Stationary mesoscale eddies, upgradient eddy fluxes, and the anisotropy of eddy diffusivity

Abstract

The mesoscale eddies of which parameterization is needed in coarse‐resolution ocean models include not only the transient eddies akin to baroclinic instability but also the stationary eddies associated with topography. By applying a modified Lorenz‐type decomposition to the eddy‐permitting Southern Ocean State Estimate, we show that the stationary mesoscale eddies contribute a significant part to the total eddy kinetic energy, eddy enstrophy, and the total eddy‐induced isopycnal thickness and potential vorticity fluxes. We find that beneath middepth (about 1000 m) the upgradient eddy fluxes, or so‐called “negative” eddy diffusivities, are mainly attributed to the stationary mesoscale eddies, whereas the remaining transient eddy diffusivity is positive, for which the Gent and McWilliams (1990) parameterization scheme applies well. A quantitative method of measuring the anisotropy of eddy diffusivity is presented. The effect of stationary mesoscale eddies is one of major sources responsible for the anisotropy of eddy diffusivity. We suggest that an independent parameterization scheme for stationary mesoscale eddies may be needed for coarse‐resolution ocean models, although the transient eddies remain the predominant part of mesoscale eddies in the oceans.

Isopycnal eddy mixing across the Kuroshio Extension: Stable versus unstable states in an eddying model

“…Using the tracer‐based least‐square estimation approach of Lu et al . [], one would be able to estimate eddy diffusivities using velocity and temperature observations from KESS. Our work provides mixing estimates in the two states in a much larger spatial domain.…”

Erosion of a California Undercurrent eddy by bottom topography

“…However, the flux direction in the heat flux time series changed when the shape of the eddy changed. Upgradient heat flux zones have been associated with stabilization points in the mean flow [ Waterman et al ., ; Waterman and Jayne , ; Lu et al ., ]. The negative heat flux values that were observed at the top of the promontory matched the location where the eddy decayed, which suggests that the topographic destabilization of transient eddies may enhance the California Current and California Undercurrent at this location.…”

Evolution of Submesoscale Ageostrophic Motions Through the Life Cycle of Oceanic Mesoscale Eddies

“…Considering that the ageostrophic velocity component of the mesoscale eddies in gradient‐wind balance, or by cyclogeostrophic effect (Penven et al, ), can also be contained in , a high‐pass filter is used to eliminate this balanced ageostrophic component in . Constrained by the Lagrangian drifter observation, only temporal filter rather than temporal‐spatial filter (Lu et al, ) can be used. As the typical time scale of the oceanic submesoscale processes is about 1–10 days, the cutoff period of our high‐pass filter is set to 7 days, which is longer than most of the unbalanced submesoscale ageostrophic motions and, at the same time, substantially shorter than the typical evolution time scale of mesoscale eddies (see supporting information Text S3 for sensitivity of this cutoff period).…”

“…a . Constrained by the Lagrangian drifter observation, only temporal filter rather than temporal-spatial filter (Lu et al, 2016) can be used. As the typical time scale of the oceanic submesoscale processes is about 1-10 days, the cutoff period of our high-pass filter is set to 7 days, which is longer than most of the unbalanced submesoscale ageostrophic motions and, at the same time, substantially shorter than the typical evolution time scale of mesoscale eddies (see supporting information Text S3 for sensitivity of this cutoff period).…”