19990929: DCGRIB

>From: Brett Newkirk <address@hidden>
>Organization: .
>Keywords: 199909291659.KAA21002

>Hi Steve:
>       Thanks for the reply about tiled grids. I have some questions
>about dcgrib: Is there a particular order in which the tiled grids must be
>decoded? Also can dcgrib handle more than one file at a time?
>        I have two tiles of ECMWF data that meet together over the
>Pacific Ocean, all in GRIB format and each file corresponds to one
>variable at one level for one tile. Is it possible to put all the
>tiles/variables together and put then into one GEMPAK file? Thanks for
>your help and your time-
>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
>


Brett,

The order that the tiles is not important since that data is written to the
file for each sector, and read in again when the next tile arrives so that the
new tile can be added to the data block and written back out. The projection
created is that of the finished global grid, and not for any one tile.

Dcgrib knows about the complete grid information for the ECMWF, AVN and MRF data
on the idd, so that it can generate the full grid projection even though
the grib block only defines a single tile.

Here are the pqact.conf entries for the ECMWF data on the IDD:
HRS     ^H[HPT][A-L][A-Z][0-9][0-9] ECMF ([0-3][0-9])([0-2][0-9])
        PIPE    /usr/local/ldm/decoders/dcgrib -d data/gempak/logs/dcgrib.log
                -g /home/gempak/NAWIPS-5.4/gempak5.4/tables
                PACK data/gempak/hds/YYYYMMDDHH_ecmf1.gem
HRS     ^H[UV][A-L][A-Z][0-9][0-9] ECMF ([0-3][0-9])([0-2][0-9])
        PIPE    /usr/local/ldm/decoders/dcgrib -d data/gempak/logs/dcgrib.log
                -g /home/gempak/NAWIPS-5.4/gempak5.4/tables
                PACK data/gempak/hds/YYYYMMDDHH_ecmf2.gem

Steve Chiswell