Multi-source remote sensing assessment of spatiotemporal heterogeneity in soil moisture in Central Asia

Central Asia, as one of the world’s typical arid and semiarid regions, experiences soil moisture variations that profoundly impact regional ecological security and sustainable agricultural development. This study employs ERA5 reanalysis data and soil moisture data derived from the SiTHv2 model, integrating trend analysis and abrupt change detection methods, to reveal the spatiotemporal evolution patterns of soil moisture at different depths across Central Asia from 1982 to 2022. The results indicate the following. (1) Overall soil moisture across the region exhibits a notable downward trend, with the mid-depth layer (ERA5 data: 7-28 cm, Slope=-4.354×10-4) exhibiting a steeper decline than the surface layer (ERA5 data: 0-7 cm, Slope=-4.217×10-4). (2) Spatially, soil moisture exhibited a “higher in the north, lower in the south” distribution pattern. Soil moisture loss was more severe in the arid and semiarid zones. Based on ERA5 data, the proportion of areas with notable surface soil moisture decline reached 52.6%. As soil depth increased, the region with notable moisture decline gradually expanded eastward, with the proportion of areas showing notable decline in deep soil layers increasing to 58.3%. (3) Interannual variability in soil moisture differed across arid zones and land types. From arid to humid regions, interannual fluctuations in surface soil moisture gradually diminished. Irrigated agricultural areas exhibited relatively stable surface soil moisture changes, whereas natural vegetation zones exhibited increased decline rates and higher dispersion. This study aims to provide scientific support for precise drought risk prevention and control, as well as adaptive ecosystem management in Central Asia.