Vertical Mixing Can Both Induce and Inhibit Submesoscale Frontogenesis

Daniel Dauhajre1,2, Kaushik Srinivasan2, M. Jeroen Molemaker2, Jonathan Gula3,4, Delphine Hypolite2, James McWilliams2, Roy Barkan2,5, William Young6
1Department of Geography, University of California, Santa Barbara
2Department of Atmospheric and Oceanic Sciences, University of California, Los Angeles
3Univ. Brest, CNRS, Ifremer, IRD, INRIA, Laboratoire d’Océanographie Physique et Spatiale, IUEM,
4Institut Universitaire de France
5Department of Geosciences, Tel Aviv University
6Scripps Institution of Oceanography, University of California, San Diego

Submesoscale fronts and filaments perpetually arise in the surface boundary layer with horizontal scales of 0.1–1 km and localized overturning circulations that can, among other consequences, significantly restratify the mixed layer. This talk will demonstrate how vertical boundary layer turbulence (vertical mixing by eddies smaller than the submesoscale) regulates the local circulation, lifecycle, and thus the consequences of submesoscale fronts (and filaments). Past studies separately demonstrate that vertical mixing can either sharpen (frontogenesis) or weaken (frontolysis) submesoscale fronts. However, these studies invoke competing interpretations that offer little consensus on the problem. Here, a suite of idealized simulations of 2D fronts demonstrates that vertical mixing can induce either frontogenesis or frontolysis; one surprising result is that weak mixing favors frontogenesis, contrary to the prevailing understanding. A competition in the buoyancy dynamics (cross-front advection versus vertical diffusion) controls regime transitions, despite the role of vertical momentum mixing in generating and sustaining the ageostrophic secondary circulation that sharpens the front. Application of two scalings to quantify the competition between cross-front buoyancy advection and vertical diffusion identifies practically equivalent parameters that map regime transitions: $\rm{Ro^2 / Ek}, \rm{Ro/Ek^{1/2}}$, where the Rossby number ($\rm{Ro}$) measures the frontal strength and the Ekman number ($\rm{Ek}$) measures the vertical mixing intensity. While these results can better inform the necessary parameterization of submesoscale fronts in global models, future work should interrogate the validity of this idealized framework, which unrealistically assumes that vertical mixing does not evolve. This idealized modeling work also motivates new questions regarding the sensitivity of simulated submesoscale flows to the choice of boundary layer turbulence parameterization in regional models.