Evaluation and Hydrological Application of a Data Fusing Method of Multi-Source Precipitation Products-A Case Study over Tuojiang River Basin

Precipitation is an essential driving factor of hydrological models. Its temporal and spatial resolution reliability directly affect the accuracy modeling. Acquiring accurate areal precipitation needs substantial ground rainfall stations in space. In many basins, are sparse uneven, so real-time satellite products (SPPs) have become important supplement to ground-gauged (GGP). A multi-source fusion method suitable for Soil Water Assessment Tool (SWAT) model has been proposed this paper. First, multivariate inverse distance similarity (MIDSM) was search optimal representative points GGP SPPs sub-basins. Subsequently, correlation-coefficient-based weighted average (CCBWA) presented applied calculate fused product (FMSPP), which combined multiple products. The effectiveness FMSPP proven over Tuojiang River Basin. case study, three were chosen as sources, namely Climate Forecast System Reanalysis (CFSR), Tropical Rainfall Measuring Mission Project (TRMM), Estimation from Remotely Sensed Information using Artificial Neural Network Data Record (PERSIANN-CDR). evaluation indicators illustrated that could capture occurrence events very well, with a maximum Probability Detection (POD) Critical Success Index (CSI) 0.92 0.83, respectively. Furthermore, its correlation GGP, changing range 0.84–0.96, higher most sub-basins on monthly scale than other SPPs. These results demonstrated performance best compared original Finally, SWAT found effectively drive contrast single source. manifested highest modeling, Coefficient Determination (R2) 0.84, Nash Sutcliff (NS) Percent Bias (PBIAS) only −1.9%.