README for PREPBUFR2LITTLER converter, version 0.9 (prerelease)

#################################################################################
#  Decode all information in a bufr file and write to little_r file
#  Input:  NCEP prepbufr observation
#  Output: prepbufr2littleR.txt which has prepbufr obs and QC in little_r format 
#          pbufr.table which is the table of prepbufr obs MNEMONIC
#  Usage:  convert_prepbufr2littleR.exe  
#          It looks for prepbufr obs with name 'bufrfile' and write to little_r
#
#  Ruifang Li 02/01/2011
#  Ruifang Li 08/30/2011 Added PW for GPSIPW type
!#################################################################################


Compiling:

Edit the "compile" script to set the parameters appropriate for your machine.

This package is written in Fortran 90, and so a Fortran 90 compiler is required. 
Set "FC=your_compiler" to set your compiler. The package has been tested with 
gfortran and pgf90, but will likely work with others. You may need to change FLAGS
depending on your platform and choice of compiler.

Either link/copy the NCEP BUFR library file (libbufr.a) into this directory (can be 
found in the compiled WRFDA code under WRFDA/var/external/bufr/) or set "LIBS" equal
to "-Lfulllibbufrdirectory". So if your libbufr.a file is found in , for example, 
/home/user/WRFDA/var/external/bufr, set LIBS like this:

LIBS=-L/home/user/WRFDA/var/external/bufr

After all this is set, run "compile". Compilation should take only a few seconds.



Running:

To run the executable, link or copy your NCEP PREPBUFR file into this directory with 
the name "bufrfile". Then run the executable. It may take several minutes depending
on the number of observations. When it is complete, you should see one or more
LITTLE_R format files split up by observation type, with the file extention ".txt".

