[MITgcm-devel] Re: [MITgcm-cvs] MITgcm/pkg/mom_vecinv CVS Commit

[Dimitris Menemenlis]

Alistair, does your modified scheme include Baylor's divergent flow 
addition?  Seems to do wonderful things to grid scale noise in W, see 
attached figure.

For 1/16th-deg 80S-80N isotropic grid:
useFullLeith=.TRUE.
viscC4Leith=0.1,
viscC4Leithd=0.1,
deltaT=60
based on attached figure, I think I can probably double time step.  No 
other restrictions on viscosity.

For 1/8th-deg 80S-80N isotropic grid:
viscC4Leith=0.1,
viscA4GridMin=0.9e-4,
viscA4GridMax=1.2e-2,
deltaT=200

For 1/4th-deg 80S-80N isotropic grid:
viscA4GridMin=8.e-4,
viscA4GridMax=8.e-3,
viscC4Leith=0.3,

For 18-km cube sphere I typically use:
viscA4GridMin=3.e-4,
viscA4GridMax=4.e-3,
viscC4Leith=0.2,
deltaT=1200,

I also have a low-viscosity 18-km cube sphere integration ongoing with:
viscC4Leith=0.0075,
deltaT=200,

D.

-- 
Dimitris Menemenlis <menemenlis at jpl.nasa.gov>
Jet Propulsion Lab, California Institute of Technology
MS 300-323, 4800 Oak Grove Dr, Pasadena CA 91109-8099
tel: 818-354-1656;  fax: 818-393-6720
-------------- next part --------------
A non-text attachment was scrubbed...
Name: FIG.jpg
Type: image/jpeg
Size: 116915 bytes
Desc: not available
URL: <http://mitgcm.org/pipermail/mitgcm-devel/attachments/20050325/53cc3870/attachment.jpg>