Hi everyone,

  Oops, I made a mistake in sending out my mail.  I sent an old message
to everyone on the wgrib email list.  I should have send the following:






                                 July 6, 1995



                     A short update on the status of wgrib.



                      Current version of wgrib is 1.3.4
                          new features over 1.3.3:
                     check for a bug in NMC grib headers


The current version of wgrib can be obtained by anonymous ftp from 
wesley.wwb.noaa.gov (/pub/wgrib/all.c).  I would also like to draw your 
attention to the directory /pub/wgrib.scripts.  There you will find

   swgrmap        a script file to make GrADS control files
   wwgrmap        a script file to make GrADS control files for time series
   ts2synoptic    a script combine time series into one file that is GrADS 
                    compatible

More details can be obtained by reading the comments in these shell scripts.


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


The current grib encoder that is used at NMC contains a bug which affects the 
NMC fortran and GrADS grib decoders (doesn't affect wgrib).  The bug shows up 
in the 1 degree x 1 degree operational grib files, and a few of the Reanalysis 
products (mainly cloud, soil and isentropic analyses).  All the Reanalysis 
products that do not have a bitmap are unaffected (majority).  If the bug 
occurs, the fortran and GrADS will put garbage into the last 1 or 2 grid 
points when they read the file.  Since these points are either at 90S or very 
close to it, they will not affect most users.  This bug does not affect wgrib.
The existence of the flawed headers can be determined by using scanning the 
files with wgrib v1.3.4.  For more information, and a program to fix headers, 
see /pub/fix_grib_bds.tar.gz on wesley.wwb.noaa.gov.

Some Reanalysis products have been affected by this bug.  Since the derived 
fields were computed with code that used the NMC grib decoders, they could 
have been computed with bad data at/near the South Pole (only if they have 
a bitmap).  

The Reanalysis CD-ROM will be delayed in order to fix-up this problem.





                            Wesley Ebisuzaki
                            Climate Analysis Center
                            NMC

                            (send requests to be on the mailing list to)
                            wd51we@sun1.wwb.noaa.gov