Abstract: Aiming at the conflict between the modeling accuracy and the computation time cost faced by existing urban pluvial flood models, a hierarchically nested multi-scale model technique is proposed according to the spatial and temporal characteristics of urban pluvial flooding.Based on a fully distributed model, the fine mesh of the surface model is nested with a larger scaled computation mesh named as the super mesh.The fine mesh is only used to simulate the surface flood where it happens, while the super mesh is used to simulate rainfall runoff where the surface flood doesn't happen yet.The fine mesh and the super mesh switch each other automatically when it is necessary during the simulation.This hierarchically nested multi-scale modeling technique is applied to simulation on the flood event happening under Chengzhuangzi railway bridge in Beijing on August 12, 2020.The case study shows that the proposed hierarchically nested multi-scale modeling technique can save plenty of computation time while assuring a comparable simulation result.This study will improve modeling application to large-scale urban pluvial floods.