add diagnostic for the "actual" sea ice increment

[shlyaeva]

Add a capability to compare "soca" increment generated by var with "soca"-type increment (i.e. full ice concentration) that was actually added to the CICE restarts. It would be useful for testing and diagnostics.

[shlyaeva]

Computing just the increments is non-trivial because:

• soca analysis is at the middle of the window, and so soca analysis increment is with respect to the background in the middle of the window
• when we run soca2cice, we use background (restart file) at the beginning of the window, and update it to be as close as possible to the soca analysis (at the middle of the window). The increment computed there would then include changes from both the analysis and the 3hr forecast.

So, to compute the increment, we'd need to compute diff between soca2cice output analysis file and the middle-of-the-window background.
Notes on possible implementation:

• copy soca2cice analysis in addition to analysis in https://github.com/NOAA-EMC/global-workflow/blob/2595e607fd905306939166a5e550e60683091bb0/ush/python/pygfs/task/marine_analysis.py#L387-L389
• add computing the difference between soca2cice analysis and background either to finalize script or to the vrfy (may complicate vrfy code unnecessarily, so finalize might be a better place)
• add config for plotting to

  GDASApp/utils/soca/fig_gallery/exgdas_global_marine_analysis_vrfy_manual.py

  Lines 141 to 149 in aeabb9b

  	plotConfig(grid_file=grid_file,
  	data_file=os.path.join(comout, f'{RUN}.t'+cyc+'z.ice.incr.nc'),
  	lats=np.arange(-60, 60, 10),
  	variables_horiz={'aice_h': [-0.2, 0.2],
  	'hi_h': [-0.5, 0.5],
  	'hs_h': [-0.1, 0.1]},
  	colormap='seismic',
  	projs=['North', 'South'],
  	comout=os.path.join(comout, 'vrfy', 'incr'))] # sea ice increment

  , using a different comout for soca2cice increments.

[shlyaeva]

Note: for experiments that ran before and didn't copy soca2cice analyses, it may be worth adding a script that can subtract the background from the total ice concentration from cice analysis restarts.

[shlyaeva]

Development in:

• soca Add optional increment output in Soca2Cice JCSDA-internal/soca#1114
• jcb-gdas NOAA-EMC/jcb-gdas@develop...feature/output_postproc_iceinc
• gdasapp develop...feature/output_postproc_iceinc
• global-workflow shlyaeva/global-workflow@35c4ccb

I'll wait till soca PR is merged, and gdasapp tags are updated to include it before opening the PRs in the noaa-emc repos.

[RussTreadon-NOAA]

@shlyaeva and @guillaumevernieres :
The sorc/soca hash in GDASApp branch feature/stable-nightly has been updated to soca @ f60b0e2. GDASApp g-w based ctests passed. Thus, the changes associated with soca PR #1114 exist and are functional in a GDASApp branch.

We can open a PR to merge feature/stable-nightly into GDASApp. This will bring the updated soca hash into GDASApp develop.

[shlyaeva]

@RussTreadon-NOAA perfect, thank you for the heads up! merging feature/stable-nightly would be great; let me know if you'd like me to open the PR.