[cam-users] Restart Run - Courand

Wed Mar 17 20:56:27 MST 2004

Dear Phil, 

Thank you for your explanation.

I followed CAM User Guide Examples. Example 1 ran successfully, but example 3 
(restart run) run didn't complete. I didn´t modify the CAM source Code. The 
changes I have made were: adding "mss_irt = 0" in the namelist (Jim Rosinski's 
suggestion) and changing the path definition at CAM restart point flie. After 
that, I could start the "restart run", but it ran correctly only until timestep 
100:

NSTEP =     100   8.863219306789937E-05  6.850236144683818E-06 252.674  9.84642
E+04  2.470325627315098E+01 0.99  0.20
 NSTEP =     101   8.863111229146920E-05  6.843568489257052E-06 252.675  9.84643
E+04  2.470482532316608E+01 0.99  0.19
COURLIM: *** Courant limit exceeded at k,lat=  1 10 (estimate =  0.999), solutio
n has been truncated to wavenumber  41 ***
 *** Original  Courant limit exceeded at k,lat=  1 10 (estimate =  0.999) ***
 NSTEP =     102   8.862973662740181E-05  6.837205409114171E-06 252.677  9.84643
E+04  2.470712312424124E+01 1.00  0.19

 When the "restart case" is running the model write this mensage:

>max rss=0 shared mem=0 unshared data=0 unshared stack=0

This happens just after it read the "restart files".

I will look for at subroutine tphysbc. 

More one detail, I use :

MPSTKZ      = 148M
stack limits= unlimited

Do you see any problens with this configuration ?

Thank you again,

Best Regards,

Marcio.

> Dear Marcio.
> 
> The model is actually stoping in subroutine "findsp" found in the file
> cldwat.F90.
> 
> It is stopping in a routine that I wrote, attempting to determine the
> wet bulb temperature through an iterative procedure. The procedure
> proceeds until the wet bulb temperature and corresponding saturation
> mixing ratio have converged to at least 4 significant digits.  It is
> very unusual for the model to stop in this procedure. We have run the
> routine for simulations over thousands of years in the last 5 years.
> 
> The diagnostics are indicating that the initial guess for the
> iteration had temperature and mixing ratio of approximately 209 and 3.e-2
> respectively and that the iteration gave up when it got to a
> temperature of 263 and 2.e-2. The initial value of that mixing ratio
> is pretty strange. It is a value that would be more characteristic of a
> mixing ratio at a temperature near the surface in the tropics (eg 30
> g/kg) not of a parcel near -60C.
> 
> The likely explanation is that there is a problem somewhere else in
> the model that has produced a situation that is so non-physical that
> findsp cannot find a viable wet-bulb temperature. 
> 
> I would focus on that point (column = 12, level = 8) within subroutine
> tphysbc and figure out what generated the unusual combination of
> temperature and mixing ratio.
> 
> This sort of situation usually happens when one modifies the model in
> some unusual way, and it has begun to fail by drifting into an
> entirely unusual situation. Alternatively, you might have set up the
> restart incorrectly.
> 
> Regards
> 
> Phil Rasch
> 
> 
> On Tue, Mar 16, 2004 at 09:02:11PM -0300, cataldi at coc.ufrj.br wrote:
> > Hi All,
> > 
> > When I run CAM 2.0 in "restart run" I find problems related with the
> courand 
> > parameter that stop the execution of the model (below).
> > 
> > ..  
> > ** Original  Courant limit exceeded at k,lat=  1 10 (estimate =  1.108)
> ***
> >  NSTEP =     196   8.846274340589037E-05  6.654129814775965E-06 252.721  
> > 9.84643E+04  2.470731866012359E+01 1.11  0.24
> >   findsp not converging at point i, k            12            8
> >   t, q, p, enin     209.9547719743370        3.8770276331074847E-002
> >     8567.746034482248         320831.0644319958
> >   tsp, qsp, enout     263.6827235411741        2.1032348772033886E-002
> >     320831.0860996427
> >  ENDRUN IS BEING CALLED
> >   
> > Would anybody be able to helps me in that subject?
> > 
> > Thank you in advance,
> > 
> > Best Regards,
> > 
> > Marcio Cataldi
> > 
> > Federal University of Rio de Janeiro
> > 
> > _______________________________________________
> > cam-users mailing list
> > cam-users at ucar.edu
> > http://mailman.ucar.edu/mailman/listinfo/cam-users
> 
> -- 
> Phil Rasch, Climate Modeling Section, National Center for Atmospheric
> Research
> Mail     --> P.O. Box 3000, Boulder CO 80307  
> Shipping --> 1850 Table Mesa Dr, Boulder, CO 80305
> email: pjr at ucar.edu, Web: http://www.cgd.ucar.edu/cms/pjr Phone:
> 303-497-1368, FAX: 303-497-1324 
> 