A new framework for tracking flash drought events in space and time

Flash drought is a space–time phenomenon with rapid intensification nature that poses a series of challenges for early warning systems and drought relief. Traditional works do not consider the space–time dynamic processes of flash droughts, unable to provide important information such as how fast an event spreads in space that hampers governors and stakeholders from making timely drought mitigation operations. Here we introduced a novel framework for tracking flash droughts that fully accounts of their dynamic space–time behavior, with focus upon the instantaneous intensification/recovery rate (IIR/IRR) and spatial propagation. Flash drought events were defined by a space–time coherent set of grids where pentad-scale standardized evapotranspiration deficit index is below a prescribed threshold. The drought development/recovery stage is identified in terms of intensity, and IIR/IRR between two consecutive drought patches is determined based on the variable motion relationship of speed-time process from the physics perspective. A single space–time drought event is extracted when the duration, averaged intensification rate and IIR all reach prescribed standards. Using daily meteorological station data and daily root zone (0–100 cm) gridded soil moisture data, the framework is demonstrated by analyzing the flash droughts in the Pearl River basin over China from 1960 to 2015. Results indicate that the framework can well capture space–time structure of flash droughts including the severity and dynamic spatial propagation. Most of the identified flash drought events last 5–6 pentads but affect over half of the basin, and the top seven events have affected over 90% of the basin which intensify in two pentads showing rapid intensification nature. Moreover, flash droughts are associated with precipitation, humidity, temperature, and sunshine duration. The framework is conducive to better understanding of flash drought processes that help guide effective monitoring and development of early warning systems.