A Review On Applications of Urban Flood Models in Flood Mitigation Strategies

Natural Hazards (2021) 108:31–62
https://doi.org/10.1007/s11069-021-04715-8
REVIEW ARTICLE
A review on applications of urban flood models in flood

mitigation strategies
Wenchao Qi1,2 · Chao Ma1,2 · Hongshi Xu3 · Zifan Chen1,2 · Kai Zhao1,2 · Hao Han4
Received: 4 January 2021 / Accepted: 23 March 2021 / Published online: 30 March 2021
© The Author(s), under exclusive licence to Springer Nature B.V. 2021
Abstract
As a result of climate change, urban areas are increasingly vulnerable to flooding, which
can cause devastating effects, both in terms of loss of life and property. Therefore, an accu-
rate assessment of urban flood processes and improved pre-disaster mitigation strategies in
threatened areas is essential. Urban flood modeling enables users to understand, assess, and
forecast flood conditions as well as their impact. For effective flood modeling, especially
in highly urbanized floodplains, model selection based on the contextual requirements is
a challenge. This review provides a systematic overview of the application of urban flood
modeling approaches from the perspective of urban flood strategy design sequences. In this
review, recent research advances are presented as well as suggestions for future research
topics that will improve the availability and reliability of urban flood modeling methods.
The results of this review will help urban flood managers and potential users balancing
model complexity and needs while undertaking effective flood modeling tasks.
Keywords Flood mitigation strategies · Forecasting system · Optimal operation · Real-time

control · Urban flood
1 Introduction
Presently, regarding climate change and increasing urbanization, flooding is one of the
most challenging issues facing cities (Yin et al. 2015; Dash and Punia 2019; Diakakis
et al. 2019; Li et al. 2019; Xu et al. 2019a, b). Between 1995 and 2015, climate disasters
claimed 606,000 lives, with an average of 30,000 deaths/year, wherein 47% of these events
were floods (Gran and Ramos 2019; CrED and UnIsDr 2015). Urban flooding is the result
* Chao Ma
[email protected]
1
State Key Laboratory of Hydraulic Engineering Simulation and Safety, Tianjin University,
Tianjin 300072, China
2
School of Civil Engineering, Tianjin University, Tianjin 300072, China
3
School of Water Conservancy Engineering, Zhengzhou University, Zhengzhou 450001, China
4
State Key Laboratory of Eco‑Hydraulics in Northwest Arid Region of China, Xi’an University
of Technology, Xi’ an 710048, China
13
Vol.:(0123456789)
32 Natural Hazards (2021) 108:31–62
of primarily natural and anthropogenic conditions. Natural causes of flooding are gener-
ally the result of climate change (Jamshed et al. 2020), whereas anthropogenic causes are
enhanced by urbanization (Waghwala and Agnihotri 2019). The main causes and mecha-
nisms of urban flooding are shown in Fig. 1. Urban megacities, particularly in late urbani-
zation countries, tend to be densely populated and many impervious areas undergoing
rapid growth, usually with insufficient infrastructure, leading to greater urban flooding and
environmental degradation (Mercado et al. 2020; Sakieh 2017). This trend is projected to
continue in the coming years (Lo et al. 2015; Salman and Li 2018). Therefore, urban flood
hazard prevention and mitigation strategies have been increasingly recognized as critical
issues of urban planning and management (Baldassarre et al. 2010; Amoako et al. 2019).
Increasing the incidence of urban flooding highlights the need for reliable flood miti-
gation strategies that can minimize the losses of lives and property (Koc and Isik 2020;
Li et al. 2019; Moghadas et al. 2019). For example, sustainable stormwater management
(SSM), which is an innovative stormwater management strategy that mitigates urban flood
issues, has been widely adopted in urban floodplains worldwide. Several SSM strategies
have been developed in different countries, for example, low-impact development (LID)
in the USA, the Active, Beautiful, Clean (ABC) Waters Program in Singapore, Water Sen-
sitive Urban Design (WSUD) in Australia, Low Impact Urban Design and Development
(LIUDD) in New Zealand, and the Sponge City (SC) in China. Whatever the name, the
ideas and concepts of SSM strategies proposed are very similar and all referring to the bal-
ance between hydrological cycle and their impact on the urban watershed (Buurman and
Padawangi 2018). Environmental, landscaping, and economic gains highlight the advan-
tages of SSM strategies, controlling not only the volume but also the peak flow, frequency,
and quality of runoff (Moura et al. 2015; Yazdanfar and Sharma 2015; Luan et al. 2019;
Wang et al. 2019). SSM strategies can reduce the impacts of post-development and main-
tain site’s pre-development hydrological features on urban areas. Despite an increasing
understanding and awareness of SSM strategies, the transition to a more sustainable urban
flood mitigation solutions has been slow (Elliott and Trowsdale 2007).
Existing research methods to reduce the impact of urban flooding can be divided into
two categories: qualitative and quantitative approaches (Wu et al. 2020). The qualitative
analyses mainly depend on expert knowledge and experience to assess the risk of urban
flooding, such as the multi-criteria assessments based on the analytic hierarchy process
Fig. 1 Main causes of urban flooding
13
Natural Hazards (2021) 108:31–62 33
(Moghadas et al. 2019) and the comprehensive evaluation methods based on fuzzy math-
ematics (Tang et al. 2018; Wu et al. 2020). However, these results are often influenced by
the subjectivity of the experts. Conversely, quantitative analyses typically involve estab-
lishing urban flood models based on measurable data and have been widely adopted in
urban flood mechanism analyses and process simulations. An efficient urban flood model
is an essential tool of stormwater administration and flood forecasting to sustainable urban
development. The physical process of urban flooding is accurately simulated by these mod-
els based on detailed hydraulic and hydrological data. In addition to the conventional urban
flood model, with the development of artificial intelligence technology, advanced deep
learning technologies have been successfully applied for the hazard and risk prediction in
environmental sciences. The most remarkable advantage of deep learning models is that
they can analyze the complex association patterns hidden in the database (Bui et al. 2020).
Hence, over the last decade, extensive machine learning techniques, such as support vec-
tor machines (Yan et al. 2018; Tehrany et al. 2015), random forest (Sadler et al. 2018), and
artificial neural networks (Berkhahn et al. 2019), have been widely applied for urban flood
risk assessments and mitigation strategies. An urban flood model is an effective tool for the
planning, operation, and design to flood mitigation strategies (Zhang and Pan 2014), and
the availability could act to encourage wider uptake for flood mitigation solutions, such as
SSM. Urban flood models can also demonstrate outcomes that can be used for education
and policy development, and make the application and design of flood mitigation strate-
gies more efficient. The greatest challenge is how to translate highly complex flood pro-
cesses into a computerized system instruction that allows for the evaluation of mitigation
strategies at a range of scales applicable to urban flood management. Numerous studies
have shown great interest in the optimization of stormwater facilities in small-scale control
through modeling evaluations, including the type, size, layout, combination way, number,
and location allocation of components (Bahrami et al. 2019; Giacomoni and Joseph 2017;
Zhang et al. 2013; Zhou et al. 2019). As a result, a great deal of research has combined
multi-objective optimization algorithm with urban flood models to compare and evaluate
different configurations of SSM measures and their effects. These studies aimed to help
designers explicitly develop a cost-effective urban flood mitigation strategy.
Urban flood models have undergone significant improvement over the past decades.
These theories and disciplines have been further popularized and applied in urban flood
hazard mitigation strategy assessments, which help to understand and control flood dis-
asters, while providing a certain scientific reference for government decision making (Su
et al. 2019; Zhang and Pan 2014). In particular, researches of flood modeling and map-
ping have become more widespread and interdisciplinary as a result of the participation of
hydrologists, geographers, geologists, and other experts. Nevertheless, there remain limi-
tations and challenges regarding the comprehensive understanding of various dimensions
of flood hazard mitigation strategies because of the rapidly evolving, diverse, and frag-
mented nature of urban flooding causes. As a result, there is a high uncertainty regarding
the implementation process of urban mitigation strategies. Furthermore, there has been no
systematic review of research on the applications of urban flood models using flood mitiga-
tion strategies. Therefore, this work aims to review the role of urban flood models in terms
of their impacts in the field of flood mitigation strategies. Herein, a detailed overview is
presented based on the urban flood mitigation strategy design sequence. The remainder of
the paper is organized as follows:
The remainder of the paper is organized as follows: Sect. 2 provided a comprehen-
sive analyzes of the main features of the existing modeling methods. Section 3 high-
lights the application of flood models in urban flood forecasting systems. The modeling
13
for designing of the urban flood mitigation strategies is discussed in Sect. 4. Section 5
discusses some model-based control methods for flood mitigation measures. Crucial
challenges and findings for urban flood models and relevant application to urban flood
mitigation strategies are depicted in Sect. 6. Finally, Sect. 7 summarizes the conclusions
obtained from the study.
2 Urban flood models
2.1 Classification of urban flood models
A model understood a simplified representation of real-world scenario, and urban flood

models are mainly used to predict flood scenario behaviors and analyzed various urban
hydrological processes (Devia et al. 2015). Numerous models developed by govern-
ment departments, academic institutions, and engineering consultants are literally
that are capable of simulating water quantity and quality in an urban watershed. This
review does not aim to rank and summarize all the current modeling approaches for
urban floods, but rather determine the general current situation by means of several rep-
resentative models. The goal of this review is to identify the challenges of in terms of
uncertainty of modeling detail and constantly growing data requirements. For an analy-
sis of the characteristics of particular models, consult Elliott and Trowsdale (2007). A
single concise classification of an integrated urban flood model is difficult to formulate
because of its complexity and broad scope. However, it is possible to identify the mode-
ling mechanism, spatial extent, calculation algorithm, dimensionality, and mathematical
complexity in urban flood modeling. This classification (Fig. 2) is intended to explore
the methodologies and main assumptions involved in developing these models. There-
fore, in this study, we systematically enumerate a comparison of representative urban
flood models according to the calculation algorithm (Zhang and Pan 2014).
Fig. 2 Classification scheme of flood modeling methodologies
13
2.2 Models elaboration based on calculation algorithm
2.2.1 Hydrological model
Hydrological models are vital components and essential tools for water resource, environ-
mental planning, and management (Abdulkareem et al. 2018). Urban hydrological mod-
els are primarily developed to evaluate the impact of climate change, urbanization, and
make predictions of flood forecasting on the urban areas. Hydrological models are based
on watershed hydrology, wherein cities are considered as small-and medium-scale basins
and divided into several catchments. Thus, computational methods for catchment runoff
generation and concentration are the same as those for basins. Water transfer from catch-
ments to city outlets occurs via a drainage system. In the simulation of urban flooding,
hydrological models use rainfall as an input and output the simulation results. Because of
less data requirement and high computational efficiency, these models, including the Soil
and Water Assessment Tool (SWAT), the Hydrologic Engineering Center’s-Hydrologic
Modeling System (HEC-HMS), and MIKE System Hydrological European (MIKE SHE),
are widely used in urban flood simulation and planning. These models can accurately esti-
mate the discharge of the drainage system, so as to provide solutions for most of the actual
demands of the design and implementation. Hydrological models are generally classified
based on the model parameters and input, and the extent of physical principles adopted.
Moradkhani and Sorooshian (2009) divided the models into event-based and continuous
models. The former produces only output over a specific period of time, while the latter
produces a continuous output. Abdulkareem et al. (2018) classified the models as empirical
models, conceptual methods, or physically based models.
Since the hydrological models cannot simulate the entire hydrological cycle, strict
boundary assumptions, such as incoming surface runoff, are required. Further, it is unfea-
sible for hydrological models to give a detailed description for large urban catchments.
The reason can be attributed to urban-catchment hydrology diverges from classical urban
hydraulic concepts, mainly because of scale issues. Moreover, based on the spatial aggre-
gation levels, hydrological models range between fully distributed, semi-distributed, and
lumped distributed (Hou et al. 2018). In most cases, the spatial features of basins and pre-
cipitation, and preserving the water balance in urban catchments can be considered in a
distributed model (Nourani et al. 2014; Smith et al. 2012). However, overall performance
of the lumped model is better than the distributed model (Nguyen et al. 2016). Further,
distributed models generally include spatial variability, whereas a lumped model does not
consider the spatial distribution of the input (Abdulkareem et al. 2018).
Under the background of rapid urbanization, the underlying city surface changes daily,
and the micro-features of the city and the blockage of the rainwater outlet easily affect
the water flow path, making it difficult to obtain the flow yield of each urban catchment
area according to the empirical hydrology method. Meanwhile, to obtain more accurate
simulation results, the research area must be further subdivided, which incurs difficulties
in the calibration of the model parameters. Most importantly, because this type of model
is employed for the generalized processing of cells, the node will not be able to provide
model flooding dynamic processes, such as the changing process of the surface water depth
and velocity, which is particularly important for inner city management, such as traffic
channels.
The spread of geographic information system (GIS) technology is definitely one of
the principal drivers in the development of spatially distributed hydrological models.
13
Currently, nearly all types hydrological models benefit from spatial data analysis. GIS tech-
nologies are playing a significant role in model coupling and identifying hydrological unit
responses and hydrological elements. To date, more than 80% of the urban flood modeling
approaches use GIS technologies (Salvadore et al. 2015).
2.2.2 Hydrodynamic model
Compared with the watershed flood process, the urban flood process is greatly affected
by topography, and the hydrodynamic model shows a stronger advantage in considering
the hydraulic characteristics of urban drainage systems, river channel systems, streets, and
other micro-topographies. Therefore, many researchers have used the hydrodynamic model
as a powerful tool for urban flood analysis.
Hydrodynamics models based on the shallow water equation (SWEs) have a clear physi-
cal meaning (Leandro et al. 2011; Su et al. 2019). As a result, the SWEs play an impor-
tant role in simulating hydrodynamics process in various surface flows (Hou et al. 2013;
Su et al. 2019). The process of solving SWEs is complicated, especially when they are
applied to complicated real world topographies (Su et al. 2019). Finite element and finite
volume frameworks are always used to solve SWEs numerically. In particular, Godunov-
type schemes finite volume framework has become popular in solving SWEs (Hou et al.
2013; Liang and Borthwick 2009; Liang and Marche 2009). Based on the Godunov-type
scheme, the complex shallow water flow phenomena such as shock-type flows, transcriti-
cal flows, and moving wet-dry interface of a water wave front can be well simulated and
solved (Liang and Marche 2009; Liang et al. 2010). Unstructured finite-volume schemes
can insure irregular geometries of natural water areas to furnish resolution of localized
flow structures, and conserve mass and momentum locally (Liu et al. 2013, 2015). Com-
pared with structured finite-volume methods, unstructured finite-volume methods have a
large gradient (Liu et al. 2015). However, several persistent problems exist in practice, such
as wet/dry fronts solving, shock capturing and source terms (Song et al. 2011; Liu et al.
2015).
Considering all characteristics of urban flood, to provide a reliable flooding forecast-
ing in complex urban areas, high-resolution numerical simulations are necessary to resolve
complicated urban topographic features, such as streets, traffic channel, and buildings,
whereas the high computational burden of full hydrodynamic models has greatly limited
their application to real-time flood modeling (Hu and Song 2018). To overcome this prob-
lem, different strategies are applied to solve the SWEs, including the: cellular automata
approach (Dottori and Todini 2011), adaptive meshes methods (Hou et al. 2013; Hu and
Song 2018), and speeding-up strategies. For example, Castro et al. (2011), de la Asuncion
et al. (2013), Lacasta et al. (2014), Vacondio et al. (2014), Liang and Smiht (2015), Hu and
Song (2018) utilized GPU (graphics processing unit)-based technology to overcome the
computational efficiency for fast flood modeling methods. Liang and Smith (2015) dem-
onstrated the potential for generalized methods applicable to adopt GPU and CPU (central
processing unit) coprocessors. Furthermore, some simplified modeling approaches have
been adopted to improve the computational efficiency. In addition, a diffusion wave approx-
imation which is one of the most significant methods in simplified modeling approaches
has been widely established. It simplifies the solving process by assuming that the resist-
ance forces and gravitation are always in equilibrium, and ignoring the acceleration effect.
Most hydrodynamic modeling methods are adopted by 1D, 2D, or 1D/2D schemes. A
1D model can save simulating time and require less modeling information, while a 2D
13
model produces more accurate results, but requires more time and higher terrain accuracy
(Morales-Hernandez et al. 2013). Therefore, to improve modeling accuracy and improve
computing time, a coupled 1D/2D model has been established recently and successfully
adopted to complicated urban watershed (Gejadze and Monnier 2007; Leandro et al. 2009;
Marin and Monnier 2009; Finaud-Guyot et al. 2011; Morales-Hernandez et al. 2013). Cou-
pling modeling method can be divided into external coupling, internal coupling, and full
coupling (Morita and Yen 2002; Lian et al. 2007). External coupling methods are the sim-
plest and most common in most researches. Results from one or more of these models
are adopted as the input information to other models (Lian et al. 2007). Internal coupling
methods generally share information and boundaries, and via iterative processes to achieve
the information sharing at the boundaries. The significant advantage of internal coupling
methods is that can obviously reduce computing time. Full coupling methods generally
involve reformulating the governing equation for coupled model (Lian et al. 2007). Some
authors (Morales-Hernandez et al. 2013; Chen et al. 2017) proposed the main river net-
work and sewer systems in urban areas that can be predicted by 1D model. Pipe and chan-
nel flow can be governed by the Saint–Venant equations (Liu et al. 2015; Zhang and Pan
2014). Furthermore, a 2D model is adopted to calculate the water level and flow velocity
by solving 2D shallow water equations (Liu et al. 2015). These models are linked by a weir
equation, in which the volume of flow from the 1D domain to the 2D domain is determined
by the water level difference. Another form to couple 1D/2D hydrodynamic models is to
average the 2D terms along the cross sections and imposing continuity at the interfaces. As
a result, this method achieves a transformation of 2D quantities into 1D quantities; a sub-
domain iterative procedure is subsequently performed to solve the coupled 1D/2D problem
(Miglio et al. 2005; Morales-Hernandez et al. 2013). Some recent studies proposed a cou-
pled method by stitching the center of the 2D grid cell by considering the numerical fluxes
of each model (Li et al. 2009; Bladé et al. 2012). As of late, coupling 1D and 2D mod-
els has become a popular method in urban flood research. For example, Hsu et al. (2000)
introduced a coupled 1D/2D model for urban flood process simulation. Chen et al. (2018)
outlined a new hydrodynamic model that couples the stormwater management model
(SWMM, 1D model) and the shallow water model (SWM, 2D model) and adopted the vali-
dated model to the Haikou City to identify the urban flood response to rainstorm patterns.
Wu et al. (2017) developed a coupled model that couples SWMM and LISFLOOD-FP to
simulate urban flood process.
2.2.3 Simplified models
Hydrodynamic and hydrological modeling generally requires various types data, such as
infiltration conditions, sewer conveyance data, and topography data, some of which may
be difficult to obtain. Generally speaking, it can be difficult to establish a physically based
models because of complex boundary conditions and intricate drainage systems. In addi-
tion, iterative calculations in urban flood process simulations are complex and require
substantial time to execute. All of these factors have greatly restricted the widespread
application of the large hydrodynamic and hydrological models previously mentioned.
Consequently, a simplified model as another valid simulation method has been introduced
in some researches. This model has sometimes been referred to as a 0D model (Pender
2006). Teng et al. (2017) presented a detailed overview of relative weaknesses and merits
of simplified models. Zhang and Pan (2014) divided the model into two categories based
on whether it can output the flood process. One category can simulate flood process based
13
on the Saint–Venant equations, and the other category is based on the cellular automata
method. Teng et al. (2015) constructed a simplified model, which can be adopted rapid
assessment of flooding in large floodplains to support flood forecasting and warnings. The
model combines the LiDAR DEM and the river stage to simulate the floodplain inunda-
tion. With the use of non-detailed pipe network data and publicly available elevation data,
Chen et al (2009) developed a simplified GIS-based flood model, which can greatly shorten
the simulation time, but the model is required in flat areas.
Due to these types, flood models do not consider physical process simulation and can
produce approximate predictions of final flood extent and water levels, with obvious com-
putational time benefits compared to large hydrological and hydrodynamic models. A sim-
plified model is most beneficial for that flood modeling tasks that have low demands on
accuracy of flow dynamics and do not require velocity and flood process output. Simu-
lation time-saving indicated that simplified model is more apposite for large-scale water-
sheds, and the dynamic effects are insignificant. However, simplified models also do not
represent flow dynamics and most models do not predict the velocities, which makes them
unsuitable for flood modeling tasks in urban areas with complex topography. The devel-
opment of simplified models mainly involves two categories (Teng et al. 2017): (1) solv-
ing the shortcomings of lacking representation of velocity and mass conservation, and (2)
using remote sensing data to achieve model assimilation to improve the prediction accu-
racy of simplified models.
3 Quantitative forecasting for urban flood in a modeling perspective
Urban areas worldwide are facing an increasing flood risk. Flood forecasting system
(Fig. 3) is an appropriate option to deal with the growing frequency of flood disaster. Cur-
rently, various models are capable to simulate urban flood process. As decision-making
tools, their main role is to adopt available data by executing simulations to help the under-
standing of flood in urban watershed. Hydrological and hydrodynamic models for rain-
fall–runoff process and river flow routing simulations have been widely adopted within
flood forecasting systems to mitigate the enormous influence of floods in urban water-
shed. In this section, the application of flood models in urban flood forecasting systems is
discussed.
3.1 Classification of urban flood forecasting system based on modeling methods
Flood forecasting systems could be classified in many ways. It can be classified according
to the items that constitute the standard architecture of flood forecast systems, including the
level of pretreatments, the technology of data collection, and the type of source for climate
forecasting. Herein, we focus on the application of the urban flood model in the forecast-
ing system and suggest a two-type classification as follows (Henonin et al. 2013): (1) pre-
simulated forecasting system with scenario built from previous urban flood simulations. (2)
Real-time forecasting system with online and real-time hydraulic models.
3.1.1 Pre‑simulated forecasting system
In the pre-simulated forecasting system, rainfall forecasting is adopted as input for sce-
nario selection, but a pre-study scheme that combines the modeling simulations has been
13
Fig. 3 Conceptual diagram of integrated urban flood forecasting and warning modeling framework
establish in the scenario. The estimation of rainfall design value and return is considered
to be of great significance in pre-simulated flood forecasting system. A frequency analysis
is an efficient design tool that reduces the cost of the project due to the employment of the
forecasting technique. It also plays a significant role in estimating corresponding event and
their magnitudes the return periods, thereby helping to balance the economic and hydro-
logic evaluations of flood mitigation strategies (Yin et al. 2015). Globally, there have been
developed various methods for frequency analyses. Some official flood distributions have
been recognized by specific countries, such as generalized extreme value and generalized
logistic distribution in the UK, log Pearson type III in the USA, and Pearson-III distribu-
tion in China (Yazdi et al. 2016). Multi-variable probability methods, such as gamma dis-
tribution (Skaugen 2007), exponential (Bacchi et al. 1994), Copula-based (Lian et al. 2013;
Xu et al. 2014), normal distribution, meta-Gaussian (Kelly and Krzysztofowicz 1997), and
Gumbel-mixed (Yue et al. 1999), are often adopted to study joint probability. For example,
Yin et al. (2013) investigated future scenarios of climate change and storm induced flood
of Shanghai City by adopting Pearson-III distribution. Lian et al. (2013) established the
joint probability of tidal level and rainfall based on the optimal copula in Fuzhou City.
Zheng et al. (2013) investigated the dependence between rainfall and tide level by applied a
bivariate logistic threshold-excess model.
3.1.2 Real‑time forecasting system
A real-time flood forecasting system should be adopted online rainfall data as an input; it
must comply with some real-time standards, such as rapid transfer/display of the simula-
tion results and short simulation time (Henonin et al. 2013). The accuracy of flood simula-
tion results is one main issue for this system, which is greatly influenced by the modeling
13
calibration, real-time controlled device status, and anomalous changes in drainage system.
In view of the more detailed of the urban flood model, the more difficult it is to update.
Therefore, an appropriate balance must be determined between the accuracy of flood simu-
lation results and the complexity of modeling. Moreover, effective maintenance is a critical
issue for a real-time flood forecasting system (Henonin et al. 2013). Therefore, it is better
to keep the flood model simple, so that it is easier to update and maintain, instead of using
a complicated model, because lack of effective maintenance will face a greater risk of rapid
obsolescence (Henonin et al. 2013).
3.2 Performance evaluation of urban flood models for extreme flood forecasting

system
As a typical non-structural flood mitigation measure, the purpose of flood forecasting sys-
tem is to prevent repetitive flood damage (Choo et al. 2019). Various types of urban flood
models have recently been adopted for flood forecasting. For example, Li et al. (2014)
adopted the data assimilation approach and support vector machine (SVM) to simulate the
flood process for real-time flood forecasting. Lin et al. (2013) established a flood forecast-
ing model; SVM was applied to predict the control point flood forecasting module to gen-
erate 1–3 h lead time flooding maps. Yagi and Shiba (1999) reproduced a flood process by
combining the SVM with Mike FLOOD. Bertsch et al. (2017) developed an urban flood
forecasting framework by the 1D and 2D flood simulation of CityCAT. Jhong et al. (2018)
successfully established a flood forecasting framework by combining the multi-objective
genetic algorithm with SVM. A few urban flood forecasting systems based on modeling
methods are listed in Henonin et al. (2013).
An effective flood forecasting system is increasingly important for reducing flood-
related property and casualty losses. However, there are many uncertainties in the flood
forecasting system (such as the model accuracy and input data), resulting in differences in
time, flood volume, and spatial extent between predicted the flood process and the actual
flood process. Two main factors affect the quality of a flood forecasting system, including
(1) the treatment and representation of rainfall data quality forecasting, and (2) appropriate
model selection for rainfall–runoff simulation (Roodsari et al. 2019).
• Urban flood model quality
Physically driven models and data-driven models are widely used for forecasting systems,
of which each has advantages and limitations. A 1D urban flood model is an effective method
for understanding and managing drainage network processes, and 1D urban flood model simu-
lations are normally fast and stable (Liu et al. 2015; Morales-Hernandez et al. 2013; Finaud-
Guyot et al. 2011). But, 1D urban flood model would be greatly limited when the network
overflows occur on the city surface. Modeling study the urban flood process requires a 2D
model. To date, when the impact of the drainage network must be considered, it is recom-
mended to use a one-dimensional/two-dimensional coupling method to describe the interac-
tion between the storm drainage network and the urban surface. While physically driven mod-
els are accurate tools for flood forecasting system, the large input requirements are required,
subjecting them to the known problems of over-parameterization (Beven 2006; Roodsari et al.
2019). Additionally, it is required a high simulation time to the execution of physically based
urban flood models. Data-driven models have been widely adopted as the model basis of flood
forecasting systems due to their flexibility. However, data-driven models adopt mathematical
13
equations analyzing the input and output time series instead of simulating physical processes
(Nourani et al. 2014; Roodsari et al. 2019).
Furthermore, flow measurement at the outlet or inlet of the catchment area and drainage
network is greatly helpful to enhance and validate the flood process estimation of the urban
flood model. However, there are few spatial and temporal verification data in urban watershed.
Fortunately, recent technology such as crowdsourcing method has led to new sources of data
to validate flood process maps (Poser and Dransch 2010; Bhola et al. 2019).
• Rainfall data quality
The quality of rainfall forecasting has a significant impact on the urban flood forecasting
system (Henonin et al. 2013; Rene et al. 2013). A rain gauge is the most adopted direct rainfall
measurements. Rain gauge is reliable for rainfall point measurements, while it cannot capture
rainfall in space and time. A weather radar is another popular method of rainfall monitoring
and forecasting due to it has a wider coverage area than rain gauges. In addition, the weather
radar can generate continuous, urban-scale, and real-time data, there is a great interest in rain
data obtained from radar measuring for urban flood models execute in operational flood fore-
casting system. However, it is not reliable for a local level rain forecasting and monitoring
by adopted a national radar. Thus, combining the micro-scale radar with mesoscale and mac-
roscale radar is likely a sustainable solution to obtain real-time rainfall data (Henonin et al.
2013). Therefore, before using radar data as a rainfall source for urban flood models calibra-
tion, careful selection and validation are necessary.
4 Model for designing of the urban flood mitigation strategies
In the last two decades, the number of studies on designing urban flood control measures and
their impact on urban flooding has greatly increased, making it difficult to follow all develop-
ments. Some researchers found that urban flood measures function differently relying on their
scale of implementation (Demuzere et al. 2014). For example, Fiori and Volpi (2020) showed
that LID measures are more effective in reducing the peak flow in a local area, and its effec-
tiveness will great decreases as the catchment size increases. Therefore, in this section, the
design of flood control strategies under different watershed scales is analyzed, and the follow-
ing classification is suggested by Haghighatafshar et al. (2017):
(1) Microscale, at a single flood mitigation measures scale (e.g., green roof, rain garden,
and grassed swale) was investigated under site-specific conditions.
(2) Mesoscale, wherein a group of several flood mitigation measures are implemented at
a catchment/community scale.
(3) Macroscale, wherein urban flood mitigation strategies are placed into a broader context
of stormwater management framework (such as SSM) and are studied from a concep-
tual framework perspective at the city-scale.
13
4.1 Flood measures designing under microscale condition
4.1.1 Gray infrastructure
Traditional stormwater systems mainly rely on drainage systems, sometimes called the
gray infrastructure (Zischg et al. 2017). Gray infrastructure (Fig. 4) is a well-designed
system that can route stormwater directly to downstream water bodies or be collected by
detention facilities (Prudencio 2018). Herein, drainage systems are of particular interest.
In urban watershed, stormwater is mainly collected by a drainage system. In the tradi-
tional drainage systems, rainwater system and wastewater pipes are usually used in com-
bination, but modern drainage systems usually adopt separate paths, which can quickly
eliminate runoff to prevent flooding.
Designing a drainage system is the first step in evaluating its performance. Diam-
eter of drainage pipe is usually chosen based on pre-selected slope; the constraining
condition normally should consider the peak flow and velocity. Traditionally, the peak
flow estimate is relied on the empirical formulas. It is obvious that the combined effect
of climate change and urbanization will have a significant impact on drainage systems.
Although many researchers have investigated the impact of urbanization on climate
change and how to incorporate the flood mitigation, few studies consider both impacts
simultaneously. Obviously, if these two driving factors are not considered, the planning
framework of the drainage systems will not be able to adapt to future changes. Com-
pared with traditional design methods, modeling designing considers some factors such
as the interaction between the environment, drainage systems, and society. In most cit-
ies, a drainage system is designed based on models that permit for a risk of flooding
with a return period of 2a–100a relying on the size of the catchment and the monetary
values. In addition, some components characteristics, such as buried depth of pipes and
conduit size are designed to satisfy certain constraints and achieve design objectives.
Some commercial packages, such as synchronous optimization and simulation of the
urban wastewater system (SYNOPSYS), WEST, and integrated catchment simulator
(ICS), are proposed to simulate water quality and quantity (Butler and Schuetze 2005;
Rauch et al. 2002). With the development of rapid computer advancements, some con-
tinuous models suitable for urban watershed including the SWMM, King County Runoff
Time Series (KCRTS), Quality–Quantity Simulator (QQS), Storage, Overflow, Runoff
Model treatment (STORM), Hydrologic Simulation Program-FORTRAN (HSPF), and
Distributed Routing Rainfall–Runoff Model (DR3M) were found in studies (Jackson
et al. 2001; Zoppou 2001).
Fig. 4 Examples of gray infrastructure measures for stormwater management
13
An equally important consideration that is of continued interest to researchers is the

optimization of sewer drainage systems based on urban flood models coupled with an opti-
mization algorithm. Cembrano et al. (2004) established an optimal combination of net-
work control to reduce the risk of flooding and prevent overflows based on an optimization
model. In addition, the operation process of the drainage system is usually combined with
real-time control (RTC) methods to achieve the maximum flood control effect. For exam-
ple, Lowe et al. (2016) investigated an optimization model based on stochastic forecast
optimization model for RTC of urban drainage systems to minimize the combined sewer
overflow (CSO). Generally speaking, a RTC operation process is one of the most advanta-
geous solutions to reduce the overflow and frequency in combined sewers (Duchesne et al.
2001; Lowe et al. 2016) and has been widely discussed in studies (Pleau et al. 2000; Garcia
et al. 2015; Pleau et al. 2005; Vezzaro and Grum 2014).
4.1.2 Green infrastructure for stormwater on flood management
Various green infrastructure measures (Fig. 5) have been adopted to minimize urban flood
hazards, such as permeable pavements, green roofs, and rain gardens. Sustainable storm-
water measures aim to control pollutant load and flood risk through decentralized and
small-scale control measures (Freni et al. 2010; Beecham et al. 2012). Herein, we only
considered the effect of these structural measures on urban flood reduction. The studies
on modeling assessment sustainable stormwater measures can be divided into external and
internal factors (Sohn et al. 2019). The capacity of sustainable stormwater facilities to store
and release stormwater during storm events are generally determined by the internal struc-
ture, including soil composition, facility scale, layer structure, and vegetation type. Mean-
while, external factors of sustainable stormwater measures can be classified by climate
characteristics, such as storm events change from high to low intensity, from long to short
duration, and from hot to cold of weather condition (Sohn et al. 2019). For example, based
on a hydrodynamic model, Hou et al. (2020) simulated a single rain garden and found that
Fig. 5 Examples of green infrastructure measures for stormwater management
13
its runoff-control rate was 53.02–100% with a return period of 0.5a–5a. Gagliano et al.
(2016) adopted a dynamic model (DesignBuilder) evaluate the performance of the green
roof during the summer period. Liu and Chiu (2019) adopted the SWMM to evaluate the
hydrological performance of green roofs under various rainfall conditions.
4.2 Flood measure designing under mesoscale perspective
The efficacy of flood mitigation strategies regarding the mitigation of urban floods at the
catchment/community scale remains a complex question. The purpose of this section is to
provide insights into flood mitigation strategies at the catchment scale, as guided by two
main research methodologies.
4.2.1 Coupled urban flood models with Optimizing methods
Optimizing methods commonly used in research articles include genetic algorithm (GA),
particle swarm optimization (PSO), simulated annealing (SA), ant colony optimization
(ACO), artificial bee colony (ABC), harmony search (HS), and cuckoo search (CS). Gener-
ally speaking, a mathematical optimization model aims to obtain the “best” result for its
objective function without violating constraints. The mathematical optimization model can
usually be defined as Shishegar et al. (2018):
max(min)f (x),
gi (x){≤, =, ≥}bj ∀i ∈ I,
xj ∈ S ∀j = 1, 2, ..., n,
where x is a set of decision variables, f(x) is the objective function, and gi(x) represents all
the functions that, together with bi, the boundaries, and S, the set constraints on x, deter-
mine the problem constraints. Note that a problem can have more than one objective func-
tion. In this case, the objective mathematical optimization model is to satisfy all constraints
(Shishegar et al. 2018).
A complete urban flood model offers more detailed simulations and presents results for
the entire urban flood process. This helps to quantify the flood water depth, inundation
extent, duration, and water flow velocity. Optimizing methods can solve nonlinear and com-
plex problems as well as deal with incomplete data (Shishegar et al. 2018). The urban flood
model coupled with optimization method is an extremely important tool for urban rainwa-
ter management. In addition, the optimization method provides a mechanism to automati-
cally execute a series of systematic models and find the optimal solution from a series of
possible results. These results are mathematically defined by the objective function. It is of
increasing interest to employ an optimization analysis coupled with an urban flood model
to examine the role of flood mitigation strategies developed in the recent studies. In the
planning of urban flood mitigation strategies, the optimization aim is to search for a trade-
off between the goals of flood mitigation, sustainability, and cost-effectiveness. Exploring
this trade-off through a multi-objective optimization and analysis has been widely studied.
For example, optimizing stormwater facilities includes the type, size, layout, combination
way, number, and location allocation of its components. Note that the main constraint fac-
tors for stormwater facility optimization include three themes (Bakhshipour et al. 2019):
13
(1) design objectives, such as the reduction of runoff and peak flow volume reductions; (2)
site characteristics, such as degree of urbanization, topography, climate features and land
use; and (3) cost of stormwater facilities.
Furthermore, we distinguished the optimization of urban flood mitigation strategies
based on the following perspectives: (i) the stormwater management approach, which can
be gray stormwater facility (such as drainage system and storage tanks) optimization, green
stormwater facility (such as bioretention, porous pavements, green roofs, detention/reten-
tion basins, wetlands, and vegetated swales, which are the basic technologies of all SSM
practices) optimization and coupled gray with green stormwater facilities optimization. (ii)
A control method that can be static or dynamic, and (iii) objective benefits, either eco-
nomic or environmental. Table 1 summarizes mitigation strategy optimization studies and
their modeling characteristics.
Optimization methods allow for the identification of an optimal solution set from a large
number of results. However, there remain some research gaps. First, an optimization model
is a “black-box” approach, which lowers the confidence of city planners in the optimization
results. Second, optimization often leads to non-unique solution sets. Third, most of the
previous studies have focused on coupled simulation–optimization methods, which nor-
mally require huge computational burden (particularly for 2D flood inundation modeling).
Thus, we must design a more efficient way to evaluate future designs.
4.2.2 Scenario analysis based on urban flood models
The shortcomings of optimizing methods have led many researchers to turn to scenario
analysis methods for optimal flood measure design. Compared with optimizing models,
which are objective driven approaches, scenario analysis methods are driven by a set of
influencing factors, wherein each planning scenario is often designed based on certain pre-
requisites. From the perspective of flooding, environmental, and technical severity, Dong
et al. (2017) designed three system configurations including green roofs, permeable pave-
ments and storage tanks, and their resilience to extreme rainfall was compared. Casal-Cam-
pos et al. (2015) investigated more planning scenarios by integrating regret-based methods
to evaluate the relative performance of strategies in multiple impact categories and their
robustness. However, the quality of scenario assumptions greatly influences the reliability
of scenario analysis (Urich and Rauch 2014; Zischg, et al. 2017). In addition, the inability
to identify all potential scenarios, scenario analysis does not seek the most cost-effective
solutions, which often result in schemes far from pareto optimality (Liu et al. 2016b; Xu
et al. 2017).
4.3 Flood measures designing under macroscale perspective
In this study, modeling assessments at the urban catchment scale were found to stress
the theoretical framework of stormwater management. Traditional stormwater manage-
ment methods sought to eliminate stormwater from a flooding site as soon as possi-
ble and then store large flood volume in downstream facilities to control the peak flow.
Traditional flood risk management methods are the source of severe challenges facing
urban watershed, and their inherent characteristics make them unsuitable for solving
future sustainability issues. More recently, the new focus approaching sustainability has
been studied and techniques of SSM (such as LID measures) are widely recommended
and applied in many parts of the world. SSM methods emphasize a comprehensive
13
46
Table 1 Summary of stormwater management optimization model characteristics found in the studies
Reference Optimization objects Flood mitigation measures Optimization algorithm Model
13
Type Location Size Combi-
nation
way
Bahrami et al. (2019) √ √ Bioretention cells and porous pavement GA SWMM

Bakhshipour et al. (2019) √ √ √ √ Rain barrels, infiltration trenches and Hanging garden algorithm SWMM
sewer system
Damodaram and Berglund (2013) √ √ Rainwater harvesting and permeable GA SWMM HEC-HMS
pavement
Giacomoni and Joseph (2017) √ Green roofs and permeable pavement Nondominated Sorting GA SWMM
Huang et al. (2018) √ Green roof and bioretention cell SA SWMM
Liang et al. (2019) √ Bioretention cell, permeable pavement GA SWMM
Liu et al. (2016a) √ √ Grassed swale, green roof, bioreten- Multi-algorithm genetically adaptive L-THIA-LID
tion, etc. multiobjective
Liu et al. 2016b √ √ √ Retention pond, rain barrel, porous Multi-algorithm genetically adaptive L-THIA-LID
pavement, etc. multiobjective
Tao et al. (2014) √ Stormwater detention GA SWMM
Xu et al. (2019a, b) √ √ √ Porous pavement, bioretention, rain GA L-THIA-LID
barrel/cistern
Xu et al. (2017) √ Permeable pavement, concave-down GA SWMM
greenbelt
Zeng et al. (2020) √ √ Bioretention cell, permeable pavement, GA SWMM
green roof
Zhang et al. (2013) √ √ √ Bioretention, porous pavement and Epsilon-non-dominated sorting GA SWMM
green roof
Zhou et al. (2019) √ √ √ Pipe system, permeable pavement, GA SWMM
green roof, etc.
solution within the urban landscape, emphasizing source control in the urban area to
minimize changes in the urban water cycle. Note that the terminology varies in differ-
ent regions, but the designs have similar philosophies. Further, the concept of resilient
cities has proposed in several studies. Compared with SSM methods, the flood control
concept of resilient cities not only focuses on pre-disaster defense but also emphasizes
the resistance to disturbances and the speed of restore to counterpoise. Ahern (2011)
indicated that the adaptability is the basis for discussing resilience, and the concept of
sustainability is evolving from fail-safe goals to the concept of resilient systems. The
development and evolution of these terms are shown in Fig. 6.
Resilience and sustainability are a wide and complicated concept that are difficult to
express quantitatively. From a pragmatism perspective, few mathematical models have
tangible capabilities to quantify resilience. Although the design and implementation of
city-wide scale SSM retrofits are complex, the basic components applied in flood miti-
gation measures (such as LID measures) are seemingly similar. These studies mainly
compare the performance before and after construction.
5 Optimal operation of the flood measures based on a modeling

method perspective
Land cover and climate change can make significant changes in the characteristics of
urban areas, which can have significant impact on runoff. Current flood management
practices, such as detention/retention basins, do not adapt well to these continuous
changes. Therefore, adaptive flood measures operation must be integrated into an urban
flood strategy (Bilodeau et al. 2019; Shishegar et al. 2019). In general, the operation of
flood measures can be divided into static and RTC (Bilodeau et al. 2019). Further, the
literature divides operation of the flood measures into RTC strategies and static model
strategies (Garcia et al. 2015). This section discusses some model-based operation
methods for flood mitigation measures.
Fig. 6 Emergence of new water management concepts over time
13
5.1 Static control operation
In most cases, the outlet of the discharge remains fixed, and the discharge rate of storm-
water only depends on the change of the water level, which is known as static control. In
general, once green facilities are designed, their operation management is static, whereas
in gray facilities, such as pump stations and drainage systems, two control operations can
be adopted to determine the most effective operation method. Traditionally, static control
operation approaches are classified as optimization and simulation methods.
Traditionally, the major measures to prevent flood hazards are pumping stations, storm
flood detention ponds, and drainage systems. Regarding the static control of these meas-
ures, previous researches usually adopted simulation methods to establish flood models
and discussed the operating results of various models. Some examples of these researches
are presented herein. Hsu, et al. (2000) executed an urban flood model that combines sewer
model with a 2D diffusive overland-flow model to simulate urban flood caused by over-
loading of pumping stations and drainage systems. Chang et al. (2008) adopted the fuzzy
neural network method to the urban flood control system, and the pump and gate opera-
tion records are used as the model training sample input. The main advantage of the fuzzy
neural network includes two categories, namely (1) complete the complex conversion
from artificial intelligence to fuzzy logic operating system, and (2) effectively solve highly
nonlinear control problems. Based on the counter-propagation fuzzy neural network and
adaptive neuro-fuzzy inference system, Chiang et al. (2011) established an efficient and
accurate pump station operation model, the simulation results proved the applicability and
reliability of adaptive neuro-fuzzy inference system in the automatic control of urban flood
mitigation strategies. Recently, optimization methods have been widely adopted in flood
mitigation analysis. For example, Yagi and Shiba (1999) combined the GA with fuzzy
logic control to improve pump operations. Yazdi et al. (2016) considered the randomness
of rainfall events and designed the long-term operation rules of the pumping station by
coupling the mathematical model with the new hybrid HS. The results show that, com-
pared with the current operating rules of pumping stations, it has higher efficiency in flood
control and pump performance.
5.2 Real‑time control operation
5.2.1 Definition of Real‑time control operation
The static operation of flood measures (such as stormwater detention basin, urban drainage
system and flood pumping stations) can be changed to dynamic operation by controlling
the real-time approach. A RTC operation is generally based on pre-established rules such
as weather and/or hydraulic conditions, which can control the discharge rate, filling and
emptying rate, and storage capacity in real time (Bilodeau et al. 2019). RTC is an online
method, and wherein the important process variables, such as rainfall process and system
variables, are continuously monitored, and controllable elements (such as gates, pumps,
and weirs) are operated to achieve the optimal use of flood mitigation measures. For effec-
tive stormwater management and flood mitigation, a multi-component stormwater control
system in Fig. 7 involves the incorporation of pumping station, detention pond and drain-
age pipes. The control gates can regulate the flow through inflow and outflow informa-
tion measured by sensors. The pump station can divert excess rainwater that exceeds the
13
Fig. 7 Conceptual diagram of RTC operation for stormwater management measures
drainage capacity of the drainage pipes to avoid overflow. The role of the detention pond
can be used to reduce the peak flow and recycle excess stormwater. In the whole system,
a RTC strategy can be applied according to the manager’s stormwater management goal.
An effective operation system of flood mitigation measures that considers predicted
data, in addition to historical data, can adapt to varying environmental conditions. There-
fore, compared with static operation strategies, employing RTC strategies for flood meas-
ures provides flexibility to urban stormwater management systems. Some commercial
flood models such as Personal Computer Storm Water Management Model (PCSWMM),
INFOWORKS, and MOUSE, simulate local or extended reactive control at various types
controllable elements, thereby allowing users to assess the potential for RTC. When cou-
pled with urban flood modes, RTC has been widely discussed as an adaptive operation
method for flood mitigation in the studies.
5.2.2 Modeling method of flood mitigation measures for RTC operation
The existing researches of flood mitigation measure based on RTC operation can be
divided into two basic control patterns: (1) rule-based methods typically allow the opera-
tor easy to understand, considering the possible scenarios that may occur during the flood
period, and formulate appropriate control rules to determine control actions, and (2) inte-
grating optimization techniques into control of set points, wherein the method based on
the optimization algorithm usually requires more simulation time and constraint expression
of flood measures, but does not rely on expert knowledge. These algorithms can generate
control optimal operation that produce effective mitigation flood performance. In general,
integrating control rules at the outlet of the flood mitigation measures to enable it to adapt
to weather change. However, the integration of optimization technology into the scheme of
controlling the set points provides a more dynamic solution to flood mitigation strategies
that are applicable to measures of different scales and types. The objects of RTC are deten-
tion pond, sewer drainage systems, and flood pumping stations, and its control elements are
pumps, gates, weirs, and pipe outlets.
In addition, in consideration of global and local goals of RTC operation require a differ-
ent number of system information. The centralized solution needs to use more information
about the entire system, so it can achieve both local and global goals. Conversely, when
the system only needs to consider local goals, a non-centralized control strategy should
be selected. Furthermore, a centralized solution requires a more expensive communication
13
50
13
Table 2 Summary of application of RTC based on urban flood models found in the studies
References Control measure RTC operation objectives Control pattern Model tool
Rjeily et al. (2018) Urban drainage system Increasing the retention tank capacity Optimization-based EPA-SWMM
Darsono and Labadie (2007) Urban drainage system Minimizing the pollution impacts of CSO Optimization-based UNSTDY
Jafari et al. (2020) Urban drainage system Mitigating urban flood Optimization-based EPA-SWMM
Jafari et al. (2018a) Urban drainage system Flood mitigation efficiency Optimization-based EPA-SWMM
Pleau et al. (2005) Urban drainage system Improving runoff quantity and TSS concentration Optimization-based SWIFT
Jafari et al. (2018b) Pumping station Mitigating urban flood Optimization-based EPA-SWMM
Bilodeau et al. (2019) Detention pond Delaying basin peak flows and improving water quality Rule-based PCSWMM
Di Matteo et al. (2019) Detention pond Reducing flood peak flows Optimization-based EPA-SWMM
Gaborit et al. (2013) Detention pond Mitigating the urban runoff and improving water quality Rule-based EPA-SWMM
Sharior et al. (2019) Detention pond Improving basin water level, detention time and water quality Rule-based EPA-SWMM
Shishegar et al. (2018) Detention pond Reducing peak flows and improving the water quality Optimization-based PCSWMM
structure than a non-centralized control method. In Table 2, several urban flood models
adopted in the operation of control schemes are listed, all of which are able to execute RTC
operation. This section summarizes the research of urban flood models applied to the RTC
operations of urban flood measures to alleviate urban flooding.
Over the last few decades, RTC operations to urban flood mitigation measures have
been widely investigated. The operation of existing flood control measures benefited from
the RTC operation because (Borsanyi et al. 2008): (1) most flood mitigation measures are
designed based on static rules. However, a flood mitigation strategy is usually executed
under a dynamic loading condition, and (2) climate change has increased the frequency and
intensity of urban flood; it is necessary for flood mitigation measures to the adapt a new
hazard scenario.
• Optimization of operational management practices for drainage systems
RTC operation has been widely adopted in combined sewer systems to alleviate harm-
ful CSO incidents. Jafari et al. (2018a) presented two operations of RTC approach and
long-term optimal operation in urban drainage systems for flood mitigation; the results
inferred that the RTC operation is superior to long-term optimal operation. Pleau et al.
(2005) revealed that RTC optimal operation led to the reduction in overflow of up to 60%
in storm events compared with a static control method. Beeneken et al. (2013) discussed
the application of RTC operation in urban drainage system using the Dresden City drainage
system in Germany as a real-world implementation operation example. Darsono and Lab-
adie (2007) presented an optimal RTC framework of combined sewer system, which has
an effective adaptive learning ability to satisfy the time constrains of RTC operation. Jafari
et al. (2018b) compared the performances of single-period and multi-period in deriving
real-time optimal operating to a drainage system. As previously stated, there exist various
control objectives when applying an RTC approach to a drainage system, including:
• Minimize urban flood risk,

• Minimize the CSO,
• Minimize operation costs,
• Minimize water pollution,
• Better balance flows at wastewater treatment plants.
• Optimization of operational management practices for the detention pond
A stormwater detention pond is devised to capture stormwater to delay peak flows

(Sharior et al. 2019), especially during large storm events. As events of this magni-
tude seldom occur, these detention ponds commonly have unutilized storage poten-
tial (Lee and Li 2009; Park et al. 2014). A major disadvantage of detention ponds is
that they can hardly provide detention for smaller storm events. RTC operation can be
regulated over time to adapt detention ponds performance by opening and closing the
outlets of detention ponds as well as operating the stormwater pumps in accordance
with set operating rules. Further, it is becoming increasingly popular to reduce redun-
dant storage (Gaborit et al. 2013). Several studies have concentrated on an RTC oper-
ation concerning stormwater detention basin management. For example, in an early
study of rule-based control systems, Jacopin et al. (2000) successfully demonstrated
that RTC operation can both increase detention time to promote sedimentation during
13
small storm events and maintain the maximum capacity for flood mitigation during
large storm events. Gaborit et al. (2013) adopted control rules of rainfall process and
pond depth to reduce the discharge rate and maximize the hydraulic detention time
of the retention pond. Shishegar et al. (2019) proposed a RTC optimal framework of
detention pond to reduce peak flows and improving the quality of outflow, the results
indicated that a peak flow reduction from 73 to 95% compared with static control
method. Rohrer and Armitage (2017) compared the performance of decentralized and
centralized detention basin systems, finding that decentralized detention basin systems
maximize the available storage within a catchment, and showing that harvesting storm-
water in a centralized manner is more economically attractive despite its lower storm-
water yields. As previously stated, various control objectives exist when applying RTC
approach to a detention pond, including:
• Maximize the detention time of stormwater,

• Minimize the hydraulic shocks induced to the receiving water bodies,
• Minimize peak flows of urban flood,
• Delay peak flows of urban flood.
• Optimization of operational management practices for pumping stations
A pumping station is extremely important in urban flood mitigation strategies, and its
function is generally to transfer stormwater through rivers to downstream. When the
water level of the river rises and the drainage of stormwater has been hindered, the
gate of urban drainage system would be closed to prevent the river water from flowing
into the urban watershed. Under this condition, urban stormwater can only be drained
by pumping. Thus, the utilization of pumping station is crucial to flood mitigation
when the drainage gate is closed (Hsu et al. 2013). Traditionally, the decision making
on the operation of the pumping station is based on the subjective formulation of the
current water level observed in the front pool. Such solutions may be subject to errone-
ous and uncertainty probability (Hsu et al. 2013). Applications of the RTC operation
have been widely investigated, demonstrating that real-time pumping station operation
for flood mitigation are essential. For example, Hsu et al. (2013) compared the perfor-
mance of two RTC optimal operation of pumping station for urban flood mitigation.
Jafari et al. (2018b) developed a RTC multi-period optimization method to find the
best operating strategy for the pump and gate at each decision time during the storm
event. In addition, in order to reduce the simulation time of the model, a new model
was presented in which the number of decision variables was reduced by half com-
pared with the previous methods. Wei et al. (2014) presented a two-stage RTC optimal
operation of pumping station to predict the desired pump flow and the corresponding
optimal combination of active pumps. Various control objectives exist when applying
RTC to pumping stations, including (Hsu et al. 2013):
• Minimize the front-pool water level of the pumping station during a storm event,
• Minimize the absolute difference between the final and target front-pool water
level,
• Minimize the number of open and close of pumping station during a storm event.
13
6 Challenges and findings
Having evaluated the literature concerning urban flood models and relevant applica-
tion to urban flood mitigation strategies, some present knowledge gaps were identified.
While we noticed an upward trend toward the application of urban flood models, there
are still many challenges in this field. The details are as follows:
(1) Based on the studies of the current urban flood models. It is a current challenge for
developers to balance accuracy, stability, and computational efficiency of urban flood
models by establishing an optimized model structure. The urban flood models have
been developed rapidly, but the reliability and efficiency of the models still need to
be improved. For example, simulation techniques of the exchange of surface flow and
sewer flow are based on local-scale simulation. Some existing urban flood models often
directly adopt simplified methods and empirical equations, which still lack accuracy.
Excessive assumptions and simplifications will lead to errors in the simulation of the
urban flood process. In addition, the simulation cost is especially important for high-
resolution modeling tasks. Despite the rapid development of accelerated methods (such
as GPU-based algorithms and cloud computing techniques), it is still impractical for
2D models to serve real-time simulations of early flood forecasting.
(2) The input data of the model (such as high-precision DEM data and short duration
rainfall) greatly affect the simulation results of the model, which will have a negative
impact on the implementation of the rain and flood management measures. Currently,
data sharing is a major obstacle in developing countries as it is often controlled by local
governments or meteorological units. Therefore, in the future model development pro-
cess, it is necessary to consider how to solve the problem of model input data, and that
some employed method can transform low-precision data into effective high-precision
model input data.
(3) A wide range of existing models rely heavily on empirical parameters. This reliance
leads to subjective factors in modeling that greatly affect the accuracy of the model.
Therefore, the application of real-time monitoring and RTC technology in the process
of model parameter selection can greatly simplify parameter selection and human-
induced parameter errors. Further, there must be more linkages to calibration tech-
niques and the model predictions should be compared with field data.
(4) The flood process analysis serves the formulation of urban flood mitigation strategies.
Future urban flood models should attempt to incorporate decision aid tools as the
current research does not list or discuss the available conceptual framework. These
research areas have been identified as productive topics for future studies.
(5) The design and implementation of city-wide flood mitigation strategies retrofitting
is complex. One such complexity is the competition for space in the urban landscape
between the frequently divergent interests of numerous stakeholders. Currently, there
is no clear modeling methodology to measure the “sustainability” of SSM measures.
Therefore, it is difficult to estimate the value of employing flood mitigation strategies.
(6) To date, most research on flood mitigation measures focuses on the modeling analysis
of specific areas. However, in the future, many cities that have not experienced floods
may become victims to flooding as a result of climate change. Therefore, it is necessary
to create a method system for flood control and drainage based on the existing research
that includes model development, urban rainwater management, and flood mitigation
13
measures. Only by forming a systematic method system for urban flooding can we
better serve the decision-making bodies that make flood control strategies.
7 Conclusions
Herein, we reviewed state-of-the-art prevailing flood modeling methods based on their

associated potential and limitations for modeling urban floods in urban watershed. We
found that there are various flood modeling methods, each of which has its own drawbacks
and advantages for applications in urban flood mitigation strategies. In consideration of
the importance of flood modeling in flood risk management, we provided a comprehensive
understanding of the main features of the existing urban flood models. Accordingly, we
identified what areas would help urban flood managers and potential users undertake effec-
tive modeling tasks, balancing both modeling requirements and their needs. Our findings
are as follows:
• Urban flood modeling and simulation play a significant role in mitigating flood haz-
ards. A wide range of existing models, including conventional stormwater models, are
able to predict urban flood processes. However, selecting the appropriate model for a
specified purpose is one of the most essential elements in urban flood modeling. Some
models are better for urban watershed dealing with small sub-catchments, while others
are better for large catchments. A 1D hydraulic model is normally very fast and stable,
while this model becomes limited because of sewer network overflows and city surface
flow. A 2D hydrodynamic model was proven to be useful and accurate in the simula-
tion of urban flood modeling tasks, but the high-resolution DEM input requires a long
computation time. In addition, another aspect to should be considered is the availability
and usability of the model. The cost of commercial models is often high, and their
availability is limited. Conversely, the open-source models require only nominal cost,
but the user needs more technical knowledge to execute the models.
• Urban flood model is a key element in flood forecasting systems. Herein, a classifica-
tion was presented based on urban flood models used as a study tool or as a real-time
tool. Urban flood modeling is considered to be the core of flood forecasting systems,
and it relies on a proper database, such as DEM and rainfall data. Models selection
should be based on the level of investment needed as well as on flood context and
actual forecasting requirements. A 1D urban flood model is an incontestably valuable
tool for understanding and managing drainage network processes, and 1D simulations
are normally very fast and stable. Meanwhile, 2D hydrodynamic models are accurate in
representing urban flooding, but their long computation times constrain them to real-
time simulated scenarios.
• In the process of urban flood mitigation strategies, a model is usually used to evaluate
and optimize structural measures. In general, urban flood mitigation measures not only
rely on gray infrastructures but also green infrastructure. Gray infrastructure mainly
includes the optimization of urban drainage systems and stormwater detention ponds,
whereas green facilities primarily consider location, size, layout, and a combination of
different types of measures. However, there are few studies that have coupled gray and
green facilities, which is also a trend of urban flood mitigation strategy optimization.
Further, urban flood models combined with optimization algorithms have also been
used in urban flood processes and urban flood mitigation.
13
• Adaptive flood measures should be integrated into urban flood strategies. The opera-
tion of flood measures can be divided into static control and RTC operation. In gen-
eral, once the green facilities are designed, their operation mode is in a static state,
whereas for gray facilities, such as pump stations and drainage systems, two control
operations can be adopted. Regarding the design and simulation task models based on
static control, time is not an essential factor as these models do not require online com-
putation. Meanwhile, for models aimed at RTC tasks, time is a critical variable because
the model must execute a large number of control actions within a short selected sam-
pling time. Compared with static operation strategies, employing RTC strategies for
flood measures provides flexibility to urban stormwater management systems. Some
urban flood models were used in the operation of control strategies perform RTC, such
as PCWMM, INFOWORKS, and MOUSE. To optimize the operation of flood control
measures and reduce the error probability, it is necessary to develop intelligent and
RTC technologies combined with urban flood models.
Acknowledgements This study is supported by the National Natural Science Foundation of China (grant
number 51679156).
Declarations
Conflict of interest The authors declare no competing financial interests.
References
Abdulkareem JH, Pradhan B, Sulaiman WNA, Jamil NR (2018) Review of studies on hydrological model-
ling in Malaysia. Model Earth Syst Environ 4:1577–1605. https://doi.org/10.1007/s40808-018-0509-y
Ahern J (2011) From fail-safe to safe-to-fail: sustainability and resilience in the new urban world. Landsc
Urban Plan 100:341–343. https://doi.org/10.1016/j.landurbplan.2011.02.021
Amoako C, Cobbinah PB, Darkwah RM (2019) Complex twist of fate: the geopolitics of flood management
regimes in Accra. Ghana Cities 89:209–217. https://doi.org/10.1016/j.cities.2019.02.006
Bacchi B, Becciu G, Kottegoda NT (1994) Bivariate exponential model applied to intensities and durations
of extreme rainfall. J Hydrol 155:225–236. https://doi.org/10.1016/0022-1694(94)90166-X
Bahrami M, Bozorg-Haddad O, Loaiciga HA (2019) Optimizing stormwater low-impact development strat-
egies in an urban watershed considering sensitivity and uncertainty. Environ Monit Assess 191:340.
https://doi.org/10.1007/s10661-019-7488-y
Bakhshipour AE, Dittmer U, Haghighi A, Nowak W (2019) Hybrid green–blue–gray decentralized urban
drainage systems design, a simulation-optimization framework. J Environ Manag 249:109364. https://
doi.org/10.1016/j.jenvman.2019.109364
Baldassarre GD, Montanari A, Lins H, Koutsoyiannis D et al (2010) Flood fatalities in Africa: from diagno-
sis to mitigation. Geophys Res Lett 37:L22402. https://doi.org/10.1029/2010GL045467
Beecham S, Pezzaniti D, Kandasamy J (2012) Stormwater treatment using permeable pavements. Water
Manag 165:161–170. https://doi.org/10.1680/wama.2012.165.3.161
Beeneken T, Erbe V, Messmer A, Reder C et al (2013) Real-time control (RTC) of urban drainage systems:
a discussion of the additional efforts compared to conventionally operated systems. Urban Water J
10:293–299. https://doi.org/10.1080/1573062X.2013.790980
Berkhahn S, Fuchs L, Neuweiler I (2019) An ensemble neural network model for real-time prediction of
urban floods. J Hydrol 575:743–754. https://doi.org/10.1016/j.jhydrol.2019.05.066
Bertsch R, Glenis V, Kilsby C (2017) Urban flood simulation using synthetic storm drain networks. Water
9:925. https://doi.org/10.3390/w9120925
Beven K (2006) A manifesto for the equifinality thesis. J Hydrol 320:18–36. https://doi.org/10.1016/j.jhydr
ol.2005.07.007
13
Bhola PK, Nair BB, Leandro J, Rao SN et al (2019) Flood inundation forecasts using validation data gen-
erated with the assistance of computer vision. J Hydroinform 21:240–256. https://doi.org/10.2166/
hydro.2018.044
Bilodeau K, Pelletier G, Duchesne S (2019) Real-time control of stormwater detention basins as an adapta-
tion measure in mid-size cities. Urban Water J 15:858–867. https://doi.org/10.1080/1573062X.2019.
1574844
Bladé E, Gómez-Valentín M, Dolz J, Aragón-Hernández JL et al (2012) Integration of 1D and 2D finite
volume schemes for computations of water flow in natural channels. Adv in Water Resour 42:17–29.
https://doi.org/10.1016/j.advwatres.2012.03.021
Borsanyi P, Benedetti L, Dirckx G, de Keyser W et al (2008) Modelling real-time control options on virtual
sewer systems. J Environ Eng Sci 7:395–410. https://doi.org/10.1139/S08-004
Bui DT, Hoang ND, Martinez-Alvarez F (2020) A novel deep learning neural network approach for predict-
ing flash flood susceptibility: a case study at a high frequency tropical storm area. Sci Toatal Environ
701:134413. https://doi.org/10.1016/j.scitotenv.2019.134413
Butler D, Schuetze M (2005) Integrating simulation models with a view to optimal control of urban waste-
water systems. Environ Model Softw 20:415–426. https://doi.org/10.1016/j.envsoft.2004.02.003
Buurman J, Padawangi R (2018) Bringing people closer to water: integrating water management and urban
infrastructure. J Environ Manag 61:2531–2548. https://doi.org/10.1080/09640568.2017.1404972
Casal-Campos A, Fu GT, Butler D, Moore A et al (2015) An integrated environmental assessment of green
and gray infrastructure strategies for robust decision making. Environ Sci Technol 49:8307–8314.
https://doi.org/10.1021/es506144f
Castro MJ, Ortega S, de la Asunción M, Mantas JM et al (2011) GPU computing for shallow water flow
simulation based on finite volume schemes. CR Mecanique 339:165–184. https://doi.org/10.1016/j.
crme.2010.12.004
Cembrano G, Quevedo J, Salamero M, Puig V et al (2004) Optimal control of urban drainage systems. a
case study. Control Eng Pract 12:1–9. https://doi.org/10.1016/S0967-0661(02)00280-0
Chang FJ, Chang KY, Chang LC (2008) Counterpropagation fuzzy-neural network for city flood control
system. J Hydrol 358:24–34. https://doi.org/10.1016/j.jhydrol.2008.05.013
Chen WJ, Huang GR, Zhang H (2017) Urban stormwater inundation simulation based on SWMM and dif-
fusive overland-flow model. Water Sci Technol 76:3392–3403. https://doi.org/10.2166/wst.2017.504
Chen WJ, Huang GR, Zhang H, Wang WQ (2018) Urban inundation response to rainstorm patterns with
a coupled hydrodynamic model: a case study in Haidian Island, China. J Hydrol 564:1022–1035.
https://doi.org/10.1016/j.jhydrol.2018.07.069
Chen J, Hill AA, Urbano LD (2009) A GIS-based model for urban flood inundation. J Hydrol 373:184–192.
Chiang YM, Chang LC, Tsai MJ, Wang YF et al (2011) Auto-control of pumping operations in sewerage
systems by rule-based fuzzy neural networks. Hydrol Earth Syst Sci 15:185–196. https://doi.org/10.
5194/hess-15-185-2011
Choo YM, Jo DJ, Yun GS, Lee EH (2019) A study on the improvement of flood forecasting techniques
in urban areas by considering rainfall intensity and duration. Water 11:1883. https://doi.org/10.3390/
w11091883
CrED, UnIsDr (2015) The human cost of weather-related disasters 1995–2015. Center for research on the
Epidemiology of Disasters (CrED) and Un office for disaster risk reduction (UnIsDr) https://www.
unisdrorg/we/inform/publications/46796.
Damodaram C, Berglund EM (2013) Simulation-optimization approach to design low impact development
for managing peak flow alterations in urbanizing watersheds. J Water Res Plan Man 139:290–298.
https://doi.org/10.1061/(ASCE)WR.1943-5452.0000251
Darsono S, Labadie JW (2007) Neural-optimal control algorithm for real-time regulation of in-line storage
in combined sewer systems. Environ Model Softw 22:1349–1361. https://doi.org/10.1016/j.envsoft.
2006.09.005
Dash P, Punia M (2019) Governance and disaster: analysis of land use policy with reference to Uttarakhand
flood 2013, India. Int J Disaster Risk Reduct 36:101090. https://doi.org/10.1016/j.ijdrr.2019.101090
de la Asuncion M, Castro MJ, Fernandez-Nieto ED, Mantas JM et al (2013) Efficient GPU implementation
of a two waves TVD-WAF method for the two-dimensional one layer shallow water system on struc-
tured meshes. Comput Fluids 80:441–452. https://doi.org/10.1016/j.compfl uid.2012.01.012
Demuzere M, Orru K, Heidrich O, Olazabal E et al (2014) Mitigating and adapting to climate change: multi-
functional and multi-scale assessment of green urban infra-structure. J Environ Manag 146:107–115.
https://doi.org/10.1016/j.jenvman.2014.07.025
Devia GK, Ganasri BP, Dwarakish GS (2015) A review on hydrological models. Aquatic Procedia 4:1001–
1007. https://doi.org/10.1016/j.aqpro.2015.02.126
13
Di Matteo M, Liang R, Maier HR, Thyer MA et al (2019) Controlling rainwater storage as a system:
an opportunity to reduce urban flood peaks for rare, long duration storms. Environ Model Softw
111:34–41. https://doi.org/10.1016/j.envsoft.2018.09.020
Diakakis M, Andreadakis E, Nikolopoulos EI, Spyrou NI et al (2019) An integrated approach of ground
and aerial observations in flash flood disaster investigations, the case of the 2017 Mandra flash
flood in Greece. Int J Disaster Risk Reduct 33:290–309. https://doi.org/10.1016/j.ijdrr.2018.10.
015
Dong X, Guo H, Zeng SY (2017) Enhancing future resilience in urban drainage system: green versus
grey infrastructure. Water Res 124:280–289. https://doi.org/10.1016/j.watres.2017.07.038
Dottori F, Todini E (2011) Developments of a flood inundation model based on the cellular automata
approach: testing different methods to improve model performance. Phys Chem Earth 36:266–280.
https://doi.org/10.1016/j.pce.2011.02.004
Duchesne S, Mailhot A, Dequidt E, Villeneuve JP (2001) Mathematical modeling of sewers under sur-
charge for real time control of combined sewer overflows. Urban Water 3:241–252. https://doi.org/
10.1016/S1462-0758(01)00037-1
Elliott AH, Trowsdale SA (2007) A review of models for low impact urban stormwater drainage. Envi-
ron Model Softw 22:394–405. https://doi.org/10.1016/j.envsoft.2005.12.005
Finaud-Guyot P, Delenne C, Guinot V, Llovel C (2011) 1D–2D coupling for river flow modeling. Cr
Mecanique 339:226–234. https://doi.org/10.1016/j.crme.2011.02.001
Fiori A, Volpi E (2020) On the effectiveness of LID infrastructures for the attenuation of urban flooding
at the catchment scale. Water Resour Res. https://doi.org/10.1029/2020WR027121
Freni G, Mannina G, Viviani G (2010) Urban storm-water quality management: centralized versus
source control. J Water Resour Plan Manag 136:268–278. https://doi.org/10.1061/(ASCE)0733-
9496(2010)136:2(268)
Gaborit E, Muschalla D, Vallet B, Vanrolleghem PA et al (2013) Improving the performance of storm-
water detention basins by real-time control using rainfall forecasts. Urban Water J 10:230–246.
https://doi.org/10.1080/1573062X.2012.726229
Gagliano A, Nocera F, Detommaso M, Evola G (2016) Thermal behavior of an extensive green roof:
numerical simulations and experimental investigations. Int J Heat Mass Transf 34:226–234.
https://doi.org/10.18280/ijht.34S206
Garcia L, Barreiro-Gomez J, Escobar E, Tellez D et al (2015) Modeling and real-time control of urban
drainage systems: a review. Adv Water Resour 85:120–132. https://doi.org/10.1016/j.advwatres.
2015.08.007
Gejadze IY, Monnier J (2007) On a 2D zoom for 1D shallow-water model: coupling and data assimilation.
Comput Methods Appl Mech Eng 196:4628–4643. https://doi.org/10.1016/j.cma.2007.05.026
Giacomoni MH, Joseph J (2017) Multi-objective evolutionary optimization and Monte Carlo simula-
tion for placement of low impact development in the catchment scale. J Water Resour Plan Manag
143:4017053. https://doi.org/10.1061/(ASCE)WR.1943-5452.0000812
Gran JA, Ramos SL (2019) Climate change and flood risk: vulnerability assessment in an urban poor
community in Mexico. Environ Urban 31:75–92. https://doi.org/10.1177/0956247819827850
Haghighatafshar S, Nordloef B, Roldin M, Gustafsson L et al (2017) Efficiency of blue-green stormwa-
ter retrofits for flood mitigation: conclusions drawn from a case study in Malm Sweden. J Environ
Manag 207:60–69. https://doi.org/10.1016/j.jenvman.2017.11.018
Henonin J, Russo B, Mark O, Gourbesville P (2013) Real-time urban flood forecasting and modelling: a
state of the art. J Hydroinf 15:717–736. https://doi.org/10.2166/hydro.2013.132
Hou JM, Guo KH, Liu FF, Han H et al (2018) Assessing slope forest effect on flood process caused by
a short-duration storm in a small catchment. Water 10:1256. https://doi.org/10.3390/w10091256
Hou JM, Liu FF, Tong Y, Guo KH et al (2020) Numerical simulation for runoff regulation in rain garden
using 2D hydrodynamic Model. Ecol Eng 153:105794. https://doi.org/10.1016/j.ecoleng.2020.
105794
Hou JM, Simons F, Mahgoub M, Hinkelmann R (2013) A robust well-balanced model on unstructured
grids for shallow water flows with wetting and drying over complex topography. Comput Methods
Appl Mech Eng 257:126–149. https://doi.org/10.1016/j.cma.2013.01.015
Hsu MH, Chen SH, Chang TJ (2000) Inundation simulation for urban drainage basin with storm sewer
system. J Hydrol 234:21–37. https://doi.org/10.1016/S0022-1694(00)00237-7
Hsu NS, Huang CL, Wei CC (2013) Intelligent real-time operation of a pumping station for an urban
drainage system. J Hydrol 489:85–97. https://doi.org/10.1016/j.jhydrol.2013.02.047
Hu XZ, Song LX (2018) Hydrodynamic modeling of flash flood in mountain watersheds based on high-
performance GPU computing. Nat Hazards 91:567–586. https://doi.org/10.1007/s11069-017-3141-7
13
Huang CL, Hsu NS, Liu HJ, Huang YH (2018) Optimization of low impact development layout designs for
megacity flood mitigation. J Hydrol 564:542–558. https://doi.org/10.1016/j.jhydrol.2018.07.044
Jackson CR, Burges SJ, Liang X, Leytham KM et al (2001) Development and application of simplified con-
tinuous hydrologic modeling for drainage design and analysis. Water Sci Appl 2:39–58. https://doi.
org/10.1029/WS002p0039
Jacopin C, Lucas E, Desbordes M, Bourgogne P (2000) Optimisation of operational management practices
for the detention basins. Water Sci Technol 44:277–285. https://doi.org/10.2166/wst.2001.0780
Jafari F, Mousavi SJ, Kim JH (2020) Investigation of rainfall forecast system characteristics in real-time
optimal operation of urban drainage systems. Water Resour Manag 34:1773–1787. https://doi.org/10.
1007/s11269-020-02528-1
Jafari F, Mousavi SJ, Yazdi J, Kim JH (2018a) Long-term versus real-time optimal operation for gate regu-
lation during flood in urban drainage systems. Urban Water J 15:750–759. https://doi.org/10.1080/
1573062X.2018.1556307
Jafari F, Mousavi SJ, Yazdi J, Kim JH (2018b) Real-time operation of pumping systems for urban flood
mitigation: single-period vs. multi-period optimization. Water Resour Manag 32:4643–4660. https://
doi.org/10.1007/s11269-018-2076-4
Jamshed A, Birkmann J, Rana IA, McMillan JM (2020) The relevance of city size to the vulnerability
of surrounding rural areas: an empirical study of flooding in Pakistan. Int J Disaster Risk Reduct
48:101601. https://doi.org/10.1016/j.ijdrr.2020.101601
Jhong YD, Chen CS, Lin HP, Chen ST (2018) Physical hybrid neural network model to forecast typhoon
floods. Water 10:632. https://doi.org/10.3390/w10050632
Kelly KS, Krzysztofowicz R (1997) A bivariate meta-Gaussian density for use in hydrology. Stoch Hydrol
Hydraul 11:17–31. https://doi.org/10.1007/BF02428423
Koc K, Isik Z (2020) A multi-agent-based model for sustainable governance of urban flood risk mitigation
measures. Nat Hazards 104:1079–1110. https://doi.org/10.1007/s11069-020-04205-3
Lacasta A, Morales-Hernandez M, Murillo J, Garcia-Navarro P (2014) An optimized GPU implementation
of a 2D free surface simulation model on unstructured meshes. Adv Eng Softw 78:1–15. https://doi.
org/10.1016/j.advengsoft.2014.08.007
Leandro J, Chen AS, Djordjevic S, Savic DA (2009) Comparison of 1D/1D and 1D/2D coupled (Sewer/
Surface) hydraulic models for urban flood simulation. J Hydraul Eng 135:495–504. https://doi.org/10.
1061/(ASCE)HY.1943-7900.0000037
Leandro J, Djordjevic S, Chen AS, Savic DA et al (2011) Calibration of a 1D/1D urban flood model using
1D/2D model results in the absence of field data. Water Sci Technol 64:1016–1024. https://doi.org/
10.2166/wst.2011.467
Lee JS, Li MH (2009) The impact of detention basin design on residential property value: case studies
using GIS in the hedonic price modeling. Landsc Urban Plan 89:7–16. https://doi.org/10.1016/j.landu
rbplan.2008.09.002
Li WF, Chen QW, Mao JQ (2009) Development of 1D and 2D coupled model to simulate urban inunda-
tion: an application to Beijing Olympic Village. Chin Sci Bull 54:1613–1621. https://doi.org/10.1007/
s11434-009-0208-1
Li XL, Lu HS, Horton R, An TQ et al (2014) Real-time flood forecast using the coupling support vector
machine and data assimilation method. J Hydroinform 16:973–988. https://doi.org/10.2166/hydro.
2013.075
Li ZM, Zhang XX, Ma YF, Feng CY et al (2019) A multi-criteria decision making method for urban flood
resilience evaluation with hybrid uncertainties. Int J Disaster Risk Reduct 36:101140. https://doi.org/
10.1016/j.ijdrr.2019.101140
Lian JJ, Xu K, Ma C (2013) Joint impact of rainfall and tidal level on flood risk in a coastal city with a com-
plex river network: a case study of Fuzhou City China. Hydrol Earth Syst Sci 17:679–689. https://doi.
org/10.5194/hess-17-679-2013
Lian YQ, Chan IC, Singh J, Demissie M et al (2007) Coupling of hydrologic and hydraulic models for the
Illinois River Basin. J Hydrol 344:210–222. https://doi.org/10.1016/j.jhydrol.2007.08.004
Liang CY, You GJY, Lee HY (2019) Investigating the effectiveness and optimal spatial arrangement of low-
impact development facilities. J Hydrol 577:124008. https://doi.org/10.1016/j.jhydrol.2019.124008
Liang Q (2010) Flood simulation using a well-balanced shallow flow model. J Hydraul Eng 136:669–675.
https://doi.org/10.1061/(ASCE)HY.1943-7900.0000219
Liang Q, Borthwick AG (2009) Adaptive quadtree simulation of shallow flows with wet-dry fronts over
complex topography. Comput Fluids 38:221–234. https://doi.org/10.1016/j.compfl uid.2008.02.008
Liang Q, Marche F (2009) Numerical resolution of well-balanced shallow water equations with complex
source terms. Adv Water Resour 32:873–884. https://doi.org/10.1016/j.advwatres.2009.02.010
13
Liang Q, Smith LS (2015) A high-performance integrated hydrodynamic modelling system for urban flood
simulations. J Hydroinform 17:518–533. https://doi.org/10.2166/hydro.2015.029
Lin GF, Lin HY, Chou YC (2013) Development of a real-time regional-inundation forecasting model for the
inundation warning system. J Hydroinform 15:1391–1407. https://doi.org/10.2166/hydro.2013.202
Liu Q, Qin Y, Zhang Y, Li ZW (2015) A coupled 1D–2D hydrodynamic model for flood simulation in flood
detention basin. Nat Hazards 75:1303–1325. https://doi.org/10.1007/s11069-014-1373-3
Liu X, Chui TFM (2019) Evaluation of green roof performance in mitigating the impact of extreme storms.
Water 11:815. https://doi.org/10.3390/w11040815
Liu Y, Zhou JZ, Song LX, Zou Q et al (2013) Numerical modelling of free-surface shallow flows over
irregular topography with complex geometry. Appl Math Model 37:9482–9498. https://doi.org/10.
1016/j.apm.2013.05.001
Liu YZ, Cibin R, Bralts VF (2016a) Optimal selection and placement of BMPs and LID practices with a
rainfall-runoff model. Environ Model Softw 80:281–296. https://doi.org/10.1016/j.envsoft.2016.03.
005
Liu YZ, Theller LO, Pijanowski BC, Engel BA (2016b) Optimal selection and placement of green infra-
structure to reduce impacts of land use change and climate change on hydrology and water quality: an
application to the Trail Creek Watershed Indiana. Sci Total Environ 553:149–163. https://doi.org/10.
1016/j.scitotenv.2016.02.116
Lo SW, Wu JH, Lin FP, Hsu CH (2015) Visual sensing for urban flood monitoring. Sensors 15:20006–
20029. https://doi.org/10.3390/s150820006
Lowe R, Vezzaro L, Mikkelsen PS, Grum M et al (2016) Probabilistic runoff volume forecasting in risk-
based optimization for RTC of urban drainage systems. Environ Model Softw 80:143–158. https://
doi.org/10.1016/j.envsoft.2016.02.027
Luan B, Xu P, Wang X, Yang XM et al (2019) Evaluating green stormwater infrastructure strategies effi-
ciencies in a rapidly urbanizing catchment using SWMM-based TOPSIS. J Clean Prod 223:680–691.
https://doi.org/10.1016/j.jclepro.2019.03.028
Marin J, Monnier J (2009) Superposition of local zoom models and simultaneous calibration for 1D–2D
shallow water flows. Math Comput Simul 80:547–560. https://doi.org/10.1016/j.matcom.2009.09.001
Mercado JMR, Kawamura A, Amaguchi H (2020) Interrelationships of the barriers to integrated flood risk
management adaptation in Metro Manila, Philippines. Int J Disaster Risk Reduct 49:101683. https://
doi.org/10.1016/j.ijdrr.2020.101683
Miglio E, Perotto S, Saleri F (2005) Model coupling techniques for free-surface flow problems: Part II. Non-
linear Anal-Theor 63:5–7. https://doi.org/10.1016/j.na.2005.03.085
Moghadas M, Asadzadeh A, Vafeidis A, Fekete A et al (2019) A multi-criteria approach for assessing urban
flood resilience in Tehran, Iran. Int J Disaster Risk Reduction 35:101069. https://doi.org/10.1016/j.
ijdrr.2019.101069
Moradkhani H, Sorooshian S (2009) General review of rainfall-runoff modeling: model calibration, data
assimilation, and uncertainty analysis. Hydrological modeling and the water cycle 63:1–24. https://
doi.org/10.1007/978-3-540-77843-1_1
Morales-Hernandez M, Garcia-Navarro P, Burguete J, Brufau P et al (2013) A conservative strategy to cou-
ple 1D and 2D models for shallow water flow simulation. Comput Fluids 81:26–44. https://doi.org/10.
1016/j.compfl uid.2013.04.001
Morita M, Yen BC (2002) Modeling of conjunctive two-dimensional surface-three-dimensional subsurface
flows. J Hydraul Eng 128:184–200. https://doi.org/10.1061/(ASCE)0733-9429(2002)128:2(184)
Moura NCB, Pellegrino JRM, Martins JRS (2015) Best management practices as an alternative for flood
and urban storm water control in a changing climate. J Flood Risk Manag 9:243–254. https://doi.org/
10.1111/jfr3.12194
Nguyen P, Thorstensen A, Sorooshian S, Hsu KL et al (2016) A high resolution coupled hydrologic-hydrau-
lic model (HiResFlood-UCI) for flash flood modeling. J Hydrol 541:401–420. https://doi.org/10.
1016/j.jhydrol.2015.10.047
Nourani V, Baghanam AH, Kisi O (2014) Applications of hybrid wavelet-artificial intelligence models in
hydrology: a review. J Hydrol 514:358–377. https://doi.org/10.1016/j.jhydrol.2014.03.057
Park D, Jang S, Roesner LA (2014) Evaluation of multi-use stormwater detention basins for improved urban
watershed management. Hydrol Process 28:1104–1113. https://doi.org/10.1002/hyp.9658
Pender G (2006) Briefing: introducing the flood risk management research consortium. P I Civ Eng-Water
Manag 159:3–8. https://doi.org/10.1680/wama.2006.159.1.3
Pleau M, Colas H, Lavallee P, Pelletier G et al (2005) Global optimal real-time control of the Quebec
urban drainage system. Environ Model Softw 20:401–413. https://doi.org/10.1016/j.envsoft.2004.
02.009
13
Pleau M, Pelletier G, Colas H, Lavallee P et al (2000) Global predictive real-time control of Quebec
Urban Community’s westerly sewer network. Water Sci Technol 43:123–130. https://doi.org/10.
2166/wst.2001.0404
Poser K, Dransch D (2010) Volunteered geographic information for disaster management with
application to rapid flood damage estimation. Geomatica 64:59–98. https://doi.org/10.5623/
geomat-2010-0008
Prudencio L (2018) Stormwater management and ecosystem services: a review. Environ Res Lett
13:033002. https://doi.org/10.1088/1748-9326/aaa81a
Rauch W, Bertrand-Krajewski JL, Krebs P, Mark O et al (2002) Deterministic modelling of integrated urban
drainage systems. Water Sci Technol 45:81–94. https://doi.org/10.1016/S0043-1354(01)00232-9
Rene JR, Madsen H, Mark O (2013) A methodology for probabilistic real-time forecasting: an urban
case study. J Hydroinf 15:751–762. https://doi.org/10.2166/hydro.2012.031
Rjeily AY, Abbas O, Sadek M, Shahrour I et al (2018) Model predictive control for optimising the operation
of urban drainage systems. J Hydrol 566:558–565. https://doi.org/10.1016/j.jhydrol.2018.09.044
Rohrer AR, Armitage NP (2017) Improving the viability of stormwater harvesting through rudimentary
real time control. Water 9:371. https://doi.org/10.3390/w9060371
Roodsari BK, Chandler DG, Kelleher C, Kroll CN (2019) A comparison of SAC-SMA and adaptive
neuro-fuzzy inference system for real-time flood forecasting in small urban catchments. J Flood
Risk Manag 12:e12492. https://doi.org/10.1111/jfr3.12492
Sadler JM, Goodall JL, Morsy MM, Spencer K (2018) Modeling urban coastal flood severity from
crowd-sourced flood reports using Poisson regression and random forest. J Hydrol 559:43–55.
Sakieh Y (2017) Understanding the effect of spatial patterns on the vulnerability of urban areas to flood-
ing. Int J Disaster Risk Reduct 25:125–136. https://doi.org/10.1016/j.ijdrr.2017.09.004
Salman AM, Li Y (2018) Flood risk assessment, future trend modeling, and risk communication: a
review of ongoing research. Nat Hazards Rev 19:04018011. https://doi.org/10.1061/(ASCE)NH.
1527-6996.0000294
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
Sharior S, McDonald W, Parolari AJ (2019) Improved reliability of stormwater detention basin perfor-
mance through water quality data-informed real-time control. J Hydrol 573:422–431. https://doi.
org/10.1016/j.jhydrol.2019.03.012
Shishegar S, Duchesne S, Pelletier G (2018) Optimization methods applied to stormwater management
problems: a review. Urban Water J 15:276–286. https://doi.org/10.1080/1573062X.2018.1439976
Shishegar S, Duchesne S, Pelletier G (2019) An integrated optimization and rule-based approach for pre-
dictive real time control of urban stormwater management systems. J Hydrol 577:124000. https://
doi.org/10.1016/j.jhydrol.2019.124000
Skaugen T (2007) Modelling the spatial variability of snow water equivalent at the catchment scale.
Hydro Earth Syst Sci 11:1543–1550. https://doi.org/10.5194/hess-11-1543-2007
Smith MB, Koren V, Zhang ZY, Zhang Y et al (2012) Results of the DMIP 2 Oklahoma experiments. J
Hydrol 418:17–48. https://doi.org/10.1016/j.jhydrol.2011.08.056
Sohn W, Kim JH, Li MH, Brown R (2019) The influence of climate on the effectiveness of low impact
development: a systematic review. J Environ Manage 236:365–379. https://doi.org/10.1016/j.
jenvman.2018.11.041
Song LX, Zhou JZ, Li QQ, Yang XL et al (2011) An unstructured finite volume model for dam-break
floods with wet/dry fronts over complex topography. Int J Numer Methods Fluids 67:960–980.
https://doi.org/10.1002/fld.2397
Su BN, Huang H, Zhu W (2019) An urban pluvial flood simulation model based on diffusive wave approxi-
mation of shallow water equations. Hydrol Res 50:138–154. https://doi.org/10.2166/nh.2017.233
Tang M, Liao HC, Li ZM, Xu ZS (2018) Nature disaster risk evaluation with a group decision making
method based on incomplete hesitant fuzzy linguistic preference relations. Int J Environ Res Pub-
lic Health 15:751. https://doi.org/10.3390/ijerph15040751
Tao T, Wang J, Xin K, Li S (2014) Multi-objective optimal layout of distributed storm-water detention.
Int J Environ Sci Technol 11:1473–1480. https://doi.org/10.1007/s13762-013-0330-0
Tehrany MS, Pradhan B, Mansor S, Ahmad N (2015) Flood susceptibility assessment using GIS-based
support vector machine model with different kernel types. CATENA 125:91–101. https://doi.org/
10.1016/j.catena.2014.10.017
Teng J, Jakeman AJ, Vaze J, Croke BFW, Dutta D, Kim S (2017) Flood inundation modelling: a review of
methods, recent advances and uncertainty analysis. Environ Model Softw 90:201–216. https://doi.org/
10.1016/j.envsoft.2017.01.006
13
Teng J, Vaze J, Dutta D, Marvanek S (2015) Rapid inundation modelling in large floodplains using LiDAR
DEM. Water Resour Manag 29:2619–2636. https://doi.org/10.1007/s11269-015-0960-8
Urich C, Rauch W (2014) Exploring critical pathways for urban water management to identify robust strat-
egies under deep uncertainties. Water Res 66:374–389. https://doi.org/10.1016/j.watres.2014.08.020
Vacondio R, Dal Palu A, Mignosa P (2014) GPU-enhanced finite volume shallow water solver for fast flood
simulations. Environ Model Softw 57:60–72. https://doi.org/10.1016/j.envsoft.2014.02.003
Vezzaro L, Grum M (2014) A generalised dynamic overflow risk assessment (DORA) for real time control
of urban drainage systems. J Hydrol 515:292–303. https://doi.org/10.1016/j.jhydrol.2014.05.019
Waghwala RK, Agnihotri PG (2019) Flood risk assessment and resilience strategies for flood risk manage-
ment: a case study of Surat City. Int J Disaster Risk Reduct 40:101155. https://doi.org/10.1016/j.ijdrr.
2019.101155
Wang C, Hou JM, Miller D, Brown I et al (2019) Flood risk management in sponge cities: the role of inte-
grated simulation and 3D visualization. Int J Disaster Risk Reduct 39:101139. https://doi.org/10.
1016/j.ijdrr.2019.101139
Wei CC, Hsu NS, Huang CL (2014) Two-stage pumping control model for flood mitigation in inundated
urban drainage basins. Water Resour Manag 28:425–444. https://doi.org/10.1007/s11269-013-0491-0
Wu XS, Wang ZL, Guo SL, Liao WL et al (2017) Scenario-based projections of future urban inundation
within a coupled hydrodynamic model framework: a case study in Dongguan City, China. J Hydrol
547:428–442. https://doi.org/10.1016/j.jhydrol.2017.02.020
Wu ZN, Zhou YH, Wang HL, Jiang ZH (2020) Depth prediction of urban flood under different rainfall
return periods based on deep learning and data warehouse. Sci Total Environ 716:137077. https://doi.
org/10.1016/j.scitotenv.2020.137077
Xu HS, Xu K, Lian JJ, Ma C (2019a) Compound effects of rainfall and storm tides on coastal flooding risk.
Stoch Environ Res Risk Assess 33:1249–1261. https://doi.org/10.1007/s00477-019-01695-x
Xu K, Ma C, Lian J, Bin L et al (2014) Joint probability analysis of extreme precipitation and storm tide
level in a coastal city under changing environment. PLoS ONE 9:e109341. https://doi.org/10.1371/
journal.pone.0109341
Xu T, Jia HF, Wang Z, Mao XH et al (2017) SWMM-based methodology for block-scale LID-BMPs plan-
ning based on site-scale multi-objective optimization: a case study in Tianjin. Front Environ Sci Eng
11:48–59. https://doi.org/10.1007/s11783-017-0934-6
Xu T, Li K, Engel BA, Jia HF et al (2019b) Optimal adaptation pathway for sustainable low impact devel-
opment planning under deep uncertainty of climate change: a greedy strategy. J Environ Manag
248:109280. https://doi.org/10.1016/j.jenvman.2019.109280
Yagi S, Shiba S (1999) Application of genetic algorithms and fuzzy control to a combined sewer pumping
station. Water Sci Technol 39:217–224. https://doi.org/10.1016/S0273-1223(99)00236-X
Yan J, Jin JM, Chen FR, Yu G et al (2018) Urban flash flood forecast using support vector machine and
numerical simulation. J Hydroinform 20:221–231. https://doi.org/10.2166/hydro.2017.175
Yazdanfar Z, Sharma A (2015) Urban drainage system planning and design: challenges with climate change
and urbanization: a review. Water Sci Technol 72:165–179. https://doi.org/10.2166/wst.2015.207
Yazdi J, Choi HS, Kim JH (2016) A methodology for optimal operation of pumping stations in urban drain-
age systems. J Hydro-Environ Res 11:101–112. https://doi.org/10.1016/j.jher.2015.09.001
Yin J, Ye MW, Yin Z, Xu SY (2015) A review of advances in urban flood risk analysis over China. Stoch
Environ Res Risk Assess 29:1063–1070. https://doi.org/10.1007/s00477-014-0939-7
Yin J, Yu D, Yin Z, Wang J et al (2013) Modelling the combined impacts of sea-level rise and land sub-
sidence on storm tides induced flooding of the Huangpu River in Shanghai China. Clim Change
119:919–932. https://doi.org/10.1007/s10584-013-0749-9
Yue S, Ouarda TBMJ, Bobée B (1999) The Gumbel mixed model for flood frequency analysis. J Hydrol
226:88–100. https://doi.org/10.1016/s0022-1694(99)00168-7
Zeng JJ, Huang GR, Mai YP, Chen WJ (2020) Optimizing the cost-effectiveness of low impact development
(LID) practices using an analytical probabilistic approach. Urban Water J 17:136–143. https://doi.org/
10.1080/1573062X.2020.1748208
Zhang G, Hamlett JM, Reed P, Tang Y (2013) Multi-objective optimization of low impact development
designs in an urbanizing watershed. OJOp 2:95–108. https://doi.org/10.4236/ojop.2013.24013
Zhang SH, Pan BZ (2014) An urban storm-inundation simulation method based on GIS. J Hydrol 517:260–
268. https://doi.org/10.1016/j.jhydrol.2014.05.044
Zheng F, Westra S, Sisson SA (2013) Quantifying the dependence between extreme rainfall and storm surge
in the coastal zone. J Hydrol 505:172–187. https://doi.org/10.1016/j.jhydrol.2013.09.054
Zhou Q, Lai Z, Blohm A (2019) Optimising the combination strategies for pipe and infiltration-based
low impact development measures using a multiobjective evolution approach. J Flood Risk Manag
12:e12457. https://doi.org/10.1111/jfr3.12457
13
Zischg J, Goncalves MLR, Bacchin TK, Leonhardt G et al (2017) Info-Gap robustness pathway method for
transitioning of urban drainage systems under deep uncertainties. Water Sci Technol 76:1727–1281.
https://doi.org/10.2166/wst.2017.320
Zoppou C (2001) Review of urban storm water models. Environ Model Softw 16:195–231. https://doi.org/
10.1016/S1364-8152(00)00084-0
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.
13

A Review On Applications of Urban Flood Models in Flood Mitigation Strategies

Uploaded by

Copyright:

Available Formats

A Review On Applications of Urban Flood Models in Flood Mitigation Strategies

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

A Review On Applications of Urban Flood Models in Flood Mitigation Strategies

Uploaded by

Copyright:

Available Formats

Natural Hazards (2021) 108:31–62

A review on applications of urban flood models in flood

Keywords Flood mitigation strategies · Forecasting system · Optimal operation · Real-time

Fig. 1 Main causes of urban flooding

2 Urban flood models

2.1 Classification of urban flood models

A model understood a simplified representation of real-world scenario, and urban flood

Fig. 2 Classification scheme of flood modeling methodologies

2.2 Models elaboration based on calculation algorithm

3 Quantitative forecasting for urban flood in a modeling perspective

3.1 Classification of urban flood forecasting system based on modeling methods

3.1.1 Pre‑simulated forecasting system

3.1.2 Real‑time forecasting system

3.2 Performance evaluation of urban flood models for extreme flood forecasting

• Urban flood model quality

• Rainfall data quality

4 Model for designing of the urban flood mitigation strategies

4.1 Flood measures designing under microscale condition

Fig. 4 Examples of gray infrastructure measures for stormwater management

An equally important consideration that is of continued interest to researchers is the

4.1.2 Green infrastructure for stormwater on flood management

Fig. 5 Examples of green infrastructure measures for stormwater management

4.2 Flood measure designing under mesoscale perspective

4.2.1 Coupled urban flood models with Optimizing methods

4.2.2 Scenario analysis based on urban flood models

4.3 Flood measures designing under macroscale perspective

Bahrami et al. (2019) √ √ Bioretention cells and porous pavement GA SWMM

5 Optimal operation of the flood measures based on a modeling

Fig. 6 Emergence of new water management concepts over time

5.1 Static control operation

5.2 Real‑time control operation

5.2.1 Definition of Real‑time control operation

Fig. 7 Conceptual diagram of RTC operation for stormwater management measures

5.2.2 Modeling method of flood mitigation measures for RTC operation

• Optimization of operational management practices for drainage systems

• Minimize urban flood risk,

• Optimization of operational management practices for the detention pond

A stormwater detention pond is devised to capture stormwater to delay peak flows

• Maximize the detention time of stormwater,

• Optimization of operational management practices for pumping stations

6 Challenges and findings

Herein, we reviewed state-of-the-art prevailing flood modeling methods based on their

Conflict of interest The authors declare no competing financial interests.

You might also like

2 Urban flood models

2.1 Classification of urban flood models

2.2 Models elaboration based on calculation algorithm

3 Quantitative forecasting for urban flood in a modeling perspective

3.1 Classification of urban flood forecasting system based on modeling methods

3.1.1 Pre‑simulated forecasting system

3.1.2 Real‑time forecasting system

3.2 Performance evaluation of urban flood models for extreme flood forecasting

4 Model for designing of the urban flood mitigation strategies

4.1 Flood measures designing under microscale condition

4.1.2 Green infrastructure for stormwater on flood management

4.2 Flood measure designing under mesoscale perspective

4.2.1 Coupled urban flood models with Optimizing methods

4.2.2 Scenario analysis based on urban flood models

4.3 Flood measures designing under macroscale perspective

5 Optimal operation of the flood measures based on a modeling

5.1 Static control operation

5.2 Real‑time control operation

5.2.1 Definition of Real‑time control operation

5.2.2 Modeling method of flood mitigation measures for RTC operation

6 Challenges and findings