[MITgcm-support] Offline tracer advection

[Neill Mackay]

Hi all,

I've been running a simulation of a tracer release experiment with the offline MITgcm using velocity fields derived from satellite altimetry in order to look at diapycnal diffusivities in the Southern Ocean. It's been flagged up that my method does not take account of time-dependent undulations in the thickness of isopycnal layers. Specifically, it seems that MITgcm solves this advection-diffusion equation:

dC/dt + divergence(vC) = RHS (1)

Where 'C' is the tracer concentration and 'v' is the velocity field. When in fact we ought to be solving:

dhC/dt + divergence(vhC) = RHS (2)

Where 'h' is the layer thickness. My question is, is it possible to get MITgcm to solve equation (2) offline if we give it 3D time-evolving fields of 'h'?

I can't find any reference to such a capability in the manual or e-mail archives so I'm guessing it's not an option, but it seemed worth checking as it would solve my current problem!

Thanks,

Neill.
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mitgcm.org/pipermail/mitgcm-support/attachments/20140124/74d87a8b/attachment.htm>