Quasi-Global (50° S–50° N) of Soil Moisture and Precipitation Extremes

Clarifying the interplay between extreme soil moisture (SM) and precipitation (P) is imperative to understand the impacts of extreme events on ecosystems in a changing climate. However, the detailed relationships, pathways, and quantitative characterization of SM-P extremes at a quasi-global (50° S–50° N) scale remain unclear. Here, we systematically evaluated the co-occurrence and temporal dependencies of SM-P extremes from 2000 to 2022, quantified their synchronous probability, used statistical modeling to reveal the directional pathways among evapotranspiration (ET), P, and SM, and detected long-term trends in P and SM extremes. Our results show a significant increase in the co-occurrence frequency of SM-P extremes globally, with strong spatiotemporal co-occurrence patterns. A lower conditional probability (62%) of extreme SM anomalies was observed within a short term (34 days) after P extremes occurred, while a significantly higher conditional probability (88%) of P extremes was found following extreme SM anomalies. Path analysis (structural equation modeling) indicates a strong direct positive pathway from P to SM, whereas SM influences P indirectly through ET. Compared to satellite-based observations, the BCC-ESM1 model within the CMIP6 framework reproduces the synchrony of SM-P extremes reasonably well, offering a feasible alternative for predicting SM-P relationships in regions lacking satellite observations and aiding future projections of their trends. Our study broadens the perspective on land–atmosphere interactions and coupling mechanisms, providing a solid theoretical basis for predicting and managing the effects of extreme events on ecosystems.