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

20001107: 20001106: GEMPAK NAGRIB and NCEP reanalysis on gaussian grid

Werner,

When I say rectangular, I mean that dx and dy stay the same over the grid.
The gaussian dy varies with latitude, so you will not be able to process them.
The point here is that there are no GEMPAK routines to interpolate from the 
gaussian 
grid points to the constant dx & dy gridpoints. The program has no trouble
reading the grib file itself since it is self describing.

If you can obtain the grib data on a CED grid, then you should be able to
decode them.

Steve Chiswell


>From: Werner Wintels <address@hidden>
>Organization: UCAR/Unidata
>Keywords: 200011072004.eA7K4W428590

>
>The NCEP reanalysis Gaussian grid is a rectangular lat-lon grid (192x94). It 
>is, however, unevenly spaced in the Y direction. 
>
>So I am trying another tack. I'm attempting to store the Gaussian grid as a
>CED grid, which is inaccurate, but will do for my data storage purposes 
>(my diagnostic programs read gempak files).
>
>
>
>I'm using the NAGRIB settings: 
>
> GBFILE = grb3d89010100
> INDXFL =
> GDOUTF = 8801.01_10.gaussian.gem
> PROJ = CED
> GRDAREA = -89;0;89;360
> KXKY = 192;94
> MAXGRD = 9999
> CPYFIL =
> GAREA = grid
> OUTPUT = T
> GBTBLS =
> GBDIAG =
>
>
>I still get an error message, although it is encouraging that the decoder 
>seems to reading the grib data:
>
> Changing center table to cntrgrib1.tbl
> Changing vertical coord table to vcrdgrib1.tbl
> Changing WMO parameter table to wmogrib2.tbl
> Changing center parameter table to ncepgrib2.tbl
> Opening the GEMPAK grid file
> GEMPAK grid file is ready... 
>       308 GRIB messages were read or scanned from the GRIB file: 
>     
>grb3d89010100
>         0 grids were written to the GEMPAK file: 
>     
>8801.01_10.gaussian.gem
>
>
>Note that I have created the gempak grid using the appropriate settings i.e.
>
> PROJ = CED
> GRDAREA = -89;0;89;360
> KXKY = 192;94
> MAXGRD = 9999
>
>
>Is the decoder attempting to interpolate the grib data? Is there any hope 
>of simply tricking the decoder into just putting the grib data directly onto 
>a CED grid? 
>
>Thanks again!
>
>
>
>On Mon, 6 Nov 2000, Unidata Support wrote:
>
>> 
>> Werner,
>> 
>> No, the decoder cannot handle the gaussian grid.
>> 
>> Since GEMPAK grid programs only handle rectangular grids, the
>> decoder would first have to interpolate the gaussian points
>> to a known projection before writing them out.
>> 
>> Steve Chiswell
>> Unidata User SUpport
>> 
>> 
>> 
>> >From: Werner Wintels <address@hidden>
>> >Organization: UCAR/Unidata
>> >Keywords: 200011062112.eA6LC5417026
>> 
>> >
>> >I am trying to convert GRIB data (NCEP reanalysis on a Gaussian grid 
>> >from the NCAR mass storage system) to gempak format using NAGRIB in GEMPAK 
>> >version 5.4. 
>> >
>> >The settings are as follows:
>> >
>> >
>> > GBFILE = grb3d89010100
>> > INDXFL =
>> > GDOUTF = 8801.01_10.gaussian.gem
>> > PROJ =
>> > GRDAREA =
>> > KXKY =
>> > MAXGRD =
>> > CPYFIL = GDS
>> > GAREA =
>> > OUTPUT = T
>> > GBTBLS = 
>> > GBDIAG =
>> >
>> >My understanding is that this should read the grib file header in 
>> >grb3d89010100, create a gempak file 8801.01_10.gaussian.gem, create a 
>> >navigation grid corresponding to the grib header, and translate the data 
>> >accordingly. 
>> >
>> >However, I get the following error:
>> >
>> >         0 GRIB messages were read or scanned from the GRIB file: 
>> >     
>> >grb3d89010100
>> >         0 grids were written to the GEMPAK file: 
>> >     
>> >8801.01_10.gaussian.gem
>> >
>> >
>> >Can the decoder handle the gaussian grid?
>> >
>> >
>> >********************************
>> >Werner Wintels 
>> >Atmospheric and Oceanic Sciences
>> >McGill University    
>> >Montreal, Canada
>> >********************************
>> >
>> 
>> ****************************************************************************
>> 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/     
>> ****************************************************************************
>> 
>
>********************************
>Werner Wintels 
>Atmospheric and Oceanic Sciences
>McGill University    
>Montreal, Canada
>********************************
>