed areas), buildings and population, resulting in a full dis-
cretization with an average size of sub-catchment smaller than 1 ha.
4.1.2. Which spatial discretization for urban catchment scale models?
All studies are based on semi-distributed or distributed models. Semi-
distributed and distributed models are dened for a catchment divided
Fig. 1. Workow of the present review.
Table 1
List of keywords used for bibliographic research.
Groups Keywords
CSO keywords “CSO mitigation” OR “outlet controls” OR “CSO impact
assessment” OR “combined sewer system”
Network keywords “CSS” OR “combined sewer system” OR “drainage systems”
OR “sustainable network” OR “sustainable urban drainage
system”
Disconnection
keywords
“decentraliz*” OR “degree of decentralization” OR
“disconnection opportunities” OR “different management
strategies” OR “low impact development” OR “LID” OR
“nature-based solutions” OR “NBS” OR “sustainable urban
drainage systems” OR “SUDS” OR “stormwater
disconnection” OR “sponge city” OR “blue-geen
infrastructure” OR “stormwater control measures” OR “water
Sensitive Urban design” OR “WSUD”
Modelling keywords “model*” OR “city scale” OR “continuous simulation” OR
“event-based model*” OR “hydraulic model*” OR
“hydrologic model*” OR “integrated model*” OR “urban
catchment” OR “urban watershed” OR “urban model*” OR
“stormwater” OR “sewer model*”
V.A. Montoya-Coronado et al.
Journal of Environmental Management 362 (2024) 121073
4
into sub-catchments having different properties (e.g., elevation, urban-
ization, soil characteristics). The size of the sub-catchment holds sig-
nicance from a modelling perspective as it describes the level of
description for the territory (Elliott and Trowsdale, 2007; El-Nasr et al.,
2005).
For the studies giving this information (n=41), the catchment size
was extracted, and the mean size of sub-catchment was calculated by
dividing the total catchment area by the number of sub-catchments. The
sample was grouped into four clusters using the Gaussian Mixture
Model, as shown in Fig. 2. These 4 groups of catchment size were
delimitated by small (44% of studies), medium-small (12% of studies),
medium-large (22% of studies), and large catchments (22% of studies),
with a respective size smaller than 500 ha, between 500 and 1000 ha,
between 1000 and 2400 ha and bigger than 2400 ha (Fig. 2).
A disparity was found in the size of the catchments with the largest
catchment being that of Torres et al. (2021) with 35000 ha and the
smallest described by Villarreal et al. (2004) with 4 ha. For small
catchments, the size of sub-catchments is smaller than 60 ha and almost
50% of the sub-catchments mean size are smaller than 2 ha. For
medium-small catchments, the mean size sub-catchments appears to be
divided into two groups, those with an average size of less than 5 ha and
those with a size of between 20 and 40 ha. For medium-large catch-
ments, sub-catchments are smaller than 20 ha. Finally, for large
sub-catchments the average size varies between 1 and 426 ha. Despite
this high variability, the size of the sub-catchments increases with the
size of the catchments.
Several studies have investigated the effect of simplifying the num-
ber of sub-catchments (Casey et al., 2015; Elliott and Trowsdale, 2007;
Ghosh and Hellweger, 2012). To the best of our knowledge, there is no
consensus on the discretization required to accurately model the
catchment hydrology. Therefore, decisions should be based on the level
of detail required for the hydrologic model, considering both the ob-
jectives of the study and the available data.
4.2. Rainfall data and temporal scale of simulations
4.2.1. Temporal scale of simulations
Temporal scale to simulate SUDS scenarios at catchment scale to
mitigate CSO can be gathered into two categories: i) continuous simu-
lations, and ii) event-based simulations. Continuous simulations, used
by 36% of authors (n=18), yield hydrological ndings without setting
the specication of initial conditions for each rain event (temperature,
soil moisture, and the lling state of the reservoir structures). 56% of
papers (n=28) opt for event-based simulations and a subset of authors,
8% (n=4), employ both approaches.
In continuous simulation studies (n=22), 10/22 of the authors
simulate a 1-year rainfall event, while the remaining half is divided,
with 5/22 simulating events of less than a year and 7/22 simulating
events lasting more than a year. On the other hand, it was observed that
among studies using future scenarios (n=10), a signicant majority, 8/
10 of studies employ continuous simulations spanning a duration of 12
months or longer. Continuous simulation is sometimes preferred as it
captures the time-varying responses of changing weather conditions and
runoff patterns, providing a more accurate representation of their
effectiveness. However, they may represent a signicant modelling
effort, and depending on the aim of the modelling study, the choice of
period may have not been representative of current events or extreme
events. Ferrans et al. (2022) showed that the proportion of continuous
simulation used in studies is increasing.
Event-based simulations (n=32) are divided into 3 categories. The
rst category consists in using design storm (i.e., ctitious rainfall event
with controlled characteristics such as hyetograph shape, maximum
mean intensity, return period). 15/32 authors using event-based simu-
lations, use a return period between 1 month and 100 years. The second
category consists in several observed rainfall events, constituting 13/32
of authors using event-based simulation. The observed rainfall events
may be simulated independently and often omit periods unrelated to the
research questions. In this context, some authors disregard dry intervals,
assuming that CSO are absent during such period. While Liao et al.
(2015) use of many rainfall events (53), half of studies use less than 7
rainfall events. Among them and not exhaustively, Sarminingsih et al.
(2019), Stovin et al. (2013) and Torres et al. (2021) have only used 3, 3
and 2 events, respectively. These authors justify their limited sample size
by specically choosing representative rainfall events based on char-
acteristics such as frequency, duration, and return time observed over an
extended period. The third category uses historical rainfall events and
has been used by 4/32 authors to investigate the trends in extreme
weather occurrence. Using several rainfall events with different char-
acteristics is recommended by Bertrand-Krajewski et al. (2008).
36% (n=18) of the authors have justied their temporal simulation
choices based on systematic data analysis. The authors’ justications
encompass four distinct categories: rstly, they consider rainfall char-
acteristics, such as rainfall depth, intensity, and return period. Secondly,
they account for anticipated future trends in rainfall patterns. Thirdly,
they evaluate the specic rainfall attributes unique to the geographical
study area, including the frequency and type of rainfall events typically
observed. Lastly, the authors assess the impact of the observed rainfall
events on CSO. The approach to temporal simulation selection reects a
comprehensive strategy by these authors to ensure the relevance of their
simulations.
4.2.2. Temporal discretization
Temporal discretization used to simulate SUDS scenarios at catch-
ment scale to mitigate CSO was mentioned by 60% of authors (n=30).
Three groups have been identied: i) authors who use a time step be-
tween 1 and 15 min (20/30), ii) an hourly time step (9/30), and iii) a
daily time scale (1/30). On the other hand, some authors use a variable
time step. For example, Tavakol-Davani et al. (2016) simulate three
years with different time step during dry weather than during rainfall
events. This seems reasonable because the dry weather processes are
slower. However, the higher the time step, the lower the peak ow and
the CSO volume. Hence, it cannot be expected to accurately quantify
CSO volume with large temporal scales (Zhu and Schilling, 1996).
5. How to model the urban catchment compartment and what is
lacking?
The conceptualization of the urban catchment and its compartments
Fig. 2. Distribution of Sub-Catchment mean sizes by catchment size group
(n=41). The rst group (44% of the studies) includes catchments smaller than
500 ha. The second group (12% of the studies) encompasses catchments
ranging between 500 and 1000 ha. The third group (22% of the studies) consists
of catchments sized between 1000 and 2400 ha. The nal group (22% of the
studies) comprises catchments larger than 2400 ha.
V.A. Montoya-Coronado et al.
Journal of Environmental Management 362 (2024) 121073
5
is a prerequisite for understanding the current state of the systems and
the consequences of SUDS implementation. Nevertheless, the internal
structure of the hydraulic-hydrological model is frequently not explicitly
mentioned. Aspects such as formulating crucial processes with equa-
tions, using numerical techniques, and establishing boundary conditions
are often overlooked (Deletic et al., 2012).
5.1. Considered urban catchment compartments
In urban areas, the hydrological compartments are the surface area,
soil, aquifer and river system. Hydraulic compartments are the networks
and the CSO. There is no simple denition of which compartment must
be integrated in an urban drainage model (Salvadore et al., 2015), but
from this literature review focusing on the objective of CSO mitigation
through SUDS, almost all the authors have at least represented the
surface, the network, the CSO and SUDS compartments (Fig. 3). Some
studies also incorporate additional compartments to enhance the
catchment representation. For instance, 54% (27/50) of the studies
include the soil compartment in their catchment models. A smaller
portion, 26% (13/50), introduce the aquifer compartment, and 8%
(4/50) consider the river system. The challenge arises in determining
which compartments need to be adequately represented to avoid
neglecting crucial processes, striking the right balance in the CSO urban
scale modelling effort. Six combinations have been observed, ranging
from the simple (using only the surface and SUDS compartment) to the
complex (surface, soil, groundwater, network, CSO, SUDS and the rivers
compartments). These combinations of compartments, which make up
the system, are highlighted in Fig. 3. Among them, three combinations
emerge as the most used. The rst one combines surface, network, CSO,
and SUDS, accounting for 42% of the reviewed studies. The second
recurring combination extends to the initial urban system by including
the soil compartment and constitutes 22% of the studies. The third
combination further incorporates the groundwater compartment, rep-
resenting an additional 24% of the studies in the review.
The selection of compartments within an urban catchment is closely
linked to the objectives of the articles, as detailed in Appendix 1. The
analysis of the ensemble of articles shows that for a singular objective (e.
g., volume mitigation), a basic conguration (e.g., surface, network,
Fig. 3. Overview of the compartments of the urban drainage system and usage trends for modelling disconnection strategies at the catchment scale. a) Presentation
of compartment types referenced throughout the ensemble of articles (n=50), along with their frequency of occurrence. b) Description of the various compartment
combinations found in the collection of articles (n=50) and their association with the employed models. c) Analysis of the different compartment combinations
identied in the articles (n=50) and their correlation with modelling objectives.
V.A. Montoya-Coronado et al.
Journal of Environmental Management 362 (2024) 121073
6
CSO, SUDS) appears to be sufcient. However, addressing dual objec-
tives (such as combining volume mitigation with cost reduction or ood
mitigation or pollutant reduction or to face climate change effects) can
be addressed through the same basic conguration. This observation
raises the question of whether the basic conguration can effectively
meet the objectives of modelling CSO in urban environment.
The relationship between the choice of system compartments was
correlated to the temporal scale of simulation (Appendix 2). It was
observed that a basic conguration is commonly employed in both
event-based and continuous temporal scales of simulation. A particu-
larly interesting observation is the tendency of studies incorporating the
soil compartment to favor event-based temporal simulations. This
preference could be attributed to the signicant impact of soil inltra-
tion and saturation (initial conditions) on the hydrograph at event scale.
Conversely, studies that include groundwater and/or river components
predominantly opt for continuous simulations. This approach seems
logical considering that groundwater and river dynamics are slower
processes, which are more effectively captured over extended time
scales.
5.2. Conceptual models for urban drainage systems
Models and their applications are developed to address specic
research and/or practical objectives guided by the objectives and
available data (Devia et al., 2015). To model CSO and the dissemination
of SUDS within the urban catchment, developers have to make two key
decisions: i) the hydrological and hydraulic key processes to simulate,
and ii) the appropriate level of complexity for the model.
5.2.1. Rainfall-runoff model
Hydrologic catchment routines are used to transform an effective
storm hyetograph into a runoff hydrograph. As illustrated in Fig. 4,
runoff can come from impervious but also from pervious surfaces. Three
major rainfall-runoff models were identied across 32% (16/50) of the
studies: (i) the rational method, utilized by 3/16 of the studies, calcu-
lates peak discharge based on the catchment area, runoff coefcient and
rainfall intensity. This equation is particularly appropriate for catch-
ments smaller than 250 ha. Beyond this size, its assumptions and
empirical constants may lead to inaccurate runoff estimations (Wang
and Wang, 2018). Montalto et al. (2007) and Rodriguez et al. (2008)
used it to calculate the peak runoff ow rate and quantify the CSO event
through a threshold equation (further details about CSO equations in
§Combined sewer overow models); (ii) the time area method, used by
4/16 of the studies, transforms an effective storm hyetograph into a
runoff hydrograph. The method accounts for translation only and does
not include storage. For example, Villarreal et al. (2004) used this
method to estimate runoff based on the available physical characteristics
of each sub-catchment; and (iii) the Curve Number runoff equation (SCS
CN), used by 7/16 of the studies, is an empirical method that expresses
the runoff volume generated by a given rainfall volume. This method is
based on daily runoff depth, land use, and hydrologic soil group data. It
is used in SWMM (Rossman and Simon, 2022). The fact that this method
is used twice as much as the previous two is presumably due to the
popularity of SWMM. It is worth emphasizing that 42% of authors using
SWMM do not explain their rainfall model, perhaps because it is
considered to be well-known.
5.2.2. Inltration and subsurface ow models
Pervious surfaces may contribute to CSO by surface runoff when they
are saturated and/or by slow inltration processes within the soil layer,
which subsequently seep into the buried sewer pipes via cracks, joints,
or defects (Fig. 4). Only 26% (13/50) of the authors detail the inltra-
tion and subsurface models they used. Three different models are set up:
i) the Horton’s equation (Horton, 1939) used by 6/13 authors; ii) the
Green-Ampt model (Green and Ampt, 1911) used by 6/13 authors as
well; and iii) the linear reservoir model employed by only one author
(Andres-Domenech et al., 2010). The other authors do not mention
consideration of this process.
Groundwater recharge and inow to the combined sewer system
(Fig. 4) were accounted by 24% (12/50) of the studies. These studies
consider the water’s ow from the surface to the aquifer. Half of them
took into consideration the ow from the surface into the aquifer and
then into the combined sewer system. Two modelling approaches were
identied: (i) the injection into the combined sewer system, adopted by
9/13 authors. The injection of ow can be a constant ow at different
points within the sewer system or variable monthly ow injection rep-
resenting seasonal variation. These ow paths, come for most of them
from Modow simulation (Harbaugh, 2005) as is the example of Fryd
et al. (2013) and Roldin et al. (2012); and (ii) the second approach, used
by 3/13 of authors, is a linear reservoir that represents the drainage of
subsurface groundwater by the sewer network and the drainage of the
subsurface groundwater by the river.
Indeed, inltration processes are sometimes neglected (in this review
by 37/50 studies) and the interaction between ow and inltration
components and the network is often ignored at the urban catchment
scale as the hydrology of urbanized zones is far from being simple: the
urban environment is highly heterogenous in terms of land use, subsoil
characteristics and other factors, which serve to inuence all hydro-
logical processes (Rodriguez et al., 2008).
Futhermore, inltration from impervious surfaces was not addressed
Fig. 4. Representation of urban drainage system processes in a combined sewer system. Inow meaning the inltration of water from surrounding soil to the
sewer pipes.
V.A. Montoya-Coronado et al.
Journal of Environmental Management 362 (2024) 121073
7
by any authors, nor was the exltration of water from the drainage
system, despite that in specic situations, these processes can play an
important role in the hydrological balance (Ragab et al., 2003; Ramier
et al., 2011).
5.2.3. Flow routing model
A classical hydraulic model conceptualizes the combined sewer
system by nodes and pipes. Often, the authors include other hydraulic
structures such as pumps and chambers. Their description and model-
ling are generally complex and require the use of partial differential
equations. However, it is not rare that ow dynamics can be simplied.
In almost half of the sample studies (22/50), six different models have
been mentioned (not mentioned for the rest of papers) to model the
transport ow: (i) the dynamic wave routine used by 9/22 of authors,
(ii) the full Saint-Venant equation used by 7/22 authors, (iii) the kine-
matic wave approach used by 3/22 of authors, (iv) the double linear
reservoir model used by Desai and Londhe (2016), (v) the Muskingum
equation used by Torres et al. (2021), and (vi) the Manning-Strickler
used on a lumped model by Fu et al. (2019).
5.2.4. Combined sewer overow models
The CSO structure is part of the conveyance compartment and should
be represented when the aim is CSO mitigation. All the authors have at
least represented one CSO structure. In this literature review two
modelling approaches were identied. The rst one, and the most
common (32% of studies, 16/50) is a threshold model. It consists of a
binary decision model guided by the maximum capacity of the sewer
drains before overow occurs. The threshold can be determined using
empirical relationship between outows from the sewer which arrives at
the CSO structure and overows: the maximum ow capacity down-
stream the CSO structure (used by 15/16 of studies). If the catchment
outow exceeds the maximum downstream capacity, the CSO spill ow
is equal to excess ow rate. The latter can be determined by the
maximum capacity of the downstream pipe (used by 12/16 of studies),
or by the maximum wastewater treatment plant (WWTP) volume ca-
pacity (used by 3/16 of studies), It is a simple volume balance equation
that neglected ow dynamics and estimated the CSO spill volume using
a volume balance equation based on the maximum WWTP volume ca-
pacity and the sewer volume produced in the catchment. The CSO
structure threshold can be also determined through a frequency analysis
(Mailhot et al., 2015; Yu et al., 2018). This process relies on observed
data to determine the CSO structure’s response based on rainfall dura-
tion, depth, and intensity values using long-term statistical methods.
However, this method was not mentioned in the study sample, probably
due to the lack of CSO data. The second modelling approach of CSO
structure (used by 5 authors) is based on the use of hydraulic structures
like orices, weirs, and pipes to simulate ow regulators mimicking the
threshold and the outow. Thus, complicated CSO structures are rep-
resented by simple hydraulic structures and related models calibrated
with eld data.
However, the approach basically used to represent CSO structures
and thus, simulating spill events is most of the time not mentioned. More
than half of the studies (58%, 29/50) do not mention how CSO struc-
tures were modelled. This omission can be attributed to the difculties
in instrumenting CSO and measuring data and to their complex hy-
draulic modelling (Montserrat et al., 2017). That is why some simpli-
cations might be made by the authors, or they might have used an
existing module for CSO in the modelling software. Such a gap raises a
crucial challenge: how to quantify the SUDS system capacity to mitigate
CSO without knowing how to properly model CSO structures?
5.2.5. Quality model: build-up, wash-off and pollutant transfer models
22% (11/50) of the studies have addressed pollution issues. Pollutant
accumulation on urban surfaces can be inuenced by preceding dry
weather periods (Walaszek et al., 2018). Frequently, data on pollutant
accumulation are directly sourced as input parameters from literature or
prior studies (Choi et al., 2019). In the sample study of this literature
review, three types of models were employed in order to assess the
quality performance of SUDS scenarios at catchment scale to mitigate
the impact of CSO: (i) the Event-Mean Concentration (EMC) model, used
by 3/11 of authors, generally reported as the mass of pollutants dis-
charged per unit volume of runoff during a storm event. Due to the
limitation in sampling programs and the random nature of rainfall
characteristics (Charbeneau and Barrett, 1998). Authors resort to using
literature data (Chen et al., 2019) while others calibrate the ECM value
with observations despite the scarcity of samples (Sarminingsih et al.,
2019); (ii) the wash-off model has been found in 2/11 studies. This
model simulates the process of pollutants being washed-off from sur-
faces, typically in urban areas due to rainfall runoff; and (iii) build-up
and wash-off models are used together by 6/11 of studies. In a rst
instance, the build-up model accumulates pollutants in the study area
during dry weather (Al Ali et al., 2017). Then, the wash-off model is
implemented.
These models have been applied to different pollutants, mainly
organic matter (biochemical oxygen demand (BOD) and chemical oxy-
gen demand (COD)), nutrients (ammonium (NH
4
), ammonia (NH
3
),
total nitrogen (TN), total phosphorous (TP)) and total suspended solids
(TSS).
5.3. Calibration and validation: challenging steps at the catchment scale
This section reviews the common practices involved in calibrating
and validating a model of urban drainage system with the objective of
modelling CSO in a context of SUDS implementation scenarios through
two questions: i) which parameters are commonly used to calibrate and
to validate the models? and ii) which metrics are selected to evaluate the
quality of the calibration/validation?
5.3.1. Parameters to calibrate the model
Model calibration involves selecting and adjusting specic parame-
ters within the model to ensure its outputs closely match observed or
reference data, thereby enhancing the model’s accuracy and predictive
capability. In Table 2, the most frequently utilized parameters for model
calibration have been categorized into surface, soil characteristics, sub-
surface characteristics, and network parameters. In the corpus of the 50
articles, it has been observed that no study has calibrated CSO param-
eters. Yet, in studies particularly focusing on CSO mitigation, it would be
an asset to consider the calibrating parameters associated with CSO
structures to better t CSO indicators.
The parameters summarized in Table 2 were compiled from 34%
Table 2
Summary of parameters identied for the calibration. The frequency of occur-
rence represents the percentage of articles, among the 34% of studies (n=17)
giving the information.
Type of parameter Physical measure Frequency of
occurrence in the
corpus
Surface parameters
(pervious and/or
impervious surfaces)
Depression storage/initial
losses
47%
Manning coefcient 35%
Impervious and/or pervious
surface area
29%
Sub-catchment width 24%
Runoff coefcient parameters 12%
Soil and sub-surface
parameters
Soil storage capacity 18%
Area contributing to rain
inltration inow to the
sewer system
12%
Sub-surface ow rate 6%
Network parameters Manning coefcient 12%
Model routing coefcient 6%
Bathymetric and backwater
curves
6%
V.A. Montoya-Coronado et al.
Journal of Environmental Management 362 (2024) 121073
8
(17/50) of the articles reviewed. On average, studies calibrated
approximately three parameters. However, there are instances, (e.g.,
Riechel et al., 2016), where a notably high number up to 89 parameters
were analyzed one at a time to determine the calibration parameters.
While this is not considered as a standard approach, it does raise the
important question of selecting the appropriate model in order to avoid
overtting. The answer to this question inherently depends on the total
number of parameters present within the model and on the subset of
parameters with a large impact on the model’s results (Perrin et al.,
2001).
Parameters chosen for calibration play a crucial role in shaping the
model’s behavior, as they directly inuence how the model responds to
input data and reects the real-world phenomena it aims to simulate.
Even if some parameters are measurable, their experimental estimation
is usually difcult at the catchment scale.
When it comes to SUDS parameters, as performance studies have
been increasing in the last years, more data is available about the design
elements (i.e., drainage area, media composition, thickness of each
media layer, surface ponding depth and underdrain size) and physical
properties (i.e., eld capacity, wilting point, hydraulic conductivity
slope, suction head). This development is highlighted by works like
those of Hou et al. (2021) and Hernes et al. (2020) who xed and cali-
brated SUDS parameters using measurements obtained from dedicated
experimental campaigns. This emerging wealth of information opens
new possibilities for improving modelling results diversity.
5.3.2. Metrics for calibration and validation phases
Calibration and validation of models typically use metrics to quantify
the model’s ability to replicate the observed/reference data. The selec-
tion of a metric (e.g., an objective function) for validating the model
makes part of the practices of modelling to understand if the goals of the
given modelling task are achieved (Jakeman et al., 2006). However,
nearly half of the studies (42%, 21/50) have indicated the objective
function they used. Most of these studies (13/21) applied the
Nash-Sutcliffe Efciency coefcient (NSE). Other objective functions
identied in the reviewed studies include the Root Mean Squared Error
(RMSE, for 6/21 of studies), the relative error (for 4/21 for studies), the
R
2
coefcient (for 3/21 of studies), as well as the Mean Absolute Error
(MAE), Mean Absolute Percentage Error (MAPE), Kling-Gupta Efciency
(KGE), and contingency table, each used by only one study. Some studies
(7/21) use more than one objective function and the one that comes up
the most is the combination of NSE and RMSE.
Within the context of the selected metrics, different variables are
considered to assess and quantify the model’s performance. The most
frequent variables were summarized in Table 3 among 58% articles of
the ensemble (n=29). Flow in the sewer system was the most frequently
used (by 11/29 of studies), followed by water levels in the sewer system
(by 8/29 of studies). The remaining 10 cases are presented in Table 3.
Furthermore, it should be noted that if the primary objective is to
mitigate CSO, it is advisable to perform model calibration utilizing ow
data obtained directly from the CSO facility’s measurements. However,
little attention has been given to calibrating the ow that is measured at
the discharges from CSO structures because of the large monitoring
effort that is required (Montserrat et al., 2017).
5.3.3. Temporal scale of dataset to calibrate and evaluate the model
Evaluation of whether a calibrated model is suited to study objectives
will be inuenced by the selection of its testing dataset. In the corpus,
there is a lack of clear indication of the choices made. Indeed, only 34%
(17/50) of the authors give the details on how they choose the time
period for calibrating and validating. 12/17 of those studies use between
1 and 10 events for calibrate and validate the model and 5/17 use be-
tween one and two years of time series and particularly focused on
complex objectives, often addressing more than two goals. It is impor-
tant to point out that for calibration and evaluation of studies on pol-
lutants (which represent 22% of the ensemble of studies (11/50)), some
studies (4/11) do not make any calibration and validation steps as there
was no available data. Thus, the modelling parameters were taken from
the literature. The barrier to properly validate the model is certainly the
hard demanding to t several parameters and amount of observed data
with 5/11 studies using less than 10 campaigns and 2/11 of them using
more than one year time series. Even if the lack of datasets is a recurrent
remark, it is important to mention that an appropriate calibration
methodology is lacking.
5.3.4. Bias and limitations for model application
Some studies do not explain any step of the calibration/validation
process (13/50) while others explicitly mention not calibrating the
model due to lack of observed data (8/50). The omission of these crucial
steps raises serious concerns. While it is understandable that data limi-
tations may complicate these processes, it prompts a critical examina-
tion of the overall rigor of these studies. Papers that focus on comparing
the performance of different SUDS scenarios omitting the calibration
and validation step could potentially compromise the reliability and
applicability of their results. Such omissions may yield results with high
uncertainties, undermining their utility in real-world decision-making.
Although some authors argue that some models do not require calibra-
tion, neglecting this step (even for physical based models), can comprise
accuracy and lead to biased conclusions. In particular if only positive
results are published.
5.4. Large scale SUDS modelling approaches
The purpose of this paragraph is to provide a summary of the
modelling approaches employed for the catchment-scale dissemination
of SUDS to mitigate CSO. Almost all of studies (n=48) describe the ap-
proaches used for modelling SUDS. Three different modelling ap-
proaches have been identied.
The multiple layers representation of SUDS, used by 19/48 of au-
thors, is commonly found in popular modelling tools such as SWMM,
Infoworks, MIKE Urban and L-THIA-LID. In this approach, a portion of
the sub-catchment is no longer directly drained by the network. Instead,
ow path rst passes through a SUDS module, which comprises several
layers. SUDS are composed by different layers (surface, soil, and storage
layers) and depending on the type of structure, the layers may vary. For
more precision about this type of SUDS modelling tools, please refer to
Lerer et al. (2015) and Kaykhosravi et al. (2018).
Modifying the ow path using storage structures (used by 15/48 of
authors) is an approach that aims to redirect the runoff from impervious
surfaces to pervious surfaces or retention facilities before diverting the
stormwater runoff to the sewer system. For example, Roldin et al. (2012)
routed the impervious runoff area to a ctive soakaway disconnected
from the sewer system.
Table 3
Summary of variables identied for evaluating the calibration and validation
step. The frequency of occurrence represents the percentage of articles among
the 58% of studies (n=29) using the parameters.
Type of parameter Variable Frequency of occurrence
Sewer Flow 38%
Water levels 28%
Peak ow 14%
Volume 7%
Peak time 3%
Velocity mean 3%
CSO volume 3%
CSO discharge 3%
Peak duration 3%
Surface Runoff volume 7%
Runoff ow 7%
Runoff duration 3%
Pollutant Concentration 7%
River Stream ow 7%
V.A. Montoya-Coronado et al.
Journal of Environmental Management 362 (2024) 121073
9
Modifying the sub-catchment characteristics (used by 14/48 of au-
thors) is a simplied approach and a global modication is to charac-
terize the catchment area in two ways. First, an equivalent area is equal
to the sum of the SUDS area. The impervious area is converted to
pervious area. The total area of the catchment is thus conserved after the
SUDS implementation (Eulogi et al., 2022; Muhandes et al., 2022;
Radinja et al., 2019; Torres et al., 2021). Second, modifying the initial
losses to mimick storage and inltration structures (Moore et al., 2012;
Stovin et al., 2013). All changes are uniformly distributed by altering
input data. For example, as rainwater tanks were not already included in
the model used by Elliott and Trowsdale (2007), the authors opt to
model them indirectly by modifying the initial loss parameter so it
represents the available space for storage.
The relationship between the type of modelling tool and the SUDS
modelling approach used is illustrated in Fig. 5. Despite the possibility of
modelling SUDS in conventional urban drainage models (SWMM, Info-
work, etc), only 40% of the studies (n=19) used the multiple layers
representation of SUDS.
During the examination of the modelling approaches, only a few
studies mention the fate of runoff after a SUDS structure becomes
saturated. Two modelling approaches were identied: (i) re-routing the
excess runoff to the sewer system (D’Ambrosio et al., 2022; Fan et al.,
2022; Roldin et al., 2012) or (ii) operating under the hypothesis of a
perfectly functioning SUDS structure where there is neither excess
runoff nor deep percolation rerouted to the sewer system (Joshi et al.,
2021; Villarreal et al., 2004). Actually, the interconnection between the
different compartments (surface, SUDS, soil, network) is often more
complex, and the impact of underground infrastructure (such as
sewerage systems and buildings) modies the ow paths of inltrated
water. This can potentially impact the ow rate in the sewer system
(Bonneau et al., 2017; Pophillat et al., 2021).
6. SUDS planning scenarios at urban catchment scale: scenario
elaboration and selection
The purpose of this section is to provide a summary of the method-
ologies used in the sample of the 50 articles for scenario elaboration and
selection for the catchment-scale dissemination of SUDS to mitigate
CSO.
6.1. SUDS scenario elaboration
To reach the objectives of CSO mitigation planning, the SUDS
implementation scenarios at catchment scale are tested. A scenario is a
catchment conguration different from the status quo situation. Several
literature reviews have analyzed the different practices guiding the
scenario elaboration (Ferrans et al., 2022; Lerer et al., 2015; Wang et al.,
2021). Three key questions are regularly addressed: i) which SUDS
technologies and combination to implement (studied by 23/50 of
authors), ii) how much and/or which size of the SUDS should be
implemented (studied by 18/50 of authors), and iii) where to implement
SUDS on the catchment (studied by 11/50 of authors).
6.1.1. Which SUDS?
Among the authors considering the “which” question (46%, n=23),
three methods were found through 11/23 authors. First, 3/11 use a
decision tree by establishing various criteria, such as local constraints
(area, drainage area, slope, soil type, groundwater level, etc.), land use
(residential, public, industrial), catchment characteristics (green spaces
and soil inltration rates, etc.), and construction operation costs. The
natural landscape features such as slope, soil type, or open green spaces,
are typically incorporated into placement purposes by either tools or
stakeholders, to determine "where" question. Second, 2/11 studies
employed a hierarchization. Third, 7/11 authors opted for a random
type of SUDS or could not provide a justication for their selection.
6.1.2. How many SUDS and which size of SUDS?
The “size” and the “number of SUDS” parameters has been gathered
due to their similarity during the modelling phase. They refer to the
fraction of the catchment area disconnected from the sewer system. The
fraction of the catchment area to be transformed into SUDS is a common
question, addressed by 36% of studies (n=18), in which different per-
centage of surface are disconnected. The scenarios leading to the ques-
tion of quantity have been distinguished into two categories. First, 13/
18 of studies, have no specic SUDS type but a maximum range of space
in which they can be deployed. Second, 5/18 of studies, have a dened
SUDS typology and a maximum space range in which they can be
deployed. For example, McGarity et al. (2017) locate a random number
of specic type of SUDS considering that the total size cannot exceed the
actual impervious area in each sub-catchment.
6.1.3. Where to implement them?
In new development areas, SUDS location are chosen during the
planication phase. The difculty remains for already existing urban
areas where available space is scarce, private or near buildings. The
criteria used to select the location is specic to each study and depends
on the urban objectives. 22% of papers (n=11), Fig. 6, have mentioned
the choice of SUDS placement based on the following criteria: specic
area (residential, industrial, commercial or other land use characteris-
tics), available space, public space, pervious area and impervious areas
to be converted in pervious. Hydrological and pollution constraints,
economical or environmental aspects were not mentioned. Among these
studies 8/11 used a random approach to select the placement of SUDS.
6.1.4. Correlation between the modelling approaches selection and
scenarios elaboration
The correlation between the development of SUDS scenarios (spe-
cically addressing the ’which’, ’where’, and ’how’ aspects) and the
Fig. 5. Sustainable Urban Drainage System (SUDS) modelling approaches and modelling tools. Categories are dened as follow: SWMM, Inforworks, MIKE Urban, L-
THIA-LID and others are the modelling tools. The modelling approaches are: the multiple layers representation of SUDS which means a detailed representation of
SUDS in an specic sub catchement, modifying the ow path using storage structures and modifying the catchment characteristics and modifying specic sub-
catchment characteristics.
V.A. Montoya-Coronado et al.
Journal of Environmental Management 362 (2024) 121073
10
conguration of the catchment system is explored in Appendix 3. This
analysis aims to discern whether the selection of system compartments
correlates with the objectives of the scenarios. Our observations indicate
that studies focusing on more than one question (which and how, or
which and where, or where and how) typically opt for a complex system
conguration that includes the surface, network, CSO, SUDS, and often
the soil and/or groundwater components. Conversely, authors who have
selected a basic conguration generally aim to address a single,
straightforward question (e.g., which).
Furthermore, the relationship between the choice of SUDS scenario
development questions and the approaches used to model them is also
explored (Appendix 4). We nd that lumped modelling approaches, such
as modifying the sub-catchment characteristics or the water ow paths,
are employed to explore scenarios with a singular objective (which’ or
’where’ or ’how’). On the other hand, more complex SUDS modelling
approaches, which offer detailed processes representation, are used for
scenarios encompassing two or more questions (e.g., which and where or
how and which).
6.2. SUDS scenario selection
6.2.1. Which typology of SUDS are frequently modelled?
A diverse range of SUDS is available to address the challenges of CSO
in urban settings (Castellar et al., 2021). These systems can be employed
either individually or in combination. A signicant majority of papers
(88%, n=44) specied the type of SUDS they modelled, while 12%
(n=6) used a lumped modelling approach which does not distinguish the
types of SUDS. This review identied 28 distinct types of SUDS in the
corpus. The ve most modelled SUDS are permeable pavements (used in
17/44 papers), green roofs (used in 14/44 papers), rain gardens (used in
13/44 papers), rain barrels (used in 12/44) and bioretention cells (used
in12/44). Notably, four out of these ve SUDS types can be modelled
using the SUDS modules in SWMM, Infoworks, and MIKE Urban. Half of
the articles focus on scenarios that consider only one SUDS type. Articles
combining two types of SUDS represent only 10% of the corpus (n=5),
while scenarios incorporating three or more SUDS account for 42% of
the corpus (n=21).
6.2.2. SUDS performance indicators for comparing modelled scenarios
Once the SUDS are implemented in the model, it is important to
assess their performance regarding CSO mitigation. This step requires to
extract some performance indicators from the model outputs. Five cat-
egories of performance indicators have been identied in the corpus
(Table 4): i) surface runoff control, ii) sewer network performance, iii)
sustainability, iv) river quality and v) economical factor. Typically,
authors have used only two categories of objectives. The most used
combinations of indicators are: sewer network performance and
economical for 14% (n=7), surface runoff control and sewer network
performance for 10% (n=5), sewer network performance and river
quality for 6% (n=3), and surface runoff control and river quality for 4%
(n=2). Only two authors used four categories (Liao et al., 2015; Mon-
talto et al., 2007), while one article encompassed all of them (Fan et al.,
2022). Two observations have been made. The rst is that other objec-
tives than the reduction of CSO are also investigated. The second is that
when sewer network indicators are not available to assess the effec-
tiveness of SUDS on CSO (due to the lack of calibration or use of this
compartment), surface runoff control indicators are used.
Another classication of indicators refers to the temporal resolution.
Two primary categories of indicators are recognized: time series-based
and event-based indicators. In studies that evaluate performance
within the context of climate change, time series-based indicators are
typically chosen. Examples of these indicators include the percentage
reduction in ood volumes, the yearly CSO duration, or the cumulative
volume or load of pollution over the specied time series (Muhandes
et al., 2022; Rodriguez et al., 2023).
Although authors have exhibited prociency in selecting perfor-
mance indicators, it is important to note that most of these selected
indicators have not undergone calibration nor validation. In addition,
caution should be exercised when selecting indicators that are often
presumed as "adequate." Lau et al. (2002) questioned the use of fre-
quency and volume as pollution indicators for receiving waters. Their
study demonstrated that, beyond a certain point of CSO volume reduc-
tion, there was no further improvement in river pollution. Thus, they
concluded that the frequency/volume of overows can serve as perfor-
mance indicators for the quality of receiving waters, provided their
signicant limitations are understood. It is therefore essential to observe
the effects of SUDS dissemination on various criteria.
6.2.3. Methods used for relevant scenarios selection
Once several scenarios have been elaborated and simulated, the user
might need to make a choice and to select or rank the different scenarios.
In the corpus, 66% of the articles (n=33) detail which method they used
to select the most suitable SUDS scenarios. Despite the diversity of
methodologies employed, three overarching methods stood out: i) the
legislation/regulation constraint, ii) the optimization function method
and iii) the ranking method.
The legislation/regulation constraint represents 18/33 of the arti-
cles. In constraint methods, scenarios are selected if they meet a specic
criterion. This criterion can be a threshold of a variable related to a local
legislation or local decisions. The criterion is often based on large pre-
vious studies considering multidisciplinary aspects. According to
regional regulations together with local studies (such as water devel-
opment and management master plan), objectives are set. Regulation
has therefore become a driver for action. The CSO frequency is the most
used variable considered when this method is applied (Gong et al., 2019;
Jean et al., 2021). An observation is that 8/33 of articles elaborate the
methods for the purpose of the very article, thus detail it. However,
10/33 did not detail how the scenarios were elaborated, as it was done in
a previous internal study, which prevents the repeatability.
The optimization function methods represent 9/33 of the articles
where the method is detailed. Optimization methods are usually used
when there are multiple objectives in the study. The most used variables
considered when this method is applied are the CSO reduction and cost
reduction. Jean et al. (2022) utilized an objective function to minimize
Fig. 6. Percentage of studies addressing specic SUDS planning scenarios
questions. ’Where?’ refers to the spatial distribution of SUDS, ’Which?’ to the
types of SUDS implemented, and ’How many?’ to the size and quantity of
SUDS deployed.
V.A. Montoya-Coronado et al.
Journal of Environmental Management 362 (2024) 121073
11
factors such as infrastructure cost, CSO volume, node overload, and use
of the downstream network capacity. Radinja et al. (2019) applied a
multi-criteria analysis based on: CSO reduction per implemented SUDS,
CSO reduction per investment, OPEX (operating expenses), feasibility,
inuence on amenity, and impact on biodiversity. Lastly, Torres et al.
(2022) employed a numerical optimization method to develop scenarios
that meet performance goals. A common thread is that they all those
stydies require the user to input a system of penalty functions to
determine which objective is prioritized. The signicance of each cri-
terion is weighted according to user preference, and an indicator is used
to dene which scenario is most appropriate.
The ranking methods represent 6/33 of the articles. All scenarios are
ranked according to the importance given to each service. For example,
Eaton (2018) uses the runoff reduction to rank the scenarios and Fan
et al. (2022) add to the latter the pollution reduction. The variables
considered when this method is applied are the runoff reduction,
pollution reduction, CSO volume reduction and cost effectiveness.
6.3. SUDS scenarios performances
Comparing the results of different SUDS scenarios is difcult due to
the specicity of each case study. Even though, 5 authors results
(Autixier et al., 2014; D’Ambrosio et al., 2023; Desai and Londhe, 2016;
Joshi et al., 2021; Stovin et al., 2013) were considered comparable as
they implement the same modelling strategy in different degrees
(Fig. 7). Those authors have modelled different types of SUDS in
different degrees of deployment. Fig. 7 presents the CSO reduction
volume (with the SUDS implemented) compared to the status quo situ-
ation, as a function of the catchment disconnected surface area. All case
studies have a performance of at least 1% volume reduction for 1%
surface area disconnected. For example, Autixier et al. (2014) deployed
rain gardens in the study catchment with a total area of 1% (3.37 ha),
2% (6.18 ha), 3% (9.05 ha) and 4% (12.24 ha), and respectively a CSO
volume reduction of 8%, 15%, 22% and 28%. When comparing the case
studies, the choice of modelling approach is an important factor. In fact,
not all the articles represent the study systems in the same way (cf. §How
to model the urban catchment compartment and what is lacking?) and
do not use the same type of rain (cf. §Rainfall data and temporal). For
example, Autixier et al. (2014) use several rainfall events while D’Am-
brosio et al. (2022) use a design rainfall event and Joshi et al. (2021)
continuous simulation. For those reasons, comparing the results of these
studies might be biased, but it is the best that could be retrieved. Event
though the difference between the case studies, all results demonstrate
an immediate effect of catchment disconnection on the CSO volume
reduction.
7. Conclusions and perspectives
This review aims at summarizing the literature on SUDS imple-
mentation modelling for CSO mitigation at the catchment scale through
50 case studies published between 1999 and 2023. This state of the art
covers hydrologic-hydraulic modelling practices, methods of SUDS
scenario design planning, and performance of modelling reported in the
corpus. The main objective is to provide stakeholders information to
help them better understand the implications and limits of modelling
choices. The present work is going further than the already existing
reviews by proposing a comprehensive review encompassing every stage
of the modelling process. This includes the construction, validation, and
calibration of the status quo of the urban system, as well as the elabo-
ration, modelling, and selection of SUDS scenarios to evaluate their
performance on CSO.
Concerning the methodologies of modelling the status quo of the
urban system, the key aspects identied are the following:
•There is an important diversity of: urban catchments sizes, ap-
proaches of modelling, implemented compartments and equations of
the system modelled for the objective of CSO mitigation with SUDS
scenarios. This diversity can be explained by the high variability in
local context including stormwater system knowledge, legislative
frameworks and resources of stakeholders.
Table 4
Performance indicators found in the corpus. Indicators are classied in ve
categories: surface runoff control, sewer network performance, sustainability,
river quality and economical objectives. All parameters were used for CSO
volume mitigation, additional objective(s) of the study that uses these param-
eters are marked by
for climate change objectives; for river pollution miti-
gation objectives; for ood mitigation objectives and for non-point source
pollution objectives.
Performance indicator category Detailed parameters
Surface runoff control (Used by
32% of the authors)
Runoff quantity control parameters:
Runoff reduction rate
Peak ow reduction
Total ow volume
Flow volume reduction
Flow duration
Reduction of CSO discharge per impervious
hectare
Non-point source pollution control
parameters:
Concentration of COD, SS, NH3–N,TP
Sewer network performance
(Used by 76% of the authors)
Flood control parameters:
Number of ooding nodes
Percentage reduction in ood volume
CSO dynamic control parameters:
Peak ow reduction
Maximum CSO ow
Flow duration
Annual mean peak
CSO quantity control parameters:
Volume reduction
Total volume
Maximum overow volume
Annual volume
Daily total volume
CSO frequency control parameters:
Reduction rate of overow frequency
Quality control parameters:
Pollutant loads (BOD, NH
4
, COD load
, SS
load , TN load, NH
3
–N
, TP , TSS, BOD)
Pollution load reduction rate
Annual Overow load (SS, COD, TP, TN
Ammonia)
WWTP control parameters:
Pollution load reduction rate (COD, NH3, SS,
TP)
Sustainability (Used by 8% of the
authors)
Maintenance cost parameters:
Maintenance cost
Average CSO cost reduction per implemented
units
Annual SUDS life cost
River quality (Used by 22% of the
authors)
Quality control parameters:
Dissolved oxygen (DO) concentrations
Frequency of critical DO conditions
Total duration of critical DO conditions pH
Pollution load (BOD, COD, TSS, oil, grease,
total coliforms)
Economical (Used by 26% of the
authors)
Construction and maintenance
parameters:
Cost of CSO reduction
Construction cost
Normalized cost index (cost per m3 of CSO
volume reduction)
Capital expenditure
Annual operational costs
Cost effectiveness analysis
Life cost per unit of a specic measure
Installation cost
Efciency cost parameters:
Average CSO reduction per investment
Costs Per Reduction (CPR) index
V.A. Montoya-Coronado et al.
Journal of Environmental Management 362 (2024) 121073
12
•In addition to the lack of clear methods for modelling the network
system, the authors do not explain in detail how they modelled all
the compartments they chose, in particular the CSO structure.
•When specied, the compartments modelled are mainly compart-
ments with a rapid hydraulic response (runoff surface, sewer
network). However, the modelling approaches of slow responses
between the soil, groundwater and the sewer system are lacking or
not mentioned.
•Deep inltration compartment and exchanges with the sewer
network and thus the impact on CSO volume is often ignored at the
urban catchment scale.
•Less than half of articles (42%) carried out validation and calibra-
tion, with limited explanation for this shortfall. It is likely attribut-
able to the challenges in instrumenting, especially CSO structures,
and the consequent scarcity of eld data: a recurring issue in this
eld.
•Although the primary objective of limiting CSO is to protect the
quality of the receiving environment, models focusing on pollution
are less common, with only 22% of studies addressing them. This
scarcity is partly due to the limited availability of pollution data
compared to hydrological data, a challenge that becomes evident
during the calibration and validation phases, where often fewer than
10 events, or sometimes none, are used.
•No methodology for assessing uncertainties was implemented in the
corpus of studies.
After having developed a hydraulic-hydrologic model of the urban
system, authors incorporate SUDS scenarios at the catchment level to
evaluate their impact on CSO mitigation:
•Three different modelling approaches were adopted for SUDS
modelling at the catchment scale. Thus, no clear modelling approach
trend could be drawn here.
•There is a diversity of performance indicators for evaluating SUDS
scenarios. The most commons are sewer network indicators which is
in accordance with the CSO mitigation objective. In addition, climate
change, river pollution mitigation, non-point source and ood miti-
gation performance indicators are also evaluated.
•Elaboration and selection of SUDS scenarios for CSO mitigation is
mainly based on local constraints.
•In all 50 articles studied, the stormwater management through SUDS
has immediate and positive effect on CSO mitigation.
•Nevertheless, the reliability of those results can be questioned
knowing that calibration/validation is barely carried out. Therefore,
this stage needs to be improved.
From the state of the art offered by this review some perspectives can
be drawn to improve current practice in modelling SUDS scenarios at
large scale for CSO mitigation:
•The lack of knowledge on CSO structures and the lack of eld
monitoring data seem to be common and hampers the modelling
process and the appropriate management of the network. We need to
improve clear practices of CSO modelling at the urban catchment
scale and the calibration and validation of the catchment system
model.
•Effort should focus on developing methods to optimize the perfor-
mance of scenarios for CSO mitigation using hydrological drivers to
improve results and optimize modelling efforts.
•Due to the lack of data to feed, validate and calibrate the models,
there is a need for parsimonious and user-friendly models to simulate
the implementation of a SUDS deployment over the long periods
required to evaluate CSO structures.
CRediT authorship contribution statement
V.A. Montoya-Coronado: Writing – original draft, Visualization,
Validation, Methodology, Investigation, Formal analysis, Data curation,
Conceptualization. D. Tedoldi: Writing – review & editing, Supervision,
Methodology, Investigation, Formal analysis, Conceptualization. E.
Lenormand: Data curation, Conceptualization, Writing – review &
editing. H. Castebrunet: Writing – review & editing, Supervision,
Project administration, Investigation, Funding acquisition, Conceptual-
ization. P. Molle: Writing – review & editing, Funding acquisition,
Conceptualization. G. Lipeme Kouyi: Writing – review & editing, Su-
pervision, Investigation, Funding acquisition, Conceptualization, Proj-
ect administration.
Declaration of competing interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Data availability
Data will be made available on request.
Acknowledgements
The authors would like to thank the Ofce Français de la
Fig. 7. SUDS performance in terms of CSO volume reduction derived from various studies for different percentages of impervious area disconnected from the
sewer system.
V.A. Montoya-Coronado et al.
Journal of Environmental Management 362 (2024) 121073
13
Biodiversit´
e, Rhˆ
one–M´
editerran´
ee–Corse and Adour-Garonne water
agencies for their nancial support to the TONIC research project. This
work was carried out with the support of the EURH2O’Lyon (ANR-17-
EURE-0018) of the University of Lyon (UdL) as part of the "Investisse-
ments d’Avenir" program managed by the French National Research
Agency (ANR) and performed in part within the framework of the OTHU
(Field Observatory for Urban Water Management – http://othu.org) and
the MULTISOURCE project, funded by the European Union’s Horizon.
H2020-EU.3.5.2. under grant agreement 101003527.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.jenvman.2024.121073.
References
Ahiablame, L.M., Engel, B.A., Chaubey, I., 2012. Effectiveness of low impact
development practices: literature review and suggestions for future research. Water
Air Soil Pollut. 223, 4253–4273. https://doi.org/10.1007/s11270-012-1189-2.
Al Ali, S., Bonhomme, C., Dubois, P., Chebbo, G., 2017. Investigation of the wash-off
process using an innovative portable rainfall simulator allowing continuous
monitoring of ow and turbidity at the urban surface outlet. Sci. Total Environ. 609,
17–26. https://doi.org/10.1016/j.scitotenv.2017.07.106.
Al Aukidy, M., Verlicchi, P., 2017. Contributions of combined sewer overows and
treated efuents to the bacterial load released into a coastal area. Sci. Total Environ.
607–608, 483–496. https://doi.org/10.1016/j.scitotenv.2017.07.050.
Al-Rubaei, A.M., Engstr¨
om, M., Viklander, M., Blecken, G.-T., 2017. Effectiveness of a 19-
year old combined pond-wetland system in removing particulate and dissolved
pollutants. Wetlands 37, 485–496. https://doi.org/10.1007/s13157-017-0884-6.
Alves, A., Sanchez, A., Vojinovic, Z., Seyoum, S., Babel, M., Brdjanovic, D., 2016.
Evolutionary and holistic assessment of green-grey infrastructure for CSO reduction.
Water 8, 402. https://doi.org/10.3390/w8090402.
Andres-Domenech, I., Munera, J.C., Frances, F., Marco, J.B., 2010. Coupling urban event-
based and catchment continuous modelling for combined sewer overow river
impact assessment. Hydrol. Earth Syst. Sci. 14, 2057–2072. https://doi.org/
10.5194/hess-14-2057-2010.
Autixier, L., Mailhot, A., Bolduc, S., Madoux-Humery, A.-S., Galarneau, M., Pr´
evost, M.,
Dorner, S., 2014. Evaluating rain gardens as a method to reduce the impact of sewer
overows in sources of drinking water. Sci. Total Environ. 499, 238–247. https://
doi.org/10.1016/j.scitotenv.2014.08.030.
Bertels, D., De Meester, J., Dirckx, G., Willems, P., 2023. Estimation of the impact of
combined sewer overows on surface water quality in a sparsely monitored area.
Water Res. 120498 https://doi.org/10.1016/j.watres.2023.120498.
Bertrand-Krajewski, J.-L., Barraud, S., Lipeme Kouyi, G., Torres, A., Lepot, M., 2008. On-
line monitoring of particulate pollutant loads in urban sewer systems : stakes,
methods, example of application. La Houille Blanche 94, 49–57. https://doi.org/
10.1051/lhb:2008039.
Bonneau, J., Branger, F., Castebrunet, H., Lipeme Kouyi, G., 2023. The impact of
stormwater management strategies on the ow regime of a peri-urban catchment
facing urbanisation and climate change: a distributed modelling study in Lyon,
France. Urban Water J. 1–18. https://doi.org/10.1080/1573062X.2023.2217809, 0.
Bonneau, J., Fletcher, T.D., Costelloe, J.F., Burns, M.J., 2017. Stormwater inltration and
the ‘urban karst’ – a review. J. Hydrol. 552, 141–150. https://doi.org/10.1016/j.
jhydrol.2017.06.043.
Botturi, A., Ozbayram, E.G., Tondera, K., Gilbert, N.I., Rouault, P., Caradot, N.,
Gutierrez, O., Daneshgar, S., Frison, N., Akyol, Ç., Foglia, A., Eusebi, A.L., Fatone, F.,
2021. Combined sewer overows: a critical review on best practice and innovative
solutions to mitigate impacts on environment and human health. Crit. Rev. Environ.
Sci. Technol. 51, 1585–1618. https://doi.org/10.1080/10643389.2020.1757957.
Casey, M.J., Stagge, J.H., Moglen, G.E., McCuen, R.H., 2015. Effects of watershed
subdivision on peak discharge in rainfall-runoff modeling in the WinTR-20 model.
J. Hydrol. Eng. 20, 04015020 https://doi.org/10.1061/(ASCE)HE.1943-
5584.0001188.
Castellar, J.a.C., Popartan, L.A., Pueyo-Ros, J., Atanasova, N., Langergraber, G.,
S¨
aumel, I., Corominas, L., Comas, J., Acu˜
na, V., 2021. Nature-based solutions in the
urban context: terminology, classication and scoring for urban challenges and
ecosystem services. Sci. Total Environ. 779, 146237 https://doi.org/10.1016/j.
scitotenv.2021.146237.
Charbeneau, R.J., Barrett, M.E., 1998. Evaluation of methods for estimating stormwater
pollutant loads. Water Environ. Res. 70, 1295–1302. https://doi.org/10.2175/
106143098X123679.
Chen, J., Liu, Y., Gitau, M.W., Engel, B.A., Flanagan, D.C., Harbor, J.M., 2019. Evaluation
of the effectiveness of green infrastructure on hydrology and water quality in a
combined sewer overow community. Sci. Total Environ. 665, 69–79. https://doi.
org/10.1016/j.scitotenv.2019.01.416.
Chocat, B., Krebs, P., Marsalek, J., Rauch, W., Schilling, W., 2001. Urban drainage
redened: from stormwater removal to integrated management. Water Sci. Technol.
43, 61–68. https://doi.org/10.2166/wst.2001.0251.
Choi, J., Na, E., Ryu, J., Kim, J., Kim, H., Shin, D., 2019. Analysis of Pollutant Build-Up
Model Applied to Various Urban Landuse. https://doi.org/10.12989/
mwt.2019.10.1.013.
D’Ambrosio, R., Balbo, A., Longobardi, A., Rizzo, A., 2022. Re-think urban drainage
following a SuDS retrotting approach against urban ooding: a modelling
investigation for an Italian case study. Urban For. Urban Green. 70, 127518 https://
doi.org/10.1016/j.ufug.2022.127518.
D’Ambrosio, R., Longobardi, A., Schmalz, B., 2023. SuDS as a climate change adaptation
strategy: scenario-based analysis for an urban catchment in northern Italy. Urban
Clim. 51, 101596 https://doi.org/10.1016/j.uclim.2023.101596.
Deletic, A., Dotto, C.B.S., McCarthy, D.T., Kleidorfer, M., Freni, G., Mannina, G., Uhl, M.,
Henrichs, M., Fletcher, T.D., Rauch, W., Bertrand-Krajewski, J.L., Tait, S., 2012.
Assessing uncertainties in urban drainage models. Phys. Chem. Earth 3–10. https://
doi.org/10.1016/j.pce.2011.04.007. Parts A/B/C 42–44.
Desai, A.A., Londhe, S., 2016. Catchment level modeling of green roofs using InfoWorks
CS. Int. J. Earth Sci. Eng. 9, 15.
Devia, G.K., Ganasri, B.P., Dwarakish, G.S., 2015. A review on hydrological models.
Aquatic Procedia, INTERNATIONAL CONFERENCE ON WATER RESOURCES,
COASTAL AND OCEAN ENGINEERING 4, 1001–1007. https://doi.org/10.1016/j.
aqpro.2015.02.126 (ICWRCOE’15).
Di Nunno, F., Granata, F., Parrino, F., Gargano, R., de Marinis, G., 2021. Microplastics in
combined sewer overows: an experimental study. J. Mar. Sci. Eng. 9, 1415. https://
doi.org/10.3390/jmse9121415.
Eaton, T.T., 2018. Approach and case-study of green infrastructure screening analysis for
urban stormwater control. J. Environ. Manag. 209, 495–504. https://doi.org/
10.1016/j.jenvman.2017.12.068.
Elliott, A., Trowsdale, S., 2007. A review of models for low impact urban stormwater
drainage. Environ. Model. Software 22, 394–405. https://doi.org/10.1016/j.
envsoft.2005.12.005.
Ellis, J.B., 2006. Pharmaceutical and personal care products (PPCPs) in urban receiving
waters. Environ. Pollution, Soil and Sediment Remediation (SSR) 144, 184–189.
https://doi.org/10.1016/j.envpol.2005.12.018.
El-Nasr, A.A., Arnold, J.G., Feyen, J., Berlamont, J., 2005. Modelling the hydrology of a
catchment using a distributed and a semi-distributed model. Hydrol. Process. 19,
573–587. https://doi.org/10.1002/hyp.5610.
Eulogi, M., Ostojin, S., Skipworth, P., Kroll, S., Shucksmith, J.D., Schellart, A., 2022.
Optimal positioning of RTC actuators and SuDS for sewer overow mitigation in
urban drainage systems. Water 14, 3839. https://doi.org/10.3390/w14233839.
Even, S., Mouchel, J.-M., Servais, P., Flipo, N., Poulin, M., Blanc, S., Chabanel, M.,
Paffoni, C., 2007. Modelling the impacts of combined sewer overows on the river
seine water quality. Sci. Total Environ. 375, 140–151. https://doi.org/10.1016/j.
scitotenv.2006.12.007.
Fan, G., Lin, R., Wei, Z., Xiao, Y., Shangguan, H., Song, Y., 2022. Effects of low impact
development on the stormwater runoff and pollution control. Sci. Total Environ.
805, 150404 https://doi.org/10.1016/j.scitotenv.2021.150404.
Fenner, R.A., 2000. Approaches to sewer maintenance: a review. Urban Water. Sewer
Systems and Processes 2, 343–356. https://doi.org/10.1016/S1462-0758(00)00065-
0.
Ferrans, P., Torres, M.N., Temprano, J., Rodríguez S´
anchez, J.P., 2022. Sustainable
Urban Drainage System (SUDS) modeling supporting decision-making: a systematic
quantitative review. Sci. Total Environ. 806, 150447 https://doi.org/10.1016/j.
scitotenv.2021.150447.
Fryd, O., Backhaus, A., Birch, H., Fratini, C.F., Ingvertsen, S.T., Jeppesen, J., Panduro, T.
E., Roldin, M., Jensen, M.B., 2013. Water sensitive urban design retrots in
Copenhagen – 40% to the sewer, 60% to the city. Water Sci. Technol. 67,
1945–1952. https://doi.org/10.2166/wst.2013.073.
Fu, X., Goddard, H., Wang, X., Hopton, M.E., 2019. Development of a scenario-based
stormwater management planning support system for reducing combined sewer
overows (CSOs). J. Environ. Manag. 236, 571–580. https://doi.org/10.1016/j.
jenvman.2018.12.089.
Ghosh, I., Hellweger, F.L., 2012. Effects of spatial resolution in urban hydrologic
simulations. J. Hydrol. Eng. 17, 129–137. https://doi.org/10.1061/(ASCE)HE.1943-
5584.0000405.
Gogien, F., Dechesne, M., Martinerie, R., Lipeme Kouyi, G., 2023. Assessing the impact of
climate change on Combined Sewer Overows based on small time step future
rainfall timeseries and long-term continuous sewer network modelling. Water Res.
230, 119504 https://doi.org/10.1016/j.watres.2022.119504.
Golden, H.E., Hoghooghi, N., 2018. Green infrastructure and its catchment-scale effects:
an emerging science. WIREs Water 5. https://doi.org/10.1002/wat2.1254.
Gong, Y., Chen, Y., Yu, L., Li, J., Pan, X., Shen, Z., Xu, X., Qiu, Q., 2019. Effectiveness
analysis of systematic combined sewer overow control schemes in the sponge city
pilot area of beijing. IJERPH 16, 1503. https://doi.org/10.3390/ijerph16091503.
Green, W.H., Ampt, G.A., 1911. Studies on soil phyics. J. Agric. Sci. 4, 1–24. https://doi.
org/10.1017/S0021859600001441.
Harbaugh, A.W., 2005. MODFLOW-2005, the US Geological Survey Modular Ground-
Water Model: the Ground-Water Flow Process. US Department of the Interior, US
Geological Survey Reston, VA, USA.
Hernes, R.R., Gragne, A.S., Abdalla, E.M.H., Braskerud, B.C., Alfredsen, K., Muthanna, T.
M., 2020. Assessing the effects of four SUDS scenarios on combined sewer overows
in Oslo, Norway: evaluating the low-impact development module of the Mike Urban
model. Nord. Hydrol 51, 1437–1454. https://doi.org/10.2166/nh.2020.070.
Horton, R.E., 1939. Analysis of runoff-plat experiments with varying inltration-
capacity. Eos, Transactions American Geophysical Union 20, 693–711. https://doi.
org/10.1029/TR020i004p00693.
Hou, X., Guo, H., Wang, F., Li, M., Xue, X., Liu, X., Zeng, S., 2020. Is the sponge city
construction sufciently adaptable for the future stormwater management under
V.A. Montoya-Coronado et al.
Journal of Environmental Management 362 (2024) 121073
14
climate change? J. Hydrol. 588, 125055 https://doi.org/10.1016/j.
jhydrol.2020.125055.
Hou, X., Qin, L., Xue, X., Xu, S., Yang, Y., Liu, X., Li, M., 2021. A city-scale fully
controlled system for stormwater management: consideration of ooding, non-point
source pollution and sewer overow pollution. J. Hydrol. 603, 127155 https://doi.
org/10.1016/j.jhydrol.2021.127155.
Jakeman, A.J., Letcher, R.A., Norton, J.P., 2006. Ten iterative steps in development and
evaluation of environmental models. Environ. Model. Software 21, 602–614.
https://doi.org/10.1016/j.envsoft.2006.01.004.
Jayasooriya, V.M., Ng, A.W.M., 2014. Tools for modeling of stormwater management
and economics of green infrastructure practices: a review. Water Air Soil Pollut. 225,
2055. https://doi.org/10.1007/s11270-014-2055-1.
Jayasooriya, V.M., Ng, A.W.M., Muthukumaran, S., Perera, C.B.J., 2020. Optimization of
green infrastructure practices in industrial areas for runoff management: a review on
issues, challenges and opportunities. Water 12, 1024. https://doi.org/10.3390/
w12041024.
Jean, M., Morin, C., Duchesne, S., Pelletier, G., Pleau, M., 2022. Real-time model
predictive and rule-based control with green infrastructures to reduce combined
sewer overows. Water Res. 221, 118753 https://doi.org/10.1016/j.
watres.2022.118753.
Jean, M.-`
E., Morin, C., Duchesne, S., Pelletier, G., Pleau, M., 2021. Optimization of real-
time control with green and gray infrastructure design for a cost-effective mitigation
of combined sewer overows. Water Resour. Res. 57, e2021WR030282 https://doi.
org/10.1029/2021WR030282.
Joshi, P., Leit˜
ao, J.P., Maurer, M., Bach, P.M., 2021. Not all SuDS are created equal:
impact of different approaches on combined sewer overows. Water Res. 191,
116780 https://doi.org/10.1016/j.watres.2020.116780.
Kay, P., Hughes, S.R., Ault, J.R., Ashcroft, A.E., Brown, L.E., 2017. Widespread, routine
occurrence of pharmaceuticals in sewage efuent, combined sewer overows and
receiving waters. Environ. Pollut. 220, 1447–1455. https://doi.org/10.1016/j.
envpol.2016.10.087.
Kaykhosravi, S., Khan, U., Jadidi, A., 2018. A comprehensive review of low impact
development models for research, conceptual, preliminary and detailed design
applications. Water 10, 1541. https://doi.org/10.3390/w10111541.
Lau, J., Butler, D., Schütze, M., 2002. Is combined sewer overow spill frequency/
volume a good indicator of receiving water quality impact? Urban Water 4,
181–189. https://doi.org/10.1016/S1462-0758(02)00013-4.
Lerer, S., Arnbjerg-Nielsen, K., Mikkelsen, P., 2015. A mapping of tools for informing
water sensitive urban design planning decisions—questions, aspects and context
sensitivity. Water 7, 993–1012. https://doi.org/10.3390/w7030993.
Liao, Z.L., Zhang, G.Q., Wu, Z.H., He, Y., Chen, H., 2015. Combined sewer overow
control with LID based on SWMM: an example in Shanghai. China. Water Science
and Technology 71, 1136–1142. https://doi.org/10.2166/wst.2015.076.
Mailhot, A., Talbot, G., Lavall´
ee, B., 2015. Relationships between rainfall and combined
sewer overow (CSO) occurrences. J. Hydrol. 523, 602–609. https://doi.org/
10.1016/j.jhydrol.2015.01.063.
McGarity, A.E., Szalay, S., Cohen, J., 2017. StormWISE model using green infrastructure
to achieve philadelphia’s CSO volume reductions at minimum cost, 334–344. htt
ps://doi.org/10.1061/9780784480632.026.
Mikovits, C., Tscheikner-Gratl, F., Jasper-T¨
onnies, A., Einfalt, T., Huttenlau, M.,
Sch¨
opf, M., Kinzel, H., Rauch, W., Kleidorfer, M., 2017. Decision support for
adaptation planning of urban drainage systems. J. Water Resour. Plann. Manag. 143,
04017069 https://doi.org/10.1061/(ASCE)WR.1943-5452.0000840.
Montalto, F., Behr, C., Alfredo, K., Wolf, M., Arye, M., Walsh, M., 2007. Rapid assessment
of the cost-effectiveness of low impact development for CSO control. Landsc. Urban
Plann. 82, 117–131. https://doi.org/10.1016/j.landurbplan.2007.02.004.
Montserrat, A., Hofer, T., Poch, M., Muschalla, D., Corominas, Ll, 2017. Using the
duration of combined sewer overow events for the calibration of sewer
hydrodynamic models. Urban Water J. 14, 782–788. https://doi.org/10.1080/
1573062X.2016.1254255.
Moore, S.L., Stovin, V.R., Wall, M., Ashley, R.M., 2012. A GIS-based methodology for
selecting stormwater disconnection opportunities. Water Sci. Technol. 66, 275–283.
https://doi.org/10.2166/wst.2012.172.
Muhandes, S., Dobson, B., Mijic, A., 2022. The value of aggregated city scale models to
rapidly assess SuDS in combined sewer systems. Frontiers in Water 3.
Muttil, N., Nasrin, T., Sharma, A.K., 2023. Impacts of extreme rainfalls on sewer
overows and WSUD-based mitigation strategies: a review. Water 15, 429. https://
doi.org/10.3390/w15030429.
Paijens, C., Bressy, A., Fr`
ere, B., Tedoldi, D., Mailler, R., Rocher, V., Neveu, P.,
Moilleron, R., 2021. Urban pathways of biocides towards surface waters during dry
and wet weathers: assessment at the Paris conurbation scale. J. Hazard Mater. 402,
123765 https://doi.org/10.1016/j.jhazmat.2020.123765.
Park, S.Y., Lee, K.W., Park, I.H., Ha, S.R., 2008. Effect of the aggregation level of surface
runoff elds and sewer network for a SWMM simulation. Desalination 226, 328–337.
https://doi.org/10.1016/j.desal.2007.02.115.
Perrin, C., Michel, C., Andr´
eassian, V., 2001. Does a large number of parameters enhance
model performance? Comparative assessment of common catchment model
structures on 429 catchments. J. Hydrol. 242, 275–301. https://doi.org/10.1016/
S0022-1694(00)00393-0.
Pophillat, W., Sage, J., Rodriguez, F., Braud, I., 2021. Dealing with shallow groundwater
contexts for the modelling of urban hydrology – a simplied approach to represent
interactions between surface hydrology, groundwater and underground structures in
hydrological models. Environ. Model. Software 144, 105144. https://doi.org/
10.1016/j.envsoft.2021.105144.
Radinja, M., Comas, J., Corominas, L., Atanasova, N., 2019. Assessing stormwater control
measures using modelling and a multi-criteria approach. J. Environ. Manag. 243,
257–268. https://doi.org/10.1016/j.jenvman.2019.04.102.
Ragab, R., Rosier, P., Dixon, A., Bromley, J., Cooper, J.D., 2003. Experimental study of
water uxes in a residential area: 2. Road inltration, runoff and evaporation.
Hydrological Processes 17, 2423–2437. https://doi.org/10.1002/hyp.1251.
Ramier, D., Berthier, E., Andrieu, H., 2011. The hydrological behaviour of urban streets:
long-term observations and modelling of runoff losses and rainfall–runoff
transformation. Hydrological Processes 25, 2161–2178. https://doi.org/10.1002
/hyp.7968.
Riechel, M., Matzinger, A., Pawlowsky-Reusing, E., Sonnenberg, H., Uldack, M.,
Heinzmann, B., Caradot, N., von Seggern, D., Rouault, P., 2016. Impacts of combined
sewer overows on a large urban river – understanding the effect of different
management strategies. Water Res. 105, 264–273. https://doi.org/10.1016/j.
watres.2016.08.017.
Rodriguez, F., Andrieu, H., Morena, F., 2008. A distributed hydrological model for
urbanized areas – model development and application to case studies. J. Hydrol.
351, 268–287. https://doi.org/10.1016/j.jhydrol.2007.12.007.
Rodriguez, M., Fu, G., Butler, D., Yuan, Z., Cook, L., 2023. Global resilience analysis of
combined sewer systems under continuous hydrologic simulation. J. Environ.
Manag. 344, 118607 https://doi.org/10.1016/j.jenvman.2023.118607.
Roldin, M., Fryd, O., Jeppesen, J., Mark, O., Binning, P.J., Mikkelsen, P.S., Jensen, M.B.,
2012. Modelling the impact of soakaway retrots on combined sewage overows in
a 3km2 urban catchment in Copenhagen, Denmark. J. Hydrol. 452–453, 64–75.
https://doi.org/10.1016/j.jhydrol.2012.05.027.
Roseboro, A., Torres, M.N., Zhu, Z., Rabideau, A.J., 2021. The impacts of climate change
and porous pavements on combined sewer overows: a case study of the city of
Buffalo, New York, USA. Front. Water 3, 725174. https://doi.org/10.3389/
frwa.2021.725174.
Rossman, L.A., Simon, 2022. Storm Water Management Model User’s Manual Version
5.1.
Salvadore, E., Bronders, J., Batelaan, O., 2015. Hydrological modelling of urbanized
catchments: a review and future directions. J. Hydrol. 529, 62–81. https://doi.org/
10.1016/j.jhydrol.2015.06.028.
Sarminingsih, A., Samudro, G., Mas’adah, A., 2019. Low impact development (LID) as an
effort to achieve a sustainable urban drainage system (SUDS). Case study: left side of
garang river segment VI semarang. E3S Web Conf. 125, 07017 https://doi.org/
10.1051/e3sconf/201912507017.
Seidl, M., Servais, P., Mouchel, J.M., 1998. Organic matter transport and degradation in
the river Seine (France) after a combined sewer overow. Water Res. 32,
3569–3580. https://doi.org/10.1016/S0043-1354(98)00169-9.
Simperler, L., Himmelbauer, P., Ertl, T., Stoeglehner, G., 2020. Prioritization of
stormwater management sites in urban areas. Journal of Environmental
Management 265, 110507. https://doi.org/10.1016/j.jenvman.2020.110507.
Stovin, V.R., Moore, S.L., Wall, M., Ashley, R.M., 2013. The potential to retrot
sustainable drainage systems to address combined sewer overow discharges in the
Thames Tideway catchment: potential to retrot SuDS in the Thames Tideway
catchment. Water Environ. J. 27, 216–228. https://doi.org/10.1111/j.1747-
6593.2012.00353.x.
Tavakol-Davani, Hessam, Goharian, E., Hansen, C.H., Tavakol-Davani, Hassan, Apul, D.,
Burian, S.J., 2016. How does climate change affect combined sewer overow in a
system beneting from rainwater harvesting systems? Sustain. Cities Soc. 27,
430–438. https://doi.org/10.1016/j.scs.2016.07.003.
Torres, M.N., Fontecha, J.E., Walteros, J.L., Zhu, Z., Ahmed, Z., Rodríguez, J.P.,
Rabideau, A.J., 2021. City-scale optimal location planning of Green Infrastructure
using piece-wise linear interpolation and exact optimization methods. J. Hydrol.
601, 126540 https://doi.org/10.1016/j.jhydrol.2021.126540.
Torres, M.N., Rabideau, A., Ghodsi, S.H., Zhu, Z., Shawn Matott, L., 2022. Spatial design
strategies and performance of porous pavements for reducing combined sewer
overows. J. Hydrol. 607, 127465 https://doi.org/10.1016/j.jhydrol.2022.127465.
Torres, M.N., S´
anchez, J.P.R., de, N., Nascimento, O., Leit˜
ao, J.P., Granceri, M., 2016.
Decision Support Tools for Sustainable Urban Drainage Systems: a Systematic
Quantitative Review, vol. 7.
Ventura, D., Ferrante, M., Copat, C., Grasso, A., Milani, M., Sacco, A., Licciardello, F.,
Cirelli, G.L., 2021. Metal removal processes in a pilot hybrid constructed wetland for
the treatment of semi-synthetic stormwater. Sci. Total Environ. 754, 142221 https://
doi.org/10.1016/j.scitotenv.2020.142221.
Villarreal, E.L., Semadeni-Davies, A., Bengtsson, L., 2004. Inner city stormwater control
using a combination of best management practices. Ecol. Eng. 22, 279–298. https://
doi.org/10.1016/j.ecoleng.2004.06.007.
Walaszek, M., Bois, P., Laurent, J., Lenormand, E., Wanko, A., 2018. Micropollutants
removal and storage efciencies in urban stormwater constructed wetland. Sci. Total
Environ. 645, 854–864. https://doi.org/10.1016/j.scitotenv.2018.07.156.
Wang, S., Wang, H., 2018. Extending the Rational Method for assessing and developing
sustainable urban drainage systems. Water Res. 144, 112–125. https://doi.org/
10.1016/j.watres.2018.07.022.
Wang, Y., Zhang, X., Xu, J., Liang, C., She, D., Xiao, Y., 2021. Evaluating effects of urban
imperviousness connectivity on runoff with consideration of receiving pervious area
properties. Urban Water J. 18, 598–607. https://doi.org/10.1080/
1573062X.2021.1918182.
Xu, Z., Wu, J., Li, H., Chen, Y., Xu, J., Xiong, L., Zhang, J., 2018. Characterizing heavy
metals in combined sewer overows and its inuence on microbial diversity. Sci.
Total Environ. 625, 1272–1282. https://doi.org/10.1016/j.scitotenv.2017.12.338.
Yu, Y., Zhang, S., An, A.K., Furumai, H., 2018. Simple method for calculating hydraulic
behavior of combined sewer overow from rainfall event data. J. Water Resour.
V.A. Montoya-Coronado et al.
Journal of Environmental Management 362 (2024) 121073
15
Plann. Manag. 144, 04018061 https://doi.org/10.1061/(ASCE)WR.1943-
5452.0000972.
Zhang, K., Chui, T.F.M., 2018. A comprehensive review of spatial allocation of LID-BMP-
GI practices: strategies and optimization tools. Sci. Total Environ. 621, 915–929.
https://doi.org/10.1016/j.scitotenv.2017.11.281.
Zhou, Q., 2014. A review of sustainable urban drainage systems considering the climate
change and urbanization impacts. Water 6, 976–992. https://doi.org/10.3390/
w6040976.
Zhu, H., Schilling, W., 1996. Simulation errors due to insufcient temporal rainfall
resolution — annual combined sewer overow. Atmospheric Research, Closing the
gap between theory and practice in urban rainfall applications 42, 19–32. https://
doi.org/10.1016/0169-8095(95)00050-X.
V.A. Montoya-Coronado et al.
