Reconstructing and downscaling terrestrial water storage to Reveal vegetation–water coupling across continuous areas of Western China-Central Asia-Western Asia under climate warming

Study area: The continuous regions of Western China-Central Asia-Western Asia (WCCAWA) feature complex topography, diverse vegetation types, and predominantly arid to semi-arid climates. This vast transboundary area is highly sensitive to global warming and serves as a critical region for understanding hydrological and ecological responses in dryland and high-elevation environments. Research focus: To overcome the limitations of discontinuous and coarse terrestrial water storage (TWS) datasets, this study reconstructed original-resolution TWS anomalies (TWSA) using the seasonal decomposition LOESS (STL) method and downscaled them to 0.1° resolution (2001–2020) with a Random Forest algorithm. The continuous, high-resolution TWSA dataset was then used to analyze spatiotemporal variations and vegetation responses across WCCAWA. New hydrological insights for the study area: The STL method effectively filled TWSA gaps (R²= 0.994, RMSE = 5.2 mm, MAE = 2.42 mm), and the downscaled data captured fine-scale heterogeneity in plateau and mountain regions. A significant TWSA decline (-0.0109 mm/month, P < 0.05) occurred in over 70% of WCCAWA, notably in the Caspian Sea basin and southern Tibetan Plateau. Faster losses in low-elevation irrigated areas and bareland indicate intensified groundwater and soil-moisture depletion, whereas stronger vegetation sensitivity at high elevations suggests tighter coupling among cryosphere dynamics, water storage, and ecosystems. These findings enhance understanding of regional water depletion and provide valuable support for water resource management and ecological restoration in arid and high-cold ecosystems.