[Met_help] MET FEEDBACK from Bob Pasken

[John Halley Gotway]

Bob,

The issue with reading the output of WRF directly isn't so much with data types.  Instead it's with handling the staggered grid and vertical coordinates that WRF uses.  MET expects data to be on a
standard de-staggered grid with pressure levels in the vertical.  We chose to read GRIB data since that's the output of WPP - which places the WRF output on a destaggered grid using pressure levels in
the vertical.

However I'm sorry to hear that you're having troubles running WPP.  I do still believe it's your best shot for preparing you data for use in MET.

Here are a couple of places you might look for information on the WPP issues you're seeing:

WRF User's Forum:
Here's a list of issues related to floating point exceptions in WPP...
http://forum.wrfforum.com/search.php?keywords=floating+point&terms=all&author=&fid[]=53&sc=1&sf=all&sr=posts&sk=t&sd=d&st=0&ch=300&t=0&submit=Search
Could you modify your "wrf_cntrl.parm" file to remove the variable that's causing the floating point exception?

WRF-NMM User's Page:
Here's a list of release notes for each release of WPP...
http://www.dtcenter.org/wrf-nmm/users/model/graphics/WPP/updates.php

Also you could try emailing wrfhelp for support.  However, since you're using an older version of WRF, I'm guessing the first thing they'd suggest would be to try updating to the WRF version 3.1
released on April 9, 2009.  It's possible that the issue your seeing may have been fixed since version 2.2.

Hope that helps.  If you are able to prepare your data for use in MET, please feel free to write met_help at ucar.edu with questions or suggestions.

Thanks,
John Halley Gotway
johnhg at ucar.edu

MET.FEEDBACK at rap.ucar.edu wrote:
> First Name = Bob
> Last Name = Pasken
> Institution = Saint Louis University
> Email = rpasken at eas.slu.edu
> MET Feedback = I have successfully compiled METv2.0 on our OpenSolaris machines. The problem now lay with trying to get WRFv2.2 data into MET. I have tried to use WPP to convert the netcdf WRF output into the grib format needed by MET. Even when using netcdf files produced on a linux based system and running WPP on the same linux system WPP fails with floating point errors. I am becoming frustrated with the MET requirement for grib formatted input files. Are there any recommendations for converting netcdf to grib other than WPP or is there beta versions of a version MET that accepts netcdf?