Research On Flood Simulation Model In The Typical Catchments Along The Middle Route Of The South-to-North Water Diversion Project

The Middle Route of the South-to-North Water Diversion Project is an infrastructure project in China of great strategic significace.There are many small and medium-sized catchments along the route,where storm and flood occur frequently.It will seriously threaten the safety of the large river-crossing structures.Therefore,the establishment of rainfall-runoff model can provide scientific suggestions for flood-control management of the project.However,due to the complexity of runoff generation mechanism and the lack of hydrological data in semi-arid and semi-humid areas,the flood simulation is difficult.In this study,the Shuangjihe catchment in semi-humid area and the Anyanghe catchment in semi-arid area are selected as two typical catchments of the project.A physically-based distributed model GBHM,a lumped conceptual model Xin'anjiang and a regionalized parameters-based conceptual model HBV-SIMREG are established in the two catchments to simulate the daily runoff and flood events,based on data collected from the rainfall gauges and meteorological stations.The results show that HBVSIMREG model has the worst performance in both catchements,which indicates that there are still some problems for the application of HBV-SIMREG hydrological model in the small and medium-sized catchements of China.The other two models show poor performance in the Shuangjihe catchment,due to the large uncertainty of restoring computation of runoff and the low spatiotemporal resolution of rainfall data.But GBHM show better results compared to Xin'anjiang model.In the Anyanghe catchment,Hengshui station which is not affected by reservoir is selected for model calibration and the rainfall data are complete,thus the simulation results are better than the Shuangjihe chatchment.As the runoff generation mechanism of Xin 'anjiang model is too simple and the spatial heterogeneity of rainfall is not taken into consideration in the lumped model,so the simulation results is worse than GBHM.The daily peak of GBHM reaches the highest grade for flood forecasting,and Nash efficiency coefficient reaches the second grade.The flood peak reaches second grade and the peak time reaches the highest grade.Based on GBHM which is driven by European numerical weather forecast data,a flood forecasting system for typical watershed in the Middle Route of South-to-North Water Diversion Project is established to forecast the hourly runoff process of typical watershed in the next three days.The system is embedded in the early warning platform for the large river-crossing structures along the route,which provides scientific guarantee for the safety of the project.Finally,the trend of annual average temperature,annual rainfall,annual evaporation and annual runoff are analyzed by GBHM driven by GCM long-term forecast data,and the design flood peak in typical catchments under global climate change are also calculated.The results show that the flood peak in 100-year return period of the Shuangjihe catchment increases by 9% and that in the Anyanghe catchment increases by 13%.In 2050,the probability of exceeding standard flood in the Shuangjihe catchment increases from 0.395 to 0.47,and that in the Anyanghe catchment from 0.395 to 0.48.Therefore,under the influence of climate change,the control measurements in both catchments need to be strengthen to ensure the safe operation of the main canal.