Research On TRMM Satellite And Ground Precipitation Data Fusion And Its Application To Runoff Simulation In Beijiang River Basin

Accurately grasping the temporal and spatial distribution characteristics of precipitation in the basin is of great significance for the simulation of watershed hydrological processes and the planning and utilization of water resources.Conventional ground rainfall stations have higher accuracy of observational precipitation data,but the spatial estimation ability is insufficient.Satellite remote sensing precipitation data can reflect the spatial and temporal distribution characteristics of precipitation in the basin,but there are some data errors.Based on this,the paper uses the Beijiang River Basin as the research area.On the basis of evaluating precipitation data accuracy of TRMM 3B42V7 satellite,the paper studies the fusion correction of satellite remote sensing precipitation and ground station observation precipitation data,and further use the fusion precipitation data to drive SWAT semi-distributed hydrological model for runoff process simulation,which provides an effective reference for improving the accuracy of TRMM satellite precipitation data and the applicability of hydrological process simulation.The main research works of the thesis are listed as follows:(1)Accuracy evaluation of TRMM satellite precipitation data in Beijiang River Basin: The comprehensive evaluation and analysis of the accuracy of TRMM 3B42V7 satellite precipitation data in Beijiang River Basin from different time and space scales shows that TRMM remote sensing precipitation has strong correlation with ground station observation precipitation data.However,the accuracy of TRMM precipitation data varies with spatial position.As the time and space scale increase,the correlation between the two types of precipitation data is gradually enhanced.(2)Precipitation data fusion of TRMM satellite and ground station in Beijiang River Basin: The Mean Bias Correction(MBC),Geographically Weighted Regression(GWR)and Double-kernel Smoothing(DS)methods were used to construct the fusion model of TRMM satellite and ground station precipitation data.On the whole,the three precipitation data fusion methods can improve the accuracy of satellite precipitation data,but the fusion precipitation data obtained by the MBC method is smoothed to the strong precipitation center in some areas,resulting in large errors in the precipitation extremes in these areas.The DS and GWR methods are closer to improve the accuracy of satellite precipitation data,and they are better than the MBC method.However,the spatial distribution characteristics of DS fusion precipitation and ground station observation precipitation data are closer and can better reflect the extreme precipitation in the basin.(3)Runoff simulation of Beijiang River Basin based on different precipitation data: Four kinds of precipitation data such as ground station observation precipitation,TRMM satellite remote sensing precipitation and GWR and DS fusion precipitation were used to drive SWAT model to simulate the runoff process of Lishi,Shijiao and Sihui hydrological sites in Beijiang River basin.The results show that the DS and GWR methods can effectively improve the hydrological process simulation accuracy of satellite precipitation data.