Investigation of the response characteristics of urban waterlogging to model rainfall patterns taking into account the spatial distribution of rainfall

Urban waterlogging is primarily caused by short-duration rainstorms, and the rainfall pattern significantly influences the impact of waterlogging disasters. This study focuses on the Fengxi New Town in the Xianyang New Area and considers the uneven spatial distribution of rainfall. Various model rainstorm processes with different concentration levels were designed based on the rainfall pattern model. The urban stormwater model GAST-SWMM was employed to simulate and analyze the waterlogging characteristics of seven model rain patterns with different return periods. The findings reveal the following: (1) Single-peak rain patterns exhibit a more pronounced waterlogging effect compared to double-peak rain patterns, with the single-peak rain pattern featuring a mid-to-late peak demonstrating the strongest waterlogging effect. Among the single-peak rain patterns, the one with a mid peak exhibits the largest inundation volume for return periods below 50 years, while the one with a late peak demonstrates the largest inundation volume for return periods exceeding 50 years. Under a 100-year return period, the inundation volume of the late-peak single-peak rain pattern increased by 6.58% compared to rain pattern IV. (2) Among the double-peak rain patterns, rain pattern V exhibits the strongest waterlogging effect, with its inundation volume increasing by 1,912.42 m³ compared to the preceding single-peak rain pattern under a 100-year return period. For return periods exceeding 20 years, the flooded area in the moderate-risk zone for rain pattern V increases by 1.81%-2.29% compared to the single-peak rain pattern with an earlier peak. (3) The occurrence time of the peak rainfall varies for each rain pattern in the high-risk zone based on the return period. Rain pattern VI consistently exhibits the peak occurrence time at the end of the rainfall, while the peak occurrence times of rain patterns II, IV, V, and VII advance with increasing return periods. The peak occurrence times of rain patterns I and III are delayed by 105 minutes and 110 minutes, respectively, as the return period increases to less than 20 years. However, for return periods exceeding 20 years, the peak occurrence times of rain patterns I and III advance to 100 minutes. This study explores the response characteristics of urban waterlogging to different model rain patterns, providing valuable insights for relevant departments involved in flood control and disaster relief efforts.