[MITgcm-support] anomolous temperature extremes

[jeff polton]

Hi,

Hopefully this is an easy one...
I'm running the MITgcm in a channel with zonally uniform surface 
heating (of 0-2 degrees on top of the initially uniform reference 
temperature of 16 degees) and zonal wind stress. The problem seems to 
be that I'm accumulating water that is beyond the bounds of 16-18 
degrees. In particular half the domain can be less than 16 degrees 
(15.95 degrees) after 50000 timesteps.

I thought that it might be something to do with the 2nd order centered 
advection scheme but the courant number is quite large, or perhaps the 
convective adjustment - so I've tried cadjfreq=-1,100000 and cadjfreq=0 
& IMPLICITDIFFUSION=.TRUE. & ivdc_kappa=1E9. These still give the same 
"leakyness". I've also tried increasing the thermodynamic diffusivities 
diffKhT and diffKzT from 1E-5 m^2 s^-1 to 1E-3 m^2 s^-1  but without 
significant improvement.

I include my "data" file just incase anyone can help,

Thanks in advance,

Jeff
-------------- next part --------------
A non-text attachment was scrubbed...
Name: data
Type: application/octet-stream
Size: 2009 bytes
Desc: not available
URL: <http://mitgcm.org/pipermail/mitgcm-support/attachments/20040913/abe740ef/attachment.obj>
-------------- next part --------------

~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~
Scripps Institute of Oceanography, UCSD,
9500 Gilman Dr., Dept. 0213, La Jolla, CA 92093-0213
Tel: (858) 534-4560. Fax: (858) 534-8045
www-pord.ucsd.edu/~jeff