gen_be_stage2

Issues with running 3DVAR.

Re: gen_be_stage2

Postby hclin » Mon Aug 18, 2008 6:57 pm

What is the output of the following command?
your_LAPACK_directory/INSTALL/testlsame

Does it show
Code: Select all
  ASCII character set
  Tests completed
hclin
 
Posts: 68
Joined: Thu Apr 24, 2008 7:21 pm

Re: gen_be_stage2

Postby beckyselin » Tue Aug 19, 2008 5:41 pm

Yes, it does. Should it not?
beckyselin
 
Posts: 13
Joined: Thu Jul 31, 2008 3:54 pm

Re: gen_be_stage2

Postby hclin » Tue Sep 09, 2008 8:32 pm

With NMC-method, background errors are estimated using forecast differences averaged over a period of time.
The forecast differences should be calculated with forecasts valid at the same time but from different initial times.
For example,
./2007060100/wrfout_d01_2007-06-02_00:00:00 is 24h forecast valid at 2007060200 with 2007060100 initial data.
./2007060112/wrfout_d01_2007-06-02_00:00:00 is 12h forecast valid at 2007060200 with 2007060112 initial data.

For those who have seen failures in gen_be_stage2, and have the following error message in gen_be_stage2.log,
that means the wrf forecasts are not prepared correctly.
hclin wrote:Does your gen_be_stage2.log also show message as below?
hnlim wrote:-------------------------------------------------------------------------------------------------------
[3] Calculate eigenvectors, eigenvalues and inverse for psi/psi covariance
Mode Eigenvalue Cumulative Variance e(k,k)
1 0.0000E+00 NaN 0.0000E+00
2 0.0000E+00 NaN 0.0000E+00
---------------------------------------------------------------------------------------------------------

If so, that means the stage0 was not run correctly (all the forecast difference values are zero).


Compiling LAPACK/BLAS with or without -r8 does not matter.
hclin
 
Posts: 68
Joined: Thu Apr 24, 2008 7:21 pm

Re: gen_be_stage2

Postby kahraman » Mon Mar 30, 2009 5:55 am

lzhang, how did you solve the problem?
I have a similar problem with gen_be. There are NaN values at stage2 log files, and errors related to eigenvalues and eigenvectors at stage 3 logs.
Is a new compilation really needed or not?
kahraman
 
Posts: 6
Joined: Sat Mar 14, 2009 6:34 am

Re: gen_be_stage2

Postby adaypr » Mon Jan 24, 2011 3:26 pm

Hello, I have been in this situation for a few days. Now I have find the solution and I think this problem wasn't reported enough. The problem was that I wasn't using the corrects wrfout files. For example:

export START_DATE=2010120712 # the first perturbation valid date
export END_DATE=2010121000 # the last perturbation valid date
export NUM_LEVELS=28 # e_vert - 1
export BIN_TYPE=5

Also you need the following files in the corrects folders:

2010120612:
wrfout_d01_2010-12-07_12:00:00

2010120700:
wrfout_d01_2010-12-07_12:00:00
wrfout_d01_2010-12-08_00:00:00

2010120712:
wrfout_d01_2010-12-08_00:00:00
wrfout_d01_2010-12-08_12:00:00

2010120800:
wrfout_d01_2010-12-08_12:00:00
wrfout_d01_2010-12-09_00:00:00

2010120812:
wrfout_d01_2010-12-09_00:00:00
wrfout_d01_2010-12-09_12:00:00

010120900:
wrfout_d01_2010-12-09_12:00:00
wrfout_d01_2010-12-00_00:00:00

010120912:
wrfout_d01_2010-12-10_00:00:00
wrfout_d01_2010-12-10_12:00:00

Be careful, the folder's name indicate the start of the simulation and the wrfout files are the outputs on this simulation. In the previous example, you have to do 7 different simulations starting in the correct date. You only need change the start and the end fields in the namelist.input
Best regards,
Aday Perera
adaypr
 
Posts: 14
Joined: Thu Dec 23, 2010 6:56 am

Previous

Return to Runtime Problems

Who is online

Users browsing this forum: No registered users and 2 guests

cron