CSIRO - ACCESS-NRI standup minutes

CMIP7 CMIP7 development
34

Meeting minutes

Date: 29/04/2025
Attendees: @tiloz , @clairecarouge , @MartinDix , @Jhan , @Aidan , @pearseb , @manodeep , @micael , @inh599 , @dougiesquire , @harshula , @anton , @paulleopardi , @RachelLaw , @alexnorton

Spinup update:

  • Jhan: latest run has crashed this weekend. Waiting to know why and make sure the interpretation of the error message is correct. It’s the first crash in 350 years, showing a stable simulation.

    • Martin: It likely just needs a restart perturbation.
    • Aidan: No point trying to chase that down now, there is also less people available to help currently. No point of worrying too much about it unless it reoccurs. Payu will save the PBS information anyway.
    • Tilo: there was a conversion issue previously, but it was likely due to an NCI issue. It was manually fixed.

  • Ian: On the status of evaluation, the surface salinity seems to equilibrate now in the runs.

    • Jhan: it looks like the surface salinity climbs up again in the run that kept going. So we need to still keep an eye out for the surface salinity.

CICE5:

  • Anton: CICE5 running thanks to Dave. Ran for 50 years and it looks ok. Dave, Anton, Siobhan and Martin are to look at the results in more detail tomorrow, anyone is welcome to join.
  • Can’t use CICE4 restarts with CICE5. A short-lived hit is expected to the water balance when we switch over. Is there a water balance script available to check? Martin has a script for CM3, which could be adapted. Anton to figure it out since no script seems to be available for the ocean water balance.
  • Do we want to move to CMOR variable names in CICE5 and MOM? Seems a good idea but it would break user scripts. Manodeep suggests using aliases in the netcdf file so the data can be accessed using the old or the new names. Unsure if that’s possible in netcdf (possible in hdf5) and that would not be supported by the models directly, it would need to be a post-processing step. Tilo: It’s best to keep consistency in the output for the current spinup so it’s easier for evaluation. Only change for the next update to the spin-up.
  • Anton will still update the configuration so it’s ready to pick up when we want to.

Other discussions

  • Ian:

    • Reminder for Pearse and Jhan to update the spinup experiment page on the forum.
    • ILAMB workflow: need to meet with claire by the end of the week to get this going.

  • Pearse:

    • Ocean BGC update seems to have worked as expected. Fluxes look much better.
    • Needs help to update the diagnostics table to add some variables. Dougie to help.

  • Ian:

    • Thanks ACCESS-NRI for the work on fixing the output post-processing. It is now working very smoothly.

Actions:

1 Like