Re: Grib to NetCDF

Hi John,

> I have a question I was hoping you or someone else at Unidata could help 
> with.  We would like to convert data provided by the North America Regional 
> Reanalysis (NARR) program under the National Centers for Environmental 
> Prediction (NCEP), which is natively in Grib, to NetCDF, mainly because we 
> are more comfortable with NetCDF compared to GRIB.  I'm sure people have 
> converted between Grib and NetCDF before, but do you know if anyone has done 
> this specifically for the NARR data?

Sorry, I don't know about the NARR data specifically, but a paper
by Glenn Rutledge 

  http://ams.confex.com/ams/pdfpapers/72204.pdf

claimed the NARR would be available by now

  The new North American Regional Reanalysis (NARR) will become
  available at the NCDC NOMADS servers in December 2003 as NCDC will
  be archiving these data. The NARR is a 32km, 45 layer, EDAS derived
  system, from 1979 thru 2003, updated monthly. Features include a new
  land surface model, direct satellite radiance retrievals, and for
  the first time in any reanalysis, precipitation. It is expected that
  this data set when available in early 2004, will be widely used
  across the weather, climate and oceanographic communities.

but I don't where to get it if it's available.

>                                  ...  Can you point us towards a tool for 
> doing the conversion?  Just browsing the web, I saw that NCL might be a good 
> option.

That's one possibility.  Another is adapting our gribtonc program,
that's part of our decoders package.  It stitches together multiple
GRIB files into a multidimensional netCDF dataset whose structure is
specified by a CDL file, and it's available from

  ftp://ftp.unidata.ucar.edu/pub/decoders/

Some helpful documentation for gribtonc and gribtocdl is here:

  http://noserc.met.no/formkonv/gribtonc.html

Other possibilities are:

 -  The Climate Data Analysis Tool (CDAT)
      http://esg.llnl.gov/cdat/

 -  GrADS, which supports reading and writing data in
    various formats.  See

      http://grads.iges.org/grads/

    for more information.  Arlindo da Silva wrote a script named
    lats4d for this purpose:

      http://www.lasg.ac.cn/atmos/gadoc/scripts/lats4d.gs

--Russ