[MITgcm-devel] cube sphere edge problem

[Dimitris Menemenlis]

Alistair, here is one more example of a cube-sphere edge
problem.  This is for an integration with

 hFacMinDr=50.,
 hFacMin=0.3,
 hFacInf=0.1,
 hFacSup=5.,
 select_rStar=2,
 nonlinFreeSurf=4,
 viscA4Grid = 2e-3,
 viscA4Max  = 4e11,
 
which crashes at day 93.  Again it's at the cube edge, away
from corners, this time between faces 1 and 6 (see first and
second attachments).  Contrary to previous instances, the local
bathymetry near the crash is unremarkable.

Looking at hi-res cube-sphere grid definition, there is a
sharp discontinuity in the first derivative at the cube edges
(see third attachement).  Could this discontinuity be the
cause of the trouble?  If so, can we generate a grid whose
first derivative is smoother?

D.

-- 
Dimitris Menemenlis <menemenlis at jpl.nasa.gov>
Jet Propulsion Lab, California Insitute 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: W0000006480_1.jpg
Type: image/jpeg
Size: 106700 bytes
Desc: not available
URL: <http://mitgcm.org/pipermail/mitgcm-devel/attachments/20040215/f8591fea/attachment.jpg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: W0000006480_6.jpg
Type: image/jpeg
Size: 111090 bytes
Desc: not available
URL: <http://mitgcm.org/pipermail/mitgcm-devel/attachments/20040215/f8591fea/attachment-0001.jpg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: FIG1.jpg
Type: image/jpeg
Size: 74375 bytes
Desc: not available
URL: <http://mitgcm.org/pipermail/mitgcm-devel/attachments/20040215/f8591fea/attachment-0002.jpg>