RE: Decoders 3.0.2

NOTE: The decoders mailing list is no longer active. The list archives are made available for historical reasons.

On Thu, 25 Mar 2004, HOETH, BRIAN R. (JSC-ZS) (LM) wrote:


Robb,

Good luck on your GRIB2 to netcdf work ... Have you ever thought of
collaborating with FSL since y'all are neighbors there in Boulder and you
both work on similar projects?  FSL is working on GRIB2 to netCDF for AWIPS.
Granted, AWIPS netcdf is slightly different than Unidata netcdf, but it
would still probably help if you collaborated?  Just a suggestion.



Brian,

I have looked at the NWS MD division GRIB2 decoder but I haven't looked at
the FSL one. Do you know who's the contact?  The GRIB2 decoders that I
wrote is in java and it's going to interface with a netcdf package written
in java. The only part to do is to connect the two API's .  At this time,
it shouldn't be hard for "regular" type gribs, it going to be more work
for the ensembles and the stitched together grids.

Robb...


Brian

> -----Original Message-----
> From: Robb Kambic [mailto:rkambic@xxxxxxxxxxxxxxxxxxxxxxxx]
> Sent: Wednesday, March 24, 2004 4:59 PM
> To: HOETH, BRIAN R. (JSC-ZS) (LM)
> Cc: decoders@xxxxxxxxxxxxxxxxxxxxxxxx; decoders
> Subject: RE: Decoders 3.0.2
>
>
> On Wed, 24 Mar 2004, HOETH, BRIAN R. (JSC-ZS) (LM) wrote:
>
> > What are you hearing as far as a target date for the NWS starting
> > regular distribution of GRIB2?  We heard a December of this
> year date,
> > but there's been no real word on whether GRIB will continue to be
> > broadcast at that time?
>
> Brian,
>
> The target date is the end of the year Nov/Dec time frame. I
> believe the NWS will have multiple feeds for awhile, the
> GRIB2 products will probably start on the new digit NOAAport
> feed. I haven't been keeping up with the details of the new
> feed  so I don't know much about it.
>
> Robb...
>
>
>
> >
> > Thanks,
> > Brian
> >
> > > -----Original Message-----
> > > From: Robb Kambic [mailto:rkambic@xxxxxxxxxxxxxxxxxxxxxxxx]
> > > Sent: Wednesday, March 24, 2004 4:07 PM
> > > To: HOETH, BRIAN R. (JSC-ZS) (LM)
> > > Cc: decoders@xxxxxxxxxxxxxxxxxxxxxxxx; decoders; Tom Yoksas
> > > (yoksas@xxxxxxxxxxxxxxxx)
> > > Subject: RE: Decoders 3.0.2
> > >
> > >
> > > On Wed, 24 Mar 2004, HOETH, BRIAN R. (JSC-ZS) (LM) wrote:
> > >
> > > > Robb:  Will the next Decoders release support GRIB2?
> > >
> > >
> > > Brian,
> > >
> > > The current status:  a grib2 (Dump program) has been written in
> > > Java. I haven't been able to work on it for 6 months, the
> plan is to
> > > add the netcdf API on top of the dump program. Hopefully
> I will be
> > > able to restart work on it soon. The NDFD grib2 files
> will be easy
> > > to create but some of the stitched grid models require more work.
> > > For sure a grib2 decoder will be available when the NWS starts
> > > regular distribution in grib2.
> > >
> > > Robb...
> > >
> > >
> > > >
> > > > Tom:  Does Unidata still offer the gribdec.tar.Z package for
> > > > converting GRIB to McIDAS GRID?  I don't see it in the
> downloads
> > > > section
> > > http://my.unidata.ucar.edu/modules.php?name=Downloads&d_op= of
> > > > the my.unidata website?  If the gribdec.tar.Z package IS still
> > > > available, are there plans for it to support GRIB2?
> > > >
> > > > Thanks,
> > > > Brian Hoeth
> > > > Spaceflight Meteorology Group
> > > > Johnson Space Center, Houston, TX
> > > > Office:  281-483-3246
> > > > Operations:  281-483-1051
> > > >
> > > >
> > > > > -----Original Message-----
> > > > > From: owner-decoders@xxxxxxxxxxxxxxxx
> > > > > [mailto:owner-decoders@xxxxxxxxxxxxxxxx] On Behalf Of Robb
> > > > > Kambic
> > > > > Sent: Wednesday, February 04, 2004 3:17 PM
> > > > > To: decoders@xxxxxxxxxxxxxxxx
> > > > > Cc: decoders
> > > > > Subject: Decoders 3.0.2
> > > > >
> > > > >
> > > > > Hiya,
> > > > >
> > > > > I'm pleased to announce the release of Decoders 3.0.2.
> > > > >
> > > > > The major feature of the release is that the data is stored
> > > > > monotonically in the netCDF model files. To accomplish
> > > this, the cdl
> > > > > valtime variable is initialized with the valid times so
> > > the gribtonc
> > > > > program knows where to enter the data into the netCDF
> > > file.  If one
> > > > > gets messages in the log file similar
> > > > > to: "...  new valoffset = 12.000000" the data has extra
> > > offsets that
> > > > > should be added to the cdl. In the cdl, the dimension
> > > valtime_offset
> > > > > should be incremented by one and the new value entered in
> > > the data
> > > > > section of the cdl. Included in the release is a new
> set of cdl
> > > > > files that have the valtime variables already
> initialized to the
> > > > > proper values.
> > > > >
> > > > > The cdl file names have changed to reflect the new model
> > > names, ie
> > > > > avn -> gfs and they also include the grid number of
> the model.
> > > > > Here's the conversion:
> > > > >
> > > > >                        avn-q   ->   gfs_211
> > > > >                        avn-x   ->   gfs_37-44
> > > > >                        ecmwf   ->   ecmwf_1-12
> > > > >                        eta     ->   eta_211
> > > > >                        mrf-a   ->   gfs_21-24
> > > > >                        mrf-e   ->   gfs_25-26
> > > > >                        ngm-q   ->   ngm_211
> > > > >                        ocean   ->   ocean_21-24
> > > > >                        ruc     ->   ruc_211
> > > > >                        ruc2    ->   ruc2_236
> > > > >                        sst-a   ->   sst_21-24
> > > > >                        sst-t   ->   sst_61-64
> > > > >
> > > > >
> > > > > Another variable forecasttime was added similar to
> the datetime
> > > > > variable in previous cdl files. Using units: YYYY-MM-DD
> > > > > hh:mm:ssZ
> > > > >
> > > > > To easy the transition to new pqact entries, the release
> > > includes a
> > > > > file gribtonc.entries with all the needed entries and
> > > it's located
> > > > > in the cdl directory or the etc directory.
> > > > >
> > > > > Because of larger GRID products, the maximum GRID size
> > > was increase
> > > > > from 100k to 200k.
> > > > >
> > > > > The metar2nc decoder was modified to make regular expression
> > > > > matching more robust.
> > > > >
> > > > > As always, if there are any question send them to
> > > > > support@xxxxxxxxxxxxxxxx
> > > > >
> > > > > Thanks,
> > > > > Robb...
> > > > >
> > > > >


==============================================================================
Robb Kambic                                Unidata Program Center
Software Engineer III                      Univ. Corp for Atmospheric Research
rkambic@xxxxxxxxxxxxxxxx                   WWW: http://www.unidata.ucar.edu/
==============================================================================
  • 2004 messages navigation, sorted by:
    1. Thread
    2. Subject
    3. Author
    4. Date
    5. ↑ Table Of Contents
  • Search the decoders archives: