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

[GEMPAK #RQD-817711]: Using dcgrib2 to decode ECMWF ensemble data



Hi Kevin,

The dcgrib2 manual explains the use of the ENSEXT and PDSEXT extensions for 
grib and grib2 files

 http://www.unidata.ucar.edu/cgi-bin/gempak/manual/decoders_index?dcgrib2 

        ENSEXT : Environmental variable for adding ensemble extensions to 
parameter names
                If ENSEXT is not set, use of PDSEXT parameter name extensions 
defaults 
                to yes for grib1 and no for grib2. If ENSEXT is set, and equal 
to 1, 
                then the extension will be added to parameter names. If ENSEXT 
is
                set, and not equal to 1, the PDS extension will not be added to 
parameter 
                names.

ENSEXT defaults to -1, and anything other than 1 will not append an extension 
to the file.  Set it to 1 and let the LDM run some more for ECMWF and see if 
the parameter name extensions are appended to the files.

Michael
Unidata




> Hi Unidata support,
> 
> I'm attempting to degrib some ECMWF ensemble data for use in gempak,
> in which each file has a specific ensemble perturbation member.
> 
> As it stands now, I'm using dcgrib2, which will take the ECMWF data
> and create a file with all perturbations for the specified times (i.e
> HGHTC000, HGHTP001, HGHTP002, HGHTP003, etc.).
> 
> When trying to isolate each perturbation, I've used  "-e ENSEXT=0" in
> the dcgrib2 command, but it appears to overwrite HGHTC000 with HGHTP001,
> and then HGHTP001 with HGHTP002, and so on until the final gempak file
> only has one perturbation (the last it processed).  Lastly, I've tried
> using the %ens% tag in the filenaming, but that does not appear to work,
> as only one file is created with %ens% (instead of P001 for example)
> in the file name.
> 
> Here's an example command:
> 
> dcgrib2 -e ENSEXT=0 ecmwf_ens_%ens%_YYYYMMDDHHfFFF.gem < T1E10190000102818001
> 
> I've done some extensive research, but have not been able to identify
> the correct syntax for decoding the ECMWF ensemble data.  Do you have
> any suggestions for using dcgrib2 (or NAGRIB) for processing ECMWF data
> with only one perturbation per file, with the specified perturbation in
> the file name?
> 
> P.S. here's an example email response, that got me close:
> 
> http://www.unidata.ucar.edu/support/help/MailArchives/gempak/msg04805.html
> 
> Thanks for your help,
> Kevin Clifford
> 
> -----------------------------------------
> Kevin Clifford - Meteorologist
> PG&E ATS - Meteorology Services
> Phone: (925) 866-5744 | Cell: (415) 606-2154
> Wx Office: (415) 223-3223 | address@hidden<mailto:address@hidden>
> 
> 
> 
> 
> PG&E is committed to protecting our customers' privacy.
> To learn more, please visit http://www.pge.com/about/company/privacy/customer/
> 
> 

Ticket Details
===================
Ticket ID: RQD-817711
Department: Support GEMPAK
Priority: Normal
Status: Open