(6) Validation of the model

In order to check the applicability of the parameters in the study area, the simulation results need to be validated. A rainfall event with a 50-year rainfall intensity was used as the boundary condition for the model to compare the distribution of the simulated flooding points in the study area with the actual flooding points. The actual distribution of the flooding points was obtained from the information on flood-prone points published by the Zhengzhou traffic department.

Comparing the historical statistical inundation points in red in Figure 2 with the inundation extent of the simulation results, it can be seen from the figure that the inundation points simulated by the model match the distribution of the actual inundation points. Most of these inundation points are located where the duration of inundation is relatively long. The model parameters are therefore considered to be appropriate and the simulation results are reliable.

**Figure 2.** Distribution of simulated results compared to measured water accumulation points.

#### 2.3.2. Scenario Setting

In this study area, several sets of rainfall scenarios were set up based on the storm intensity formula to analyse the flooding impacts of different rainfall scenarios [37]. The storm intensity equations used for the study are as follows:

$$q = \frac{7057.6(1 + 0.794 \lg P)}{(t + 25.8)^{0.948}} \tag{8}$$

where *q* is the average storm intensity in mm/min; *P* is the return period in a; *t* is the rainfall time in min.

The design recurrence period of the pipe network in the study area is 2 a. Combining the historical rainfall and flood control needs of the study area, the flooding situation of the Dongfeng canal area was simulated under the rainstorm scenario, and six rainstorm scenarios with a rainfall duration of 1 h and 2 h for 5 a, 20 a, and 50 a events were designed as the different rainfall scenario driving models. The rainfall process time interval was set to 5 min and the design rainfall process line was obtained according to the rainfall intensity formula and using the ICM design rainfall generator as shown in Figure 3.

**Figure 3.** Rainfall processes with different rainfall return periods and different rainfall durations: (**a**) Rainfall duration of 1 h rainfall process. (**b**) Rainfall duration of 2 h rainfall process.

#### 2.3.3. Flood Risk Analysis Methodology

According to the UK Environmental Protection Agency, the flood risk rate is calculated by combining two key physical variables, water depth and flow velocity. During the calculation of flood risk rates, the type of subsurface is also considered to be an important factor [38]. The calculation is as follows:

$$R\_H = h(\upsilon + 0.5) + \text{CDF} \tag{9}$$

where *RH* is the risk rate, with a scale of one; *h* is the flood inundation depth, m; *v* is the flood flow velocity, m/s; CDF is the debris factor, i.e., the increased risk factor due to debris carried in the flood.

The CDF is mainly used to increase the weight of the impact of road floats on flood risk and is often used in flood risk analysis [39,40]. The CDF is assigned to 1 if the type of bedding surface is a road or a building site and the flood velocity is greater than 2, whereas the rest are assigned to 0. The flood risk rate value *RH* was quantified for any point in the study area. In addition, the flood risk rating was divided into four zones. When *RH* is less than 0.75, it means that the area is in a low-risk zone; when Rh is between 0.75 and 1.25, it means a medium-risk zone; when Rh is between 1.25 and 2.00, it means a high-risk zone; and when Rh is greater than 2.00, it means that the current area is in a very-high-risk zone. The flood risk classification is shown in Table 2.


**Table 2.** Flood risk classification.

The methodological route of the study is shown in Figure 4.

**Figure 4.** Distribution of simulated results compared to measured water accumulation points.
