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

[CONDUIT #CQQ-962463]: CONDUIT GRIB-2's for GFS 0.5 deg



> Thanks Steve - this makes sense now that I look closer at it.
> I was looking at these files in a HEX editor and noticed there were some
> large blocks of text in them.
> I'm assuming that is normal as well.
> Kevin

Kevin,

If you are looking at files you FTP'd from ftp://tgftp.nws.noaa.gov, then
you shouldn't see large blocks of text.

If instead you are referring to files you are storing off of CONDUIT itself,
then there is a file that begins with ".status" in the data stream that is the 
inventory
of all the grib products sent for that forecast hour/product as shown here:
http://www.unidata.ucar.edu/data/conduit/ldm_idd/

If you are seeing this block of text, then you should modify your FILE action 
in pqact so that you 
don't store this text inventory with the GRIB data (its something I add to the 
datastream for QC, and 
not part of the original grib data).

You can use the pattern such as:
^.status.*ST.opnt.....
to only match the inventory
and
^/afs.*ST.opnt....
to only match the grib data itself.

Steve Chiswell
Unidata User Support



> 
> Unidata CONDUIT Support wrote:
> 
> >>Hi,
> >>I was examining the CONDUIT grib-2 files for the GFS 0.5 deg model found at
> >>ftp://tgftp.nws.noaa.gov/SL.us008001/ST.opnt/MT.gfs_CY.12/RD.20060720/PT.grid_DF.gr2/
> >>Specifically, I downloaded the file fh.0054_tl.press_gr.0p5deg_.
> >>_But within the contents of the file I found that there were also
> >>48-hour forecasts contained in this file along with the 54-hour
> >>forecasts (a little over 5 M of the total 38 M file)
> >>Is this something that is known?
> >>Thanks!
> >>Kevin
> >>
> >>
> >>
> >>
> >
> >Kevin,
> >
> >I decoded the fh.0054_tl.press_gr.0p5deg file for July 20, and in fact, all 
> >grids are F054
> >as decoded in GEMPAK (gb2ftim.c routine).
> >
> >I suspect your confusion arises due to the the way the time is presented in
> >Product Definition Template 4.8 in GRIB2 for accumulations (Precipitation, 
> >max temp etc for example).
> >PDT4.8 Provides a "End time" of the overall period (which is the 54 hour 
> >forecast time verification).
> >The "48" you will find in in the forecast time would be the time at which 
> >the accumulation, average or
> >extreme period begins since the start time of the model. The 48 by itself 
> >outside of the context
> >of the product definition template which is necessary for decoding is 
> >amorphous.
> >
> >As in the case of precipitation accumulation, the value is accumulated over 
> >the 6 hours, and that
> >value is the accumulation at the end of the period.
> >
> >I'm glad that you are looking in to this data set, as we are about to extend 
> >the period delivered to
> >CONDUIT through 180 hours (using the ftpprd ncep files rather than the tgftp 
> >file names though).
> >
> >Steve Chiswell
> >Unidata User Support
> >
> >Ticket Details
> >===================
> >Ticket ID: CQQ-962463
> >Department: Support CONDUIT
> >Priority: Normal
> >Status: Closed
> >
> >
> >
> 
> 


Ticket Details
===================
Ticket ID: CQQ-962463
Department: Support CONDUIT
Priority: Normal
Status: Open