[CIG-MC] Re: geoid scaling in CitcomS-2.2.1

Eh Tan tan2 at geodynamics.org
Tue Oct 30 11:29:55 PDT 2007

Scott King wrote:
> initial residue of momentum equation F 1.684117294e+00 9741456
> AhatP (000) after 1.88612 seconds with div/v=2.084e-05 dv/v=0.000e+00
> and dp/p=0.000e+00 for step 0
> bottom heat flux= 26.146639
> surface heat flux= 7.580164
> initial residue of momentum equation F 1.684157643e+00 9741456
> AhatP (000) after 1.87142 seconds with div/v=2.081e-05 dv/v=0.000e+00
> and dp/p=0.000e+00 for step 1
> So this is the output from the run.   Should I be worried that
> dp/p=0.000e+00 ??   How would I tell there is a problem?


dv/v and dp/p are the relative magnitude of velocity and pressure
correction between the Uzawa iterations. They are always 0 for the first
iteration. Nothing to worry about them.

Your model seems to converge very well at step 0 and 1. Does the
problematic geoid occur at these time steps? If yes, what is the min/max
of your velocity solution at these time steps? (A convenient GMT command
"minmax -H2 *velo*" can query the min/max of all velo files.) If not,
can you check the convergence history around the problematic time step?

Eh Tan
Staff Scientist
Computational Infrastructure for Geodynamics
2750 E. Washington Blvd. Suite 210
Pasadena, CA 91107
(626) 395-1693

More information about the CIG-MC mailing list