Spatiotemporal patterns of meteorological–soil moisture drought propagation on the Qinghai-Xizang Plateau, China 后印本

摘要: Drought is among the most destructive and recurrent natural disasters worldwide. In recent decades, the frequency of drought events has increased, exerting significant impacts on socioeconomic development. The propagation of meteorological drought (MD) to soil moisture drought (SMD) is a common natural process; however, its dynamics across different seasons and vegetation types on the Qinghai-Xizang Plateau, as well as the underlying meteorological driving mechanisms, remain insufficiently understood. This study utilized precipitation and soil moisture data from the European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis v5 (ERA5)-Land reanalysis dataset for the period 1982–2022. The standardized precipitation index (SPI) and standardized soil moisture index (SSMI) were employed to characterize MD and SMD, respectively. By integrating run theory with an optimal parameter geographical detector (OPGD) model, this study systematically analyzed the average duration and propagation time of MD and SMD across the Qinghai-Xizang Plateau, and quantitatively evaluated the explanatory power of various meteorological and topographical factors influencing drought propagation. The results indicated that the mean duration of SMD across the Qinghai-Xizang Plateau from 1982 to 2022 was generally longer than that of MD. Significant seasonal differences in propagation time were observed, with the average propagation time ranked as winter (21 d)>spring (14 d)>autumn (10 d)>summer (8 d). Spatial variability of propagation time was more pronounced in spring and winter than in summer and autumn. Furthermore, the analysis of driving mechanisms revealed that drought propagation from MD to SMD on the Qinghai-Xizang Plateau was primarily influenced by precipitation (relative contribution proportion of 51.9%), followed by evaporation (15.1%) and snowmelt (13.6%), with the strongest interaction effects associated with precipitation. Although the dominant factors across different vegetation types were generally consistent with those for the entire plateau, solar radiation also showed a relatively high contribution (average 13.9%) across vegetation types. In summary, this study provides a scientific basis for improving drought early warning systems and optimizing water resource management strategies.