ESCOMP / POP2-CESM

Parallel Ocean Program (POP2) in CESM
http://www.cesm.ucar.edu/models/cesm2/ocean/
4 stars 24 forks source link

Use latest version of MARBL #13

Closed mnlevy1981 closed 5 years ago

mnlevy1981 commented 5 years ago

Description of changes:

A new version of MARBL has been released to support autotrophs that are explicit calcifiers

Testing:

Test case/suite: I ran aux_pop on cheyenne and aux_pop_MARBL on both cheyenne and hobart Test status: existing tests are all bit-for-bit; I added a new test to exercise the new MARBL code, and also marbl_in has changed so NLCOMP tests fail if the ecosystem is enabled

User interface (namelist or namelist defaults) changes? Yes, for compsets using MARBL (see above)

mnlevy1981 commented 5 years ago

Note: I still need to update namelist_defaults_pop.xml to point to a new init_ecosys_init_file when running with OCN_BGC_CONFIG=cesm2.1+cocco; I will generate that file, run the test suite one more time, and then comment that this is ready to be merged.

mnlevy1981 commented 5 years ago

Testing is complete -- as expected, aux_pop generated NLCOMP failures for all tests that used MARBL-enabled compsets but all BASELINE tests passed. New initial condition file has been added to inputdata and (and has already been downloaded on hobart).

I'm unclear about the git-based process for merging and tagging: should I update ChangeLog? Are we still tagging every commit? It looks like master is a little ahead of the last tag because @alperaltuntas updated github templates, but this is the first PR since the last tag that changes source code.

alperaltuntas commented 5 years ago

Are we still tagging every commit?

My feeling is that we shouldn't necessarily tag every commit but should tag every PR to be merged. Perhaps we may delay tag creation if a PR is not so extensive and doesn't require us to run aux_pop, unless the changes need to be added to a CESM tag promptly. Not sure if such exceptions would cause impracticality. As for updating ChangeLog, I think we should update it each time creating a new tag. What do you think @mnlevy1981 @klindsay28 ?

mnlevy1981 commented 5 years ago

I like the ChangeLog and am happy to keep it alive. I'll make one more commit to this branch / PR to add some notes about the tag.

@alperaltuntas do you want me to wait for #7 before merging this? If so, I'll hold off on the ChangeLog update until after merging in the new master.