What should the 100km mom config inherit from 25km / OM4-05?

[minghangli-uni]

Issue description

Ref: https://forum.access-hive.org.au/t/nri-ocean-meeting-minutes-2025/4053/20

As discussed in the osit meeting, we plan to update the MOM config for the 100 km setup. However, we’re not yet sure which parameters can be inherited from the current 25 km configuration.

A comparison table which also includes the OM4-05 config 51e50f2, has been prepared in https://github.com/COSIMA/access-om3-doc/blob/main/tables/MOM_input/GFDL-OM4_ACCESS-OM3_nml_diff.md. This issue is intended as a space for discussion and feedback.

Issue checklist

• I have labelled this issue with the affected configurations (e.g. MC_25km_jra_ryf, [1] more options)
• I have added the issue information to the ACCESS-OM3 Mega Project board!:
  After issue creation, under Projects please set:
• Status (blocked, Todo, In Progress, Done)
• Priority (P0, P1, P2)
• Component

[1] some label options
MC_100km_jra_ryf
MC_100km_jra_iaf
MC_100km_jra_ryf+wombatlite

[chrisb13]

Thanks @minghangli-uni, I think the comparison table to look at this one. Is that right? And just to be clear, the objective is to pick changes from the dev-25km* not the release-25km*?

How does one interpret cells that have no value? I suppose this could mean i) it takes the default, ii) it's a new option that has been created by a code update.

[minghangli-uni]

I think the comparison table to look at this one.

Yes, that was already merged - I’ve updated it in the first comment. Thanks!

And just to be clear, the objective is to pick changes from the dev-25km* not the release-25km*

I've added a new one #765 (has been merged) to the dev-25km branch. It hasn’t been merged into the release branch yet.

How does one interpret cells that have no value? I suppose this could mean i) it takes the default, ii) it's a new option that has been created by a code update.

My understanding is that parameters with empty cells are dependent variables - if their parent parameter isn’t enabled, those settings aren’t applicable.

[dougiesquire]

This is a summary of changes to the 25km RYF that have not yet made it into the 100km configs:

	25km PR(s)	Change in 100km configs?	Who?	Completed?
MOM6 parameter updates from GFDL OM	Mostly done in #556 and #648, though some changes in #556 were reverted in later PRs e.g. #641	We'll need a dedicated discussion about which of these changes we want to carry across, informed by @minghangli-uni's tables linked above.	@dougiesquire to set up PR with everything, then organise discussion
Tidal dissipation parameterisation	#562	Yes	@minghangli-uni	✅ MC-RYF (#779) ✅ MC-IAF (#784) ✅ MC-RYF+WOMBATlite (#789)
Sapphire rapids and load balancing	#591, #725	Yes	@chrisb13 (load balancing should be done last)
Diagnostics on density coordinates	#622, #641 and #648	Yes	@dougiesquire	⏳ MC-RYF (#791) ⏳ MC-IAF (#794) ⏳ MC-RYF+WOMBATlite (#795)
Change EOS and prognostic temp/salt	#641	Yes	@dougiesquire	✅ MC-RYF (#787) ✅ MC-IAF(#788) ⏳ MC-RYF+WOMBATlite (#793)
WOA23 initial conditions/restoring with correct salinity	#610 and #725	Yes	@ezhilsabareesh8	✅ MC-RYF (#787) ✅ MC-IAF(#788) ⏳ MC-RYF+WOMBATlite (#793)
Update README in diagnostic_profiles	#649	Yes	@anton-seaice	✅ MC-RYF (#772) ✅ MC-IAF (#785) ✅ MC-RYF+WOMBATlite(#786)
Move CICE concatenation and datastore generation to postscript	#658	Yes	@anton-seaice	✅ MC-RYF (#772) ✅ MC-IAF (#785) ✅ MC-RYF+WOMBATlite(#786)
Set ice initial condition to none	#678	No? (See here)	-	-
Update to ACCESS-OM3@2025.08.001	#747	Yes	@anton-seaice	✅ MC-RYF (#772) ✅ MC-IAF (#785) ✅ MC-RYF+WOMBATlite(#786)
Use CORIOLIS_SCHEME = SADOURNY75_ENSTRO	#765	Yes	@dougiesquire	⏳ MC-RYF (#791) ⏳ MC-IAF (#794) ⏳ MC-RYF+WOMBATlite (#795)

[chrisb13]

@dougiesquire, my row for Sapphire rapids and load balancing has this PR. Which I think could be a typo from the previous row?

Also, I would think Sapphire rapids would be a build repo change whereas load balancing would be config (here maybe?) change? @minghangli-uni will have useful comments I'm sure on these points!

[minghangli-uni]

I've updated the related PRs to #591, #725

[dougiesquire]

Also, I would think Sapphire rapids would be a build repo change whereas load balancing would be config (here maybe?) change?

Both are config changes. The former is changing the config to run on sapphire rapids nodes

[minghangli-uni]

Sapphire rapids would be a build repo change

I’m running tests on the compiler flag changes in ACCESS-NRI/ACCESS-OM3#148. The new flags will apply only to SR, not to CL.

load balancing would be config (here maybe?) change

The run sequence should be good at the moment.

Currently I think you should only focus on the config changes, such as in config.yaml and nuopc.runconfig.

[chrisb13]

Great, thanks both. A bit more back and forth and it seems this is blocked for the time being. So I'll wait for a prompt to pick it up again.