water

Article
Two-Dimensional Hydrodynamic Simulation of the Effect of
Stormwater Inlet Blockage on Urban Waterlogging
Weiwei Guo 1 , Mingshuo Zhai 2,3 , Xiaohui Lei 2,4 , Haocheng Huang 5 , Yan Long 2,4, * and Shusen Li 6

1 College of Water Resource Science and Engineering, Taiyuan University of Technology, Taiyuan 030024, China;
[email protected]
2 School of Water Resources and Electric Power, Hebei University of Engineering, Handan 056038, China;
[email protected] (M.Z.); [email protected] (X.L.)
3 Hebei Key Laboratory of Smart Water Conservancy, Hebei University of Engineering, Handan 056038, China
4 China Institute of Water Resources and Hydropower Research, Beijing 100038, China
5 State Key Laboratory of Simulation and Regulation of Water Cycle in River Basin, China Institute of Water
Resources and Hydropower Research, Beijing 410075, China; [email protected]
6 Tongliao Water Authority, Tongliao 028000, China; [email protected]
* Correspondence: [email protected]

Abstract: The drainage capacity of stormwater inlets, which serve as the connection between the
surface and the underground drainage system, directly affects surface runoff and the drainage
capacity of underground drainage systems. However, in reality, stormwater inlets are often blocked
due to the accumulation of leaves, human waste disposal and other factors, resulting in a greatly
reduced drainage capacity of the drainage network and, in turn, urban waterlogging disasters. In view
of the problem of stormwater inlet blockage, employing a typical waterlogging point in the Lianjiang
Middle Road area of Fuzhou city as the research object, the stormwater inlet equivalent drainage
method was adopted in this paper to characterize the drainage capacity of the pipe network and
enable the control of the stormwater inlet blockage state. Coupled with the stormwater inlet drainage
equation, an improved ITF-FLOOD two-dimensional hydrodynamic model was constructed, and the
influence of stormwater inlet blockage on urban waterlogging under different rainfall return periods
was simulated and analyzed. With increasing rainfall return period, the influences of stormwater inlet
Citation: Guo, W.; Zhai, M.; Lei, X.; blockage on both the maximum area and the depth of accumulated water in the study area gradually
Huang, H.; Long, Y.; Li, S. Two-
decreased compared with those of a nonblocked stormwater inlet, and the growth proportions
Dimensional Hydrodynamic
decreased from 43.35% and 34.58% under the 1-year rainfall scenario to 3.34% and 9.76% under the
Simulation of the Effect of Stormwater
50-year rainfall scenario, respectively. However, in terms of the change in the accumulated water
Inlet Blockage on Urban Waterlogging.
level, stormwater inlet blockage will cause an increase, and the influence will always be significant.
Water 2024, 16, 2029. https://doi.org/
10.3390/w16142029
Overall, stormwater inlet blockage aggravated the waterlogging risk and the extent of waterlogging.
Therefore, the results provided a reference for the construction of an urban waterlogging model and
Academic Editor: Luca
have certain guiding significance for waterlogging prevention and control in the study area prone to
Giovanni Lanza
stormwater inlet blockage.
Received: 12 June 2024
Revised: 9 July 2024 Keywords: urban waterlogging; stormwater inlet blockage; ITF-FLOOD model; equivalent drainage
Accepted: 15 July 2024 method; stormwater inlet drainage
Published: 17 July 2024

1. Introduction
Copyright: © 2024 by the authors.
Licensee MDPI, Basel, Switzerland.
In recent years, with the development of society and the economy, the urban popula-
This article is an open access article
tion has increased, and the urbanization rate in China has continued to increase [1]. Many
distributed under the terms and
natural bedding surfaces have been transformed into hardened pavements, which, while
conditions of the Creative Commons satisfying the needs of the urban population in terms of production and living, has also
Attribution (CC BY) license (https:// generated negative impacts, such as the frequent occurrence of disasters during extreme
creativecommons.org/licenses/by/ weather events. Such consequences have also led to the frequent occurrence of heavy
4.0/). rainfall and flooding [2]. Compared with inland cities, coastal cities are more severely

Water 2024, 16, 2029. https://doi.org/10.3390/w16142029 https://www.mdpi.com/journal/water

Water 2024, 16, 2029 2 of 20

affected by summer typhoon landfalls [3–5]. Therefore, targeted research on waterlogging

is important for urban construction.
Because of the increasing frequency of urban flooding disasters, urban flood man-
agement has become a real problem. To accurately determine the cause of waterlogging
and provide an effective decision-making basis for waterlogging prevention and control,
scholars have extensively studied the causes of waterlogging. Most scholars have attributed
urban flooding to both natural and human factors [6,7]. From the perspective of natural
factors, global warming has increasingly exacerbated the heat island effect and rain island
effect, leading to the frequent occurrence of heavy rainfall and flooding disasters [8–10].
Alemaw et al. [11] constructed a Storm Water Management Model (SWMM) to detect
overloads in sewer systems during extreme rainfall events and to simulate and predict
the relationships between precipitation parameters and sewer collection system overflows.
Affected by rainstorms and floods, backward river water can cause serious waterlogging.
Liu et al. [12] established a drainage model of a pipe network and simulated the drainage
of water outlets under different river levels. Regional topographic features also affect the
magnitude of urban rainfall. Based on 20 years of urban stormwater inundation records
from 1993 to 2012 for Japan, Komori et al. [13] identified and analyzed the frequent areas
of urban rainfall-related floods in the cities of Osaka and Nagoya, Japan, by principal
component analysis of topographic features. From a human factor perspective, low con-
struction standards of urban drainage systems can aggravate the occurrence of urban
flooding events [14]. In addition, due to the accelerating urbanization process, the in-
crease in impervious surfaces is one reason for the frequent occurrence of urban flooding
disasters [15–17]. From a land use change perspective, Ahmad et al. [18] analyzed the
increase in the urbanization rate in Srinagar city from 1972 to 2014 using remote sensing
data and assessed its impacts on the extent of waterlogging inundation and the risk of
waterlogging. Wang et al. [19] explored the impacts of landscape patterns and topography
on waterlogging based on records of historical waterlogging events.
However, most surface runoff generated by rainfall flows into the underground
drainage system through stormwater inlets at the surface. The drainage capacity of
stormwater inlets, which serve as the connection between the surface and the underground
drainage system, directly affects surface runoff and the drainage capacity of the under-
ground drainage system. In addition, stormwater inlets can intercept floating pollutants
to prevent underground pipeline blockage, which can result in drainage system failure.
In practice, stormwater inlets are often blocked due to the accumulation of leaves and
human waste and other factors, which greatly reduces the drainage capacity of stormwater
inlets, resulting in serious urban waterlogging disasters. Scholars have widely investigated
stormwater inlets, and the main research methods are physical model experiments, the-
oretical analysis, and three-dimensional numerical simulations [20–22]. By establishing
a physical model experimental platform, Kim et al. [23] derived a calculation equation
for stormwater inlet interception flow, and the factors influencing the capacity of a single
stormwater inlet were explored to provide a theoretical basis for the layout and installation
of stormwater inlets. Based on theoretical analysis, Noh et al. [24] reported that different
inflow conditions can affect the discharge patterns of stormwater inlets, so they classified
the discharge patterns of stormwater inlets into weir flow and orifice flow and calibrated
the flow coefficient under these two discharge patterns. With the use of the numerical
simulation method, Martins et al. [25] revealed the three-dimensional hydraulic character-
istics of the flow field within a stormwater inlet and derived flow coefficient equations that
can be directly used as the internal boundary conditions of the dual-drainage model. In
addition, Martins et al. [26] combined physical experiments with numerical simulations to
investigate the interception performance of stormwater inlets and found that the geometric
characteristics of stormwater inlets, road characteristics and flow velocity are all influencing
factors. Rubinato et al. [27] explored the factors influencing the discharge capacity of ten
circular stormwater grates and focused on the effects of the porosity of the stormwater grate
and the depth of water in front of the grate on the discharge capacity of stormwater inlets.
Water 2024, 16, 2029 3 of 20

The results showed that the interception efficiency of stormwater inlets was positively
correlated with the water depth in front of the gate and its porosity.
Through many studies, the causes of urban waterlogging have been expressed more
systematically, and the factors influencing urban waterlogging have been more clearly
defined. Failure of the drainage system due to stormwater inlet blockage is one reason.
However, existing studies have focused on only the drainage capacity of stormwater inlets
and its influencing factors from a blockage mechanism perspective, but there are few
applications in simulation analysis of urban waterlogging. Therefore, the problem of
stormwater inlet blockage must still be investigated.
In this paper, Guohuo East Road (the middle section of Lianjiang Middle Road and
Changle Middle Road), which is a typical waterlogging point in the Lianjiang Middle
Road area of Fuzhou city, was chosen as the research object, and the ITF-FLOOD model,
which was independently developed by the China Institute of Water Resources and Hy-
dropower Research, was applied. The stormwater inlet equivalent drainage method was
adopted to characterize the drainage capacity of the pipe network and enable the con-
trol of the stormwater inlet blockage state. Moreover, coupled with the stormwater inlet
drainage equation, an improved ITF-FLOOD two-dimensional hydrodynamic model was
constructed to explore the effects of stormwater inlet blockage on urban waterlogging.
The results of this study could provide a reference for the construction of urban waterlog-
ging models and could have certain guiding significance for waterlogging prevention and
control in areas prone to stormwater inlet blockage.

2. Materials and Methods

2.1. Research Area and Data
Fuzhou is a coastal city in China. The area mainly exhibits a subtropical monsoon
climate and a subtropical maritime climate. The city has a total land area of 11,968 km2 , and
there are 107 inland rivers in the urban area. The number of monthly maximum rainfall
days in the urban area is 18 days, the average number of annual rainfall days is 149 days,
the maximum daily rainfall is 170.9 mm, the average annual rainfall is 1367.5 mm, and the
maximum annual rainfall is 2074.6 mm. The annual flood season usually lasts from April
to October, with the peak of typhoon activity occurring from July to September. Typhoons
make landfall in Fuzhou twice a year on average, and the heavy precipitation associated
with typhoons is a major cause of the frequent waterlogging in Fuzhou city.
Based on the waterlogging records for Fuzhou from 2017 to 2022 obtained from the
governmental department, Guohuo East Road (the middle section of Lianjiang Middle
Road and Changle Middle Road) was adopted as the research object, and the selected
Water 2024, 16, x FOR PEER REVIEW waterlogging point is the intersection of Guohuo East Road and Lianhui 4 of Road,
21 as shown
in Figure 1. A total of 17 waterlogging events occurred in the area, 8 of which involved
stormwater inlet blockage.

Figure1. 1.
Figure Location
Location mapmap
of theofstudy
the study
area. area.

2.2. DEM Reconstruction

2.2.1. DEM Reconstruction Process
With the continuous development of science and technology, an increasing number of
hydrodynamic models have been applied in the simulation of urban waterlogging. Basic
Water 2024, 16, 2029 4 of 20

2.2. DEM Reconstruction

2.2.1. DEM Reconstruction Process
With the continuous development of science and technology, an increasing number of
hydrodynamic models have been applied in the simulation of urban waterlogging. Basic
data, especially DEM elevation data, directly affect the accuracy of the hydrodynamic
model simulation results. Even a small deviation from the basic data may lead to a large
error between the simulation results and the actual conditions. The original raster data
exhibit low elevation accuracy, but they cover a wide range and can characterize the overall
change in mountainous terrain. In contrast, manual mapping data exhibit high accuracy
and can accurately represent the elevation change in partial areas. The disadvantage is
that their coverage is low and limited to streets and surrounding road areas. Therefore,
within the street area, the low-precision basic data were transformed into higher-precision
elevation data by using multisource data fusion, which can not only meet the accuracy
requirements of the model but also clearly reflect the spatial change trend in the surface
elevation. This method provides a reference for urban waterlogging model construction.
The basic data needed for DEM reconstruction include the original DEM elevation
raster data, land use data, and ground elevation data of inspection wells, stormwater
grates, and manholes within the pipe network. First, the street ranges in the original DEM
elevation raster data were deleted, and the retained non-street range elevation raster data
were converted into elevation points. Then, the “Selection by location” function in ArcGIS
was applied to select elevation points within the building range. The elevation of the
selected elevation point was uniformly increased by 10 m using the “Field calculator” tool,
and elevation data of the corrected building elevation points were obtained. Next, the
street elevation was processed based on manually mapped ground elevation data from
inspection wells, stormwater grates, manholes, etc. First, the “Create fishing net” tool was
used to divide the street area into 2 m × 2 m fishing nets with imaging element sizes, and
the “Spatial connection” function was then employed to combine the fishing network with
the ground elevation data of inspection wells, stormwater gates and manholes to obtain
reconstructed street elevation data. The street range elevation points and non-street range
elevation points were merged to obtain a combined complete raster, and inverse distance
weighted (IDW) interpolation was performed. Finally, a reconstructed partial terrain DEM
Water 2024, 16, x FOR PEER REVIEW
based on multisource data fusion was obtained. Figure 2 shows a diagram of5the of 21
DEM
reconstruction process. Figure 3 shows a flowchart of the DEM reconstruction process.

Figure 2.
Figure 2. Diagram
Diagramofofthe
theDEM
DEMreconstruction
reconstructionprocess (2 (2
process mm × 2×
m2fishing nets).
m fishing nets).

Non street scope Street scope

Cut out Cut out

Original DEM
Pipeline network data
elevation raster data
Create fishing net
Non street range elevation Street range elevation raster
Water 2024, 16, 2029 5 of 20
Figure 2. Diagram of the DEM reconstruction process (2 m × 2 m fishing nets).

Non street scope Street scope

Cut out Cut out

Original DEM
Pipeline network data
elevation raster data
Create fishing net
Non street range elevation Street range elevation raster
raster data data

Non street elevation points

Non building and street

Building elevation points
elevation points
Raise by 10m
Revised building elevation Revised street elevation
points points

Merge

Revised elevation points

Interpolation

Revised elevation data

Figure3.3.DEM
Figure DEMreconstruction
reconstructionflow
flowchart.
chart.

2.2.2.
2.2.2.Spatial
SpatialInterpolation
Interpolation
Spatial
Spatial interpolation aims
interpolation aims to
to obtain the functional
obtain the functional relationship
relationship that
thatbest
bestexpresses
expressesa
aseries
seriesofofknown
known spatial data, and values can be derived for any location within
spatial data, and values can be derived for any location within the region the
region
based on this functional relationship. The obtained values can be discrete points or
based on this functional relationship. The obtained values can be discrete points or
partitioned
partitioneddata.
data.The
Theground
groundelevation
elevationdata
datawere
wereused
usedas asdiscrete
discretepoints.
points.To
Toimprove
improvethe the
DEM
DEMaccuracy,
accuracy,thetheInverse
InverseDistance
DistanceWeight
Weight(IDW)
(IDW)interpolation
interpolationmethod
methodwaswasadopted
adoptedforfor
spatial interpolation.
spatial interpolation.
The
TheIDW
IDWwaswasfirst
firstproposed
proposedby byShepard,
Shepard,which
whichisisbased
basedononthe
theprinciple
principleof
ofsimilarity
similarity
and
and assumes that the sampling points will have a local influence on the valuesof
assumes that the sampling points will have a local influence on the values ofthe
the
interpolation
interpolation points, which will gradually weaken as the distance increases [28]. For
points, which will gradually weaken as the distance increases [28]. For
example, suppose there are M known sample points near an interpolation point. Then, the
influence of these known sample points on that interpolation point is inversely related to
the distance. In other words, the sample points closer to the interpolation point exhibit a
higher weight.
In the IDW interpolation method, the distance factor is accounted for, which can be
combined with the direction factor when anisotropy occurs. In terms of the distance, the
weight factor decreases with increasing distance, which can be fitted with a power function.
Because it is a global interpolation algorithm and all the discrete points participate in the
numerical calculation, there are still certain requirements for the accuracy of the dataset.
Therefore, it is necessary to determine whether the considered discrete point data are
accurate before interpolation.
In this paper, interpolation was conducted after combining two types of elevation data,
and spatial interpolation was performed by limiting the interpolation distance (assuming
that the resolution of the original DEM is D, the resolution of the constructed mesh size
d should ensure D > 2d). When there is an elevation difference between the sidewalk
and the main lane, in the traditional spatial interpolation method, the elevation of the
interpolation point will be affected by the elevations of the sidewalk and the main lane.
Moreover, the elevation of the interpolation point will generally change smoothly, yielding
a high elevation value, which cannot reflect the vertical change in terrain. In the boundary
region of the road, the effect of abrupt vertical changes on the elevation of the interpolation
points should be considered only when there are reference points both inside and outside
 mesh size d should ensure D > 2d). When there is an elevation difference between the
sidewalk and the main lane, in the traditional spatial interpolation method, the elevation
of the interpolation point will be affected by the elevations of the sidewalk and the main
lane. Moreover, the elevation of the interpolation point will generally change smoothly,
Water 2024, 16, 2029
yielding a high elevation value, which cannot reflect the vertical change in terrain. In the
6 of 20
boundary region of the road, the effect of abrupt vertical changes on the elevation of the
interpolation points should be considered only when there are reference points both
inside and outside the road. If there are only reference points inside the road within the
the road. If there are only reference points inside the road within the restricted interval, the
restricted interval, the effect of points outside the road is not considered. This problem
effect of points outside the road is not considered. This problem can be effectively solved
can be effectively solved by the fusion of partial terrain data with multisource data.
by the fusion of partial terrain data with multisource data.
2.2.3. Comparison of the DEM Reconstruction Results
2.2.3. Comparison of the DEM Reconstruction Results
Figure 4 shows the original DEM elevation data (Fuzhou, China), and Figure 5 shows
Figure 4 shows the original DEM elevation data (Fuzhou, China), and Figure 5 shows
the locally reconstructed DEM elevation data. By comparison, the building and road areas
the locally reconstructed DEM elevation data. By comparison, the building and road areas
in the original DEM elevation data are very fuzzy and difficult to distinguish, while those
in the original DEM elevation data are very fuzzy and difficult to distinguish, while those
in the partially reconstructed DEM elevation data are very obvious.
in the partially reconstructed DEM elevation data are very obvious.

Water 2024, 16, x FOR PEER REVIEW 7 of 21

Figure 4. Original DEM
DEM and
and the
the intersection
intersectionof
ofthe
theLianjiang
LianjiangMiddle
MiddleRoad
Roadand
andJinlian
JinlianRoad.
Road.

Figure 5.
Figure 5. Reconstructed DEM and the intersection
intersection of
of the
theLianjiang
LianjiangMiddle
MiddleRoad
Roadand
andJinlian
JinlianRoad.
Road.

2.3.
2.3. Methodology
Methodology
2.3.1. ITF-FLOOD Model
2.3.1. ITF-FLOOD Model
The ITF-FLOOD model is known as the Integrated Terrestrial Fluxes model, abbrevi-
The ITF-FLOOD model is known as the Integrated Terrestrial Fluxes model,
ated as InTerFlux or ITF-FLOOD, which is a hydrodynamic model independently developed
abbreviated as InTerFlux or ITF-FLOOD, which is a hydrodynamic model independently
by the China Institute of Water Resources and Hydropower Research. In this model, graph-
developed by the China Institute of Water Resources and Hydropower Research. In this
ics processing unit (GPU) parallel computing is adopted based on the Compute Unified
model, graphics processing unit (GPU) parallel computing is adopted based on the
Device Architecture (CUDA). Preprocessing is completed on the central processing unit
Compute Unified Device Architecture (CUDA). Preprocessing is completed on the central
(CPU), and the grid parameters and physical quantities are then uploaded to the GPU for
processing unit (CPU), and the grid parameters and physical quantities are then uploaded
to the GPU for iterative calculation. Upon output, the required physical quantities are
downloaded to the CPU for postprocessing. Figure 6 shows the GPU computing process.
The model can be used to simulate the whole study area with high efficiency and high
resolution. However, the ITF-FLOOD model is not sensitive to the Manning coefficient n,
and the size and time step of the model’s grid elements have a significant impact on the
 The ITF-FLOOD model is known as the Integrated Terrestrial Fluxes model,
abbreviated as InTerFlux or ITF-FLOOD, which is a hydrodynamic model independently
developed by the China Institute of Water Resources and Hydropower Research. In this
model, graphics processing unit (GPU) parallel computing is adopted based on the
Water 2024, 16, 2029 7 of 20
Compute Unified Device Architecture (CUDA). Preprocessing is completed on the central
processing unit (CPU), and the grid parameters and physical quantities are then uploaded
to the GPU for iterative calculation. Upon output, the required physical quantities are
iterative calculation. Upon output, the required physical quantities are downloaded to the
downloaded to the CPU for postprocessing. Figure 6 shows the GPU computing process.
CPU for postprocessing. Figure 6 shows the GPU computing process. The model can be
The model can be used to simulate the whole study area with high efficiency and high
used to simulate the whole study area with high efficiency and high resolution. However,
resolution. However, the ITF-FLOOD model is not sensitive to the Manning coefficient n,
the ITF-FLOOD model is not sensitive to the Manning coefficient n, and the size and time
and the size and time step of the model’s grid elements have a significant impact on the
step of the model’s grid elements have a significant impact on the model’s results. An
model’s results. An excessively large grid element size can lead to model instability, while
excessively large grid element size can lead to model instability, while an excessively long
an excessively long time step can result in inaccurate model results.
time step can result in inaccurate model results.

Figure 6. GPU
GPU computing
computing process.

The model
The model can
can capture
capture both
both surface
surface hydrodynamics
hydrodynamics and
and water
water flow
flow in
in the
the drainage
drainage
system.
system.
Channel flows are treated using a one-dimensional approach that captures the down-
stream propagation of flood waves and the response of water flow toward free surface
slopes. The continuity and momentum equations are expressed in Equations (1) and (2),
respectively:
∂Q ∂A
+ +q = 0 (1)
∂t ∂t
S0 = S f (2)
The flood inundation process can be modeled using the storage unit method. Since
the topography of a city does not significantly change and there are no surges or rapidly
changing currents during urban flooding, this method can be used to simulate the surface
water flow process with the following control equations:

∆h ∆V
= (3)
∆t ∆X∆Y
where ∆h is the unit change in the water depth (m), ∆V is the unit change in the volume
(m3 ), ∆t is the time interval (s), and ∆X and ∆Y are orthogonal directions.
The Saint–Venant inertia equation can be used for the flow of two neighboring units:

qt − gh f low ∆tSsur f
qt+∆t =   (4)
1 + gn2 |∆t |/h7/3
f low

where qt is the flow velocity at the boundary of the adjacent cell (m/s), g is the gravitational
acceleration (m/s2 ), n is the roughness, Ssur f is the water level gradient of the adjacent cell,
and h f low is the depth of water flow at the interface of the adjacent cell (m).
Water 2024, 16, 2029 8 of 20

The water flow depth h f low at the interface of adjacent units can be calculated as
follows:
h f low = max(h1 + z1 , h2 + z2 ) − max(z1 , z2 ) (5)
where h1 and h2 are the water depths of the two neighboring units (m) and z1 and z2 are
the ground elevations of the two neighboring units (m).
Soil infiltration can be calculated using the Green–Ampt method [29] as follows:
 
φs θd
f p = Ks 1 + (6)
F

where f p is the soil infiltration capacity (mm/h), Ks is the saturated hydraulic conductivity
(mm/h), φs is the soil suction at the wetting front, θd is the difference between the saturated
and initial soil water contents, and F is the cumulative infiltration (mm).
The ITF-FLOOD model comprises one control parameter file (csv format), six asc
Water 2024, 16, x FOR PEER REVIEWformat files and four csv format files, as detailed in Table 1. 9 of 21

Table 1. ITF-FLOOD model file description.

4 Number
Serial i03_demRevRv_bmh190926.asc
Example of the File Name DEMDescription
elevation data grid
5 1 i04_ldcvFineGeo_bmh190926.asc
ctrl.itf.bmh.201017-1.csv Land cover grid
Control parameter file
6 2 i05_soilPoly_bmh201015.asc
i01_mask3_geo_bmh190927.asc Soil distribution
Program-controlled grid
grid
7 3 i02_surfWatBody_bmh201014.asc
i06_soilPoly2type_bmh201015.csv Surface drainage
Soil polygon ID andgridsoil type ID
4 i03_demRevRv_bmh190926.asc DEM elevation data grid
8 5 i07_bndryWatLvl_2days_bmh201015.csv
i04_ldcvFineGeo_bmh190926.asc Time seriesLandof cover
boundary
grid water levels
9 6 i08_prec_1gauge_1day_200612.csv
i05_soilPoly_bmh201015.asc Rainfall
Soil timegrid
distribution series
10 7 i06_soilPoly2type_bmh201015.csv
i09_strmGaugePara_bmh200903_2.csv Soil polygon
Number ID and
of river soil type sections
gauging ID
8 i07_bndryWatLvl_2days_bmh201015.csv Time series of boundary water levels
11 9 i10_gaugeUpDnZonesClip_bmh190927.asc
i08_prec_1gauge_1day_200612.csv Cross-sectional gridseries
Rainfall time for river gauging
10 i09_strmGaugePara_bmh200903_2.csv Number of river gauging sections
11 i10_gaugeUpDnZonesClip_bmh190927.asc Cross-sectional grid for river gauging
2.3.2. SWMM and ITF-FLOOD Coupling Model
The SWMM has a hydrological and hydrodynamic module, which can perform one-
2.3.2. SWMM and ITF-FLOOD Coupling Model
dimensional hydrodynamic simulation calculations, but cannot perform two-dimensional
The SWMM has a hydrological and hydrodynamic module, which can perform one-
surface inundation processes. The ITF-FLOOD model has the ability to calculate two-
dimensional hydrodynamic simulation calculations, but cannot perform two-dimensional sur-
dimensional surface inundation processes, but there are certain shortcomings in one-
face inundation processes. The ITF-FLOOD model has the ability to calculate two-dimensional
dimensional hydrodynamic simulation calculations. Therefore, by coupling the SWMM
surface inundation processes, but there are certain shortcomings in one-dimensional hydro-
with thesimulation
dynamic ITF-FLOOD model, Therefore,
calculations. a comprehensive calculation
by coupling the SWMM of with
urbanthe hydrology,
ITF-FLOOD one-
dimensional
model, pipelinecalculation
a comprehensive networkof urban
hydrodynamics, and two-dimensional
hydrology, one-dimensional surface
pipeline network
hydrodynamics can be completed, thereby making a more detailed and
hydrodynamics, and two-dimensional surface hydrodynamics can be completed, therebycomprehensive
analysisa more
making of urban waterlogging
detailed situations.
and comprehensive The ofcoupling
analysis process ofsituations.
urban waterlogging the waterlogging
The
couplingprocess
coupling modelofisthe
shown in Figurecoupling
waterlogging 7. model is shown in Figure 7.

Basic data such as Basic data such as terrain, Match pipeline network
pipeline network, terrain, underlying surface, and nodes with grid units to
and underlying surface research boundaries data generate coupling files

Structure Structure

SWMM model ITF-FLOOD model Import ArcGIS to view

the maximum water
depth and spatiotemporal
Run distribution map of water
depth at each time step
SWMM and ITF-FLOOD
". rep" result file
coupling model
View
Extract Operation Results
Output
". csv" overflow
model operation model operation
sequence file

Figure7.7.SWMM
Figure SWMMand
andITF-FLOOD
ITF-FLOOD coupling
coupling process.
process.

Due to the complexity of considering bidirectional coupling in synchronization, it is

difficult to achieve in a short period of time. Therefore, the coupling model only considers
unidirectional coupling, that is, the SWMM simulates the pipeline overflow process as the
input condition for the ITF-FLOOD model to calculate the two-dimensional surface
Water 2024, 16, 2029 9 of 20

Due to the complexity of considering bidirectional coupling in synchronization, it is

difficult to achieve in a short period of time. Therefore, the coupling model only considers
unidirectional coupling, that is, the SWMM simulates the pipeline overflow process as
the input condition for the ITF-FLOOD model to calculate the two-dimensional surface
inundation process.

2.3.3. Generalized Drainage Capacity of Pipe Network

The drainage process of the pipe network was generalized by using the equivalent
drainage method for stormwater inlets, and a schematic of the equivalent drainage method
for stormwater inlets is shown in Figure 8. The method involves setting the sink term in the
continuity equation to characterize the centralized drainage process of stormwater inlets,
as expressed in Equation (7). This method accounts for the actual pipe network layout
range and the actual amount of water flow into the pipe network from stormwater inlets.
Yang et al. [30] demonstrated that this method can provide a suitable simulation accuracy
by considering the Xixian New District in Shaanxi Province as the study area.

i − f − hinlet , i f Inlet = 1
R=
i − f, i f Inlet = 0

3.6 × 106 qinlet

hinlet = (7)
Araster
where R is the net rain rate (mm/h), i is the rainfall intensity (mm/h), f is the infiltration
Water 2024, 16, x FOR PEER REVIEW
rate (mm/h), hinlet is the drainage capacity of the grid where the stormwater inlet is located
10 of 21
(mm/h), qinlet is the discharge of the stormwater inlet (m3 /s), and Araster is the area of a
single grid (m2 ).

Figure 8. Diagram of the equivalent drainage method for stormwater inlets.

2.3.4. Simulation of STORMWATER Inlet Blockage

(1) i - f - hinletinlet
Calculation equation for theRstormwater Inlet = 1 capacity
, if discharge
= 
With increasing storm flow, the depth if Inlet =rainwater
i - f ,of the confluent 0 in front of stormwater
inlets increases, and the flow pattern of stormwater inlets can be categorized into weir flow,
orifice flow, and nozzle flow according to 3the 106 qinlet capacity of the connecting pipe
.6 ×drainage
and inlet grate and the capacity of the = well. When stormwater inlets are blocked,
hinletstorm (7)
Araster
their flow discharge pattern and capacity are not easy to determine. Based on physical
model R
where tests andnet
is the therain
dimension analysis
rate (mm/h), i is method, Xiaintensity
the rainfall et al. [31](mm/h),
established an expression
f is the infiltration
rate (mm/h), hinlet is the drainage capacity of the grid where the stormwater inlet is located
(mm/h), qinlet is the discharge of the stormwater inlet (m3/s), and Araster is the area of a single
grid (m2).

2.3.4. Simulation of STORMWATER Inlet Blockage

(1) Calculation equation for the stormwater inlet discharge capacity
Water 2024, 16, 2029 10 of 20

for the drainage capacity of stormwater inlets under different blockage conditions of the
connecting pipe and rain grate, as expressed in Equations (8) and (9).

qinlet = ka∗ AuFr −0.85 (8)

−1.285CR 2 + 0.285CR + 1(Stormwater inlet blockage)


k= (9)
−0.428CR 2 − 0.572CR + 1(Connection pipe blockage)
where qinlet is the discharge of the stormwater inlet (m3 /s), k is the blockage coefficient, a∗
is the magnitude parameter, A is the net water area of the rain grate (m2 ), u is the flow rate
Water 2024, 16, x FOR PEER REVIEW 11 of 21
(m/s), Fr is the Froude number, and CR is the blockage degree.
(2) Control of the stormwater inlet blockage state
usedTo tocontrol
controlthethestormwater
stormwater inlet blockage
inlet blockagestate, a stormwater
state. When the inlet blockageinlets
stormwater state are
control
not
file, namely
blocked, the i20_block_dtkz.csv,
blockage degree CR was is 0, defined, and the blockage
and the blockage coefficient coefficient k was used
k is 1. Moreover, to
when
control the stormwater
the stormwater inlets inlet
are blockage
completely state. When the
blocked, thestormwater inlets areCRnot
blockage degree is blocked,
1, and the
blockage
blockage degree CR is 0,
coefficient k and
is 0.theThe
blockage
degreecoefficient k is 1. Moreover,
and coefficient of partial when the stormwater
blockage are set
inlets are completely
proportionally. blocked, the blockage degree C R is 1, and the blockage coefficient k is
0. The degree and coefficient of partial blockage are set proportionally.
2.4. Model Building
2.4. Model Building
2.4.1. Construction of a Two-Dimensional Hydrodynamic Model Coupled with the
2.4.1. Construction of a Two-Dimensional Hydrodynamic Model Coupled with the
Stormwater Inlet
Stormwater Inlet Discharge
Discharge
Runoff simulation
Runoff simulation in in the
the study
study area was conducted
area was conducted usingusing the
the direct
direct rainfall
rainfall method,
method,
namely, the raster-based runoff simulation method. The whole
namely, the raster-based runoff simulation method. The whole study area was study area was divided
divided
into structural
into structural computational
computational units
units according
according to to topographic
topographic rasters.
rasters. Each
Each computational
computational
unit contains
unit contains information
information such
such asas elevation,
elevation, roughness
roughness coefficient,
coefficient, infiltration
infiltration parameters,
parameters,
percentage of
percentage of the
the impervious
impervious area,
area, and
and depth
depth ofof depression.
depression. The
The study
study area
area waswas divided
divided
into 92,516 rasters, and the pixel size was 2 m × 2 m. The simulation time step was 10
into 92,516 rasters, and the pixel size was 2 m × 2 m. The simulation time step was 10 min.
min.
To ensure calculation stability, the initial surface
To ensure calculation stability, the initial surface water water depth was set to 0.03 m, and the
boundary of the study area was defined as an open boundary. The basic data (Fuzhou,
China) needed for model construction are shown in Figure 9. 9. The land use was divided
into five types, including bare land, water systems, vegetation, vegetation, roads,
roads, andand buildings.
buildings.

(a) (b)
Figure 9.
Figure 9. Basic data for
Basic data for the
the study
study area:
area: (a) land use;
(a) land use; (b)
(b) reconstruction
reconstruction of
of DEM
DEM elevation
elevation data.
data.

2.4.2. Parameter Calibration and Validation

To
To ensure that the
the model
modelexhibits
exhibitshigh
highaccuracy
accuracyand andpracticability,
practicability, it necessary
it is is necessaryto
to calibrate and verify the model parameters. Based on the initial model
calibrate and verify the model parameters. Based on the initial model parameters, the parameters,
the historical
historical rainfall
rainfall datadata
werewere input,
input, and
and themeasured
the measureddata datawere
were compared
compared with with the
simulated data. By adjusting the parameter values, the simulated results could be fitted fitted
better with the measured data.
The
The Nash–Sutcliffe
Nash–Sutcliffe efficiency coefficient (NSE)
efficiency coefficient (NSE) is
is aa commonly
commonly used used model
model accuracy
accuracy
evaluation
evaluation index that measures the consistency between simulated and measured values
index that measures the consistency between simulated and measured values
by comparing the
by comparing the process
process lines.
lines.The
Thecloser
closerthe
theNSE
NSEvalue
valueis is
toto
1, 1,
thethe higher
higher thethe degree
degree of
of agreement
agreement between
between thethe simulationresults
simulation resultsand
andthe
themonitoring
monitoring values.
values. ItIt is
is generally
generally
believed that when NSE < 0, the simulation results are completely inconsistent with the
actual situation, and it is meaningless to use the simulation results to describe the display
process. In actual parameter calibration and model validation, it is generally believed that
NSE ≥ 0.5, and the simulation results are basically consistent with the monitoring values.
For high standard models, the critical value of evaluation indicators can be further
Water 2024, 16, 2029 11 of 20

believed that when NSE < 0, the simulation results are completely inconsistent with the
actual situation, and it is meaningless to use the simulation results to describe the display
process. In actual parameter calibration and model validation, it is generally believed
that NSE ≥ 0.5, and the simulation results are basically consistent with the monitoring
values. For high standard models, the critical value of evaluation indicators can be further
increased to define simulation results that meet the requirements.
Rainfall from 12:00 on 1 August to 0:00 on 2 August 2021 was selected to determine
SWMM parameters, and rainfall data from 9:00 on 5 August to 00:00 on 6 August and from
Water 2024, 16, x FOR PEER REVIEW00:00 on 6 August to 14:00 on 7 August (Figure 10) were selected to verify the model. 12 ofThe
21
Water 2024, 16, x FOR PEER REVIEWmeasured water levels of two pipeline level measurement stations were compared 12 with
of 21
the simulated water levels of the model (the pipeline level measurement station numbers
are marked as J1 and J2). Figure 11 shows the comparison results of underground water
levels
levels during
during rainfall
rainfall from
from 9:00
9:00 on
on 55 August
August to
to 00:00
00:00 on
on 66 August,
August,andandFigure
Figure 12
12 shows
shows thethe
levels during
comparison rainfall from 9:00 on 5 August to 00:00 on 6 August, and Figure 12 shows the
comparison results of underground water levels during rainfall from 00:00 on 6 August to
results of underground water levels during rainfall from 00:00 on 6 August to
comparison
14:00 results of underground water levels during rainfall from 00:00 on 6 August to
14:00 on
on 77 August.
August.
14:00 on 7 August.

(a) (b)
(a) (b)

(c)
(c)
Figure
Figure 10.
10. Measured
Measured rainfall
rainfall process:
process: (a)(a)rainfall
rainfallprocess
processfrom
from 12:00
12:00 on
on 11 August
August2021
2021to
to 0:00
0:00 on
on 22
Figure 10.
August Measured
2021; rainfall
(b) rainfall process:
process from (a) rainfall
9:00 process2021
on 5 August fromto12:00
0:00 on
on 16 August
August 2021
2021;to(c)0:00 on 2
rainfall
August 2021; (b)
August from
2021; (b) rainfall process from 9:00 on
on 755August
August 2021
2021 to
to 0:00
0:00 on
on 66 August
August 2021;
2021; (c)
(c) rainfall
process 0:00 rainfall process
on 6 August 2021 from 9:00 on
to 14:00 August2021. rainfall
process
process from
from 0:00
0:00 on
on 66 August
August 2021
2021 to to 14:00
14:00 on
on 77 August 2021.
August 2021.

(a) (b)
(a) (b)
Figure 11. 5 August, 9:00–6 August, 00:00 rainfall well level comparison results: (a) comparison of J1
Figure 11. 5 August,
downhole 9:00–6 August,results;
00:00 rainfall well level comparison results: (a) comparison of J1
Figure 11. liquid level
5 August, simulation
9:00–6 (b) comparison
August, 00:00 rainfall well level of J2 downhole
comparison liquid
results: (a) level simulation
comparison of J1
downhole
results. liquid level simulation results; (b) comparison of J2 downhole liquid level simulation
downhole liquid level simulation results; (b) comparison of J2 downhole liquid level simulation results.
results.
Water 2024, 16,
Water 2024,
2024, 16, 2029
16, x
x FOR
FOR PEER
PEER REVIEW
REVIEW 13
1312of
of 21
of 20
21

(a)
(a) (b)
(b)
Figure
Figure 12.
12. 66 August,
August, 00:00–7
00:00–7 August,
August, 14:00
14:00 rainfall
rainfall well
well level
level comparison
comparison results:
results: (a)
(a) comparison
comparison of
of
Figure
J1 12. 6 liquid
August, 00:00–7 August, 14:00(b)rainfall well level comparison results: (a) compari-
J1 downhole liquid level simulation results; (b) comparison of J2 downhole liquid level simulation
downhole level simulation results; comparison of J2 downhole liquid level simulation
son of J1 downhole liquid level simulation results; (b) comparison of J2 downhole liquid level
results.
results.
simulation results.
From
From Figure
Figure 10, 10, itit can
can be be seen
seen that
that during
during the the rainy
rainy period
period from from 9:009:00 on on 55 August
August to to
From Figure 10, it can be seen that during the rainy period from 9:00 on 5 August to
00:00
00:00 on
on 6
6 August,
August, the
the time
time error
error of
of the
the peak
peak liquid
liquid level
level underground
underground in
in inspection
inspection well
well
00:00 on 6 August, the time error of the peak liquid level underground in inspection well
J1
J1 was
was 000 min,
min,
min, thethe error
the error
error of of the
of the peak
the peak
peak sizesize was
size was 0.250
was 0.250
0.250 m,m,
m, thethe relative
the relative
relative errorerror
error was was 0.099,
was 0.099,
0.099, andand the
and the
the
J1 was
NSE
NSE was 0.813; the peak time error of the underground liquid level in inspection well J2
NSE was 0.813; the peak time error of the underground liquid level in inspection well J2
was 0.813; the peak time error of the underground liquid level in inspection well J2
is
is 00 min,
min, the
the peak
peak size
size error
error is
is 0.112
0.112 m,
m, the
the relative
relative error
error is
is 0.052,
0.052, and
and the
the NSE
NSE is
is 0.977.
0.977.
is 0 min, the peak size error is 0.112 m, the relative error is 0.052, and the NSE is 0.977.
From
From Figure 11,
Figure 11, it
11, it can
it can
can be be
be seen
seen that during the rainy period from 00:00 on
on 666 August to 14:00
From Figure seen thatthat during
during the the rainy
rainy period
period fromfrom 00:0000:00 on August
August to to 14:00
14:00
on
on 7
7 August,
August, the
the time
time error
error of
of the
the peak
peak liquid
liquid level
level in
in inspection
inspection well
well J1
J1 was
was 10
10 min,
min, the
the
on 7 August, the time error of the peak liquid level in inspection well J1 was 10 min, the
error
error of
of the
the peak
peak size
size was
was 0.039
0.039 m,
m, the
the relative
relative error
error was
was 0.017,
0.017, and
and the
the NSE
NSE was
was 0.508;
0.508; the
the
error of the peak size was 0.039 m, the relative error was 0.017, and the NSE was 0.508;
peak
peak time error
timetime
error of
of the
the liquid
liquid level
level in
in the
the J2
J2 inspection
inspection well
well is
is 00ismin,
min, the
thethe peak
peak size
size error
error is
is
the peak error of the liquid level in the J2 inspection well 0 min, peak size error
0.041
0.041 m,
m, the
the relative
relative error
error is
is 0.020,
0.020, and
and the
the NSE
NSE is
is 0.987.
0.987. The
The model
model performs
performs well
well in
in terms
terms
is 0.041 m, the relative error is 0.020, and the NSE is 0.987. The model performs well in
of
of peak
peakoftime
terms peakerror
time timeand
error and
error peak
andsize
peak size
peak error,
size with
error, error,a
with peak
peakatime
awith peakdeviation
time deviation
time deviation of
of less
lessof than
thanless1 h
h and
1than and1h aa
peak
peak deviation
and adeviation of less
of lessof
peak deviation than
than 25%,
less25%, without
thanwithout
25%, without significant
significant deviation.
deviation.
significant At the
At theAt
deviation. same
same time,
the time,
same thethe
time, Nash
Nash
the
efficiency
efficiency coefficient,
coefficient, NSE,
NSE, is
is greater
greater than
than 0.5,
0.5, indicating
indicating
Nash efficiency coefficient, NSE, is greater than 0.5, indicating a high degree of fit in the a
a high
high degree
degree of
of fit
fit in
in the
the
simulation
simulation results
results andand aa goodgood simulation
simulation effect.effect.
effect.
Two
Two rainfall
rainfall grades,
grades, namely namely heavy heavy rain rain and
and rainstorm,
rainstorm, were were selected
selected for for coupled
coupled
model calibration
model calibration
calibration and and validation.
andvalidation.
validation.The The rainfall
Therainfall processes
rainfallprocesses
processesduring during
duringthe the
the two
two
two events
events
events are
areare shown
shown
shown in
in
in Figure
Figure
Figure 13.
13. 13. The
TheThe maximum
maximum
maximum rainfall
rainfall
rainfall intensity
intensity
intensity was
waswas 61.8
61.861.8
mm/h, mm/h,
mm/h, with
withwith a totala total
a total
rainfall rainfall
rainfall
of 45.4of
of 45.4
45.4
mm
mm
mm
for thefor the
the heavy
forheavy heavy
rainfallrainfall
eventevent
rainfall on 1 on
event on 11 September
September
September 2022,2022,
2022,
and the and the
the maximum
andmaximum maximum rainfallrainfall
rainfall intensity
intensity
intensity was
was
117 117
117 mm/h,
wasmm/h, withwith
mm/h, with
a totalaa total
total rainfall
rainfall
rainfall of 94.1 of 94.1
of mm mm
94.1 for
mmthe for the
the rainstorm
forrainstorm on 27 on
rainstorm on 27
27 April
April 2023. 2023.
April On 27On
2023. On 27
27
April
April
April 2023, debris
2023, debris
2023, debris blocked blocked
blocked the connecting
the connecting
the connecting branch
branchbranch
pipes in pipes
pipes in a total
in aoftotal
a total of 6 rainwater
of 6 rainwater
6 rainwater wells withinwells
wells
within
within
the study the study
thearea,
studyand area,
area,theand
and the
the degree
degree degree of
of blockage
of blockage blockage was
was relatively
was relatively high. high.
relatively high.
For the For the
the rainfall
Forrainfall eventevent
rainfall event
on 1
on
on 11 September
September 2022,
2022, since
September 2022,the since
actual
since the actual
actual blockage
theblockage is unknown,
blockage is
is unknown,
it was assumed
unknown, it
it was
wasthat assumed that
that the
the locations
assumed of
the
locations
stormwater
locations of stormwater
of inlet blockage
stormwater inlet
were
inlet blockage
the same
blockage were the
the same
as those
were during
same as those
as the
those during
rainfall
during the
event
the on rainfall
27 April
rainfall event
event on
2023.
on
27
27 April
April 2023.
2023.

(a)
(a) (b)
(b)
Figure
Figure 13.
13. Rainfall
Rainfall data
data for
for the
the two
two rainfall
rainfall events:
events: (a)
(a) on
on 11 September
September 2022;
2022; (b)
(b) on
on 27
27 April
April 2023.
2023.

Since the
Since the actual
theactual degree
degree
actual of
of stormwater
of stormwater
degree inlet
inlet blockage
inlet blockage
stormwater could
could not
blockage not
not be
be accurately
be accurately
could described,
accurately
described, it
it was assumed
described, was assumed
it wasthat that
the degree
assumed the
that the degree
of degree
blockageof blockage
of of of each
each stormwater
blockage stormwater inlet
inlet blockage
inlet blockage
of each stormwater was the
blockage
was
was the
samethe same
during
samebothduring both
rainfall
during rainfall
bothevents events
in events
rainfall in the
the calibration calibration process.
process. Figure
in the calibration Figure
14 shows
process. 14
Figurethe shows
shows the
14 simulation
the
simulation
simulation results
results for
for the
the maximum
maximum depthdepth ofof accumulated
accumulated water
water during
during the
the two
two rainfall
rainfall
Water 2024, 16, x FOR PEER REVIEW 14 of 21
Water 2024, 16, 2029 13 of 20

events.
resultsDue to maximum
for the the lack ofdepth
information on the measured
of accumulated water duringareatheoftwo
accumulated water,
rainfall events. Duethe
simulated results were relatively accurate compared with the field investigation
to the lack of information on the measured area of accumulated water, the simulated results results
and ponding
were disposal
relatively photos.
accurate The with
compared maximum depths
the field of accumulated
investigation results and water during
ponding the two
disposal
rainfall
photos.events were compared.
The maximum depths of During the rainfall
accumulated event
water on the
during 1 September
two rainfall2022, the were
events degree
ofcompared.
stormwater inlet the
During blockage
rainfallwas
event50%,
on 1the simulated
September maximum
2022, the degree depth of accumulated
of stormwater inlet
water was 23
blockage wascm,
50%,and the
the measured
simulated depth ofdepth
maximum accumulated water was
of accumulated 20 cm,
water was with
23 cm,anand
error
ofthe measured
15%. During thedepth of accumulated
rainfall event on 27water
Aprilwas 20the
2023, cm,degree
with an oferror of 15%. inlet
stormwater During the
blockage
rainfall
was 80%,event on 27 Aprilmaximum
the simulated 2023, the degree
depth of of stormwater
accumulated inlet blockage
water was 30wascm,80%,
andthethe
simulated maximum depth of accumulated water was 30 cm, and the
measured depth of accumulated water was 28 cm, with an error of 7.14%. The errors measured depth of of
accumulated water was 28 cm, with an error of 7.14%. The errors of the two sets
the two sets of rainfall simulation results were less than 20%, indicating that the model is of rainfall
simulation
relatively results were less than 20%, indicating that the model is relatively accurate.
accurate.

(a) (b)

(c) (d)
Figure
Figure14.14.Maximum
Maximumdepth ofof
depth accumulated
accumulatedwater
watersimulation
simulationresults
resultsfor
forthe
thetwo
tworainfall
rainfallevents:
events:(a)
nonblocked stormwater inlet on 1 September 2022; (b) stormwater inlet blockage on 1 September
(a) nonblocked stormwater inlet on 1 September 2022; (b) stormwater inlet blockage on 1 Septem-
2022; (c) nonblocked stormwater inlet on 27 April 2023; (d) stormwater inlet blockage on 27 April
ber 2022; (c) nonblocked stormwater inlet on 27 April 2023; (d) stormwater inlet blockage on
2023.
27 April 2023.

3.3.
Analysis
Analysisofofthe
theImpact
Impactof
ofStormwater Inlet Blockage
Stormwater Inlet Blockageon
onWaterlogging
Waterlogging
3.1.Different
3.1. DifferentRainfall
RainfallScenarios
Scenarios
Accordingtoto the
According the rainstorm
rainstorm intensity
intensity equation
equation for
for Fuzhou
Fuzhoucity,
city,asasexpressed in in
expressed
Equation (10), the Chicago rainfall generator was employed [32], the rainfall
Equation (10), the Chicago rainfall generator was employed [32], the rainfall duration was duration
was set to 6 h, the time interval was set to 10 min, and the rain peak coefficient was set to 0.4.
set to 6 h, the time interval was set to 10 min, and the rain peak coefficient was set to 0.4.
Six rainfall scenarios were designed, namely, 1-year, 3-year, 5-year, 10-year, 30-year, and
Six rainfall scenarios were designed, namely, 1-year, 3-year, 5-year, 10-year, 30-year, and
50-year recurrence scenarios. The rainfall events under the different scenarios are shown in
50-year recurrence scenarios. The rainfall events under the different scenarios are shown
Figure 15. The maximum rainfall intensity and total rainfall amount under the different
inrainfall
Figure scenarios
15. The maximum rainfallinintensity
are summarized Table 2. and total rainfall amount under the different
rainfall scenarios are summarized in Table 2.
2136.312(1 + 0.700LgT )
q = 2136.312（1 +0.7110.700LgT） (10)
q= (t + 7.576) (10)
（t+7.576）0.711
where q is the intensity of the rainstorm (mm/min), T is the rainfall recurrence period, and
where q is
t is the the intensity
rainfall durationof(min).
the rainstorm (mm/min), T is the rainfall recurrence period, and
t is the rainfall duration (min).
Water 2024,
Water 16, x2029
2024, 16, FOR PEER REVIEW 1514ofof 21
20

(a) (b)

(c) (d)

(e) (f)
Figure 15. The process of different rainfall return periods: (a) 1a; (b) 3a; (c) 5a; (d) 10a; (e) 30a; (f)
Figure 15. The process of different rainfall return periods: (a) 1a; (b) 3a; (c) 5a; (d) 10a; (e) 30a; (f) 50a.
50a.
Table 2. Options of the different rainfall return periods.
Table 2. Options of the different rainfall return periods.
Rainfall Duration
Serial
Serial Number Time Time Interval
Interval Maximum Rainfall
Rainfall ScenariosMaximum Rainfall
Total Rainfall
Total Rainfall
Rainfall Duration
(h) (h) (min) Rainfall Scenarios Intensity (mm/h) (mm)
Number (min) Intensity (mm/h) (mm)
1 6 10 P = 1a 89.68 69.28
1 6 10 P = 1a 89.68 69.28
2 6 10 P = 3a 119.63 92.66
2 3 6 6 10 10 P = 3a P = 5a 119.63
133.55 92.66
103.18
3 4 6 6 10 10 P = 5aP = 10a 133.55
152.45 103.18
117.78
4 5 6 6 10 10 P = 10aP = 30a 182.40
152.45 140.91
117.78
5 6 6 6
10 10
P = 30a P = 50a 196.33
182.40 151.67
140.91
6 6 10 P = 50a 196.33 151.67
3.2. Analysis of the Simulation Results
3.2. Analysis
Notably,ofstormwater
the Simulation Results
inlet blockage was set to be consistent with that during the rainfall
event on 27 April
Notably, 2023, and
stormwater the blockage
inlet influenceswasof stormwater inlet blockage
set to be consistent with on waterlogging
that during the
under the
rainfall different
event on 27 rainfall
Aprilscenarios
2023, and were
thecompared
influences andofanalyzed.
stormwater For inlet
the convenience
blockage on of
observation, the depths of accumulated water were classified into five
waterlogging under the different rainfall scenarios were compared and analyzed. For thelevels: the blue risk
level (0~10 cm),
convenience green risk level
of observation, the(10~20
depthscm), orange risk level
of accumulated water(20~30
werecm), dark orange
classified risk
into five
level (30~50 cm), and red risk level (greater than 50 cm). Figure 16 shows
levels: the blue risk level (0~10 cm), green risk level (10~20 cm), orange risk level (20~30the simulation
results
cm), of the
dark maximum
orange depth
risk level of accumulated
(30~50 cm), and redwater under(greater
risk level the different
than 50rainfall scenarios.
cm). Figure 16
shows the simulation results of the maximum depth of accumulated water under the
different rainfall scenarios.
Water 2024,
Water 16,16,
2024, x FOR
2029 PEER REVIEW 1516 of 21
of 20

(a)

(b)
Figure 16. Simulation results under different rainfall return periods with different stormwater inlet
Figure 16. Simulation results under different rainfall return periods with different stormwater inlet
blockage conditions: (a) nonblocked stormwater inlet; (b) stormwater inlet blockage.
blockage conditions: (a) nonblocked stormwater inlet; (b) stormwater inlet blockage.
Both
Boththethetotal
total area
area and the depth
and the depthof ofaccumulated
accumulatedwater waterwerewere compared
compared between
between thethe
different
different rainfall scenarios with different stormwater inlet blockage conditions, and thethe
rainfall scenarios with different stormwater inlet blockage conditions, and
results
resultsare
arelisted
listed in
in Table
Table 3.3. In the case
In the case of
ofthe
the1-year
1-yearrainfall
rainfallscenario,
scenario, thethe area
area and
and thethe
maximum
maximumdepth depth of of accumulated
accumulated waterwater due
duetotostormwater
stormwaterinlet inletblockage
blockage increased
increased by by
43.35%
43.35%and and31.58%,
31.58%, respectively, comparedwith
respectively, compared withthose
thoseassociated
associated with
with thethe nonblocked
nonblocked
stormwater
stormwater inlet. Under Underthe the3-year
3-year rainfall
rainfall scenario,
scenario, the growth
the growth rates
rates of of the
the area andarea
maxi-and
mum depthdepth
maximum of accumulated
of accumulatedwater decreased by 19.32%by
water decreased and 20.00%,
19.32% respectively.
and When the
20.00%, respectively.
rainfall
When therecurrence period wasperiod
rainfall recurrence once inwas
5 years,
oncecompared
in 5 years,with that for with
compared a rainfall
thatrecurrence
for a rainfall
period of once in 3 years, the growth rate of the area of accumulated
recurrence period of once in 3 years, the growth rate of the area of accumulated water water showed a
downward trend, at 11.85%, but the growth rate of the maximum
showed a downward trend, at 11.85%, but the growth rate of the maximum ponding depth ponding depth increased
from 20.00%
increased from to 22.22%.
20.00% Under the 10-year
to 22.22%. Under rainfall scenario,rainfall
the 10-year compared with those
scenario, under the
compared with
5-year rainfall scenario, the increases in the total area and the depth of accumulated water
those under the 5-year rainfall scenario, the increases in the total area and the depth of
due to stormwater inlet blockage slightly decreased, from 11.85% and 22.22% to 11.40%
accumulated water due to stormwater inlet blockage slightly decreased, from 11.85% and
and 19.35%, respectively. When the rainfall scenario was up to the 30-year recurrence
22.22% to 11.40% and 19.35%, respectively. When the rainfall scenario was up to the 30-
period, the maximum area and the depth of accumulated water due to stormwater inlet
year recurrence
blockage increased period, the and
by 4.04% maximum
10.53%, area and thecompared
respectively, depth ofwith accumulated waterwith
those associated due to
stormwater inlet blockage increased by 4.04% and 10.53%, respectively,
nonblocked stormwater inlet. Under the 50-year rainfall scenario, the area and the depth compared with
those associated water
of accumulated with nonblocked stormwater
caused by stormwater inlet.
inlet Underincreased
blockage the 50-year rainfall
by only 3.34%scenario,
and
the arearespectively.
9.76%, and the depth of with
Overall, accumulated
increasingwater
rainfall caused by stormwater
intensity, stormwaterinlet inlet blockage
blockage
increased
will cause by only 3.34%
an increase in bothand
the9.76%,
maximum respectively.
area and theOverall,
depth ofwith increasing
accumulated rainfall
water in
intensity,
the studystormwater
area, while the inlet blockage
growth rateswill causeshowed
basically an increase in both the
a downward trendmaximum
comparedarea withand
the depth of accumulated water in the study area, while the growth rates basically showed
a downward trend compared with the conditions of the nonblocked stormwater inlet,
indicating that with increasing rainfall intensity, the influence of stormwater inlet
Water 2024, 16, 2029 16 of 20

the conditions of the nonblocked stormwater inlet, indicating that with increasing rainfall
intensity, the influence of stormwater inlet blockage on waterlogging gradually decreased.
Therefore, we should consider the changes in the rainfall intensity in the future.

Table 3. Statistics of waterlogging under the different rainfall scenarios with different stormwater
inlet blockage conditions.

Area of Accumulated Water

Depth of Accumulated Water (cm)
Serial Rainfall (m2 )
Number Scenario Growth Growth
No Blockage Block up No Blockage Block up
Ratio (%) Ratio (%)
1 P = 1a 3580 5132 43.35 19 25 31.58
2 P = 3a 5880 7016 19.32 25 30 20.00
3 P = 5a 7124 7968 11.85 27 33 22.22
4 P = 10a 8104 9028 11.40 31 37 19.35
5 P = 30a 10,304 10,720 4.04 38 42 10.53
6 P = 50a 11,032 11,400 3.34 41 45 9.76

The impact of stormwater inlet blockage on the waterlogging conditions in the study
area was further analyzed. The area of each waterlogging level under the different scenarios
of nonblocked stormwater inlet and stormwater inlet blockage was statistically analyzed.
The results are listed in Table 4. Table 5 provides the percentage increase in the area of each
waterlogging level with stormwater inlet blockage compared with that of the nonblocked
stormwater inlet.

Table 4. Ponding area at the different levels under the different stormwater inlet blockage
conditions (m2 ).

Waterlogging Level (cm)

Rainfall Area at Each Accumulated Water Level under the Area at Each Accumulated Water Level under the
Scenarios Scenario of Nonblocked Stormwater Inlet (m2 ) Scenario of Stormwater Inlet Blockage (m2 )
<10 10~20 20~30 30~50 >50 <10 10~20 20~30 30~50 >50
P = 1a 3292 184 0 0 0 2388 2732 12 0 0
P = 3a 2608 3256 16 0 0 2008 4688 320 0 0
P = 5a 2668 4436 20 0 0 1848 4564 1548 8 0
P = 10a 2416 4808 872 8 0 1820 2920 4268 20 0
P = 30a 2832 2780 4660 32 0 2648 1876 4792 1404 0
P = 50a 3124 2300 4720 888 0 2888 1868 3572 3072 0

Table 5. Area growth ratio of each water level compared with that of the nonblocked group (%).
Waterlogging Level
<10 10~20 20~30 30~50 >50
Rainfall Scenarios (cm)
P = 1a −27.46 1384.78 - 0 0
P = 3a −23.01 43.98 1900 0 0
P = 5a −30.73 2.89 7640 - 0
P = 10a −24.67 −39.27 389.45 150 0
P = 30a −6.5 −32.52 2.83 4287.5 0
P = 50a −7.55 −18.78 −24.32 245.95 0

Tables 4 and 5 show, notably, that during the 1-year rainfall return period, the depth
of accumulated water under nonblocked stormwater inlet conditions was less than 20 cm,
even remaining less than 10 cm for the most part. The depth of accumulated water under
stormwater inlet blockage conditions was less than 30 cm and mainly ranged from 10
to 20 cm. Compared with the nonblocked stormwater inlet, stormwater inlet blockage
imposed the greatest impact on the depth of accumulated water ranging from 10 to 20 cm,
Water 2024, 16, 2029 17 of 20

and the increase ratio was 1384.78%. Under the 3-year rainfall scenario, the maximum
depth of accumulated water for the nonblocked stormwater inlet increased to 20~30 cm,
and that for both nonblocked stormwater inlet and stormwater inlet blockage mainly
ranged from 10 cm to 20 cm. Compared with the status of the nonblocked stormwater
inlet, the depth of accumulated water ranged from 20 to 30 cm, and the growth rate was
1900.00%. Under the 5-year rainfall return period, the depth of accumulated water for both
nonblocked stormwater inlet and stormwater inlet blockage was the same as that under the
3-year rainfall scenario, mainly ranging from 10 to 20 cm. Compared with the nonblocked
stormwater inlet, stormwater inlet blockage exerted the greatest impact on the depth of
accumulated water within the range of 20~30 cm, and the increase ratio was the highest,
at 7640.00%. Under the 10-year rainfall scenario, the depth of accumulated water for the
nonblocked stormwater inlet mainly ranged from 10 to 20 cm, and that for stormwater
inlet blockage largely ranged from 20 to 30 cm. Compared with that for the nonblocked
stormwater inlet, the depth of accumulated water for stormwater inlet blockage was less
than 20 cm, ranging from 10 to 20 cm. Stormwater inlet blockage imposed the greatest
impact on the depth of accumulated water within the range of 20~30 cm, and the increase
ratio was 389.45%. Up to the 30-year rainfall return period, the depth of accumulated water
for both the nonblocked stormwater inlet and stormwater inlet blockage mainly ranged
from 20 to 30 cm. Compared with the case of the nonblocked stormwater inlet, the depth of
accumulated water within the range of 30~50 cm was most affected by stormwater inlet
blockage, and the growth proportion reached a maximum of 4287.50%. Under the 50-year
rainfall scenario, the depth of accumulated water for both nonblocked stormwater inlet
and stormwater inlet blockage mainly ranged from 20 to 30 cm. Compared with that for
nonblocked stormwater inlet, the depth for stormwater inlet blockage decreased within the
ranges of less than 10 cm, 10~20 cm and 20~30 cm, while stormwater inlet blockage exerted
the greatest impact on the depth of accumulated water within the range of 30–50 cm, with
a growth proportion of 245.95%.
In terms of the total area of accumulated water, stormwater inlet blockage yielded the
greatest impact under the 1-year rainfall scenario, with a 43.35% increase. The influence of
stormwater inlet blockage on the area of accumulated water in the study block gradually
decreased with increasing rainfall recurrence period. Under the 50-year rainfall scenario,
stormwater inlet blockage only resulted in a 3.34% increase in the total area of accumulated
water. The impact of stormwater inlet blockage on the maximum depth of accumulated
water generally decreased with increasing rainfall recurrence period. This may occur
because the effects of the rainfall intensity and low-elevation terrain on waterlogging
increased, thus weakening the effect of stormwater inlet blockage on waterlogging. In terms
of changes within the different accumulated water level ranges, stormwater inlet blockage
caused an increase in the level of accumulated water, a reduction in the proportion of low-
risk areas, and an increase in the proportion of high-risk areas. Moreover, with increasing
rainfall recurrence period, the level of accumulated water most affected by stormwater
inlet blockage differed. Under the 1-year rainfall scenario, the level of accumulated water
most affected by stormwater inlet blockage was 10–20 cm. Under the 3-year, 5-year and
10-year rainfall scenarios, the level of accumulated water most affected by stormwater inlet
blockage was 20–30 cm, while under the 30-year and 50-year rainfall scenarios, the level
of accumulated water most affected by stormwater inlet blockage was 30–50 cm. Overall,
stormwater inlet blockage aggravated the waterlogging risk and the extent of waterlogging,
which could result in more severe waterlogging disasters.

4. Discussion
The hydraulic roughness coefficient has a significant impact on the simulation results
of the SWMM. When the hydraulic roughness coefficient increases, it indicates that the
water flow velocity per unit width decreases, the flow velocity decreases, and thus affects
the changes in water level. But for the independent ITF-FLOOD model, there was no signif-
icant impact. For the SWMM and ITF-FLOOD coupling model, the hydraulic roughness
Water 2024, 16, 2029 18 of 20

coefficient will affect the input of the ITF-FLOOD model. But stormwater inlet blockage
has a more significant impact on the simulation results, as it can lead to a decrease in the
drainage capacity of the pipeline network, a decrease in runoff, and thus affect the changes
in water level.
To alleviate the impact of stormwater inlet blockage on the risk of waterlogging,
prevention and control recommendations are proposed in terms of both stormwater inlet
installation and management measures. The specific suggestions are as follows:
(1) Replace aged or easily clogged stormwater grates
The aging of drainage facilities can affect the discharge rate of rainwater. Management
personnel can replace aged or easily clogged stormwater grates in the research area with
anti-clogging types through investigation. This type of stormwater grate can effectively
intercept debris, garbage, and other blockages, thereby reducing the degree and probability
of stormwater inlet blockage. In addition, the intercepted blockages are stored on the surface
of the stormwater inlet for easy maintenance and treatment by management personnel.
(2) Implement a stormwater inlet subcontracting system
The large number and wide distribution of stormwater inlets are important factors
leading to the neglect of stormwater inlet management. Moreover, some businesses or
households randomly discard oil and debris in stormwater inlets, which greatly reduces
the drainage efficiency. Given these problems, it is recommended that the front door re-
sponsibility system be implemented in areas where stormwater inlets are often blocked. For
example, the supervision and inspection of stormwater inlets should be subcontracted to
nearby businesses or households. During periods without rain, the responsible businesses
or households are responsible for supervising the behavior of randomly throwing waste
and other debris into stormwater inlets, assessing the status of stormwater inlets before
rainfall onset, and reporting severe stormwater inlet blockage to the management depart-
ment, which will then perform timely dredging work. At the same time, some incentive
measures can be adopted to ensure smooth implementation.
(3) Combining other measures to address accumulated water
The simulation results showed that under high-intensity rainfall, although the stormwa-
ter inlets were not blocked, waterlogging in the study area was severe. Therefore, under
high-intensity rainfall, it is not enough to implement a single treatment measure for dredg-
ing rainwater orifices. Notably, it is necessary to combine other drainage measures to
prevent and control waterlogging in the study area to effectively reduce the risk of water-
logging disasters.

5. Conclusions
In this paper, the drainage capacity of the pipe network was generalized by using
the equivalent drainage method for stormwater inlets. By coupling the stormwater inlet
drainage equation, the ITF-FLOOD two-dimensional hydrodynamic model was improved
to construct a coupled waterlogging model of the typical waterlogging point of Guohuo
East Road in the middle Lianjiang Road area of Fuzhou city. The effect of stormwater inlet
blockage on waterlogging under different rainfall scenarios was simulated and analyzed by
setting the stormwater inlet blockage state, which provided a reference for the construction
of an urban waterlogging model and has certain guiding significance for waterlogging
prevention and control in the study area prone to stormwater inlet blockage. The results
indicated that both the total area and the maximum depth of accumulated water caused by
stormwater inlet blockage increased by 43.35% and 31.58%, respectively, compared with
those associated with the nonblocked stormwater inlet under the 1-year rainfall scenario.
When the rainfall intensity was once in fifty years, the total area of accumulated water
due to stormwater inlet blockage increased by only 3.34% compared with the nonblocked
stormwater inlet, and the maximum depth of accumulated water increased by only 9.76%.
With increasing rainfall recurrence period, the effects of stormwater inlet blockage on the
Water 2024, 16, 2029 19 of 20

area and the maximum depth of accumulated water gradually decreased, but the effect of
stormwater inlet blockage on the level of accumulated water increased. Overall, stormwater
inlet blockage aggravates the risk level and extent of waterlogging, resulting in severe
waterlogging disasters.

Author Contributions: Conceptualization, W.G.; Validation, W.G. and M.Z.; Data Curation, W.G. and
M.Z.; Writing—Review and Editing, W.G.; Supervision, W.G.; Funding Acquisition, W.G.; Method-
ology, M.Z.; Formal Analysis, M.Z., X.L. and H.H.; Writing—Original Draft, M.Z.; Investigation,
X.L.; Resources, H.H. and S.L.; Project Administration, Y.L. All authors have read and agreed to the
published version of the manuscript.
Funding: This research was funded by Hebei Natural Science Foundation, grant number [E2021402039],
Hebei Natural Science Foundation, grant number [E2024402142] and the Water Conservancy Technol-
ogy Promotion Program of Shanxi province, grant number [2021LS011].
Data Availability Statement: The data included in this study are available from the corresponding
author on reasonable request.
Conflicts of Interest: The authors declare no conflicts of interest.

References
1. Qi, G.Z.; Wang, Z.B.; Wei, L.J.; Wang, Z.X. Multidimensional effects of urbanization on PM2.5 concentration in China. Environ. Sci.
Pollut. Res. 2022, 29, 77081–77096. [CrossRef] [PubMed]
2. Duan, C.Y.; Zhang, J.Q.; Chen, Y.N.; Lang, Q.L.; Zhang, Y.C.; Wu, C.Y.; Zhang, Z. Comprehensive Risk Assessment of Urban
Waterlogging Disaster Based on MCDA-GIS Integration: The Case Study of Changchun, China. Remote Sens. 2022, 14, 3101.
[CrossRef]
3. Yu, H.F.; Zhao, Y.L.; Fu, Y.C.; Li, L. Spatiotemporal Variance Assessment of Urban Rainstorm Waterlogging Affected by Impervious
Surface Expansion: A Case Study of Guangzhou, China. Sustainability 2018, 10, 3761. [CrossRef]
4. Chen, X.S.; Jiang, S.L.; Xu, L.S.; Xu, H.G.; Guan, N.N. Resilience assessment and obstacle factor analysis of urban areas facing
waterlogging disasters: A case study of Shanghai, China. Environ. Sci. Pollut. Res. 2023, 30, 65455–65469. [CrossRef] [PubMed]
5. Xie, S.; Liu, W.; Yuan, Z.; Zhang, H.Y.; Lin, H.; Wang, Y.Q. Integrated Risk Assessment of Waterlogging in Guangzhou Based on
Runoff Modeling, AHP, GIS and Scenario Analysis. Water 2022, 14, 2899. [CrossRef]
6. Qi, W.C.; Ma, C.; Xu, H.S.; Zhao, K. Urban flood response analysis for designed rainstorms with different characteristics based on
a tracer-aided modelling simulation. J. Clean. Prod. 2022, 335, 131797. [CrossRef]
7. Zhang, Q.F.; Wu, Z.F.; Zhang, H.; Dalla Fontana, G.; Tarolli, P. Identifying dominant factors of waterlogging events in metropolitan
coastal cities: The case study of Guangzhou, China. Environ. Manag. 2020, 271, 110951. [CrossRef]
8. Merz, B.; Blöschl, G.; Vorogushyn, S.; Dottori, F.; Aerts, J.C.J.H.; Bates, P.; Bertola, M.; Kemter, M.; Kreibich, H.; Lall, U.; et al.
Causes, impacts and patterns of disastrous river floods. Nat. Rev. Earth. Environ. 2021, 2, 592–609. [CrossRef]
9. Bryndal, T.; Franczak, P.; Kroczak, R.; Cabaj, W.; Kołodziej, A. The impact of extreme rainfall and flash floods on the flood risk
management process and geomorphological changes in small Carpathian catchments: A case study of the Kasiniczanka river
(Outer Carpathians, Poland). Nat. Hazards 2017, 88, 95–120. [CrossRef]
10. Zhao, X.F.; Li, H.M.; Cai, Q.; Pan, Y.; Qi, Y. Managing Extreme Rainfall and Flooding Events: A Case Study of the 20 July 2021
Zhengzhou Flood in China. Climate 2023, 11, 228. [CrossRef]
11. Alemaw, B.F.; Tafesse, N.T. Urban Stormwater and Sewerage Modelling: An Approach for Peak Runoff and Volume Assessment.
J. Water Resour. Prot. 2021, 13, 855–880. [CrossRef]
12. Liu, Y.; Zhang, S.H.; Su, H.D.; Wang, D.; Gu, Q.H. Modelling and Application of Urban Drainage Based on Mike Urban Model.
IOP Conf. Ser. Earth Environ. Sci. 2020, 474, 062003. [CrossRef]
13. Komori, D.; Nakaguchi, K.; Inomata, R.; Oyatsu, Y.; Tachikawa, R.; Kazama, S. Topographical Characteristics of Frequent Urban
Pluvial Flooding Areas in Osaka and Nagoya Cities, Japan. Water 2022, 14, 2795. [CrossRef]
14. Zhang, Q.F.; Wu, Z.F.; Tarolli, P. Investigating the Role of Green Infrastructure on Urban WaterLogging: Evidence from
Metropolitan Coastal Cities. Remote Sens. 2021, 13, 2341. [CrossRef]
15. Wu, J.S.; Sha, W.; Zhang, P.H.; Wang, Z.Y. The spatial non-stationary effect of urban landscape pattern on urban waterlogging: A
case study of Shenzhen City. Sci. Rep. 2020, 10, 7369. [CrossRef]
16. Zhao, Y.J.; Xia, J.; Xu, Z.X.; Zou, L.; Qiao, Y.F.; Li, P. Impact of Urban Expansion on Rain Island Effect in Jinan City, North China.
Remote Sens. 2021, 13, 2989. [CrossRef]
17. Hu, M.C.; Zhang, X.Q.; Siu, Y.L.; Li, Y.; Tanaka, K.; Yang, H.; Xu, Y.P. Flood Mitigation by Permeable Pavements in Chinese
Sponge City Construction. Water 2018, 10, 172. [CrossRef]
18. Ahmad, T.; Pandey, A.C.; Kumar, A. Evaluating urban growth and its implication on flood hazard and vulnerability in Srinagar
city, Kashmir Valley, using geoinformatics. Arab. J. Geosci. 2019, 12, 308. [CrossRef]
Water 2024, 16, 2029 20 of 20

19. Wang, L.Y.; Li, Y.; Hou, H.; Chen, Y.; Fan, J.J.; Wang, P.; Hu, T.G. Analyzing spatial variance of urban waterlogging disaster at
multiple scales based on a hydrological and hydrodynamic model. Nat. Hazards 2022, 114, 1915–1938. [CrossRef]
20. Zhang, K.; Li, J.K.; Ning, L.Z.; Xie, W.F.; Li, Y.F.; Ullah, Z.; Peng, K. Three-Dimensional Numerical Simulation Research on
Discharge Characteristics of Storm-Drain Inlet in Low-Lying Areas. Water Resour. Manag. 2023, 37, 5273–5287. [CrossRef]
21. Yu, S.; Zhang, S.H.; Di, S.C.; Zhou, X.; Ge, J.; Xiao, B.Q. Effect of combined flat and vertical renovation of urban road gullies on
surface water flooding prevention based on physical model experiments. J. Hydraul. Eng. 2022, 53, 854–861.
22. Harada, S. Application of Porous Concrete Infiltration Techniques to Street Stormwater Inlets That Simultaneously Mitigate
against Non-Point Heavy Metal Pollution and Stormwater Runoff Reduction in Urban Areas: Catchment-Scale Evaluation of the
Potential of Discrete and Small-Scale Techniques. Water 2023, 15, 1998. [CrossRef]
23. Kim, J.S.; Kwak, C.J.; Jo, J.B. Enhanced method for estimation of flow intercepted by drainage grate inlets on roads. J. Environ.
Manag. 2021, 279, 111546. [CrossRef] [PubMed]
24. Noh, S.J.; Lee, S.; An, H.; Kawaike, K.; Nakagawa, H. Ensemble urban flood simulation in comparison with laboratory-scale
experiments: Impact of interaction models for manhole, sewer pipe, and surface flow. Adv. Water Resour. 2016, 97, 25–37.
[CrossRef]
25. Martins, R.; Leandro, J.; Carvalho, R.F.D. Characterization of the hydraulic performance of a gully under drainage conditions.
Water Sci. Technol. 2014, 69, 2423–2430. [CrossRef] [PubMed]
26. Martins, R.; Rubinato, M.; Kesserwani, G.; Leandro, J.; Djordjević, S.; Shucksmith, J.D. On the characteristics of velocities fields on
the vicinity of manhole inlet grates during flood events. Water Resour. Res. 2018, 54, 6408–6422. [CrossRef]
27. Rubinato, M.; Lee, S.; Martins, R.; Shucksmith, J.D. Surface to sewer flow exchange through circular inlets during urban flood
conditions. J. Hydroinform. 2018, 20, 564–576. [CrossRef]
28. Santoso, B.P.; Yanto, M.; Apriyono, A. Inverse distance weighting interpolated soil properties and their related landslide
occurrences. MATEC Web Conf. 2018, 195, 03013. [CrossRef]
29. Tügel, F.; Hassan, A.; Hou, J.; Hinkelmann, R. Applicability of Literature Values for Green–Ampt Parameters to Account for
Infiltration in Hydrodynamic Rainfall–Runoff Simulations in Ungauged Basins. Environ. Model. Assess. 2022, 27, 205–231.
[CrossRef]
30. Yang, D.; Hou, J.M.; Li, B.R.; Li, D.L.; Li, Y.Q.; Fu, D.Y.; Ji, G.Q. Generalized simulation of drainage process in urban networks
without data. J. Water Res. Water Eng. 2020, 31, 139–144.
31. Xia, J.Q.; Cheng, Q.; Dong, B.L.; Zhang, X.L. Experimental study on the effect of clogging on the inlet discharge capacity. Adv.
Water Sci. 2020, 31, 843–851.
32. Chen, J.; Li, Y.; Zhang, C. The Effect of Design Rainfall Patterns on Urban Flooding Based on the Chicago Method. Int. J. Environ.
Res. Public Health 2023, 20, 4245. [CrossRef] [PubMed]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.