The NCEP reanalysis grib data that you can get from NCAR already comes in
synoptic format (all variables together at one time). It's just that the
reanalysis data provided by the NOAA
Climate Diagnostic Center (CDC) has been
repackaged by variable to simplify climate studies. The CDC also get their
data from NCAR.
The NCAR Archive is
available in GRIB format and as a university you can join and get access,
otherwise it appears to cost about $30 per gigabyte. You would need to
access the 6-hrly analysis archive (2.5 deg grid, 17 pressure levels) which
is called "ds090.0". The data files would have a name starting with pgb.f00
followed by YYMMDDHH. Each grib file contains all the variables required by
Hysplit and should be readable by the ARL converter without modification. Their
on-line directory shows which tar volume contains the files you need. For
instance if you are interested in October of 1983, the pgb files can be
found on volume A04814. You may want to try to get a sample file from them
for testing (pgb.f0071123118) from the NCAR ftp site.
The syntax of the Hysplit reanalysis data grib decoder would be ncr2arl
[file] [lat] [lon], where file is the name of the grib file, lat and lon are
the center of a 100x100 grid that is extracted from the grib file. The
output goes to a file called Data.ncr. You can use the utility programs
display.exe and profile.exe in the \metdata directory to view the data.ncr
file. The program processes only one time period. To run Hysplit you will
need to put multiple time periods into one file. You can use the copy, cat,
or type commands to append multiple files into one: e.g. "type file2 >>file1".
The best way to process multiple files is to write a batch file
or script to handle the process.
Roland Draxler