# Using observations of SSH, SST and air-sea turbulent flux of heat to find the depth of the ocean that interact with the atmosphere

**CoAuthors**

**Event: **2019 Ocean Surface Topography Science Team Meeting

**Session: **Science II: Large Scale Ocean Circulation Variability and Change

**Presentation type: **Type Poster

The canonical view of air-sea interaction in the mid-latitudes is that atmospherically driven turbulent fluxes of heat drive SST changes and that these SST (sea surface temperature) anomalies are then damped by the atmosphere. With a focus on the North Atlantic, we examine the lagged correlations between monthly and interannual SST from OISST with OAFLUX turbulent flux to assess where this paradigm holds. We cluster the structure of the lagged correlations using Kmeans clustering and find three regimes, one near the Gulf Stream, one in the subtropical gyre interior, and one in the subpolar gyre. Repeating the analysis using SSH, we find similar classification regions.

We use two different idealized one-dimensional models to understand the physical processes that control the lagged correlation structures found in the cluster analysis. The regime found within the Gulf Stream peaks near zero lag (with positive surface flux indicating cooling of the ocean) and shows locations where ocean heat transport anomalies control SST and air-sea interaction. In the subtropical interior, the lagged correlations are asymmetric about zero lag, with positive correlations when Q leads and negative correlations when SST leads; atmospheric forcing dominates following the canonical view of air-sea interaction. In the subpolar gyre, the atmosphere forces changes in SST with no significant correlation when Q leads, and a negative correlation when SST leads. In this case, the atmosphere forces SST anomalies, with little feedback to the atmosphere.

Using the lagged correlation structures, we define a feedback parameter that quantifies the strength of the surface flux anomaly for a given SST anomaly with units of Watts/m2/degree C. The results are consistent with studies using in situ based data sets, although within the western boundary current regions, the feedback is up to twice as big as in earlier analyses. We also perform the same feedback analysis using SSH (sea surface height) to define a feedback with units of Watts/m2/cm.

We find the ocean depth that interacts with the atmosphere by assuming that SSH anomalies result from thermal expansion over a depth H with thermal expansion coefficient given by the climatological SST. H is found by taking the ratio of the SST feedback to the SSH feedback divided by the thermal expansion coefficient. This results in an H that is larger than the mixed-layer depth in many regions, particularly in the North Atlantic Current, and in the return flows of the Northern and Southern Recirculation gyres. In these regions, heat must be converged from the side to feed the heat released to the atmosphere.

We use two different idealized one-dimensional models to understand the physical processes that control the lagged correlation structures found in the cluster analysis. The regime found within the Gulf Stream peaks near zero lag (with positive surface flux indicating cooling of the ocean) and shows locations where ocean heat transport anomalies control SST and air-sea interaction. In the subtropical interior, the lagged correlations are asymmetric about zero lag, with positive correlations when Q leads and negative correlations when SST leads; atmospheric forcing dominates following the canonical view of air-sea interaction. In the subpolar gyre, the atmosphere forces changes in SST with no significant correlation when Q leads, and a negative correlation when SST leads. In this case, the atmosphere forces SST anomalies, with little feedback to the atmosphere.

Using the lagged correlation structures, we define a feedback parameter that quantifies the strength of the surface flux anomaly for a given SST anomaly with units of Watts/m2/degree C. The results are consistent with studies using in situ based data sets, although within the western boundary current regions, the feedback is up to twice as big as in earlier analyses. We also perform the same feedback analysis using SSH (sea surface height) to define a feedback with units of Watts/m2/cm.

We find the ocean depth that interacts with the atmosphere by assuming that SSH anomalies result from thermal expansion over a depth H with thermal expansion coefficient given by the climatological SST. H is found by taking the ratio of the SST feedback to the SSH feedback divided by the thermal expansion coefficient. This results in an H that is larger than the mixed-layer depth in many regions, particularly in the North Atlantic Current, and in the return flows of the Northern and Southern Recirculation gyres. In these regions, heat must be converged from the side to feed the heat released to the atmosphere.