Hotspots of Monthly Land Precipitation Variations Affected by SST Anomalies

Abstract

Abstract Climate predictability at seasonal to interannual time scales is mainly associated with sea surface temperature anomalies (SSTAs). How to quantitatively assess the impact of SSTAs on climate variability and predictability is an unresolved topic. Using a novel metric [bulk connectivity (BC)], the integrated influences of global SSTAs on precipitation anomalies over land are examined in observations and compared with Atmospheric Model Intercomparison Project (AMIP) simulations in 1957–2018. The hotspots of the land precipitation variation affected by global SSTA are identified, and the seasonality is evaluated. Such hotspots indicate the regions of land precipitation predictability caused by SSTAs. The hotspots are observed in the Sahel region in September–March, in the Indochina Peninsula in April and May, and in southwestern United States in December–March, which are mostly linked to the influence of El Niño–Southern Oscillation (ENSO). The overall impact of SSTAs on land precipitation is larger in the Southern Hemisphere than in the Northern Hemisphere. The spatial variations of BC and hotspots in the observations are partially reproduced in the AMIP simulations. However, an individual run in the AMIP simulations underestimates the integrated influence of global SSTA on land precipitation anomalies, while the ensemble mean amplifies the integrated influence, and both show a challenge in capturing the seasonality of the SST influence, particularly the time of the strongest impact. The results of the BC metric can serve as a benchmark to evaluate climate models and to identify the predictability sources.