Page 1 of 1

Change in GFS and GDAS 1/2 degree files

Posted: July 26th, 2017, 9:16 am
by barbara.stunder
Coinciding with NCEP's upgrade to the global forecast model (GFS) last week:

Archive files at: ftp://arlftp.arlhq.noaa.gov/archives/gdas0p5/ (named <date>_gdas0p5)

Forecast files at:
http://nomads.ncep.noaa.gov/pub/data/nc ... plit.<date>
ftp://arlftp.arlhq.noaa.gov/forecast/<date>
named hysplit.tCCz.gfs0p5f, where CC=00, 06, 12, or 18

Effective beginning with the archive file 20170719_gdas0p5,
(1) there is no vertical velocity, and
(2) the data are short-term GFS forecasts instead of GDAS
because NCEP is no longer outputting vertical velocity to the files which are used to create these hybrid-level global files in ARL/HYSPLIT-format and because ARL is no longer processing NCEP GDAS files.

For these *_gdas0p5 and *.gfs0p5f files, HYSPLIT will automatically compute vertical velocity from the divergence when the vertical motion is set to "data" (0), and the following message will be in standard output and the WARNING file:
*WARNING*: metini - vertical velocity missing, switching to divergence

There is no change in the one-degree pressure-level files:
GDAS archive - ftp://arlftp.arlhq.noaa.gov/archives/gdas1/
GFS forecasts - same URLs as for half-degree, files named hysplit.tCCz.gfsf and hysplit.tCCz.gfsa

If there are questions, please post to the Forum section "General questions on meteorological data needed for HYSPLIT" viewforum.php?f=25