[Wrf-users] converting nmm outputs to grads

[Steve]

Elif --

Check the total size of your concatenated grib file.  If it is over 2 GB 
(2**31-1 or 2147483648 bytes), grads and the associated utilities will 
probably not work correctly on the file.  I have had similar problems on 
my linux system with some large-grid WRF grib files.

As an alternative to concatenating the files, look at the grads "options 
template" directive for the grads control file.  Grads can access a set 
of files with a time-based naming convention as if they were one large 
file.  You modify the dset and options lines on the ctl file, then run 
gribmap.  Be sure to look at the verbose gribmap log to verify that it 
is processing the data correctly.  You can then open the ctl file in 
grads and view all the files.  You can find more info templates in the 
grads documentation at http://grads.iges.org/grads/gadoc/templates.html.


Steve Masters  smasters at cfl.rr.com
Melbourne, FL

> Message: 1
> Date: Sat, 19 May 2007 20:43:38 -0400 (EDT)
> From: "Elif SERTEL" <esertel at envsci.rutgers.edu>
> Subject: [Wrf-users] converting nmm outputs to grads
> To: wrf-users at ucar.edu
> Message-ID:
> 	<4637.172.23.42.253.1179621818.squirrel at webmail.envsci.rutgers.edu>
> Content-Type: text/plain;charset=iso-8859-1
> 
> 
> Dear Wrf users,
> 
> I have a problem while converting model outputs into grads format.
> I have my NMM outputs namely wrfnmm00.tm00, wrfnmm06.tm00 ....
> wrfnmm480.tm00. This data belongs to 1-20 August 2004, and it has 6 hr
> interval.
> 
> I used cat command and make a big grib file from my 60 model outputs.
> 
>       ....cat wrfnmm*.tm00 case5.grb
> 
> Then I used grib2ctl function to obtain .ctl file of my grib data.
> 
>       .....grib2ctl.pl -verf case5.grb>case5.ctl
> 
> I checked the .ctl file but I got 2 hour intervals between output data
> instead of 6 hours. I changed that value to 6 hours.
> 
>      .....tdef 127 linear 00Z01aug2004 6 hr
> 
> Then, I used gribmap command and have .idx file.
> 
>      .....gribmap -i case5.ctl
> 
> When I tried to plot my results in GRADS, I found out that I did not have
> any data after t=44 (2004-08-11-18:00)(no data between t=44 and t=60).
> 
> To be sure about if the model output has any problem, I looked each
> wrfnmm*.tm00 separately, but they seem OK.
> 
> I guess there is a problem during the conversion procedure. I need all of
> my outputs in one file, otherwise it is hard to plot them. What should I
> do?
> 
> Best wishes,
> Elif.
> 
>