Spatiotemporal Characteristics and Driving Factors of Snow Drought on the Mongolian Plateau

Abstract Snow drought exerts widespread impacts on ecosystems and water resources. However, its spatiotemporal variability and key drivers in cold and arid regions remain poorly quantified. In this study, we used the ERA5-Land reanalysis to construct a snow water equivalent index to characterize snow drought. Its spatiotemporal variations were assessed using Theil-Sen slope estimation and the Mann-Kendall test, while variations in its driving factors were examined using Geographical Detector, variance partitioning analysis, cross-wavelet transform, and wavelet coherence analysis. From 1951 to 2024, snow drought intensified across the Mongolian Plateau, with the strongest intensification occurring in the central and southern arid and semi-arid regions. Snow drought frequency and severity declined in the early decades but increased synchronously after 2000. High frequency emerged along the northeastern margin, while high severity conditions expanded across the northeastern plateau and western mountains. Before 2000, relative humidity was the dominant control. After 2000, humidity and radiation remained primary drivers, whereas the influences of temperature and vapor pressure deficit strengthened. Topographic effects were consistently weak. Driver importance differed markedly among climate zones. El Niño-Southern Oscillation (ENSO) showed the strongest linkage with snow droughts, with coherence mainly occurring after 2000. These findings provide a scientific basis for regional ecological security planning and adaptive water resources management.