Re: Converting from netCDF to grib 218

• To: Brian Etherton <betherto@xxxxxxxxxxxxxxx>
• Subject: Re: Converting from netCDF to grib 218
• From: Robb Kambic <rkambic@xxxxxxxxxxxxxxxx>
• Date: Fri, 4 Apr 2003 09:20:52 -0700 (MST)

Brian,

Thanks for the information. I've communicated with Jason a time ago, he
sounds knowledgable about grib. I'm wondering if he will update his
converter to the grib2 format? I'll enter this  message into the mail
archives for future reference.

Thanks,
Robb...


On Fri, 4 Apr 2003, Brian Etherton wrote:

Hello there Robb,

It appears that someone has infact made the conversion from a NetCDF file
into a Grib file.  I haven't seen the results yet, but the fellow is Jason
Burks <Jason.Burks@xxxxxxxx>

The SOO at NWS Miami said that Jason took our LAPS generated NetCDF file,
and made a grib file out of it.  So alas - someone has done it!

More when I know of it,
Brian E.

----- Original Message -----
From: "Robb Kambic" <rkambic@xxxxxxxxxxxxxxxx>
To: "Brian Etherton" <betherto@xxxxxxxxxxxxxxx>
Cc: "decoders" <support-decoders@xxxxxxxxxxxxxxxx>
Sent: Tuesday, December 10, 2002 6:11 PM
Subject: Re: Converting from netCDF to grib 218


> Brian,
>
> Actually I don't know of any s/w to create GRIB files from netCDF.  That
> isn't to say that there's not any out there or it can't be done.  One can
> dump a netCDF file to stdout using a netCDF routine called ncdump.  One
> would have to install netCDF and udunits, not a major under taking. Since
> the netCDF output from ncdump is usually large, it would probably be
> better to modify ncdump to write to a grib file. The hard part would be to
> write the grib file.  The Unidata decoders package reads GRIB and writes
> netCDF, so one would have to reverse the routine, probably not a easy
> task. gribtonc reads the GRIB file over a template and then it extracts
> the info. The gribdump routine reads the GRIB and writes the output to
> stdout. Probably the reverse of gribdump would be the better choice. One
> would have to install the decoders package for these routines.  I looked
> on the web, there are GRIB encoding packages available, never used any so
> I can't comment. That's probably a better way to go instead of trying to
> read the ~10k lines code of gribdump.
>
> Some info about GRIB encoding, many more on the web :
> http://rsd-www.nrl.navy.mil/7214/data_fmt.html#grib
>
> You might want to look at the output of the eta model using gribdump, so
> you are aware of the format and parameters.
>
>
> I would appreciate hearing about your findings. I've heard of the
> netCDF-GRIB request a couple of times and never heard of the result.  Also
> you might want to send your problem to decoders@xxxxxxxxxxxxxxxxx mail
> list. You have to join to send the message.
>
> The decoders page has all the needed info.
>
> http://www.unidata.ucar.edu/packages/decoders
>
> Good luck,
> Robb...
>
>
>
> On Tue, 10 Dec 2002, Brian Etherton wrote:
>
> > Hello there Robb,
> >
> > My name is Brian Etherton, post-doc at the University of Miami.  I am
working in collaboration (COMET proposal) with the NWS to initialize the
workstation ETA model using the LAPS analysis produced at NWS Miami.  At
present, the LAPS output is in netCDF form.  The ETA model, as near as I
know, is initialized using the grib 218 format.
> >
> > So, my task is to convert from one to the other, and to have the correct
variables in the grib file (which I reckon may not be the same as the LAPS
analyses, certainly, the vertical coordinate will differ).
> >
> > I am kind of new to this whole game, so any insights you have would be
great.  There seems to be lots of ways to go from grib to netCDF, but as of
yet, I haven't found a way to do the reverse.  My guess is that I will have
to code this all up myself, which I can do if given a pretty clear road map
of how to go about doing it.
> >
> > Thanks for any and all help,
> > Brian Etherton
>

==============================================================================
Robb Kambic                                Unidata Program Center
Software Engineer III                      Univ. Corp for Atmospheric Research
rkambic@xxxxxxxxxxxxxxxx                   WWW: http://www.unidata.ucar.edu/
==============================================================================