clairecarouge
(Claire Carouge, ACCESS-NRI Land Modelling Team Lead)
2
6/11/2024
Apologies: Tammas, Arnold Agenda:
What are stand-up meetings (Aidan)
Schedule the stand-up meetings (Claire)
Goal for ESM1.6-land development (Tilo)
Current status of development (Rachel)
Upcoming development work (Claire)
Notes:
Get everyone in the room that needs to be there, not everyone required every day. Focus on solving problems. Used to make decisions on blockers and get action items. Put agenda and notes on the forum. Also a status update.
Start with 9.30am on Monday mornings
Changes around the landuse change scheme - improvements in pools initialisation, additional wood harvest.
Bringing changes from CABLE3 and AM3
Bug fixes and documented issues
Changes happening in the atmosphere as well by Harun and Arnold.
Changes to atmospheric coupling
More configurations for ESM1.6
Deployment repository for ESM1.6 - need to update from ESM1.5
Pool initialisation and wood harvest: Tammas working on it, getting it in ESM1.6
CABLE3 harmonisation between all the coupled and standalone versions: Rachel, Jhan
CABLE as a library: Ben
Update CABLE to CABLE3 for ESM1.6 build - get all development to merge into UM7 main.
Blockers:
payu: what is the most effective way to build the code while working with payu. Need more configurations soon (4xCO2).
easier to use SVN UM checkout and git clone CABLE in that checkout to bring CABLE in.
Actions:
add “ESM1.6 FastTrack” milestone to relevant CABLE issues (all)
GitHub project board for ESM1.6 - same board as ocean (Claire)
Create chat for this group to be used for questions to the group (Aidan)
Make 4xCO2 issue in ESM1.6 config repo (Rachel)
Add wiki to document how to run ESM1.6 for developers (Aidan)
Update description of ESM1.6-CABLE3 PR (Jhan)
Send meeting invite (Claire)
Add information of issue labelling in ESM1.6 wiki once Aidan has started it (Claire)
1 Like
clairecarouge
(Claire Carouge, ACCESS-NRI Land Modelling Team Lead)
3
2 build processes: clone spack and building the code locally (documented - see above) or pushing a branch to the repo and using the CD to build the code (to be documented). No recommended option yet. Second option better for sharing. Jhan found second method easier.
Aidan
(Aidan Heerdegen, ACCESS-NRI Release Team Lead)
5
Apologies for missing the meeting. I created the initial dev branch, but I have not closed the issue, as it needs to be updated when the next release of ESM1.5 is out (in a week or two).
clairecarouge
(Claire Carouge, ACCESS-NRI Land Modelling Team Lead)
6
Apologies: Rachel and Tilo (1st aid training) Agenda:
Status of action items
Longer runs to look at (@RachelLaw ) [Have 100 year runs for ESM1.5 like and Medlyn/rootbeta. Testing Haverd2013 but doesn’t look like it will help with NEP seasonality]
How to contribute to ESM1.6 development? (@clairecarouge )
Notes:
Jhan and Spencer to figure out the problem with running Jhan’s branch with ESM1.6 through payu. Once this is working, Jhan is to have a PR for his UM7 branch (18-esm15-cable3) to merge in main. This will be the base for additional changes needed to the interface between UM and CABLE3.
Status on the library: it is important to know where the separation is between the library and the UM.
There is a new --library flag available to build.bash that will compile the science code (plus a few additional required files) into libcable_science.a. This library file, will need to be tested for compilation with the UM.
As per the PR, there are a handful of files from other directories offline/params/util that are USEd by the “science” code, which is why the library build requires them.
Feedback welcome.
clairecarouge
(Claire Carouge, ACCESS-NRI Land Modelling Team Lead)
8
Longer runs to look at (@RachelLaw ) [Have 100 year runs for ESM1.5 like and Medlyn/rootbeta. Testing Haverd2013 but doesn’t look like it will help with NEP seasonality]
get 18-esm15-cable3 UM7 branch working in a payu run (@Jhan@spencerwong )
Contributing to ESM1.6-land (@clairecarouge ): new branch names coming up for UM7. Brand new README with some instructions.
Notes:
When should we have a ESM1.6 build with both changes to land-atmosphere and ocean?
It shouldn’t be a problem for the science to do it now. And in terms of timeline it would be preferable to have all the components now.
Need a default configuration to go with it as generic tracers need new input files.
Should we merge the stand-ups with ocean from next week?
Library: CABLE build can now create a library. Need to add the option to the CABLE spack package. Then need to merge with the interface from Jhan’s work. And then need to test in an ESM build.
CMT: looking good need to incorporate with latest CABLE version
Long runs: options for the root distribution but not really a resolution yet.
Wood thinning inputs: Spencer working with Tammas to get these in vk83
Tasks:
Create default Spack package for ESM1.6 with all the current updates.
Create default configuration for ESM1.6 with all the inputs for generic tracers. Need to check with ocean folks.
Need a permanent dev PR in the deployment repository to correspond to the dev branches in UM7. (model release)
contact Chris about merging the stand-ups for next 2 weeks. (@clairecarouge ) Advice group on what is decided.