Greetings. After much trial and error over the past few weeks and with
guidance from NCEP (thanks to Brill, Klein, Ebisuzaki, and Iredell) we
were able to develop a method using available tools to pull in the UKMET
grib files, process them with copygb and wgrib, and then pump them through
nagrib to get gempak grids.
Recall that these files can not directly be nagribbed properly (from
ftp://tgftp.nws.noaa.gov/SL.us008001/ST.opnl/MT.bmoav_CY.CC/RD.YYYYMMDD/PT.grid_DF.bb
). This is because of the following:
These are thinned gribs (number of x points decreases as you head toward
the poles)
For a given fhr and field and level (like f12 height 500mb) you get EIGHT
tiles with the SAME name in the SAME grib file.
This implies one must take extra steps for a given fhr, field, and level
to figure out how to recognize the differences between each tile, stitch
them together, and create a contiguous "non thinned" grib field so NAGRIB
can chew on it.
Fun.
So the approach we took to fully automate this process was the following:
Loop through each fhr -- pull data from ftp site
Use wgrib (not wgrib2) to detect what parameters and levels exist in the
pulled grib file
For each parameter/field at the given fhr use wgrib to recognize which
records contain the tile over the globe you wish to stitch
Use copygb to convert and merge (stitch) the thinned grib tiles into an
NCEP grid of your choice
nagrib into a gempak grid
Soak, rinse and repeat
copygb comes with WRF EMS but if you don't have it copygb is available at
http://www.nco.ncep.noaa.gov/pmb/codes/nwprod/util/sorc/copybg.fd/ (be
sure to read the documentation)
We recommend you manually try out the process and test it gives you what
you think it should before writing a script to automate.
But the pieces are here and it does work.
Pete
From: pmanousos@xxxxxxxxxxxxxxxxxxx
To: gembud@xxxxxxxxxxxxxxxx
Date: 09/25/2013 01:34 PM
Subject: [gembud] Degribbing Ukmet Grib1 from tgftp site
Sent by: gembud-bounces@xxxxxxxxxxxxxxxx
Greetings. We are receiving an NOBNDS - No grid points in graphics area
[DG -40] Subset grid bounds cannot be computed message from gdplot3 on
nagribbed ukmet files retrieved from the tgftp server.
We suspect it has something to do with the grib files and know just enough
about grib to get by, but not enough to address this issue properly. It
seems like some key piece of info in the grib message is missing causing
GEMPAK to fail.
We have done searching on google, gembud, and experimented unsuccessfully
with cnvgrb, dcgrib and ultimately get the same null result.
If anyone is successfully converting ukmet grib files from the tgftp
server into GEMPAK grids (that are plottable in gdplot3) we would
appreciate any advice.
For reference here is how we get from the grib file to grid file:
Pull
ftp://tgftp.nws.noaa.gov/SL.us008001/ST.opnl/MT.bmoav_CY.CC/RD.YYYYMMDD/PT.grid_DF.bb/fh.0012_t1.hight
(where CC is the current cycle and YYYYMMDD the present date)
Degrib using nagrib (output file is test.grd -- using GDS in CPYFIL and
maxgrids=4000)
gdinfo on test.grd shows what you would expect (proper date/times, levels,
gvcords, gridnames)
Plotting in gdplot3 yields the NOBNDS error
Pete ----------------------------------------- The information contained
in this message is intended only for the personal and confidential use of
the recipient(s) named above. If the reader of this message is not the
intended recipient or an agent responsible for delivering it to the
intended recipient, you are hereby notified that you have received this
document in error and that any review, dissemination, distribution, or
copying of this message is strictly prohibited. If you have received this
communication in error, please notify us immediately, and delete the
original message._______________________________________________
gembud mailing list
gembud@xxxxxxxxxxxxxxxx
For list information or to unsubscribe, visit:
http://www.unidata.ucar.edu/mailing_lists/
