Updating to new version- associated changes

[Whyborn]

We are going to be moving to version 2025.11.000 of ESM1.6. Do we want to change the name of the current spin-up branches to November? Are there any configuration file changes that also need to come in?

• Removal of cable_user%CABLE_RUNTIME_COUPLED from cable.nml

[anton-seaice]

• Remove melt diagnostic #242
• CMIP7 diagnostics available in CICE5 / ESM1.6 #248

Are we updating the restart files (other than #248) - we need to fix the dates in the restart files, they are currently inconsistent between the atmosphere and ocn/ice

[MartinDix]

Changes to L_VOLCTS flag and removal of the volcanic forcing input file following https://github.com/ACCESS-NRI/UM7/pull/167

[Whyborn]

From the CABLE 4 meeting:

• Check the CABLE namelist is correct
• pftlookup.csv, pft_params.nml, phenology file
• JULES namelist
• Update restart

[manodeep]

Migrating my comments from #259 (as suggested by @anton-seaice )

config.yaml

• Switch to payu/1.2 from release (not prerelease)
• Explicitly load OpenMPI module

Input files

• if using upstream oasis3_mct, then the coupler/namelist file will need to be updated

[blimlim]

• Start with test-PIspinup-Oct-B
  - [ ] Take BGC restart from Pearse's run, combine with rest of restart from end of Oct-B (if possible) (see below)
• Updated bgc parameters from @pearseb
• xnslope parameter changes Use pft file compatible with xnslope change #256

[blimlim]

From @pearseb:

... That reinforces to me that we should only go with the slightest of tweaks, and critically, that we may not need to adjust the current OctB simulation with BGC fields from my test run. Simply adjusting the sinking rate down on the current OctB run should do the trick!

The grey is the way to go. Spencer, the field_table parameters wdetbio and wcaco3 should be changed to these values:

wdetbio = 2.778e-4
wcaco3 = 1.100e-4

[blimlim]

I've started a draft PR #268 to combine the changes

@Whyborn do you have details on the changes that need to go in for these?

• Check the CABLE namelist is correct
• pftlookup.csv, pft_params.nml, phenology file
• JULES namelist
• Update restart

[Whyborn]

The only thing that's needed for the configuration to run I think is the removal of cable_user%CABLE_RUNTIME_COUPLED from cable.nml. The other changes I would have to defer to @JhanSrbinovsky and @NortonAlex for precisely which files to grab, but the config will run without those updates I think.

[JhanSrbinovsky]

We have new "pft", "pft lookup" files you can find here:

/scratch/public/jxs599/blimlim/

We are working on a restart to go in as well with an updated PFT map, resurrecting some patches of dirt to hopefully increase the dust picked up from Aust.

[blimlim]

We are working on a restart to go in as well with an updated PFT map, resurrecting some patches of dirt to hopefully increase the dust picked up from Aust.

@JhanSrbinovsky, @rml599gh do you know if these PFT and restart changes will impact the model performance/optimisation? I.e. Will they be needed in the configuration that @manodeep uses for the optimisation work?

[JhanSrbinovsky]

very very very unlikely. the fractions of some bare-ground tiles (only in Aust) are increased. The numbers in the memory locations may change, but there are unlikely to be any extra memory grabs.

Also, this will likely be ready/tested by the end of today :)

Samme for the PFT changes BTW

[JhanSrbinovsky]

@blimlim - sorry there is an updated PFT lookup table:

/scratch/public/jxs599/blimlim/pftlookup_cable3_OctB_update.csv

IF you choose to use this extended name, we re going to need to change the name of the file that cable.nml is pointing to. Ridiculous perhaps

currently cable.nml is pointing to

 casafile%cnpbiome='INPUT/pftlookup_cable3_1213_exp6_c4crop.csv' 

The PFT params file in here is the right version.

The bare-ground fraction was updated in a restart from year 510 in the Oct-B spinup and is running in a parallel run now upto ~670 years at:

/g/data/p66/jxs599/ESM16/PAYU/Dev/Oct-B-combined

where -combined refers to combined PFT stuff from above and new increased dirt restart. The PFT params/lookup changes were being assessed separately in AMIP runs but I was using the wrong lookup table - so they are currently running again. Its almost unthinkable that these won't be adopted and given there is no appetite to pursue altering the veg map further - I guess that one is baked in as well.