Variability and its Climate Impacts of Atlantic Warm Pool in CMIP6 Simulations

The variability of the Atlantic Warm Pool (AWP) significantly influences Western Hemisphere climate and tropical Atlantic hurricane activity. Previous evaluations of AWP simulations in CMIP3 and CMIP5 revealed that the cold bias in the AWP was linked to local cloud feedbacks. In the present study, using historical runs from CMIP6, we find that the AWP cold bias has been dramatically reduced compared to CMIP3 and CMIP5 models. The weak correlation between cloud properties and the cold bias suggests that local cloud feedbacks no longer play a modulating role. Following the approach of the authors' previous study, Atlantic Warm Pool variability is assessed through spectral analysis, interannual and multidecadal variability, and comparisons of remote connections with El Niño-Southern Oscillation (ENSO) and the North Atlantic Oscillation (NAO) against observations. The influence of the Atlantic Warm Pool is evaluated in terms of continental precipitation over North America and vertical wind shear in the tropical Atlantic. Results reveal discrepancies among climate models, with better performance in simulating vertical wind shear than precipitation. On one hand, most major models successfully capture the mechanism by which Atlantic Warm Pool warming reduces vertical wind shear and increases tropospheric moist static instability, thereby favoring hurricane formation. On the other hand, they fail to reproduce the opposing patterns of North American precipitation anomalies induced by Atlantic Warm Pool warming across interannual versus decadal-to-multidecadal timescales, as shown in observations. This evaluation enhances our understanding of AWP simulations and informs the application of climate models for seasonal forecasting of Atlantic Warm Pool.