Multiple Data Products Reveal Long-Term Variation Characteristics of Terrestrial Water Storage and Its Dominant Factors in Data-Scarce Alpine Regions

As the “Water Tower of Asia” and “The Third Pole” world, Qinghai–Tibet Plateau (QTP) shows great sensitivity to global climate change, change in its terrestrial water storage has become a focus attention globally. Differences multi-source data different calculation methods have caused uncertainty accurate estimation storage. In this study, Yarlung Zangbo River Basin (YZRB), located southeast QTP, was selected as study area, with aim investigating spatio-temporal variation characteristics (TWSC). Gravity Recovery Climate Experiment (GRACE) from 2003 2017, combined fifth-generation reanalysis product European Centre for Medium-Range Weather Forecasts (ERA5) Global Land Data Assimilation System (GLDAS) data, were adopted performance evaluation TWSC estimation. Based on ERA5 GLDAS, balance method (PER) summation (SS) used estimate storage, obtaining four sets TWSC, which compared derived GRACE. The results show that estimated by SS based GLDAS is most consistent time-lag effect identified PER respectively, 2-month 3-month lags. Therefore, further explore long-term temporal spatial evolution YZRB. During period 1948–2017, showed significantly increasing trend; however, an abrupt detected around 2002. That is, trend before 2002 (slope = 0.0236 mm/month, p < 0.01) but decreasing ?0.397 after Additional attribution analysis conducted, indicating that, snow equivalent, soil moisture dominated TWSC. terms distribution, large heterogeneity, mainly middle reaches high intensity human activities Parlung Basin, distributed glaciers. obtained can provide reliable support technical means exploring mechanism data-scarce alpine regions.