Parameterisation of the eddy-eddy and eddy-meanfield Interactions in the large eddy simulation of oceanic circulations

This paper presents parameterisations of the subgrid eddy-eddy and eddy-meanfield interactions in a baroclinic oceanic simulation representative of the Antarctic Circumpolar Current. High resolution benchmark simulations are undertaken using a quasi-geostrophic spectral spherical harmonic code of la...

Full description

Bibliographic Details
Main Authors: Kitsios, Vassili, Frederiksen, Jorgen S., Zidikheri, Meelis J.
Other Authors: Johansson, A. V.
Format: Article in Journal/Newspaper
Language:English
Published: University of Poitiers 2013
Subjects:
Online Access:https://research.monash.edu/en/publications/987bdef2-d6e1-47cd-92c7-6dfdbaebbc48
https://researchmgt.monash.edu/ws/files/242290326/242230625_oa.pdf
http://www.scopus.com/inward/record.url?scp=85034032852&partnerID=8YFLogxK
http://www.tsfp-conference.org/proceedings/2013/v1/env1b.pdf
Description
Summary:This paper presents parameterisations of the subgrid eddy-eddy and eddy-meanfield interactions in a baroclinic oceanic simulation representative of the Antarctic Circumpolar Current. High resolution benchmark simulations are undertaken using a quasi-geostrophic spectral spherical harmonic code of largest wavenumber T = 504. The eddyeddy interactions are represented by both stochastic and deterministic parameterisations, with model coefficients determined from the benchmark simulation truncated back to the large eddy simulation (LES) truncation wavenumber TR < T. Coefficients of the deterministic eddy-meanfield model are determined by a least squares regression method. Truncations are repeated for various TR, with the dependence of the coefficients on the resolution TR identified. Kinetic energy spectra from the LESs agree with the benchmark simulation. Scaling laws governing the subgrid coefficients are presented, which remove the need to generate the subgrid coefficients from benchmark simulations.