[MITgcm-devel] newest code giving nans

[Martin Losch]

run it with seaice and the nans will go away (as usual) (o:
M.

On Jul 29, 2009, at 6:24 PM, Matthew Mazloff wrote:

> Hello,
>
> Just updated the newest code (older one was a couple months old) and  
> ran it (with no sea-ice) and got nans at time-step 1.
>
> One clue is that the time_tsnumber was not properly  
> incrementing...for example, from timestep 0 to 1 STDOUT wrote:
>
>
> (PID.TID 0000.0001) %MON vort_a_mean                  =    
> 7.7907401369469E-05
> (PID.TID 0000.0001) %MON vort_a_sd                    =    
> 7.5333276615471E-06
> (PID.TID 0000.0001) %MON vort_p_mean                  =    
> 8.1856709185886E-05
> (PID.TID 0000.0001) %MON vort_p_sd                    =    
> 3.4115931535963E-05
> (PID.TID 0000.0001) %MON surfExpan_theta_mean         =    
> 7.5944713547936E-06
> (PID.TID 0000.0001) %MON surfExpan_salt_mean          =    
> 4.5528081632941E-06
> (PID.TID 0000.0001) //  
> =======================================================
> (PID.TID 0000.0001) // End MONITOR dynamic field statistics
> (PID.TID 0000.0001) //  
> =======================================================
> cg2d: Sum(rhs),rhsMax =   1.45330722567394E-07  4.35515400006567E-01
> (PID.TID 0000.0001) //  
> =======================================================
> (PID.TID 0000.0001) // Begin MONITOR dynamic field statistics
> (PID.TID 0000.0001) //  
> =======================================================
> (PID.TID 0000.0001) %MON time_tsnumber                 
> =                     0
> (PID.TID 0000.0001) %MON time_secondsf                 
> =                   NaN
> (PID.TID 0000.0001) %MON dynstat_eta_max               
> =                   NaN
> (PID.TID 0000.0001) %MON dynstat_eta_min               
> =                   NaN
> (PID.TID 0000.0001) %MON dynstat_eta_mean              
> =                   NaN
> (PID.TID 0000.0001) %MON dynstat_eta_sd                
> =                   NaN
> (PID.TID 0000.0001) %MON dynstat_eta_del2              
> =                   NaN
> (PID.TID 0000.0001) %MON dynstat_uvel_max              
> =                   NaN
>
> However the model ran the full simulation and properly incremented  
> many things (e.g. dayrec) and the timer stats looked fine at the end  
> of the run.
> and notice initial (timestep 0) cg2d numbers are fine
>
> So any idea what may have happened?  Could this possibly be a TAF  
> thing...
>
> Thanks
> Matt
> _______________________________________________
> MITgcm-devel mailing list
> MITgcm-devel at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-devel