[MITgcm-support] CheapAML problems with non-zero wind

[Patrick Watts]

Hi,

Sorry for the slow reply. I've done a few runs with increased cheapaml_ntim, reduced cheapaml_kdif, and the advanced advection scheme; unfortunately I've either had the same problem I was having, or I appear to push things too far.

Will play around more and let you know if anything changes. Thanks for the help!
Patrick
________________________________________
From: Deremble, Bruno [bderemble at fsu.edu]
Sent: 12 May 2014 15:23
To: mitgcm-support at mitgcm.org
Subject: Re: [MITgcm-support] CheapAML problems with non-zero wind

Hi Patrick,

I wonder if this is related to a CFL criteria.
try to increase cheapaml_ntim (in &CHEAPAML_CONST) in data.cheapaml (the number of atmospheric time steps per ocean time step).
I also recommend  to use an advanced advection scheme (option useFluxLimit=.TRUE., in section &CHEAPAML_PARM02 )
last, you can try to adjust the diffusion coefficient cheapaml_kdiff (&CHEAPAML_CONST) (set by deffault to 1e4). Given your super high resolution, I would lower it.

hope this helps
bruno

_______________________________________________
MITgcm-support mailing list
MITgcm-support at mitgcm.org
http://mitgcm.org/mailman/listinfo/mitgcm-support