Table Of Contents

This Page

03.11.2014: Notes from AGU Ocean Sciences 2014

On Reckinger et al.’s talk “The anisotrophy of mesoscale eddy-induced diffusion”

Reference to Fox-Kemper et al. (2013)’s paper.

They agree tat you need to increase the zonal component of diffusion to reproduce observed tracer. What is the methodology then? Diffusion my be tracer dependent and/or spatially-varying. Ad hoc method rather than physically based?

Some thoughts

  • It seems that you need to parameterize the along-PV diffusion induced by eddies (even in a model that is too coarse meridionally to resolve any potential zonal jets) while leaving the model alone with respect to the eddy-induced cross-PV diffusion as long as the tracer in question is forced at large-scale and eddies do not modulate this forcing (like near the surface).
  • Can I translate the cross-PV velocity component into a diffusivity coefficient (see my note on the literature of eddy-mean flow interaction)? If yes, apply this to the OFES model and calculate the equivalent diffusivity below and across the Kuroshio, Gulf Stream, ACC, etc. I will have to assume that the OFES model satisfies the LPV balance (or I may have to run a simulation that does satisfy the LPV balance). There is also the issue of ensemble averaging.