[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

19991013: 19991011: ECMWF tiles

Brett,

I'm glad that you are able to use your data now as you had wished.
The dcgrib patch is completely interchangeable with the previous version.
Grid 255 will still use the gds to define the grid, and will
only create the projection defined in the gds- except if
it happens to be the grid 255 from the ECMWF, and the gds
defines a CED projection, and the number of grid points
for the tile is 60x61. If those three conditions exist,
then it will create the global CED grid and use the grids
as tiles, otherwise everything falls through.

The only down side is that at this point, you have no option to
just decode your tiles into 60x61 grids. They will always create
a 240x61 grid, even if you don't have any other tiles. That
doesn't prevent you from using the data, but it is a limitation
that at some point I will have to provide a way for the user to
specify if a larger grid section should be used- or just
the individual tile.

At any rate, if you chose to install the new version in the gempak
source tree, just replace the entire contents of the dcgrib
directory with the new source, and make sure there are no other
object (.o) files in that directory from a previous build
that could have used the old array sizes defined in geminc.inc.


Steve Chiswell
Unidata User Support







>From: Brett Newkirk <address@hidden>
>Organization: .
>Keywords: 199910131627.KAA11578

>Hi Steve:
>       I tried the patch for dcgrib that you had made and it does
>the job very well. I made my own dcgrib executable and tried it that way
>by just copying the entire ldmbridge directory over. I just wanted to ask,
>is this patch ok for the entire system (it wont do anything different for
>the grid 255-use gds grid)-or is it better if I just run my own version
>offline so no one gets an unexpected change in how dcgrib works. Thanks
>very much for what you have done- I greatly appreciate it.
>Brett
>
>Brett Newkirk    E-MAIL: address@hidden
>Office: ATG 424  Atmospheric Science/Geophysics Building
>Mailing Address: Department of Atmospheric Sciences, University of Washington
>                 Box 351640
>                 Seattle, WA 98195-1640
>Office Phone: (206) 685-2183
>
>On Mon, 11 Oct 1999, Unidata Support wrote:
>
>> 
>> Brett,
>> 
>> I have placed a tar file of dcgrib in the gbuddy ftp account under
>> nawips-5.4/patches/dcgrib_991011.tar.Z
>> 
>> This version of dcgrib will create a northern hemeisphere CED grid
>> which can be comprised of 4 tiles. Your tiles C and D will be
>> written to the appropriate row offsets in the grid.
>> 
>> I have tested the decoder on the 2 gribs you sent me and compared
>> them against the grid that NAGRIB produces as well.
>> 
>> Since the two UWND grids you sent me were for different times (eg C = f000
>> and D=f012), I did not have a single image to look at, but the
>> data decoded into the individual tiles does correctly overlay with the
>> individual files created by NAGRIB.
>> 
>> Please let me know if this accomplishes what needs to be done.
>> 
>> To install,
>> download into $NAWIPS and unpack with:
>> zcat dcgrib_991011.tar.Z | tar xvf -
>> 
>> Then build with:
>> cd $NAWIPS/unidata/ldmbridge/dcgrib
>> make clean
>> make all
>> make install
>> make clean
>> 
>> 
>> It is very important to run the "make clean" since I have changes the
>> sizes of data arrays in geminc.inc, so you don't want any old objects around
> .
>> 
>> Steve Chiswell
>> Unidata User Support
>> 
>> >From address@hidden  Tue Oct  5 10:17:31 1999
>> Received: from drizzle.atmos.washington.edu (drizzle.atmos.washington.edu [1
> 28.95.175.15])
>>      by unidata.ucar.edu (8.8.8/8.8.8) with ESMTP id KAA29325
>>      for <address@hidden>; Tue, 5 Oct 1999 10:17:31 -0600 (MDT)
>> Organization: .
>> Keywords: 199910051617.KAA29325
>> Received: from localhost (bnewkirk@localhost)
>>      by drizzle.atmos.washington.edu (8.9.1a/8.9.1) with ESMTP id JAA12916
>>      for <address@hidden>; Tue, 5 Oct 1999 09:17:30 -0700 (PDT)
>> Date: Tue, 5 Oct 1999 09:17:30 -0700 (PDT)
>> From: Brett Newkirk <address@hidden>
>> To: Unidata Support <address@hidden>
>> Subject: Re: 19990929: ECMWF data 
>> In-Reply-To: <address@hidden>
>> Message-ID: <Pine.SOL.4.10.9910050916100.12854-100000@drizzle.atmos.washingt
> on.edu>
>> MIME-Version: 1.0
>> Content-Type: TEXT/PLAIN; charset=US-ASCII
>> 
>> Steve:
>>      Sorry about that- I have placed the two grib files C_TILE.grb and
>> D_TILE.grb in the ~gbuddy/incoming directory. If there is anything I can
>> do here-please let me know-Thanks very much-
>> Brett
>> 
>> 
>> Brett Newkirk    E-MAIL: address@hidden
>> Office: ATG 424  Atmospheric Science/Geophysics Building
>> Mailing Address: Department of Atmospheric Sciences, University of Washingto
> n
>>                  Box 351640
>>                  Seattle, WA 98195-1640
>> Office Phone: (206) 685-2183
>> 
>> 
>> 
>> ****************************************************************************
>> Unidata User Support                                    UCAR Unidata Program
>> (303)497-8644                                                  P.O. Box 3000
>> address@hidden                                   Boulder, CO 80307
>> ----------------------------------------------------------------------------
>> Unidata WWW Service                        http://www.unidata.ucar.edu/     
>> ****************************************************************************
>> 
>