loading page

A method for applying lateral surface eddy diffusion in ocean models with a general vertical coordinate
  • +2
  • Gustavo M. Marques,
  • Andrew E. Shao,
  • Scott D. Bachman,
  • Gokhan Danabasoglu,
  • Frank O. Bryan
Gustavo M. Marques
NCAR

Corresponding Author:[email protected]

Author Profile
Andrew E. Shao
Canadian Centre for Climate Modelling and Analysis
Author Profile
Scott D. Bachman
NCAR
Author Profile
Gokhan Danabasoglu
National Center for Atmospheric Research (NCAR)
Author Profile
Frank O. Bryan
National Center for Atmospheric Research (UCAR)
Author Profile

Abstract

The mixing of tracers by mesoscale eddies, parameterized in many ocean general circulation models (OGCMs) as a diffusive process, contributes significantly to the distribution of tracers in the ocean. In the ocean interior, such processes occur mostly along the direction parallel to the local neutral density surface. However, near boundaries, small-scale turbulence breaks this constraint and the mesoscale transport occurs mostly along a plane parallel to the boundary (i.e., laterally near the surface of the ocean). Although this process is easily represented in OGCMs with geopotential vertical coordinates, the representation is more challenging in OGCMs that use a general vertical coordinate, where surfaces can be tilted with respect to the horizontal. We propose a method for representing the diffusive lateral mesoscale fluxes within the surface boundary layer of general vertical coordinate OGCMs. The method relies on regridding/remapping techniques to represent tracers in a geopotential grid. Lateral fluxes are calculated in this grid and then remapped back to the native grid, where fluxes are applied. The algorithm is implemented in an ocean model and tested in idealized and realistic settings. Lateral diffusion reduces the vertical stratification of the upper ocean, which results in an overall deepening of the surface boundary layer depth. Although the impact on certain global metrics is not significant, enabling lateral diffusion leads to a small but meaningful reduction in the near-surface global bias of potential temperature and salinity.