Impacts of the North Atlantic Oscillation on Sea Surface Temperature on the Northeast US Continental Shelf

This research aims to understand the role of the North Atlantic Oscillation (NAO) in modulating sea surface temperature (SST) on the Northeast US Continental Shelf (NES). The NAO is the dominant large-scale atmospheric oscillation in the North Atlantic and has profound effects on water temperatures in the North Atlantic. Waters on the NES originate primarily from Scotian Shelf Water at the surface and Labrador Slope Water and Warm Slope Water at depth through the Northeast Channel. By using a high-resolution SST dataset, we found that the correlation between the NAO and annual mean SST in the Gulf of Maine (GOM) is significant and negative at lag of four years. Further spatial correlation analysis shows that the NAO influences SST in the GOM primarily through advection of SSW or shelf water at the surface from the Labrador Sea. Cross-correlation analysis was also applied between the NAO and SSTs in other subregions of the NES (Georges Bank, Southern New England, and Mid-Atlantic Bight), but no statistically significant relationships were found at any lags. Different from temperature at depth in the GOM that is positively influenced by the NAO with a lag of two years, we concluded that the NAO has a significant negative effect on SST in the GOM four years later, while its effects on SSTs in the other three subregions of the NES are negligible. The four-year lagged relationship we found between the NAO and annual mean SST in the GOM provides a robust empirical method to predict the effect of the NAO on annual mean SST in the GOM four years in advance. Furthermore, we built a statistical prediction model for NES SST based on the conclusion that NES SST is significantly affected by along-shelf water advection from the Scotian Shelf. Correlation analysis shows that the predicted NES SST at two years lead time is significantly correlated with observation, indicating that this statistical model can be used to provide robust NES SST prediction at least 2 years in advance.