Search

Submesoscale processes have been extensively studied in observations and simulations of fronts. Recent idealized simulations show that submesoscale instabilities also occur in baroclinic mesoscale cyclones and anticyclones. The instabilities in the anticyclone grow faster and at coarser grid resolution than in the cyclone. The instabilities lead to larger restratification in the anticyclone than in the cyclone. The instabilities also lead to changes in the mean azimuthal jet around the anticyclone from 2-km resolution, but a similar effect only occurs in the cyclone at 0.25-km resolution. A numerical passive tracer experiment shows that submesoscale instabilities lead to deeper subduction in the interior of anticyclonic than cyclonic eddies because of outcropping isopycnals extending deeper into the thermocline in anticyclones. An energetic analysis suggests that both vertical shear production and vertical buoyancy fluxes are important in anticyclones but primarily vertical buoyancy fluxes occur in cyclones at these resolutions. The energy sources and sinks vary azimuthally around the eddies caused by the asymmetric effects of the Ekman buoyancy flux. Glider transects of a mesoscale anticyclone in the Tasman Sea show that water with low stratification and high oxygen concentrations is found in an anticyclone, in a manner that may be consistent with the model predictions for submesoscale subduction in mesoscale eddies.

A signature of submesoscale flows in the upper ocean is skewness in the distribution of relative vorticity. Expected to result for high Rossby number flows, such skewness has implications for mixing, dissipation, and stratification within the upper ocean. An array of moorings deployed in the Northeast Atlantic for 1 year as part of the experiment of the Ocean Surface Mixing, Ocean Submesoscale Interaction Study (OSMOSIS) reveals that relative vorticity is positively skewed during winter even though the scale of the Rossby number is less than 0.5. Furthermore, this skewness is reduced to zero during spring and autumn. There is also evidence of modest seasonal variations in the gradient Rossby number. The proposed mechanism by which relative vorticity is skewed is that the ratio of lateral to vertical buoyancy gradients, as summarized by the inverse gradient Richardson number, restricts its range during winter but less so at other times of the year. These results support recent observations and model simulations suggesting that the upper ocean is host to a seasonal cycle in submesoscale turbulence.