[MITgcm-devel] Re: arctic with thsice

[Martin Losch]

Hi Dimitris,

I have played a little with evp-discretisations and I would like you  
to test them on your grid as soon as you get a chance. Please use the  
attached file and tell me what's happening (I basically tried to  
remove as much averaging as possible, in particular dividing by  
averaged quantities, a side effect is one exchange less). It should  
work with the latest code.


Martin
-------------- next part --------------
A non-text attachment was scrubbed...
Name: seaice_evp.F.gz
Type: application/x-gzip
Size: 3943 bytes
Desc: not available
URL: <http://mitgcm.org/pipermail/mitgcm-devel/attachments/20070429/ee63fe0f/attachment.gz>
-------------- next part --------------

On 27 Apr 2007, at 10:37, Dimitris Menemenlis wrote:

> Martin, I did not mean it as a direct comparison.  The point of  
> figure was that thSIceAdvScheme=SeaiceAdvScheme=1 appears to be too  
> smooth for high-res integrations.
>
> I'll try a few more EVP integrations.  The ones that I tried from  
> pickups crashed but I have not looked in detail what happened.
>
> D.
> _______________________________________________
> MITgcm-devel mailing list
> MITgcm-devel at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-devel