Spatial heterogeneity of flash drought events in Central Asia and their link with large-scale circulation
摘要
Central Asia (CA) is characterized by arid and semiarid climates and heavy dependence on agriculture and water resources for sustenance and economic stability. As a result, the region faces heightened vulnerability to the impacts of climate change associated with global warming. This vulnerability is exacerbated by the increasing occurrence of flash droughts, which significantly harm the local ecological balance and hinder economic progress. Therefore, this study used the Evaporative Demand Drought Index (EDDI) and Soil Moisture Index (SMI) to investigate flash drought variability and the impacts of teleconnection indices (e.g., Atlantic Multidecadal Oscillation [AMO], Arctic Oscillation [AO], North Atlantic Oscillation [NAO], Niño Index [Niño 3.4], and Pacific Decadal Oscillation [PDO]) in CA from 1948 to 2022. We found that evaporative demand flash droughts (EDFDs) and soil moisture flash droughts (SMFDs), which were particularly significant in the northern and southeastern CA (i.e., event frequency of > 30), exhibited similar spatiotemporal distributions, showing synchronous variations between decreasing flash drought events and period prolongation. Furthermore, the true positive rate of EDFDs was a reliable indicator for the development of SMFDs in CA. Seasonal analysis of EDFDs and SMFDs revealed that the peak drought period in CA occurred from May to July, with EDFD demonstrating a more pervasive and sustained impact than SMFD. EDFD and SMFD in CA exhibited the strongest correlation with the Niño 3.4 index (p < 0.05). Furthermore, EDFD was predominantly associated with the NAO (p < 0.05), whereas SMFD showed a more pronounced connection with the AMO and AO (p < 0.05). These findings can aid in developing early warning systems for flash droughts, improving irrigation practices, and formulating sustainable water resource management policies in CA.