Test case for SSMIS.

test data (UPP BUFR file for DMSP F17, full orbit):
  NPR.TDUP.SB.D18193.S1054.E1238.B6029697.NS

Shell script: run_ssmis_upp.sh
==============================

This has 4 parts:

1) Perform averaging and create an averaged UPP BUFR file (same functionality as
the old SSMIS_PP averaging module). Namelist ssmis_uppaveraging_main.nl.

2) map to a user-defined grid using namelist ssmis_remapping_main.nl.

3) Create hdf5 versions of the original and averaged UPP files using namelist
ssmis_bufr2hdf5_main.nl.

4) Create quick-look images of the original and averaged UPP data, using
mwipp_quicklook.py. By default, channel 17 (91 GHz) is plotted.

The python plotting script uses Cartopy: see https://scitools.org.uk/cartopy/docs/latest/

Note that F17 SSMIS has some problems with channels 1 to 4, which you can see if
you plot the unaveraged data.

Input data file (full orbit):
10647298 NPR.TDUP.SB.D18193.S1054.E1238.B6029697.NS

Top-level namelists:
ssmis_uppaveraging_main.nl, ssmis_remapping_main.nl, ssmis_bufr2hdf5_main.nl

Lower-level namelists:
mwipp_bufr.nl  ssmis_averaging.nl  user_grid.nl

User grid file:
t255_n128_lsm.grib

Output data files:
11889587 NPR.TDUP.SB.D18193.S1054.E1238.B6029697.NS.h5
 9959944 NPR.TDUP.SB.D18193.S1054.E1238.B6029697.NS_AV.bufr
11203765 NPR.TDUP.SB.D18193.S1054.E1238.B6029697.NS_AV.h5
  558358 NPR.TDUP.SB.D18193.S1054.E1238.B6029697.NS_AV_grid.bufr
 1129479 NPR.TDUP.SB.D18193.S1054.E1238.B6029697.NS_AV_grid.h5

Output image files:
  423342 NPR.TDUP.SB.D18193.S1054.E1238.B6029697.NS.h5_17.png
  377957 NPR.TDUP.SB.D18193.S1054.E1238.B6029697.NS_AV.h5_17.png
  428051 NPR.TDUP.SB.D18193.S1054.E1238.B6029697.NS_AV_grid.h5_17.png

Approximate run times:
Part 1: 19 sec (BUFR read; average; BUFR write)
Part 2: 36 sec (BUFR read; average; re-map; BUFR write; HDF5 write)

You can inspect your output data files using tools such as h5dump (part of hdf5)
and bufr_dump (part of ecCodes), and compare with equivalent outputs in the
"reference_products" directory.

You should expect the output brightness temperatures to be consistent with the
reference outputs, to approximately 0.01K.
