Beyond simple flash drought detection: An operational index to analyse the development speed of droughts at global scale

Abstract

Research interest on flash droughts has recently risen due to the challenges posed on drought early warning and management systems. Since the main characteristic of flash drought is a rapid initial development, we first implemented a novel index capturing this feature, and then tested it against different existing ones. The proposed index does not aim at capturing only flash droughts, but it can be used to characterize the initial development speed of all types of droughts. A selected set of events were classified using an expert-based semi-quantitative approach and used to evaluate the indices. The main finding points to the Initial Development Rate in the first 3 dekads (about 30 days) of the event (IDR3) as a robust metric. A global analysis of the index highlights: 1) south-eastern Asia and the Amazon basin as hotspots with faster mean development rates; 2) Australia and the western US as areas characterized by slow events, on average. Additionally, our analysis identifies a strong seasonal component in the IDR3, with some clear relationships with climatic and environmental factors such as annual average precipitation, temperature, soil moisture, and vegetation mass. High soil moisture content and air temperature, and low vegetation amount, seem to be among the main variables controlling the speed of development. Following these results, the IDR3 seems to be a suitable index for drought forecasts aiming at anticipating the occurrence of rapid developing droughts.