Arrest of Frontogenesis by Submesoscales and Turbulence
Abigail S. Bodner1, Baylor Fox-Kemper1, Luke P Van Roekel2, James C McWilliams3 and Peter P Sullivan4, (1)Brown University, Providence, RI, United States, (2)Los Alamos National Laboratory, Los Alamos, NM, United States, (3)University of California Los Angeles, Atmospheric and Oceanic Sciences, Los Angeles, CA, United States, (4)National Center for Atmospheric Research, Mesoscale Microscale Meteorology, Boulder, CO, United States
Abstract:
Theoretical work on frontogenesis has been useful in understanding why there are so many fronts and filaments in the ocean and atmosphere, but it has been less successful in predicting the scale at which these fronts will appear in realistic environments. Observations have shown that many ocean fronts occupy the submesoscale range of scales--roughly 100 m to 10 km--yet models of fronts almost always arrive at fronts whose width is determined by the grid scale or sub-grid parameterizations rather than resolved phenomena. Thus, the discretized form of the singularities of frontogenesis theory arrive in models as well. Such singularities are an unphysical outcome, and do not resemble observations without unrealistically large diffusivity and viscosity. In order to better understand the mechanisms that are able to arrest frontogenesis, a suite of simulations spanning the submesoscale and into the boundary layer turbulence scale are carried out and analyzed. Frontogenesis occurs near the ocean surface where mixing and stratification are complex, thus we use a simplified model but a variety of realistic surface forcings to isolate control parameters and study their contributions to arrest. It is found that a variety of boundary layer processes--winds and waves, convection, and mixed layer instabilities--may compete with frontogenesis under the correct forcing parameter range. Some of the scaling laws leading to successful arrest, as well as simulations behaving arrested, will be presented. Future work will improve submesoscale-permitting modeling, as well as parameterizations of frontogenetic processes in coarse resolution models.
