Folks-
This is a tad specialized, but perhaps there is someone who has already
dealt with the problem. We are attempting to read bufr files generated
from the North American Regional reanalysis (NARR) and get them into
gempak format.
I can read bufr files from the NAM(Eta) using namsnd.
However, when I attempt to read similar bufr files from the NARR, I've
been getting a succession of errors to work through related to
differences in the packing file and slight differences in the bufr
files. The only difference according to the bufr_table.dump files are:
1) STNM is 17 bits in the NARR and 20 bits in the NAM
2) NAM has an extra variable in the HEADR: 64 bit RPID
3) NARR does not have the IMXR variable in the sounding data
4) Plus a bunch of differences in the surface grids that I can defer for
now.
If I hack a bit on the packing file, I can get a gempak file to be
created, but then get an error message related to a wrong latitude,
which is suspicious since then CLAT variable follows immediately after
the STNM, RPID combo. Various attempts gets me to an invalid dimension
size error as well.
Does this info elicit any ideas as to what may be haywire? I can provide
more details if someone is interested.
Thanks
John
_____________________
John Horel
Director, NOAA Cooperative Institute for Regional Prediction
Professor, Department of Meteorology
Office: 486 INSCC
Mail: 135 South 1460 East Rm 819
Salt Lake City 84112
Phone: (801) 581-7091
Fax: (801) 581-4362
Email: jhorel@xxxxxxxxxxxx
