Reevaluating hydrological dynamics in the Yangtze River basin and the Pearl River basin using GRACE TWSA series reconstructed with gap-filling method

Yangtze River basin and Pearl River basin. This study introduces a SPI-based gap-filling method to address missing data in GRACE terrestrial water storage anomaly (TWSA) time series. Building on this approach, it reevaluates hydrological dynamics in the Yangtze and Pearl River basins, highlighting the inaccuracies and limitations in hydrological variation analysis caused by GRACE gaps. The results indicate that gaps in the GRACE TWSA time series cause an underestimation (7 % and 8 %) of the correlation between precipitation and TWSA, introducing biases in long-term trends (0.54 Gt/yr and 0.37 Gt/yr) and seasonal amplitude estimates (2.14 Gt and 0.65 Gt) for the Yangtze and Pearl River basins, and reducing accuracy in detecting extreme hydrological events. Employing a coupled meteorological-hydrological framework, the SPI-reconstructed TWSA series ensures reliable gap-filling with performance comparable to machine learning (R²=0.79 for Yangtze River Basin; R²=0.88 for Pearl River Basin), enabling accurate hydrological reevaluation in these basins and sub-basins. • Incomplete GRACE TWSA time series limits understanding of basin hydrological dynamics. • The strong correlation between multi-scale SPI and TWSI reveals a significant relationship between precipitation and TWSA. • Proposed GRACE TWSA gap-filling method relies solely on precipitation data.