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

Re: source code gribinsert

Sure.

I'll package up some instructions and the code for you.

Steve Chiswell
Unidata User Support


On Fri, 2007-02-02 at 15:25 -0500, Toviessi,Joseph-Pierre [CMC] wrote:
> Hi Steve
> 
> Would it be possible for us to get the source code for the program 
> "gribinsert" so that we can use it to break apart our grib2 files into 
> individual records ?
> 
> Thanks
> 
> Joseph-Pierre Toviessi
> 
> Centre Météorologique Canadien
> 2121 route Transcanadienne, Dorval Québec H9P 1J3
> 
> 
> -----Message d'origine-----
> De : Justin Cooke [mailto:address@hidden] 
> Envoyé : 01 February, 2007 17:00
> À : Poulin,Lewis [CMC]
> Cc : Toviessi,Joseph-Pierre [CMC]; Steve Chiswell; Joe Carr
> Objet : Re: Request for TIGGE contact person
> 
> Lewis,
> 
> We use a program called "gribinsert" that is provided to us by UCAR's 
> Steve Chiswell (address@hidden). Gribinsert breaks apart the 
> pgrb2a and pgrb2c files into individual parameters and injects each one 
> into our LDM queue. ECMWF is an allowed subscriber to that LDM and they 
> just key off of the parameters that they need.
> 
> I will check with Joe about the document you attached, I'm personally 
> not familiar enough with the GRIB2 and GRIB1 differences in our GEFS 
> datasets to give any clarification.
> 
> Justin
> 
> Poulin,Lewis [CMC] wrote:
> > Thanks Justin. 
> >
> > I was able to download your TIGGE samples including the 
> > /pub/data1/nccf/com/tigge files referred to below. We'll have a look at 
> > them.
> >
> > When I did a quick first read of the TIGGE transfer protocol, I was under 
> > the impression each variable had to be sent in its own stand alone file? 
> > Does NCEP send files like those in pgrb2a to tigge? Or do you have another 
> > process that extracts the records from the pgrb2a files and sends each 
> > record individally to tigge?
> >
> > Finally, since you refer to the pgrb2a files, I attach an email I sent to 
> > Joe Carr and Dick Wobus yesterday about a clarification for documentation. 
> > Not sure if you can offer any insights on this. If you have a 
> > clarification, you may want to include Joe and Dick in your reply. Thanks.
> >
> > Thanks
> >
> > Lewis
> >
> > -----Original Message-----
> > From: Justin Cooke [mailto:address@hidden] 
> > Sent: 29 January, 2007 16:31
> > To: Poulin,Lewis [CMC]
> > Cc: Toviessi,Joseph-Pierre [CMC]
> > Subject: Re: Request for TIGGE contact person
> >
> > Hi Lewis,
> >
> > We send two types of ensemble files to TIGGE via LDM, which together 
> > represent all of the data that is expected.
> >
> > One type is available to the public on our anonymous FTP server
> > ftpprd.ncep.noaa.gov:
> >
> > /pub/data/nccf/com/gens/prod/gefs.YYYYMMDD/CC/pgrb2a
> >
> > The files are named with the member number (MM) at the beginning, cycle in 
> > the middle (CC), and valid forecast hour at the end (HHH).
> >
> > gepMM.tCCz.pgrb2afHHH
> >
> > The second set is currently not publicly available but I've put a set of 
> > all members for today's (01/29) 00Z run valid for forecast hour 12.
> >
> > /pub/data1/nccf/com/tigge
> >
> > The files are similarly named, but instead of an "a" there is a "c" 
> > before the valid forecast hour:
> >
> > gepMM.tCCz.pgrb2cfHH
> >
> > Let me know if you have any problems pulling the files,
> >
> > Justin
> >
> > Poulin,Lewis [CMC] wrote:
> >   
> >> Hi Justin..
> >>
> >> Joe Carr suggested I can contact you for information on TIGGE data 
> >> transfers.
> >>
> >> We are also working at setting up a data transfer from the Canadian 
> >> Meteorological Center (CMC) to TIGGE archives.
> >>
> >> I was curious to see if there was a way we could obtain one timestep of 
> >> sample NCEP files that you send to TIGGE? You could just put them on an 
> >> ftp site somehwere and we could pick them up?
> >>
> >> This will allow us to continue testing our evolving GRIB2 encoder/decoder 
> >> and give us a reference point for the preparation of our files. 
> >>
> >> We are also using info from the TIGGE project web page to help in the set 
> >> up of our system.
> >>
> >> Thanks
> >>
> >> Lewis
> >>
> >> Lewis Poulin
> >> Centre Météorologique Canadien
> >> 2121 route Transcanadienne, Dorval Québec H9P 1J3   
> >> address@hidden
> >> Tel:(514) 421-7266   Fax/Télécopieur:(514) 421-4679
> >>
> >>
> >> -----Original Message-----
> >> From: Joe Carr [mailto:address@hidden]
> >> Sent: 22 January, 2007 12:55
> >> To: Poulin,Lewis [CMC]
> >> Cc: Justin Cooke; Brent Gordon
> >> Subject: Re: Request for TIGGE contact person
> >>
> >> Lewis,
> >>
> >> Please work with Justin on this.
> >>
> >> Thanks,
> >> Joey
> >>
> >> Poulin,Lewis [CMC] wrote:
> >>   
> >>     
> >>> Hi Joe
> >>>
> >>> We are setting up a mecanism to transfer some of our data to the 
> >>> TIGGE project.
> >>>
> >>> A colleague who works in data distribution is looking for someone 
> >>> with experience sending files to TIGGE using the LDM.
> >>>
> >>> Since NCEP is already sending TIGGE data, would there be someone at 
> >>> NCEP that my colleague can touch base with to discuss a few technical 
> >>> issues with ?
> >>>
> >>> Thanks for any leads you can offer.
> >>>
> >>> Regards
> >>>
> >>> Lewis Poulin
> >>> Centre Météorologique Canadien
> >>> 2121 route Transcanadienne, Dorval Québec H9P 1J3 
> >>> address@hidden
> >>> Tel:(514) 421-7266   Fax/Télécopieur:(514) 421-4679
> >>>
> >>>  
> >>>
> >>>     
> >>>       
> >>   
> >>     
> >
> >
> >   
> >
> > ------------------------------------------------------------------------
> >
> > Subject:
> > Documentation for NCEP pgrb2a file content
> > From:
> > "Poulin,Lewis [CMC]" <address@hidden>
> > Date:
> > Mon, 29 Jan 2007 14:53:15 -0500
> > To:
> > "Joe Carr" <address@hidden>
> >
> > To:
> > "Joe Carr" <address@hidden>
> > CC:
> > "Poulin,Lewis [CMC]" <address@hidden>, "Toviessi,Joseph-Pierre 
> > [CMC]" <address@hidden>, "Richard Wobus" 
> > <address@hidden>
> >
> >
> > Hi Joe
> >
> > A quick GRIB2 question concerning NCEP GRIB2 data. 
> >
> > We are testing our decoding of NCEP GRIB2 data of the type 
> > gepNN.tccz.pgrb2a252 found at: 
> >      
> > ftp://ftpprd.ncep.noaa.gov/pub/data/nccf/com/gens/prod/gefs.20070129/00/pgrb2a/
> >
> > While reading the documentation at, for example:
> >   http://www.nco.ncep.noaa.gov/pmb/products/gens/gepNN.tccz.pgrbaf252.shtml
> >
> > We note the pgrbafhhh files should contain 51 records for non zero hours. 
> > This is indeed the case for the few grib1 files I tested found in 
> > ftp://ftpprd.ncep.noaa.gov/pub/data/nccf/com/gens/prod/gefs.20070129/00/pgrba/
> >
> >
> > However, when processing pgrb2a files of the type 
> > (../pgr2ba/gepNN.tccz.pgrb2ahhh)
> > Taken from      
> > ftp://ftpprd.ncep.noaa.gov/pub/data/nccf/com/gens/prod/gefs.20070129/00/pgrb2a/
> > It appears those grib2 files contain only 43 records and not 51 records 
> > like their grib1 counterparts. 
> >
> > Is this correct?
> >
> > Do you have another documentation page for the content of grib2 datasets if 
> > they differ from the grib1 datasets?
> >
> > At this stage we are trying to focus on an NCEP GRIB2 dataset (pgrb2a?) 
> > that has all the NAEFS variables so that we can get those working before 
> > expanding to a broader range of variables.
> >
> > Thanks
> >
> > Lewis
> >
> > Lewis Poulin
> > Centre Météorologique Canadien
> > 2121 route Transcanadienne, Dorval Québec H9P 1J3   
> > address@hidden
> > Tel:(514) 421-7266   Fax/Télécopieur:(514) 421-4679
> >
> >   
> 
> 
-- 
Steve Chiswell <address@hidden>
Unidata