*3.3. Urban Flood Risk Analysis*

The flood risk rate analysis method proposed by the UK EPA has been used to express the flood risk rate as a combination of two key physical quantities, namely water depth and flow velocity. The type of subsurface is also considered to be an important factor influencing the flood risk rate. Finally, an RH value characterising the flood risk was calculated to characterise the degree of flood risk.

Figure 10 shows the distribution of the flood risk in the study area, simulated under the different scenarios. It can be seen that the distribution of risk is dominated by low-risk

areas with very few very high-risk areas. The medium- and high-risk areas are mainly located on the banks of rivers, in built-up areas, and on low-lying urban terrain.

**Figure 10.** Inundation flood risk for different scenarios in the study area.

A detailed analysis of the change in the flood risk area during the rainfall return period for the 5 a, 20 a, and 50 a scenarios is presented based on the simulation results in Table 6. For a rainfall duration of 1 h, the low-risk areas were 7195.39 ha, 7058.04 ha, and 6959.16 ha and the medium-risk areas were 51.44 ha, 87.18 ha, and 101.74 ha, respectively. At a rainfall time of 2 h, the low-risk areas were 7148.54 ha, 6981.60 ha, and 6740.90 ha; the medium-risk areas were 62.38 ha, 107.33 ha, and 125.89 ha; the high-risk areas were 112.65 ha, 225.05 ha, and 430.37 ha; and the very-high-risk areas were 14.75 ha, 24.33 ha, and 41.16 ha, respectively.


**Table 6.** Area statistics for inundation flood risk rates for different scenarios (ha).

The shift in the rainfall return periods from the 5 a to the 50 a risk zones was then analysed under the same rainfall calendar. For a rainfall duration of 1 h, the shifts from low- to medium-, high-, and very-high-risk zones were 99.19 ha, 134.43 ha, and 2.61 ha, respectively; from medium- to high- and very-high-risk zones were 48.08 ha and 0.80 ha, respectively; and from high- to very-high-risk zones was 13.01 ha. With a rainfall duration of 2 h, the shifts from the low-risk zone to the medium-, high-, and very-high-risk zones were 124.84 ha, 277.79 ha, and 5.01 ha, respectively; from the medium-risk zone to the highand very-high-risk zones were 60.30 ha and 1.02 ha, respectively; and from the high-risk zone to the very-high-risk zone was 20.38 ha. The combination of the risk distribution

ranges in Figure 10 indicates that the upgraded risk areas are mainly located on both sides of the river and in the study area, which are low-lying and have poor drainage capacity.

There are also some characteristics of the change in the flood risk zones from a 1 h to a 2 h rainfall ephemeris for the same rainfall return period. When the rainfall return period was 5 a, 34.58 ha and 12.32 ha of low-risk zone converted to medium- and high-risk zones, respectively, and 0.05 ha of medium-risk zone converted to low-risk zone, 23.45 ha to high-risk zone, and 0.14 ha to very-high-risk zone. When the rainfall return period was 20 a, the areas of low-risk zone converted to medium- and high-risk zones were 58.31 ha and 18.02 ha, respectively, and the area of medium-risk zone converted to high-risk zone was 38.39 ha. When the rainfall return period was 50 a, the areas transformed from low-risk areas to medium- and high-risk areas were 104.15 ha and 113.66 ha, respectively, and the area transformed from medium-risk area to high-risk area was 79.99 ha. The area transformed from medium-risk zone to high-risk zone was 79.99 hectares. Combined with Figure 10, it appears that the risk areas close to the river are more likely to be upgraded, and the high-risk areas are also generally concentrated near the river.

The above results show that for the same rainfall return period, the area covered by low-risk areas decreased with increasing rainfall calendar time, and most of the reduced low-risk areas transformed into medium- and high-risk areas. The area covered by all the risk classes, except for low-risk areas, increased with increasing rainfall calendar hours. For the same rainfall calendar time, the area covered by low-risk areas decreased with increasing rainfall return periods. The area covered by all risk classes, except for low-risk areas, increased with increasing rainfall return periods. The above analysis indicates that increases in the rainfall return period and rainfall duration will result in more severe flooding. It is also important to focus on high- and very-high-risk areas when undertaking flood management and when flooding occurs. It is important to consider that the flood risk varies with the duration and intensity of rainfall ephemeris. It is therefore also important to focus on flooding in low-risk areas that could easily convert into high-risk areas.

#### **4. Conclusions**

In this study, a 1D/2D coupled urban flood model was constructed using InfoWorks ICM software based on data pertaining to pipe networks, inspection wells, roads, water systems, land uses, and elevations in the Dongfeng Canal area of Zhengzhou. Six scenarios were set up according to different rainfall recurrence periods and rainfall ephemeris using the Zhengzhou storm intensity formula. According to the simulation results, the flood inundation depth, inundation extent, duration of inundation, flood flow, and drainage system load were analysed. Finally, the flood risk was quantified and spatially analysed using the flood risk rate analysis method proposed by the UK EPA. The main findings of this study are as follows.

(1) This study uses the InfoWorks ICM model to construct a coupled hydrologicalhydraulic model for a small urban watershed area. The two-dimensional hydraulic processes of the urban one-dimensional pipe network, river, and surface are coupled and the model effects are compared using the flood inundation locations of historical precipitation. The results show that the model is effective in simulating rainfall and flooding in the Dongfeng Canal area. The model can be applied to the analysis of flood risk.

(2) According to the simulation results, inundation depths in the study area are mainly 0∼0.3 m, followed by 0.3∼0.5 m. Inundation is mainly concentrated in areas near rivers and low-lying areas of road topography. The extent of inundation in the study area at a 2 h rainfall ephemeris is greater than that at a 1 h rainfall ephemeris, and the extent of inundation tends to increase with increasing recurrence periods. The extent of inundation at <1 h and >4 h appears to decrease with increasing rainfall ephemeris, indicating that the longer the rainfall ephemeris, the faster the study area drains. The distribution of flood velocities shows a tendency to increase with increasing return periods, with higher velocities being distributed mainly in drainage inlets near the main drainage network and densely built-up areas.

(3) The proportion of nodal overflows occurring increases with increasing rainfall return periods and rainfall ephemeris. The pipe network in the study area is generally overloaded under the different scenarios, with flow overloading dominating and the length of the overloaded pipe network increasing with the rainfall return period. The length of the bathymetric overload pipe network increases with increasing rainfall return periods, whereas the length of the flow overload pipe network decreases with increasing rainfall return periods.

(4) The distribution of the flood risk in the study area is dominated by low-risk areas, with very few very high-risk areas. The medium- and high-risk zones are mainly located on both sides of the river in built-up areas and low-lying urban areas. The simulation results under the different scenarios show that the areas of medium-, high-, and very-high-risk zones increase with the increasing rainfall return period and rainfall duration. The area of low-risk zones decreases with the increasing rainfall return period and rainfall duration, and most of the reduced low-risk zones are transformed into medium- and high-risk zones. Increases in rainfall return periods and rainfall durations can lead to more severe flooding. It is therefore important to focus on high- and very-high-risk areas, as well as low-risk areas that can easily transform into high-risk areas when managing floods when flooding occurs.

This study analyses the flood risk situation in the study area in various aspects and from various perspectives. It can provide a reference for the prevention and control of flooding and the formulation of flood countermeasures in the area and help to improve the management of flood risks in the urban construction process. At the same time, the model lacks further validation due to limited measurement data. The construction of the flood risk indicators is relatively singular and does not take into account more comprehensive factors such as socio-economic factors. Therefore, a more detailed collection of hydrological and socio-economic data in the study area to further improve the accuracy of the model and construct a more comprehensive urban flood risk system will be the focus of our work in the next phase.

**Author Contributions:** Data curation, methodology, software, visualization, and writing—original draft, H.W. and L.Z.; Conceptualization, funding acquisition, and project administration, supervision, H.W.; investigation, H.W., L.Z. and J.L; resources, H.W. and J.L.; writing—review and editing, H.W. and J.L. All authors have read and agreed to the published version of the manuscript.

**Funding:** This publication is supported by the National Natural Science Foundation of China (No. 51979107, No. 51909091).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** The authors are grateful to the editors and anonymous reviewers for their insightful comments and helpful suggestions.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Shengqi Jian 1, Aoxue Wang 1, Chengguo Su 1,\* and Kun Wang <sup>2</sup>**


**Abstract:** Reference evapotranspiration (*ET*0) is an integral part of the regional hydrological cycle and energy balance and is extremely sensitive to climate change. Based on temperature data from 24 global climate models (GCMs) in the Coupled Model Intercomparison Project Phase 6 (CMIP6), this study developed a multi-model ensemble based on delta statistical downscaling with multiple interpolation methods and evaluation indicators to predict the spatial and temporal evolution trends of *ET*<sup>0</sup> in the Yellow River Basin (YRB) under four emission scenarios (SSP126, SSP245, SSP370, and SSP585) for the near- (2022–2040), mid- (2041–2060), and long- (2081–2100) term future. Results demonstrate that regional data generated based on delta statistical downscaling had good simulation performance for the monthly mean, maximum, and minimum temperatures in the YRB, and the developed multimodel ensemble had better simulation capability than any single model. Compared to the historical period (1901–2014), the annual *ET*<sup>0</sup> showed a highly significant increase for different future emission scenarios, and the increase is faster with increasing radiative forcing. The first main cycle of *ET*<sup>0</sup> change was 52, 53, 60, and 48 years for the SSP126, SSP245, SSP370, and SSP585, respectively. *ET*<sup>0</sup> in the YRB had positive values for EOF1 under all four emission scenarios, responding to a spatially consistent trend across the region. Compared to the historical period, the spatial distribution of *ET*<sup>0</sup> under different future emission scenarios was characterized by being larger in the west and smaller in the east. As the radiative forcing scenario increased and time extended, *ET*<sup>0</sup> significantly increased, with a maximum variation of 112.91% occurring in the western part of the YRB in the long-term future under the SSP585 scenario. This study can provide insight into the water cycle patterns of watersheds and scientific decision support for relevant departments to address the challenges of climate change.

**Keywords:** reference evapotranspiration; CMIP6; delta statistical downscaling; Hargreaves model; Yellow River Basin; EOF analysis
