Simulation and Evaluation of Collapsible Risk of Low Impact Development Rainwater System in Collapsible Loess Area

Abstract To assess the risk levels of collapse for various low-impact development (LID) measures used in constructing stormwater systems in collapsible loess areas, the high-efficiency and high-precision urban stormwater model (GAST) was used to analyze the risk of waterlogging. Combined with the time of waterlogging accumulation, a method is proposed to assess the collapse risk level of various LID measures by considering the correlation between the collapsibility coefficient and soil parameters. This method was then used to analyze the changing patterns of settlement risk levels in grassed swales and rain gardens under different rainfall return periods (P1=20a, P2=50a, P3=100a) and water accumulation durations (T1=0.5 day, T2=5 day). The findings indicate a direct correlation between the rainfall return periods and the level of risk for collapse. At P3=100a, 50% of the collapse risk level in the study region reaches level III. Nevertheless, when the water accumulation duration is 5 days, the percentage of places classified as category III increases to 75%. Although rain gardens are better than grass swales at storing water, they also increase the risk of collapse. Therefore, it is essential to carefully monitor the durability of rain gardens during long periods of frequent rainfall. The evaluation approach proposed in this research demonstrates a 75% accuracy rate, confirming it as an initial and reliable instrument for evaluating the risk associated with building low-impact development stormwater systems in collapsible loess areas. Moreover, it can be employed to conduct additional investigations of the collapsibility impacts of LID measures in places with collapsible loess.