NOAA / NWS's Meteorological Development Laboratory
Degrib (NDFD GRIB2 Decoder)
Subject: Degrib 1.60
Date: Tue, 24 Aug 2004 14:38:22 -0400
From: Arthur Taylor
You all have the unique distinction of being registered on my degrib
(aka NDFD_GRIB2Decoder) mailing list (all 488 of us) (I'd like to
welcome: all_full_of_feathers, r.love, peter, don, ewelsh, sean, BReed,
stephen.richart, ed, ptzscheu, moore, jmcamarillo, keren.smith,
mrappleb, giuliani, farr, ghough, bob.stadum, g960102, rcarpenter,
mark.mitchell, orson, rdunsky, jglaves, mmaciver, s.boettjer,
amirnourian, mikebobbitt, seth, dcoaster, jay_ganz, graham.sinden,
popeza2000, jacob_klee, dennis.hull, cowie, markt, dcoon,
marty.middleton, deetotten, angel, jbrame5, fasfadfsfa, ramon.williams,
wetter3, Lee-White, STILL46203, riachard_woolley, damian, donm,
syoung30, pangae1, robertmoore, johnson_abb, cloayza)
Please let me know (email me) if you want to unregister. The latest
version of the program is on:
----------- General News.
1) The first part of the NDGD (National Digital Guidance Database) is
now available for download. It currently consists of ozone model
forecasts for the eastern part of the CONUS (continental United States).
2) The Pacific Region is now (as of 8/23/2004) available in GRIB2 form
from the public ftp and http sites. It currently consists of a Guam
sector and a Hawaii sector. Note: the Guam sector currently does not
have QPF forecasts or Snow forecasts, but does have the rest of the NDFD
variables. The Hawaii sector does not have QPF forecasts, but does have
Snow and the rest of the NDFD variables.
----------- News about this release:
I have just finished putting together version "1.60" (last one that I
announced was "1.53" on 6/18/2004).
This release is mainly focused on being able to support:
A) the NDGD,
B) the Pacific Region Sectors (Hawaii, Guam), and
C) to handle some changes to the weather keys.
1) The NWS has started providing NDGD which consists of model guidance,
as opposed to the NDFD which consists of the forecasts generated by NWS
forecasters. To support this, degrib has changed some of its directory
structure. Previously it downloaded all data to
"degrib/data/grib2/<sector>". Now since we have two different data
sets, it uses "degrib/data/ndfd" and "degrib/data/ndgd". In addition
the download parts of "tkdegrib" and "tcldegrib web.tcl" needed to be
updated to go to the correct urls:
http://weather.noaa.gov/pub/SL.us008001/ST.expr/DF.gr2/DC.ndfd/ for NDFD
and http://weather.noaa.gov/pub/SL.us008001/ST.expr/DF.gr2/DC.ndgd/ for
2) The NWS has started providing Hawaii and Guam. To support this,
degrib had a configuration option which could be "enabled".
Unfortunately, when activating the option, a bug in version 1.53 was
detected. The bug was that it would occasionally use a piece of code
that didn't handle the mercator projection correctly. I have now fixed
that piece of code, but will be investigating it further.
Please note that the "Generate Images" button does not currently work
with either the NDGD, or the Pacific Regions of the NDFD.
3) The NWS has increased the possibilities for weather. In particular
they introduced in "IFPS-16", the following:
New Coverages: "Pds" == "Periods of", "Frq" == "Frequent", "Inter" ==
"Intermittent", "Brf" == "Brief".
New Hazards: "OGA" == "On Grassy Areas", "Dry" == "dry", "TOR" ==
"Tornado", "Primary" == "Highest Ranking", "Mention" == "Include
weather.c has now been updated to handle these. In addition some code
was added to hopefully enable degrib to gracefully handle new weather
codes, coverages, intensities and hazards, that may be introduced in the
4) The WMO (World Meteorological Organization) has now made JPEG2000 and
ZLIB compression schemes official. When they did that, they stated that
the template numbers should be 40, 41 instead of the experimental
numbers of 40000, and 40010. "degrib" should now be able to handle
both the new and the old template numbers.
5) The WMO (World Meteorological Organization) clarified the units of
the values in code table 3.2. I have made the required adjustments.
Note: this doesn't affect the NDFD, or NDGD.
6) Added a -refTime option. This allows one to find the earliest
reference time in a file of GRIB messages. This is useful, if you want
to know how old the data is, and the file time stamp is not updating.
Previously, one could parse the -I (inventory) output, but the -refTime
option may make it easier.
7) Added a -stdout option. In the past when one wanted detailed meta
data information about a GRIB message, one could use "degrib <file> -C
-msg 1" which would create a meta (.txt) file. The problem with
creating the meta (.txt) file is that one may not have permission to
create files in that particular directory. With the "-stdout" (or "-out
stdout") option, the meta data goes to the screen instead of to file.
8) Added the ability to work with a "tar" file containing several GRIB
files. Degrib can now inventory and extract GRIB messages directly from
the "tar" file, as long as the file is not compressed, and none of the
GRIB files in the "tar file" have been compressed. This can be useful
when dealing with archives of GRIB messages.
9) For Template 4.9, which is the "Probability forecasts" template, it
now defaults to units of % instead of the units based on the particular
variable. This is so that a probability forecast of "rain greater than
1 inch" has units of % rather than inches.
10) Added some GRIB1 support for the AirForce. Specifically, in GRIB1,
one needs to look up the type of variable in a parameter table. I
taught degrib about the NWS parameter tables, but didn't have access to
the AirForce parameter tables until recently. I've also made degrib
complain a little more if it can't find the parameter table. If it
does, please pass me the parameter table that I should use.
11) Adjusted the GRIB1 radEarth code to default to 6371.2 km if the data
originated at NCEP. This is because NCEP used 6371.2, but was forced by
the GRIB1 format to encode 6367.47. Previously I had required people to
remember that, and use the -radEarth option to over ride it. I have now
changed it so that if it is given a GRIB1 NCEP message, it should
default to 6371.2 (unless someone over rides it with the -radEarth option).
12) Several cosmetic adjustments to "tkdegrib"
----------- Personal News:
I got engaged, on Aug. 12, on the floor of the old U.S. Senate chamber! :-)
Please let me know if you have problems.