Disentangling the Uncertainty in Flash Drought Definition and the Role of Near Real Time Vegetation Information

Flash droughts (FDs) are droughts with rapid onset and intensification. Considerable uncertainty in our understanding of FD exists in part because of uncertainty in the physical drivers of the events, and because, like other forms of drought, there are many ways to define their onset, intensity, and termination. Here, we examine both the role of FD definition and the influence of vegetation on identification and characteristics of historical FDs in the United States. First, we compare different existing FD definitions, including those based on rapid soil moisture decline, and those based on changes in US Drought Monitor (USDM) categories. Differences in FD detection between the soil moisture- and USDM-based definitions are considerable and reflect the fact that the USDM based definition integrates information from various hydrological sources. As part of this analysis, we propose a new USDM-based FD definition that differs slightly from those in the literature and that borrows components from existing soil moisture-based FD definitions. Specifically, flash droughts are defined when a location in category D0 or D1 experiences a two or more-category degradation within two weeks, which then persists for at least another two weeks, and ends when there is a one category improvement in the USDM index above the category associated with the FD condition. Unlike previous USDM-based FD definitions, our suggested definition identifies only those events that begin from non-drought conditions and specifies FD duration. Next, we investigate the role of near real time vegetation information in land surface models on the detection of FD events. For this purpose, historical (1981-2017) FD characteristics across the U.S. are assessed based on soil moisture information from the outputs of two high-resolution (1/8 degree) Noah-MP LSM (Noah-Multiparameterization Land Surface Model) simulations: one that uses the prognostic vegetation scheme, and one that uses assimilated leaf area index (LAI) from satellite data (Mocko et al., 2021). The results show that considering updated LAI can lead to changes in FD characteristics which are mainly visible in irrigated regions. Together, the findings of this work help to reconcile differences in FD definition and shed light on the processes that enable more accurate FD detection.