Performance of Five Satellite‐Based Precipitation Products in Capturing Extreme Precipitation and Streamflow Over 280 Catchments Across Mainland China

ABSTRACT Satellite‐based precipitation estimates (SPEs) play a crucial role in characterising extreme precipitation and streamflow, which are essential for flood risk management and water resource planning. However, substantial discrepancies remain among existing SPE products. This study systematically evaluates the performance of five widely used SPEs, including CHIRPS V2, CMORPH BLD, PERSIANN CDR, TMPA 3B42v7, and MSWEP V2, in capturing extreme precipitation and streamflow across 280 catchments in mainland China during 1998–2020. Six extreme precipitation indices (RX1day, RX5day, R95p, R20mm, CWD, and CDD) and three extreme streamflow indices (Qx1d, Qx3d, Qx5d) were analysed using CC, RMSE, and PBIAS. Streamflow was simulated using the lumped Xinanjiang (XAJ) hydrological model and evaluated against observations through NSE and PBIAS. Results show that MSWEP V2 and CMORPH BLD, which incorporate daily gauge‐based corrections, consistently outperform the other three products in reproducing daily and extreme precipitation. When used to drive the XAJ model, these two products yield streamflow simulations comparable to those driven by gauge‐based precipitation, demonstrating their suitability for hydrological applications. Spatially, both precipitation estimation and streamflow simulation achieve higher accuracy in humid and semi‐humid regions with dense gauge networks, whereas performance deteriorates in arid and mountainous regions characterised by sparse observations and complex topography. All SPEs tend to underestimate extreme streamflow magnitudes. Nevertheless, precipitation errors are partly attenuated during the rainfall–runoff transformation, leading to improved hydrological consistency. Overall, this study highlights the robustness of MSWEP V2 and CMORPH BLD for hydrological modelling and underscores the need to further refine satellite retrieval algorithms to enhance performance in topographically complex and data‐scarce regions. The findings provide practical guidance for the selection and application of SPEs in extreme precipitation analysis and hydrological applications.