ACCESS-ESM1.6 development

CMIP7 CMIP7 development
202

Minutes from today’s standup, please correct/amend.

@RachelLaw @inh599 @Jhan I flagged a few spots where it would be great if you can fill in the blanks. (I gather @inh599 has already done this.)

See you all next week. And for the lucky few, enjoy AMOS!

203

Could I please add this to the agenda for next Tuesday:

  • Updating to MOM5 built with CMake (some context here)
204

Could I please add an item for today’s agenda:

  • Updates on oneAPI compiler switch - details in a table at the top comment in this PR
205

Minutes from today’s standup, please correct/amend.

Two small requests

  1. @RachelLaw (anyone) can you please add the discussed 4xCO2 link if you can find it.
  2. @paulleopardi I missed one of your comments please clarify.
206

@blimlim - payu (from */release/ , or */prerelease/) doesnt like “access-esm1.6” ?

File “/g/data/vk83/apps/base_conda/envs/payu-1.1.6/lib/python3.10/site-packages/payu/experiment.py”, line 165, in init_models
ModelType = model_index[m_config[‘model’]]
KeyError: ‘access-esm1.6’

207

Hi @Jhan. Based on the /g/data/vk83/apps/ path it looks like this would be the release version of payu. ESM1.6 is currently only working with the prerelease payu, which includes the required driver updates.

208

hi @blimlim, what I mean is that it doesnt work all the way with either. the model: in the config.yaml users access-esm1.6. It seems to need cloning with release/ but then you have to switch modules to prerelease/.

209

Ah interesting, do you mean cloning the dev-preindustrial+concentrations branch from the github repository?

210

Si. clone with release. unload/load execute with prerelease.

211

Hi @Jhan, could I get you to paste the commands that are leading to the issue?

It looks like it should work when using the prerelease module the whole way through:

module use /g/data/vk83/prerelease/modules
module load payu
payu clone https://github.com/ACCESS-NRI/access-esm1.6-configs -B dev-preindustrial+concentrations test-run
cd test-run
payu run

Thanks!

212

Thanks @blimlim. Same. Except I call the branch and directory the same name.

e.g.

payu clone GitHub - ACCESS-NRI/access-esm1.6-configs: Standard ACCESS-ESM1.6 configurations released and supported by ACCESS-NRI -B dev-amip -b ESM1.6_AMIP_2 ESM1.6_AMIP_2

I tried with prerelease yesterday and it worked once and then failed once. I just tried it again and all 3 worked with pre-release all the way - However I have a new problem. I started the run in ESM1.6_AMIP_1 and considered I wouldn’t need to specify -n XXX as it is probably configured to run for the length of the forcing AND if it only ran a year then I would just restart it from where it left off. But it crashes at dt=2.

I am happy to just start again in ESM1.6_AMIP_2 but what is approriate length (-n) to match forcing?

213

That’s correct, you should be able to run in separate 1 year chunks (in fact the run length needs to be 1 year for the land use change script to work). I’ll double check the dev-amip branch to see if it’s working.

Did you happen to change either the restart file or the land use change file? A crash at the start of the second year could suggest a problem when the land use change is applied to the restart file at the end of the firs run.

214

@Jhan and @spencerwong could I ask for 2 things:

  • do you mind if I move the question and resolution discussion to a separate topic to avoid email spamming all the people in the esm16-dev group? This is developing in more than a quick “1 question, 1 reply” type of problem.
  • @Jhan , when you ask for help, if you could make an effort to explain more clearly what the problem is and what you are doing that results in the problem that would be great. That would avoid us having to spend time in back and forth with you to try and get the details needed. Information like copy/paste of the commands used, the path you are working from etc. is always useful.
1 Like
215

Minutes from today’s standup, please correct/amend.

@MartinDix @paulleopardi I flagged two spots where it would be great if you can fill in the blanks.

216

I should have said during the meeting that the ACCESS-NRI working groups’ projects have an SU allocation. Please consider using these for the ESM1.6 development and evaluation while CSIRO figures out its allocations.

217

Re: p66 compute. I just got this message from NCI (Daohua Bi)…

Compute allocation for p66 at NCI

1 July at 10:25 AM
Dear all,
You may have noticed there is no compute resource allocated for p66 (and other CSIRO projects) this quarter. We believe this is a confusion due to the new IM&T structure that kicks in today. No one has provided any information about the change we are facing in this quarter. I hope this is just a temporary chaos and will be solved shortly, and we can still get the resource we need as before.
Cheers,
Dave