compare_paramfiles tool

[samsrabin]

Description of changes

Adds a utility to compare parameter files. Compares both "raw" values and values after masking and scaling (applying _FillValue to convert missing values to NaN, multiplying by scale_factor, and adding add_offset).

Example usage; results truncated:

file0=/glade/campaign/cesm/cesmdata/cseg/inputdata/lnd/clm2/paramdata/clm_params78pftModDates.c130821.nc
file1=/glade/campaign/cesm/cesmdata/cseg/inputdata/lnd/clm2/paramdata/ctsm60_params.c251124.nc

tools/param_utils/compare_paramfiles $file0 $file1

File 0: /glade/campaign/cesm/cesmdata/cseg/inputdata/lnd/clm2/paramdata/clm_params78pftModDates.c130821.nc
File 1: /glade/campaign/cesm/cesmdata/cseg/inputdata/lnd/clm2/paramdata/ctsm60_params.c251124.nc

Variable(s) present in File 0 but not File 1:
   cn_s1_bgc
   cn_s2_bgc
   ...

Variable(s) present in File 1 but not File 0:
   C2_liq_Brun89
   FUN_fracfixers
   ...

...

allconss:
   Values differ:
      [pft 71 (miscanthus)] 0.0 → 2.0 (raw)
                            nan → 2.0 (masked/scaled)
      [pft 72 (irrigated_miscanthus)] 0.0 → 2.0 (raw)
                                      nan → 2.0 (masked/scaled)
      [pft 73 (switchgrass)] 0.0 → 2.0 (raw)
                             nan → 2.0 (masked/scaled)
      [pft 74 (irrigated_switchgrass)] 0.0 → 2.0 (raw)
                                       nan → 2.0 (masked/scaled)

...

atmch4:
   Attribute(s) present in File 1 but not File 0:
       _FillValue: nan
   Dimension names differ: File 0: ['allpfts']
                           File 1: []

baset:
   Attribute(s) present in File 1 but not File 0:
       _FillValue: nan
   Values differ:
      [pft 71 (miscanthus)] 0.0 → 8.0
      [pft 72 (irrigated_miscanthus)] 0.0 → 8.0
      [pft 73 (switchgrass)] 0.0 → 8.0
      [pft 74 (irrigated_switchgrass)] 0.0 → 8.0

...

Specific notes

Contributors other than yourself, if any: None

CTSM Issues Fixed (include github issue #): None

Are answers expected to change (and if so in what way)? No

Any User Interface Changes (namelist or namelist defaults changes)? No

Does this create a need to change or add documentation? Did you do so? Yes; yes.

Testing performed, if any: All unit and system tests pass. Results on real paramfiles look good.

• clm_pymods test suite: Passed at /glade/derecho/scratch/samrabin/tests_1219-073947de

[slevis-lmwg]

@samsrabin thank you for this impressive PR! The compare_paramfiles tool is much more informative than cprnc. Your implementation looks very well thought out and you put together a nice list of unit and sys tests.

@samsrabin and I went over this in detail and @samsrabin took notes.