Investigating the Characteristics and Drivers of Slow Droughts and Flash Droughts: A Multi-Temporal Scale Drought Identification Framework

Abstract

Global climate change has altered the characteristics of conventional drought events, with an increasing number of Slow droughts (SD) rapidly transitioning into Flash droughts (FD). This study introduces a novel multi-temporal scale drought identification framework (MTSDIF) that classifies historical agricultural drought events into three types: SD, FD, and Slow-to-Flash Drought (SFD). Based on the MTSDIF, the GLDAS-Noah root zone soil moisture dataset was used to analyze the spatiotemporal characteristics, evolution, and driving factors of multi-temporal scale droughts in China. Our study confirms the effectiveness of the proposed MTSDIF in classifying droughts with different onset speeds (SD, FD, and SFD). The results indicate that, from 1980 to 2020, the three types of drought events in China exhibited short-term, medium-term, and long-term periodic oscillations. Before 2000, SD events were the predominant type of agricultural drought in China, but post-2000, the areas affected by FD and SFD have been continuously expanding. Compared to SD, key meteorological elements influencing FD and SFD show anomalies exceeding 0.5 times the standard deviation. In the southeastern regions of China, areas with human-impacted soils, leached soils, and incept soils exhibit a higher response frequency to FD. Sea surface temperature indices, including the interannual El Niño-Southern Oscillation in the Pacific and interdecadal variations such as the +PDO and −AMO, significantly influence the occurrence of FD in the monsoon regions of China (p < 0.01). Together, the results highlight the necessity of understanding the disparities and consistencies in land-atmosphere-ocean mechanisms behind drought events with varying onset speeds.