A Multidimensional Hotspot Assessment of Long-Term Terrestrial Water Storage Anomaly Change Across China

Terrestrial water storage anomaly (TWSA) provides an integrated measure for characterizing long-term variations in terrestrial water storage under the combined influences of climate variability and human activities. However, conventional assessments of long-term TWSA change have relied predominantly on linear trends, which may not fully capture concurrent shifts in interannual variability or the occurrence of wet and dry extremes. Using three GTWS-MLrec reconstructions (CSR, JPL, and GSFC) and their ensemble mean, this study investigates long-term TWSA changes across China during 1961–2020. An SED-based hotspot framework, adapted from previous climate-hotspot studies, was applied to integrate changes in mean state, interannual variability, and the frequencies of wet and dry extremes into a unified statistical metric of hotspot intensity. Results reveal strong regional heterogeneity, with pronounced water-storage declines concentrated in the North China Plain, while more heterogeneous changes occur in northwestern and southwestern China. The multidimensional analysis identifies three major statistical hotspot regions: the northwestern CB, the North China Plain, and the SWB. The major hotspot locations remain broadly consistent across the three reconstructions and under alternative normalization schemes, although the magnitude and spatial contrast of SED vary with the selected normalization factor. Basin-scale comparison further reveals regional differences in the component indicator most closely associated with the spatial distribution of SED. Mean-state decline and increasing dry-extreme frequency show relatively strong spatial associations with SED in several northern basins, whereas variability-related changes show stronger spatial associations in the SWB and YZRB. These results provide a multidimensional statistical characterization of long-term TWSA change and identify regions in which changes in mean state, interannual variability, and the occurrence of wet and dry extremes are spatially concentrated.