Intraseasonal‐to‐seasonal evolution of soil moisture‐based droughts and floods in observation‐based datasets and Coupled Model Intercomparison Project Phase 6 models

Abstract

AbstractDroughts and floods are extreme events that have tremendous socioeconomic impact. Many studies have investigated their possible variations under the effects of climate changes, while their intraseasonal‐to‐seasonal characteristics have received less attention. Based on observation‐constrained monthly soil moisture (SM) datasets since 1980, we analysed the spatiotemporal characteristics of global SM at the intraseasonal‐to‐seasonal time scales and the associated influencing factors. The results were then used to evaluate the performance of simulations by Coupled Model Intercomparison Project Phase 6 (CMIP6) models. It is found that the frequencies of extreme wet events are slightly higher than those of extreme dry events, but their spatial distributions are similar and are both affected by precipitation variability. Extreme events tend to have longer durations than nonextreme events, and their global mean durations are 4–5 months. Globally, dry events persist longer than wet events. Maximum SM anomalies are strongly affected by earlier precipitation and this effect is stronger in a transitional climate. The onset and peak times of SM anomalies in a year occur most frequently near the peak time of precipitation rather than that of SM. Approximately half of CMIP6 models overestimate the durations of both dry and wet events and approximately one third of models only underestimate wet events durations. Different versions of the same model produce similar event durations. The findings of this study enhance understanding of the intraseasonal‐to‐seasonal variabilities of global droughts and floods and could provide guidance for future study of land–atmosphere interactions and prediction of extreme events.