[Wrf-users] Re: NaN values

[Andrea Hahmann]

Dear Jorge and Goran

This type of behavior is quite common in atmospheric models. Some
synoptic situations are just more unstable than others (maybe the wind
direction in relation to topography, intense convection, etc.). And as
pointed out by Jorge, might respond differently to different
parameterizations.  Usually decreasing the time step would make the
particular simulation complete.  Other times smoothing the terrain
helps. 

Regards,
Andrea

On Mon, 2008-12-15 at 15:58 -0300, Jorge Alejandro Arevalo Borquez
wrote:
> I've got the same problem using some parametrizations on a Itanium
> based cluster. But, i couldn't solve it. When i use diferent
> parametrizations all be right.
> 
> Could you send more specifications? namelist.input and machine
> architecture.
> 
> Regards
> 
> 2008/12/15 Goran Gasparac <gorangas at gmail.com>
>         Regarding to earlier mail..
>         I also checked metgrid and ungrib logs, and everything is
>         correct..
>         
>         Also, didn't change anything in configuration,and today WRF is
>         running OK..
>         Very mysterious..
>         
>         BR
>         Goran
>         
>         
>         2008/12/15 Goran Gasparac <gorangas at gmail.com>
>         
>         
>                 Dear all,
>                 
>                 I'm running WRF - ARW with GFS (0.5 degree, 1 week)
>                 with files gfs.t00z.master.grbf* . 
>                 Yesterday (14.12) my WRF run finished forecasting with
>                 a lot's of NaN.
>                 Same happened at friday 05.12.
>                 In 14.12 WRF run, NaN values appeared from 7th term to
>                 end,and in 05.12 WRF run, they appeared 21st term,also
>                 to the end (1 term equals 3 hour period). 
>                 
>                 I suppose it is not because of my WRF configuration
>                 because it is working for about 3 months with same
>                 configuration, and this kind of problem never happened
>                 before.
>                 
>                 Did anybody had a same problem, or have an
>                 idea/suggestion what is wrong?
>                 It cross my mind, that maybe there was bug in GFS
>                 data??
>                 
>                 When ploting data (from 14.12 WRF run) with NCL, i get
>                 an error in terminal:
>                 (0)     Working on time: 2008-12-14_18:00:00
>                 fatal:Subscript out of range, error in subscript #0
>                 fatal:An error occurred reading P
>                 
>                 (2008-12-14_18:00:00 equals 7th term)
>                 
>                 Appreciate any help,
>                 Best regards
>                 Goran
>         
>         
>         
>         -- 
>         Goran
>         
>         _______________________________________________
>         Wrf-users mailing list
>         Wrf-users at ucar.edu
>         http://mailman.ucar.edu/mailman/listinfo/wrf-users
>         
> 
> 
> 
> -- 
> Atentamente
> Jorge Arévalo Bórquez
> _______________________________________________
> Wrf-users mailing list
> Wrf-users at ucar.edu
> http://mailman.ucar.edu/mailman/listinfo/wrf-users
-- 
Andrea Hahmann, PhD.
Senior Scientist
Phone direct +45 4677 5471
andrea.n.hahmann at risoe.dk 

Wind Energy Division
Risø National Laboratory for Sustainable Energy
Technical University of Denmark
Building 118, P.O. Box 49
DK-4000 Roskilde, Denmark
Tel +45 4677 4677
Fax +45 4677 5083