[Mitgcm-support] Global Model and External Fields

mitgcm-support at dev.mitgcm.org mitgcm-support at dev.mitgcm.org
Wed Jul 9 15:47:41 EDT 2003


Hi,
I have been setting up the model for a global ocean run, and would ideally 
like to include most of the arctic ocean. I have been using a 2 degree 
model extending up to 87 degrees north at the north edge of the most 
northern oceanic cell. There is a polar island further north.

So far I can run the model with a momentum time step of 360s ok, or 
1800seconds with zonal filtering. Larger time steps lead to numerical 
instability. Do you have any advice on setting up the model for this kind 
of run? We would ideally like to include the arctic ocean, but if this 
means such a very short timestep we will have to omit it, as we are aiming 
to do long climate integrations. Does the ECCO run extend further north 
than 79 degrees? Could I have a copy of the topography, and parameters 
(timesteps, viscosities etc.) used for comparison.

Am I likely to run into problems using zonal filtering with ocean runs & 
the adjoint - I could not see a verification exp set up to use zonal 
filtering with oceans.

Also the external fields_load routine can cause the model to crash by 
attempting to load a record larger than the number of forcing periods from 
the forcing files. This can occur if the time is not a multiple of the time 
step - if the model is restarted with a different time step. Why is integer 
math used in this routine - purely for speed? as this seems to be causing 
the problem.

The loading of the EmPmR field for records other than #1 in 
external_fields_load has also been commented out with 'Cfixed' in the 
source repository, is this intentional?

Could you give me another example of using the adjoint, other than the vero 
experiment, please, and any other information on setting up the adjoint.

I hope this will not be too much effort for you.
Thanks

John


---------------------------------------------------------------------
John Stark                           Email 
jods at soc.soton.ac.uk
Applications Programmer              Tel +44 (0)23 8059 6571
Southampton Oceanography Centre
---------------------------------------------------------------------




More information about the MITgcm-support mailing list