CoMorph further KGO-preserving refactoring

[MichaelWhitall]

PR Summary

Sci/Tech Reviewer: AlisonStirling
Code Reviewer: Ed Hone (@EdHone)

The bulk of the refactoring of CoMorph in the comorph_dev branch was lodged in the lfric_apps trunk in Issue #178 / PR #292. But some more KGO-preserving refactoring has been done in the comorph_dev branch after the branch lodged for #292 was split-off from it. We now need to lodge these extra refactoring changes. We also need to add more of CoMorph's tune-able constants to the LFRic namelist.

For a detailed description of the changes, see the Issue page: #713.

Since we've altered the namelists, this PR requires an upgrade macro to apply the namelist changes to existing apps, and a test branch in-which the upgrade macro has been applied to rose-stem: comorph_refact2 -> comorph_refact2_test

closes #713
blocks #776

Code Quality Checklist

• I have performed a self-review of my own code
• My code follows the project's style guidelines
• Comments have been included that aid understanding and enhance the readability of the code
• My changes generate no new warnings
• All automated checks in the CI pipeline have completed successfully *

* The check-cla test is failing, even though I have not modified the contributors.md file in this PR (and my name is already added to it on the trunk). This test just seems to be broken.

Testing

• I have tested this change locally, using the LFRic Apps rose-stem suite
• If any tests fail (rose-stem or CI) the reason is understood and acceptable (e.g. kgo changes) NA
• I have added tests to cover new functionality as appropriate (e.g. system tests, unit tests, etc.) NA
• Any new tests have been assigned an appropriate amount of compute resource and have been allocated to an appropriate testing group (i.e. the developer tests are for jobs which use a small amount of compute resource and complete in a matter of minutes) NA

Note: while all the comorph apps in the lfric_apps rose-stem tests preserved KGO exactly (as did all the UM rose-stem tests with the same changes to comorph in that model), when I tested the same changes in UM N96 AMIP runs (outside of rose-stem) there was in-fact a change in answers for the CoMA9_tb configuration. The difference appears to be insignificant / science-neutral, and is presumably due to a change in compiler optimisation behaviour (the AMIP run uses different compiler optimisation settings to the rose-stem tests). More detail of the tests performed and the results can be seen here:
Testing CoMorph refactoring and bug-fixes

trac.log

Test Suite Results - lfric_apps - comorph_refact2_test/run3

Suite Information

Item	Value
Suite Name	comorph_refact2_test/run3
Suite User	michael.whitall
Workflow Start	2026-09-09T14:30:19
Groups Run	developer', 'lfric_atm_nwp_ex1a_extra', 'lfric_atm_nwp_azspice_extra

Dependency	Reference	Main Like
casim	MetOffice/casim@2026.07.1	True
jules	MetOffice/jules@2026.07.1	True
lfric_apps	MichaelWhitall/lfric_apps@comorph_refact2_test	False
lfric_core	MetOffice/lfric_core@60fc29a	True
moci	MetOffice/moci@2026.07.1	True
SimSys_Scripts	MetOffice/SimSys_Scripts@77a5166	True
socrates	MetOffice/socrates@2026.07.1	True
socrates-spectral	MetOffice/socrates-spectral@2026.07.1	True
ukca	MetOffice/ukca@9fc2b6d	True

Task Information

✅ succeeded tasks - 1309

Security Considerations

• I have reviewed my changes for potential security issues
• Sensitive data is properly handled (if applicable) NA
• Authentication and authorisation are properly implemented (if applicable) NA

Performance Impact

• Performance of the code has been considered and, if applicable, suitable performance measurements have been conducted

No change in model performance expected.

AI Assistance and Attribution

• Some of the content of this change has been produced with the assistance of Generative AI tool name (e.g., Met Office Github Copilot Enterprise, Github Copilot Personal, ChatGPT GPT-4, etc) and I have followed the Simulation Systems AI policy (including attribution labels)

A Met Office Github Copilot Enterprise AI agent was used in visual studios code to do the following:

• Extract the relevant existing changes from the UM comorph_dev branch into a smaller separate UM branch containing only the KGO-preserving refactoring changes (however the agent made many errors so heavy manual intervention has been required after running the agent).
• Migrate the existing comorph namelist meta-data from the UM into the lfric_apps convection meta-data (but again many manual corrections were required).

Documentation

• Where appropriate I have updated documentation related to this change and confirmed that it builds correctly

Porting comorph documentation to lfric_apps is still tbd, see #359.

PSyclone Approval

• If you have edited any PSyclone-related code (e.g. PSyKAl-lite, Kernel interface, optimisation scripts, LFRic data structure code) then please contact the TCD Team NA

Sci/Tech Review

• I understand this area of code and the changes being added
• The proposed changes correspond to the pull request description
• Documentation is sufficient (do documentation papers need updating)
• Sufficient testing has been completed

(Please alert the code reviewer via a tag when you have approved the SR)

Code Review

• All dependencies have been resolved
• Related Issues have been properly linked and addressed
• CLA compliance has been confirmed
• Code quality standards have been met
• Tests are adequate and have passed
• Documentation is complete and accurate
• Security considerations have been addressed
• Performance impact is acceptable

[AlisonStirling]

Sci tech review approved

[DrTVockerodtMO]

Adjoint config changes are fine, thank you!

[MichaelWhitall]

Rose-stem test results / trac.log output on the PR summary now updated with a run from the latest branch revision; all tests still pass :)

[AlisonStirling]

Sci/Tech reviewed after additional code added since 21st Aug 2026

[github-actions]

⚠️ Hello MichaelWhitall!

Your CLA signature was found on the base branch, but you appear to have modified the CONTRIBUTORS.md file in this PR.

Please do not edit the CONTRIBUTORS.md file. If you have already signed the CLA, revert changes to the file and your signature will be picked up.

[MichaelWhitall]

Sorry just added a last-minute extra commit:

• A few more trivial code-style changes.
• Added an accidentally-missed unit test for the in-plume microphysics implicit phase-change solver (this was added to the comorph_dev branch a while ago where it was used to debug a failure in there once; would like to port this across just in case we need to use it again).

[AlisonStirling]

Sci /tech approved extra submission from Friday 11th September.

[MichaelWhitall]

James Bruten (@james-bruten-mo), cjohnson-pi, Hacka Fett (@christophermaynard), Steven Sandbach (@ss421), Matt Shin (@matthewrmshin), any chance you can look at this PR Re code-owner approval? For most of you, its just that it modifies an example configuration file you own to keep it consistent with the namelists / meta-data. Among other things, this PR moves some comorph convection scheme settings from the "convection" namelist to a new "comorph" namelist. It turns out none of the example configurations use comorph, so rather than adding the comorph namelist to the .nml files, the correct change was just to delete the existing comorph settings (they would've been trigger-ignored in an equivalent rose-app so technically shouldn't been added to these example config files in the first place).

Note I'd like to get this PR lodged soon if possible as it blocks another PR...

Cheers!
Mike

[MichaelWhitall]

Ed Hone (@EdHone), this PR is ready for code-review to begin :)

Note this one blocks PR #776 which is also aimed at the next release so it'd be nice to get it lodged relatively soon if that's possible? (still pending some code-owner approvals but I've sent a reminder to the affected code-owners just now...)

Cheers!
Mike

[James Bruten (james-bruten-mo)]

Macro approved

[Steven Sandbach (ss421)]

It's because of the namelist updates to applications/jedi_lfric_tests/*/configuration.nml, there's an equivalent set of namelists in metoffice/lfric-jedi and metoffice/jjdocs which need to be kept consistent with these ones.

Does this mean that pretty-much any lfric_apps PR which updates the namelists has to have a linked JEDI PR?? And its just because of the "example configuration" .nml files which annoyingly the upgrade macro doesn't update?

If that's so, surely we ought to urgently either: a) Add some technical wizardry to make the apply_macros.py script apply the macro changes to the "example configuration" files even though they aren't in rose, or b) Migrate all the example configurations to rose so that the upgrade macros update them in the usual way.

Otherwise the overheads from you having to update them in JEDI every time someone adds something to the lfric_apps namelist is going to get silly!

We store namelist configurations in the JEDI testing and these are based on the configurations in LFRic which is ultimately stored in the rose configuration files. Its easier to see what has changed in the namelist files but these should mirror the ones in the rose configuration. I agree that having a mechanism to automatically do that via the rose configs but we are not able to do that just yet so we have this pain for now.

From what I understand in the PR, it seems that while the namelist files have been updated that is related to changes that had not been applied previously and not this PR?

[MichaelWhitall]

From what I understand in the PR, it seems that while the namelist files have been updated that is related to changes that had not been applied previously and not this PR?

Thanks; the changes to the example configurations in this PR actually just undo the changes from a previous PR which unnecessarily (technically wrongly) added the convection namelist items resdep_precip_ramp and dx_ref to all the example configurations. I now realise that PR shouldn't have added them because they would've been trigger-ignored in those configurations so wouldn't have been added to the namelist file by rose.

I had to remove resdep_precip_ramp and dx_ref from the example configs to get the tests from this PR to pass, because this PR moves those items out of the convection namelist, so the namelist read fails if it finds them in there.

Also, I now realise most of the example config files (and those in JEDI) don't use most of the physics. As long as namelist / meta-data changes only affect things that would be trigger-ignored anyway when not using the relevant physics, then no update to the example configs in JEDI will be needed.

Cheers!
Mike

[Steven Sandbach (ss421)]

Thanks. Dan has checked and everything works for us :-)