Spatio-temporal patterns of vertical velocities in the Agulhas Current Retroflection from two years of SWOT altimetry

Coadou-Chaventon Solange1,2, Siegelman Lia3, Carli Elisa4, Swart Sebastiaan1,5, Schubert René6, Speich Sabrina2
1Department of Marine Sciences, University of Gothenburg, Gothenburg, Sweden
2LMD/IPSL, ENS, PSL University, École Polytechnique, Institut Polytechnique de Paris, Sorbonne Université, CNRS, Paris, France
3Scripps Institution of Oceanography, UC San Diego, La Jolla, CA, USA
4European Space Agency (ESA)—ESRIN, Frascati, Italy
5Department of Oceanography, University of Cape Town, Cape Town, South Africa
6Ocean Dynamics, GEOMAR Helmholtz Centre for Ocean Research Kiel, Kiel, Germany

The new Surface Water and Ocean Topography (SWOT) satellite extends the observed spectrum of sea-surface height (SSH) variability down to the submesoscale. It provides an unprecedented potential for reconstructing the three-dimensional ocean circulation, including submesoscale currents, which hold most of the vertical eddy kinetic energy and are thus critical for the vertical exchanges of properties between the surface and the ocean interior.

Using SWOT observations during the fast-repeat phase of the mission in the eddy rich Agulhas Current Retroflection region, we show that the representation of SSH frontal gradients are greatly enhanced compared to coarser-resolution altimetry products, being on average 28% stronger. This improved resolution of SSH anisotropy, curvature, and frontal gradients enable a more accurate reconstruction of vertical velocities ($w$) down to 1,000 m, using the effective surface Quasi-Geostrophic framework. The realism of this methodology is assessed against the 1/60° (~2 km) INALT60 ocean model, finding a spatial correlation of 0.6. SWOT-derived w reveal pronounced high-frequency variability, with events lasting several days to one week during which $w$ variance more than doubles. The Agulhas Retroflection emerges as a hotspot of enhanced $w$ (~300 m day$^{-1}$), driven by strong horizontal strain. Ongoing work will use these reconstructed w fields to assess seasonal and spatial variations of vertical heat fluxes in order to connect this to different dynamical regimes (e.g., mesoscale-submesoscale, topographic interactions) in the region.