Provinces of air-sea interaction in the North Atlantic Ocean

One metric evaluating the role that the ocean plays in driving atmospheric variability is the relative importance of ocean advective processes in controlling changes in SST (sea surface temperature) and Q (surface turbulent heat flux) anomalies. We examine the spatial structure of lagged correlations of monthly and interannual SST with Q to identify provinces of air-sea interaction within the North Atlantic. We use two conceptual models to guide the analysis. In Model 1 (Barsugli and Battisti, 1998) the atmospheric boundary layer temperature (TA) is strongly coupled to SST resulting in small air-sea temperature differences and weak feedback from SST to Q. In Model 2, TA is independent of SST such that there is strong feedback from SST to Q. The resulting symmetry properties of the lagged correlation between SST and Q depends on the relative strength of the atmosphere to ocean noise. K-means clustering of the lagged correlations between monthly OISST and OAFLUX turbulent fluxes is employed to identify groups of points within the North Atlantic with similar symmetry properties of the lagged correlations. One cluster appears north of the Gulf Stream; there the lagged correlations peak at zero lag with positive SST linked to heat flux out of the ocean (Model 2 with strong ocean forcing). A second cluster appears in the subtropical interior where the lagged correlations are antisymmetric about zero lag (Model 2 with weak ocean forcing). A third cluster appears in the subpolar gyre; there the lagged correlations are at a minimum when the atmosphere leads indicating that the atmosphere controls the SST anomalies but there is little local feedback (Model 1 with weak ocean forcing). On interannual time scales, the first cluster expands to both sides of the Gulf Stream and to the northeast Atlantic. The first cluster identifies regions where ocean heat transport convergence variations drive air-sea interactions, suggesting these locations in the ocean could be important in driving atmospheric variability and extremes.