Evaluation of Flash Drought Identification with Machine Learning Techniques, Part 3: Global Perspectives

Flash droughts (FDs) are droughts that can develop over a rapid time scale (~1 month). Focus on FDs over the recent years has uncovered several key variables in driving FD events, the most common of which is soil moisture, evapotranspiration, and potential evapotranspiration. Recent studies have also begun to develop ways to quantify FD events, resulting in the development of multiple FD identification methods. These FD identification methods have been used to examine their climatology, on both the regional and global scales. In addition, the first two parts of this studied began to investigate the ability of machine learning (ML) techniques to represent FD events on the regional scale for the contiguous United States, and wherein some of the ML methods were able to represent the climatology of FD events. This third and final part will expand on the work in parts 1 and 2 by extending their investigation into a global analysis, using the ERA5 reanalysis dataset for 1979 – 2021. In this study, ML techniques, both sklearn and deep learning (DL) techniques, will be applied on the global scale to represent FD events. In addition, this study can also provide a limited comparison between multiple FD identification methods in global context, and examine varying societal responses to FD events via case studies.