Dynamical mapping of SWOT: performances from real observations

The Surface Water and Ocean Topography (SWOT) mission offers two-dimensional measurements of Sea Surface Height (SSH), capturing scales of a few tens of kilometers and enabling the study of previously unobserved short mesoscale dynamical structures. However, the mission faces technical challenges in maximizing scientific benefits, particularly during the science phase with its 21-day repeat orbit, which limits observations of small-scale structure evolution over time.

To address the challenge of high spatial and low temporal samplings, we propose an original dynamic interpolation scheme that we call 4Dvar-QG. This innovative method combines a weakly constrained, reduced-order, 4-dimensional variational scheme with a quasi-geostrophic model. The weak constraint of the quasi-geostrophic model on the inversion procedure ensures that the estimated maps closely match the observations while preserving the space-time continuity of the reconstructed structures.

The 4Dvar-QG method is applied in the North Atlantic Ocean on a constellation of real conventional altimeters and SWOT, during both the SWOT’s fast sampling and science phases. Performance evaluations are conducted through Observing System Experiments, utilizing independent data (such as altimetric and drifter data) as ground truth and comparing results to operational products like the Multiscale Interpolation Ocean Science Topography product (MIOST). The 4Dvar-QG method significantly improves the mapping of short energetic structures, reducing the root mean square error by up to 50% and increasing the effective resolutions by up to 30% compared to MIOST, while maintaining good reconstruction of large-scale and/or low energetic structures.