Space-time variability in extreme drought statistical characteristics

Droughts occur at larger spatial and longer temporal scale than most hydroclimatic processes. More than for other natural hazards, a thorough understanding of the spatio-temporal dynamics of droughts is essential in monitoring, projecting, and adapting to future drought conditions. However, the long characteristic time scale of droughts severely limits their observation in the historical record, hampering our ability to track how extreme drought events evolve in space and time. This study investigates the statistical characteristics of extreme droughts using simulations from the Paleoclimate Modelling Intercomparison Project Phase 4 (PMIP4) and the Coupled Model Intercomparison Project Phase 6 (CMIP6). By analyzing historical and paleo-hydrological data, we assess the frequency, intensity, and duration of extreme drought events over multiple geographic locations and across different time scales. An advanced non-asymptotic statistical approach, which explicitly separates intensity and occurrence of the process, is employed to capture the variability and the frequency of extreme drought characteristics in space and in time. Our findings reveal significant regional differences in extreme drought properties, with pronounced variations across different climate states and time periods.