Overturning of Mixed Layer Eddies in a Submesoscale-Resolving Simulation of the North Atlantic

Le résumé fourni par la source

Abstract To study submesoscale instabilities in the ocean mixed layer, this study uses the novel Icosahedral Nonhydrostatic (ICON)-Submesoscale Telescope (SMT) configuration that exploits a telescoping grid refinement to achieve a horizontal resolution finer than 1 km over wide areas of the North Atlantic. The model’s ability to simulate mesoscale-to-submesoscale turbulence is validated by comparing spatial power spectra of sea surface temperature and height with satellite data and a 10-km eddy-resolving simulation. We find more realistic variability in the refined grid simulation compared to the coarser simulation over a wide range of scales, including the mesoscale eddy regime. Furthermore, the high resolution permits submesoscale baroclinic instabilities at ocean fronts and we observe strong frontal overturning and restratification. Overturning rates are diagnosed from eddy buoyancy flux and mean front characteristics such as horizontal and vertical density gradients. To accurately capture the vertical extent of mixed layer eddy instabilities, commonly used threshold algorithms for identifying the mixed layer depth must be modified. We compare spatial and time filtering approaches for estimating submesoscale eddy fluxes and find qualitative similarity, although time filtering yields stronger fluxes. The diagnosed overturning rates are compared to two submesoscale baroclinic instability parameterizations. Both capture overturning magnitude at ocean fronts within an order of magnitude but overestimate it at eddy rims. Comparing submesoscale eddy fluxes in the entire study area shows two different regimes where the parameterizations slightly differ in the ability to capture the diagnosed eddy fluxes. Significance Statement In this study, we use the Icosahedral Nonhydrostatic (ICON) model with a novel configuration that allows to have a flexible horizontal resolution. With that configuration, we achieve a horizontal resolution finer than 1 km over large parts of the North Atlantic. This allows us to study so-called submesoscale ocean eddies that occur on small spatial scales at upper-ocean density fronts. Based on a comparison with high-resolution satellite data, we can show that our configuration better captures the variability on small scales compared with coarser configurations. Furthermore, we quantify how rapidly the submesoscale eddies turn over the density fronts and we evaluate how well parameterizations would be able to capture such a submesoscale eddy overturning. These findings advance understanding of submesoscale dynamics, their role in ocean energy transfer, and mixing.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.