Abstract's details

Provinces of Air-Sea Interaction

LuAnne Thompson (University of Washington, United States)


Paige Lavin (University of Washington, United States); Maike Sonnewald (Princeton University, United States)

Event: 2020 Ocean Surface Topography Science Team Meeting (virtual)

Session: OSTST Closing Plenary Session

Presentation type: Type Forum only

Contribution: PDF file


Using analysis of the temporal relationship between sea surface temperature (SST), sea level anomaly (SLA), and turbulent flux of heat (Q), we investigate the relative role of the atmosphere and ocean in controlling air-sea interaction in the North Atlantic Ocean on monthly to interannual time scales. Here, we use SLA as a proxy for upper ocean heat content. We introduce a modification of a stochastic air-sea interaction model of air-sea interaction created by Frankignoul et al (1998) to frame the analysis. We add a low frequency noise forcing term that represents ocean heat transport convergence anomalies. The stochastic model demonstrates that the symmetry around zero lag depends on the relative strength of atmospheric and oceanic forcing with a more symmetric lagged correlation indicating that ocean processes dominate, while a deeper mixed layer gives more persistence.

To identify the spatial provinces of air-sea interaction and the dominant processes that control air-sea interaction in each region using surface observations alone, we employ k-means clustering of lagged correlations of SST(SLA) with Q. We specify three clusters and interpret the lagged correlations by the results of the stochastic model. We find clear demarcation of regions with differing controls on air-sea interaction, one where ocean heat transport convergence anomalies dominate the control of air-sea fluxes (Gulf Stream and Recirculation Gyres), another where the atmosphere drives and then damps SST(SLA) anomalies (the Subtropical Interior), and a region where the atmosphere drives SST(SLA) anomalies that are not damped locally (Subpolar Gyre). We perform the analysis on both monthly and interannual time scales and find more robust features at interannual time scales, with SLA providing more detailed spatial demarcation than does SST. These results provide direct evidence for the potential of the use of SLA in air-sea interaction studies.
LuAnne Thompson
University of Washington
United States