A GRACE-based framework for tracking flash drought hotspots and spatiotemporal evolution

Flash droughts cause serious damage to ecosystems and human societies due to their rapid onset and intensification. Their fast changes pose significant challenge to existing drought warning and monitoring systems. Although previous studies have focused on meteorological, ecological and soil moisture indicators for assessing flash droughts, an effective quantitative indicator capturing the direct manifestation of flash droughts—specifically, the rapid decline in terrestrial water storage (TWS)—remains unavailable. To address this gap, we propose a pentad-scale hydrological flash drought identification framework based on a daily-scale reconstructed TWS anomaly dataset derived from Gravity Recovery and Climate Experiment (GRACE) observations. We further analyze global flash drought hotspot regions, their spatiotemporal evolution, and key drivers from 1979 to 2018. Our findings reveal that, on a spatial scale, flash drought hotspots are primarily concentrated in humid and semi-humid climate zones. On a temporal scale, the impact of flash droughts has intensified in regions such as Northern Europe, Northern Asia, Southeast Asia, and South Asia, whereas the Amazon, East Africa, and West Africa exhibit a decreasing trend. In other study regions, no significant changes in flash drought conditions are observed. At the global scale, energy-related factors (including temperature and solar radiation) contribute significantly more to flash drought development than water-related factors (such as precipitation). Additionally, approximately 48 % of flash droughts worldwide evolve into long-term droughts, with this transition occurring primarily during the vegetation growing season in humid regions. The hydrological flash drought identification framework proposed in this study effectively addresses gaps in existing monitoring systems, providing a crucial scientific basis for drought early warning and disaster mitigation.