NAO’s Southern Center Position Shapes the North Tropical Atlantic SST Response across CMIP6 Models

Abstract The sea surface temperature (SST) in the north tropical Atlantic (NTA) influences precipitation patterns and extreme events across key regions of South America, as well as the frequency of hurricanes landing along America’s east coast. It can also remotely modulate the East Asian summer monsoon by atmospheric teleconnections. Previous studies found that the winter North Atlantic Oscillation (NAO) can significantly impact NTA SST on interannual time scales. However, how well-advanced coupled climate models capture these connections and associated physical processes is unknown. Our analysis, utilizing reanalysis products and historical simulations from 33 coupled climate models involved in the Coupled Model Intercomparison Project phase 6 (CMIP6), shows that the majority of these models effectively simulate the spatial structure of the NAO, especially for its southern part, and realistically simulate the negative (positive) SST anomalies in the NTA region in response to NAO forcing in its positive (negative) phase. Notably, each model’s accuracy in capturing the connection between the NAO and NTA SST anomalies depends largely on its ability to reproduce the location of the NAO’s southern center. When the models position the southern center more westward (eastward) than the observations, it results in a significantly stronger (weaker) SST anomaly response. This is attributed to the more (less) pronounced reactions of surface wind speed and shortwave radiation, which are influenced by the wind–evaporation and cloud–radiation effects, respectively, with the wind–evaporation effect as the primary driver. Additionally, the simulated eastward shift of the NAO’s southern center is linked to a significant overestimation of the North Atlantic midlatitude westerly jet in these models. These results facilitate improving the models’ simulation of air–sea interaction processes over the North Atlantic and understanding the origin of the north tropical Atlantic SST variability that influences global climate variations. Significance Statement The purpose of this article is to investigate the ability of Coupled Model Intercomparison Project phase 6 (CMIP6) models in simulating the impact of the winter North Atlantic Oscillation (NAO) on sea surface temperature (SST) anomalies over the northern tropical Atlantic (NTA) on interannual time scales. We find that the accuracy of each model is influenced by the simulation of the longitudes of the NAO’s southern center: A westward shift results in a stronger SST response, whereas an eastward shift leads to a weaker SST anomaly. Further analysis reveals that both the wind–evaporation effect and cloud–radiation effect play an important role in driving the distinct NAO–NTA linkages among these models, with the former effect playing a leading role. Also, the simulated eastward (westward) shift of the NAO’s southern center is associated with a greater (lesser) overestimation of the North Atlantic midlatitude westerly jet compared with observations in these models. These findings improve our understanding of the simulation of air–sea interaction processes over the North Atlantic among CMIP6 models, which is beneficial for enhancing the accuracy of climate predictions in the North Atlantic and surrounding areas.