Enable EnKF-only for atmosphere

[bhuang95]

This PR works along with the following three dependent PRs to enable the EnKF-only configuration for the atmosphere within the global workflow (see detailed description NOAA-EMC/global-workflow#4345)

Dependencies:
-NOAA-EMC/global-workflow#4345
-NOAA-EMC/GDASApp#2010
-NOAA-EMC/jcb#32

Resolve

• Enable EnKF-only for atmosphere global-workflow#4339

[DavidNew-NOAA]

@bhuang95 FYI I'm planning on merging jcb-gdas into the gdasapp repo soon, so that unless jcb-gdas PRs get merged soon, they will need to become gdasapp PRs. This PR will probably need to be merged into NOAA-EMC/GDASApp#2010 after that happens. Just letting you know.

[CoryMartin-NOAA]

@DavidNew-NOAA I think this is safe to merge soon, but we need to test it and coordinate it with Bo's GDASApp PR

[DavidNew-NOAA]

@CoryMartin-NOAA Ok, good to know

[CoryMartin-NOAA]

@bhuang95 can you either resolve the conflicts or add me as a collaborator on your fork so I can do it

[bhuang95]

@bhuang95 can you either resolve the conflicts or add me as a collaborator on your fork so I can do it

@CoryMartin-NOAA I just added you as collaborator in my jcb, jcb-gdas, GDASApp and global-workflow repo. I am resolving some conflicts in global-workflow and will update these four repos soon. Thanks.

[RussTreadon-NOAA]

@bhuang95 : What is the status of this PR? We can not move forward with GDASApp PR #2031 until this PR is either (a) merged into jcb-gdas develop or (b) closed without action.

Tagging @DavidNew-NOAA and @CoryMartin-NOAA for awareness.

[bhuang95]

@bhuang95 : What is the status of this PR? We can not move forward with GDASApp PR #2031 until this PR is either (a) merged into jcb-gdas develop or (b) closed without action.

Tagging @DavidNew-NOAA and @CoryMartin-NOAA for awareness.

@RussTreadon-NOAA I think this PR is ready to go. The failure of CI test may be because this PR is on my fork not NOAA-EMC as answered by @CoryMartin-NOAA for my JCB PR. Otherwise, could you help take a look and see what further changes are needed? It's very minor changes here.

[RussTreadon-NOAA]

@bhuang95 : Let me install your sequence of g-w, GDASApp, and jcb-gdas PRs on Gaea C6 and run GDASApp ctests to see what happens.

[DavidNew-NOAA]

@bhuang95 Will merging these changes break anything GDASApp without NOAA-EMC/GDASApp#2010 and NOAA-EMC/global-workflow#4345 being merged? The issue is that we're trying to merge the jcb-gdas repository into gdasapp fairly soon, and we're not accepting any new jcb-gdas PRs, so development for anything that requires changes to JCB templates is on hold. The process of coordinating the merger of 3 PRs can take a while. One option is to merge jcb-gdas into gdasapp, and then incorporate your change in this PR into NOAA-EMC/GDASApp#2010 .

[bhuang95]

@bhuang95 : Let me install your sequence of g-w, GDASApp, and jcb-gdas PRs on Gaea C6 and run GDASApp ctests to see what happens.

Thanks, @RussTreadon-NOAA FYI, I updated the g-w, GDASApp, and jcb-gdas PRs earlier this week and they passed the EnKF-only test using real observations on Ursa.

[RussTreadon-NOAA]

@bhuang95 : The following branches are behind their corresponding authoritative develop

• g-w feature/enkf_only_dev is 1 commit behind develop
• GDASApp feature/enkf_only_dev is 2 commits behind develop

I can't fully test your changes until all your branches are up to date with the current head of their respective authoritative develop.

[bhuang95]

@bhuang95 Will merging these changes break anything GDASApp without NOAA-EMC/GDASApp#2010 and NOAA-EMC/global-workflow#4345 being merged? The issue is that we're trying to merge the jcb-gdas repository into gdasapp fairly soon, and we're not accepting any new jcb-gdas PRs, so development for anything that requires changes to JCB templates is on hold. The process of coordinating the merger of 3 PRs can take a while. One option is to merge jcb-gdas into gdasapp, and then incorporate your change in this PR into NOAA-EMC/GDASApp#2010 .

@DavidNew-NOAA I don't think merging the changes in this PR will break anything in GDASApp without merging the other two PRs for GDASApp and GW. There two major changes.

1. Add {% if atm_enkfonly | default(false) %} block to distinguish the increment files. By default it is false and it will use the original increment filename, so it won't affect other components in jcb-gdas and GDASApp.
2. Remove directory observations/atmosphere-lgetkf. I am not sure if it is tested anywhere in GDASApp. If it is needed, I can simply put it back in this PR so it won't affect GDASApp either.

The change in this PR and GDASAPP is minor. But I am okay to close this PR and do a merge later if preferred.

[DavidNew-NOAA]

Thanks @bhuang95 . I would say the easiest thing to do is keep the fv3jedi_fv3inc_lgetkf.yaml.j2 changes, since it's obvious from your comment that they won't break anything, but revert the deletion of the observation files, and we can delete them later, if not in your GDASApp PR.

[DavidNew-NOAA]

Thanks @bhuang95 for the changes. I feel confident that we're safe merging this PR. Will do so now. cc @RussTreadon-NOAA