[Met_help] PREPBUFR data

[John Halley Gotway]

Mike,

To my knowledge, NCEP doesn't make an archive of PrepBufr data readily available.

However, I am aware of some PrepBufr data available via NCAR-CISL.  Try this website: http://search.dss.ucar.edu, and enter "prepbufr" in the search box.

There should be links to 3 datasets.  To access them, you'll need to register (which is free):

(1) NCEP ADP Global Upper Air and Surface (PREPBUFR format) Observations, daily, April 2008 - Continuing (ds337.0)
(2) NCEP ADP Eta/NAM Subset Surface Observations, daily, October 2000 - continuing (ds461.1)
(3) NCEP ADP Global Surface Observations, daily, April 2000 - continuing (ds461.0)

(1) contains GFS prepbufr data, and I tried downloading one and running it though PB2NC.  It worked fine.  The issue is that it's only archived back to 4/2008.
(2) and (3) contain BUFR (not prepbufr) data that doesn't appear to work well with PB2NC.  But these are archived back to 2000.

Please take a look to see if (1) would meet your needs.  If not, I could look into what it would take to get PB2NC to work on the datasets in (2) and (3).

I do know that NCEP does have this PrepBufr data sitting on their machines going back many, many years.  To my knowledge though, it's just not easily, publicly available prior to 4/2008.  However, I
could be wrong.  There may be additional data sources out there that I'm not aware of.

Regarding the timing info, the PrepBufr observations will be grouped into set chunks of data - hourly, 3-hourly, or 6-hourly.  You'll need to interrogate the files to figure out exactly what range of
observations times it contains.  I believe that all of the files from the same PrepBufr source will follow the same convention.  Please note that you can use the "-valid_beg" and "-valid_end" command
line options for PB2NC to explicitly define the range of observation times that you'd like retained in the output NetCDF file.

Hope that helps.

John

Michael D Mcatee wrote:
> I'm looking to use MET to verify WRF model forecasts for the Western US 
> and Eastern Pacific for April of this year.  Do you know of an archive of 
> PREPBUFR data?  Also in testing MET I used NAM PREPBUFR data that I 
> obtained from the NCEP site you link to on the MET webpage.  This data 
> would probably meet my needs but there did not appear to be an archive, 
> only the last couple of days, and its not clear to me for the files I did 
> use what observation times are in a file.   For example for today there 
> are the following files listed among others
> 
> ndas.t00z.prepbufr.tm03.nr 
> ndas.t00z.prepbufr.tm06.nr 
> ndas.t00z.prepbufr.tm09.nr 
> ndas.t00z.prepbufr.tm12.nr 
> 
> I did download and convert some for these files (for 12Z on 20080905)  to 
> NetCDF using pb2nc and dumped  headers/variables using ncdump to find out 
> observation times but is there something in the file name that will tell 
> be this or give me an indication of the range of  observation times in the 
> file?  I assume the 00z means 00Z cycle for the date of the directory but 
> what does the 03, 06,09 etc mean? I would like to verify daily 36 hours 
> forecasts so I'll need to collect files that contain the data I need and 
> label them appropriately. 
> 
> 
> Thanks in advance for the help
> 
> Mike
> 
> ___________________________
> Michael D. McAtee, Ph.D.
> Senior Project Engineer, Meteorological Satellite Systems
> The Aerospace Corporation
> 101 Nelson Drive
> Offutt, AFB, NE 68113-1023
> email:  michael.mcatee at aero.org
> Phone: (402) 292-1017
> Fax: (402) 291-3137
> 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Met_help mailing list
> Met_help at mailman.ucar.edu
> http://mailman.ucar.edu/mailman/listinfo/met_help