| The rapid development of global society and economy,the rapid progress of urbanization and human activities have caused many natural disasters.Am,ong them,water disasters such as dam break floods,mountain floods and urban floods caused by extreme rainfall weather are extremely serious.Such floods are sudden and unpredictable,and may destroy the downstream facilities that can release pollutants.Pollutants evolve downstream along with the floods and aggravate the damage caused by the floods.In cities,runoff scoured the transport of non-point source pollutants on the surface,and non-point source pollutants entered urban lakes and other landscape water bodies along with the runoff,causing water bodies to deteriorate and seriously affecting urban water landscape and water environment.In order to efficiently and accurately simulate and predict such flood events and the water environment problems caused by them,the numerical model has high use value and potential,and provides strong technical support for water disaster water environment management.In this paper,the high-resolution numerical model of the hydrodynamic and pollutant transport process is used to discretize the two-dimensional shallow water equation and the pollutant transport equation under the finite volume framework of the Godunov format,and the HLLC approximate Riemann solver is used to calculate the hydrodynamics and materials of the cell grid interface.For the transport flux,the MUSCL slope-limited linear reconstruction and the Runge-Kutta time integration method are used to achieve the second-order accuracy of time and space.At the same time,the model introduces GPU parallel computing technology to greatly improve the computational efficiency of the model.An ideal ground-type calculation example is used to verify the accuracy of the model,which can effectively reduce numerical dissipation and false numerical oscillations,and has good numerical harmony.The classic complex terrain example is used to simulate the evolution process of uniform concentration dam-break flow.The results show that the model can accurately capture the alternate change process of the dry and wet interface under complex terrain conditions.It is consistent with the results reported by previous reports,but the calculation efficiency of the model is better than that of the literature.Finally,the Toce River dam break and Malpasset dam break cases with actual topography were used to test the flood-driven downstream point source release pollutants with the flood evolving downstream.The simulation calculation results show that the point source pollutant discharge can lead to large-scale water bodies.Pollution and quickly affect downstream water bodies.At the same time,the same flood events are simulated on different resolution grid terrains by using different performance GPU and CPU computing models.The results show that the high performance GPU computing technology can achieve large scale and high efficiency computation in high resolution terrains while ensuring the simulation accuracy.Taking part of the core area of the sponge city of Fengxi New City in Xixian New District as the research object,the process of urban surface overflow and the transport process of surface non-point source pollutants are simulated and calculated.The Green-Ampt infiltration model is used to describe the process of soil moisture infiltration.First,the model is calibrated and verified using measured rainfall data,measured infiltration data,and high-resolution DEM data from drone aerial surveys.The simulation results show that in the study area All water accumulation points(a total of 7)are accurately simulated,and the relative error between the water accumulation range and the measured range of the main points(4 locations)is less than 10.38%,which proves that the model can accurately simulate the surface flow process.At the same time,the simulation calculation of the migration process of surface non-point source pollutants with rainfall runoff shows that effective measures can be taken to reduce the entry of non-point source pollutants into urban landscape water bodies by mastering the surface runoff path,and effectively guide the development of runoff pollution control and engineering measures. |