Fix issues with multiple ice sheets

[billsacks]

Description of changes

Fixes a few issues with running multiple ice sheets (i.e., multiple glc components):

(1) Fixes accumulation of topography with multiple ice sheets (fixes #229)

(2) Removes the need to set num_icesheets explicitly in the namelist; instead, determines this based on the number of elements in the mesh_glc namelist variable

(3) Removes a buggy double loop over ice sheet instances for smb renormalization

Specific notes

Contributors other than yourself, if any: @mvertens was the main contributor; I just made some final changes & testing

CMEPS Issues Fixed (include github issue #):

• Resolves ESCOMP/CMEPS#229

Are changes expected to change answers?

• [ ] bit for bit
• [ ] different at roundoff level
• [x] more substantial
  • Substantial changes to glc -> lnd topo values, even for runs with a single ice sheet. However, changes are only in land grid cells outside the glc domain (more precisely, outside the ice mask); in these areas, lnd ignores the glc -> lnd values. So, even though cpl hist files change substantially, this has no impact on the evolution of the model.

Any User Interface Changes (namelist or namelist defaults changes)?

• [ ] Yes
• [x] No

Testing performed if application target is CESM:(either UFS-S2S or CESM testing is required):

• [ ] (recommended) CIME_DRIVER=nuopc scripts_regression_tests.py
  • machines:
  • details (e.g. failed tests):
• [ ] (recommended) CESM testlist_drv.xml
  • machines and compilers:
  • details (e.g. failed tests):
• [ ] (optional) CESM prealpha test
  • machines and compilers
  • details (e.g. failed tests):
• [x] (other) please described in detail: CISM's aux_glc testing
  • machines and compilers: cheyenne: intel & gnu; izumi: nag & pgi
  • details (e.g. failed tests): all tests pass, though with answer changes as noted above; this testing included a few tests with multiple ice sheets, including a new test type that compares a multiple ice sheet run with a single ice sheet run and verifies that the two give bit-for-bit the same answers for the ice sheet in common between those runs.

Testing performed if application target is UFS-coupled:

• [ ] (recommended) UFS-coupled testing
  • description:
  • details (e.g. failed tests):

Testing performed if application target is UFS-HAFS:

• [ ] (recommended) UFS-HAFS testing
  • description:
  • details (e.g. failed tests):

Hashes used for testing:

• [x] CESM:
  • repository to check out: https://github.com/escomp/cism-wrapper/
  • branch: multiple_instances_v2
  • hash: fcf5f4b0792870e26b61cb3c71671ca986911fdb
• [ ] UFS-coupled, then umbrella repostiory to check out and associated hash:
  • repository to check out:
  • branch:
  • hash:
• [ ] UFS-HAFS, then umbrella repostiory to check out and associated hash:
  • repository to check out:
  • branch:
  • hash:

[billsacks]

@mvertens or others reviewing this: A few notes:

(1) I recommend viewing diffs with whitespace diffs hidden: one commit in particular introduced a lot of whitespace changes because of the removal of a level of looping, so it's easier to see what changed if you hide the whitespace diffs

(2) I have run CISM's aux_glc testing on these changes; I'm not sure if you want some other testing run

(3) I see there are merge conflicts; let me know if you want me to help resolve them

[billsacks]

I'm working on the merge conflicts and will push soon....

[billsacks]

Okay, I fixed the merge conflicts and just pushed the result. That was one ugly set of merge conflicts. For one conflict, I swear that git just threw up its hands in despair: what it gave me even outside the conflict region was completely wrong and it took me a while to figure out what was going on. But I think I have things right now: I did a careful comparison of the diffs from master to this branch before and after the merge, and the diffs look effectively the same now. I'm going to run some tests overnight to double-check the merge and will post the results back here in the morning.

[billsacks]

I reran the aux_glc testing with the latest version of this branch, post-merge, and compared with my previous set of testing. Tests passed and were bit-for-bit – other than the ones I referenced in #243 , and other than one FIELDLIST diff that I assume is expected from the update from cmeps0.13.30 to latest (the new version has time_bnds in the cpl hist file, the old did not).

So it seems like the hairy merge probably went okay.