[RegCNET] ~~~~~~~~~~~~about regcm model error~~~~~~~~~~~~~~~~

bixq bixq at ictp.it
Mon Jun 29 13:23:47 CEST 2009


Hi, Suman:

On Mon, 29 Jun 2009, Suman Maity wrote:

> i am facing problem at the time of regcm running.i have taken grid
> resolution=30 km.my domain is indian domain( 30E-120Eand 15S-45N).i
> want to run the model from 15th march to 31th may.terrain and icbc
> step is going fine.when i am running the model then it stops after one
> day giving the error "floating exception".i have run the model almost
> thousand times but it is giving the same problem.
Once there is NaN, it's no useful to keep running jobs. You have to go
back

1, back to Run, edit regcm.in, reducing the basic timestep dt;

2, back to Terrain, reconsider your domain setting, and make "the map 
factor fields as close to 1 as possible" by tuning the map projection,
and the two ture latitutes (if you use the LAMCON).

3, for south Asia domain, the NORMER is not a good option if your domain
extended 40N north, you'd better use LAMCON or ROTMER.

Regards,
Xunqiang Bi

> I have also
> identified where the error occurs.it is in tend.f.the model stops at
> the time of pressure tendency calculation.i also think the formula
> used for this calculation have some problem.i have taken radfrq  =
> 30.,abemh   =    18.,abatm   =   120. and dt      =   60..i am
> attaching the model output file from which you can check where is the
> problem.it is on the step ktau=2449.in this regcm.out,so many lines
> are there that is introduced by me for identifying the problem.how i
> can run the model....any type of suggestion will be highly
> solicited....
> it's urgent...............
>
> suman maity
> research scholar,
> CORAL
> IIT,Kharagpore.
> India.
>

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
   Dr. Xunqiang Bi         email:bixq at ictp.it
   Earth System Physics Group
   The Abdus Salam ICTP
   Strada Costiera, 11
   P.O. BOX 586, 34100 Trieste, ITALY
   Tel: +39-040-2240302  Fax: +39-040-2240449
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~



More information about the RegCNET mailing list