Skip to content

v12 shocmf updates#1201

Merged
sdrabenh merged 12 commits intofeature/sdrabenh/gcm_v12from
feature/narnold1/GCMv12rc22_shocmf_updates
Mar 26, 2026
Merged

v12 shocmf updates#1201
sdrabenh merged 12 commits intofeature/sdrabenh/gcm_v12from
feature/narnold1/GCMv12rc22_shocmf_updates

Conversation

@narnold1
Copy link
Copy Markdown
Contributor

@narnold1 narnold1 commented Feb 4, 2026

SHOCMF updates. Adds v2.0 of double gaussian PDF, removes CLASP and LTS options from EDMF, various stability improvements and code cleanup. 0-diff with exception of turb_internal_rst: new internals were added, used by SHOC only.

…laces LTS with dynamic pressure decel, switches to version 2 of ADG PDF, and general adjustments.
@narnold1 narnold1 requested a review from a team as a code owner February 4, 2026 15:44
@narnold1 narnold1 added the 0 diff The changes in this pull request have verified to be zero-diff with the target branch. label Feb 4, 2026
@mathomp4
Copy link
Copy Markdown
Member

mathomp4 commented Feb 5, 2026

@narnold1 GNU has an issue but I think it's been fixed in upstream v12. When ready, you can press the "Update branch" button and the CI should be happy.

@mathomp4
Copy link
Copy Markdown
Member

@narnold1 The build error seems to be saying something is missing in the call to partition_dblgss. There aren't enough arguments in the call.

The call does:

​            call partition_dblgss(fQi,          &
                                  TEn,          &
                                  QVn,          &
                                  QCn,          &
                                  0.0,          &
                                  ZL,           &
                                  PL*100.,      &
...

but the subroutine has:

subroutine partition_dblgss( fQi,          &  ! IN
                              tabs,         &
                              qwv,          &
                              qc,           &  ! OUT
                              qsat,         &
!                              qi,           &
                              omega,        &  ! IN
                              zl,           &
                              pval,         &

So it's either missing qsat or omega. Something is being passed in as 0.0 but I'm not sure what.

@narnold1
Copy link
Copy Markdown
Contributor Author

Yes, "qsn" should be passed in after QCn. omega is passed in as 0.

@sdrabenh
Copy link
Copy Markdown
Collaborator

My tests reverting the negative protection change show this PR passes:

  1. All regression tests
  2. 0-inc replay test
  3. All restarts from AMIP and Replay modes are 0-difference EXCEPT:
    • turb_import_checkpoint long_name changed from "surface geopotential height" to "surface geopotential"
    • turb_internal_checkpoint added variables: MFAW, QTFLXMF, SLFLXMF

@sdrabenh sdrabenh merged commit a66f688 into feature/sdrabenh/gcm_v12 Mar 26, 2026
15 checks passed
@sdrabenh sdrabenh deleted the feature/narnold1/GCMv12rc22_shocmf_updates branch March 26, 2026 18:55
@sdrabenh sdrabenh restored the feature/narnold1/GCMv12rc22_shocmf_updates branch March 26, 2026 19:40
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

0 diff The changes in this pull request have verified to be zero-diff with the target branch.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants