Reading GFS GRIB2 files: issues with gribmap

Daniele Gandini Daniele.Gandini at ARPA.PIEMONTE.IT
Tue Jul 22 03:58:36 EDT 2008


I think you have to separate the original file in two ones,
the first from 0 to 180hr and the second from 192 to 384 hr.

You can manually change time increment in ctl file produced
by g2ctl if isn't ok and then you can run gribmap.

Bye

Daniel


----- Original Message -----
From: "Gilmour, Isla" <Isla.Gilmour at DESHAW.COM>
To: <GRADSUSR at LIST.CINECA.IT>
Sent: Tuesday, July 22, 2008 9:33 AM
Subject: Reading GFS GRIB2 files: issues with gribmap


Using GrADS (v2.0.a2), wgrib2 (v1.7.7c) , g2ctl (v0.0.3h) and with perl
installed, on UNIX (SunOS i386)

Having downloaded grib2 GFS files of the form gfs.tccz.pgrb2.fxxx.grib2
from NCEP
(as described at http://www.nco.ncep.noaa.gov/pmb/products/gfs/ )
I created a ctl file using g2ctl.  This looks relatively ok (see end of
email for header), other than the tdef which is listed as 1mo rather
than something more reasonable like 3h or 12h (the forecast leadtime
varies from 3hrly 0-180h, then 12hrly out to 384h) so that's the second
question... can GrADS ctl files deal with non-uniform time steps?

Leaving tdef aside, the next step is to use gribmap (the one which comes
with grads-2.0.a2) , but this failed:

> g2ctl -0 /data/weather/gfs/20080714/00/gfs.t00z.pgrbf > gfs.ctl
> gribmap -i gfs.ctl
Open Error: Data file type invalid
--> The invalid description file record is:
--> dtype grib2
The data file was not opened.
Is there an alternative version of gribmap to use for GRIB2 files?  Or a
workaround which preserves the data in grib2 format?  (Or are these the
"strange" gfs files mentioned in a previous thread?)
Any help would be appreciated.
Isla Gilmour

 ----------------------------------------

gfs.ctl header

dset /data/weather/gfs/20080714/00/gfs.t00z.pgrbf84.grib2
index /data/weather/gfs/20080714/00/gfs.t00z.pgrbf84.grib2.idx
undef 9.999E+20
title /data/weather/gfs/20080714/00/gfs.t00z.pgrbf84.grib2
*  produced by g2ctl v0.0.3h
* griddef=1:0:(360 x 181):grid_template=0: lat-lon grid:(360 x 181)
units 1e-06 input WE:NS output WE:SN res 48 lat 90.000000 to -90.000000
by 1.000000 lon 0.000000 to 359.000000 by 1.000000 #points=65160

dtype grib2
ydef 181 linear -90.000000 1
xdef 360 linear 0.000000 1.000000
tdef 1 linear 00Z14jul2008 1mo
* PROFILE hPa
zdef 26 levels 100000 97500 95000 92500 90000 85000 80000 75000 70000
65000 60000 55000 50000 45000 40000 35000 30000 25000 20000 15000 10000
7000 5000 3000 2000 1000
options pascals
vars 139
[list of vars, which tally with wgrib2 listing]



More information about the gradsusr mailing list