基于等体积法的暴雨淹没快速模拟分析及应用OA

The calculation of storm-induced inundation serves as a critical foundation for urban flood prevention and disaster mitigation.Under changing environmental conditions,the patterns of runoff generation and flow concentration are altering,thereby increasing the demand for timely simulation of storm-induced inundation.Given constraints in data availability and computational resources,this study developed a rapid simulation model for storm-induced inundation based on the equal volume method.The model operated through three modules:① High-resolution DEM data were employed to identify sub-basins,confluence nodes,and downstream flow paths within the watershed.By considering the presence of small depressions in the original DEM,selective filling was performed by applying volume and/or depth thresholds.The filtered depressions were designated as potential inundation areas.Ultimately,a regional drainage network was generated,comprising the delineated sub-basins,downstream flow paths,and confluence nodes.② The variable infiltration capacity(VIC)hydrological model was implemented to simulate both the water balance and energy balance within the hydrological cycle.This model integrated factors such as soil properties,vegetation characteristics,and meteorological data(e.g.,precipitation,temperature,and wind speed)to compute key hydrological fluxes on the land surface,including runoff and evapotranspiration.③ The equal volume method was applied to calculate the final accumulated water volume(VW)within each sub-basin.This volume was subsequently compared against the maximum storage capacity(VD)of the depression in that sub-basin.If VW exceeded VD,the surplus flow was routed along the predetermined flow path to the downstream sub-basin depression.Conversely,if VW was less than or equal to VD,water was retained within the current sub-basin depression.Upon completion of these computations for the entire watershed,the final storm-induced inundation areas and corresponding water depths were identified.A case study in the Qingxiu District of Nanning City demonstrates that the proposed model outperforms the CaMa-Flood model in both simulation performance and computational time.The model achieves a probability of detection(POD)of 91%and computes results 23 times faster than the CaMa-Flood model.Furthermore,the study reveals that the volume threshold significantly influences model outcomes.Variations in the volume threshold lead to differences in the number of model nodes.As the threshold increases,the inundation water level in the downstream areas of the study site rises,and the spatial distribution of simulated inundation is markedly affected.Both simulation accuracy and computational duration are influenced by the volume threshold.Specifically,increasing the threshold accelerates model execution but reduces computational accuracy.An increase in the number of confluence nodes,however,contributes to enhanced model performance.The simplified model developed in this study simulates storm-induced inundation at a lower computational cost and captures the spatial dependencies of the inundation process with higher efficiency,offering a novel approach for urban flood prevention and mitigation.