The twenty-year record of satellite sea level along with high quality surface heat fluxes allow an evaluation of processes controlling the interaction between oceanic heat storage and surface heat fluxes on interannual times scales. Using gridded sea level from AVISO as a proxy for upper ocean heat content along with the surface turbulent heat flux from OAFlux, we evaluate the lagged correlations between interannual surface turbulent heat fluxes and sea level variability in the subtropical North Atlantic. Previous work has shown that lagged correlations between SST (sea surface temperature) and surface turbulent heat flux are generally antisymmetric about zero lag with negative correlations when SST leads and positive correlations when SST lags. This relationship indicates that surface heat flux force the SST variability, and at later times the SST is damped by the surface fluxes. In contrast, the lagged correlation between SSH (sea surface height) anomalies and the turbulent flux of heat show a distinctly asymmetric relationship about zero-lag. Throughout much of the region the correlations are negative when SSH leads and are not significant when SSH lags. The Gulf Stream region sea level anomalies show strong predictive skill for surface heat flux anomalies indicating a strong local feedback to the atmosphere. The lack of significant correlation when SSH lags surface flux indicate that in this region heat content anomalies are generated by oceanic heat transport convergence.

We also examine the linkages between stored heat and cloud cover in the Gulf Stream region. Minobe and colleagues have shown that the seasonal cycle of sea surface temperature drives surface wind convergence and precipitation with signatures in cloud cover in the region. Mid-level cloud fraction was shown to be linked to surface wind-convergence driven by SST gradients.

We show here that year-to-year changes in cloud fraction from ISCCP (International Satellite Cloud Climatology Project) can be linked to both surface fluxes and sea level variations in the Gulf Stream region. We find that in January and February, mid-level cloud fraction anomalies are tightly coupled to surface heat fluxes anomalies. In addition, we find that SSH leads mid-level cloud fraction in December and January by as much as 6 months. This study shows direct evidence for an atmospheric response to heat content changes in the Gulf Stream region and that the changes in the atmosphere can be predicted several seasons in advance by sea level anomalies.