Skip to content

2025‐02‐26

Jump to bottom
Teagan King edited this page Feb 26, 2025 · 1 revision

Attendees: Mike Levy, Teagan King, Dave Bailey, Cecile Hannay, Dani Coleman, Brian Medeiros

Walk through CESM workflow: env_postprocessing.xml, case.cupid

Any ADF-specific configuration values we should change?

  • BM: Can user specify different ADF template file? ML: We should add this in.

DC: Can we use a flag in ADF command to run MDTF? We could have default in template file with MDTF off, and give it a flag to turn MDTF on (diag_mdtf_info:mdtf_run). ML: We should also add mdtf_codebase_path since this has a hardcoded path to glade. Or maybe include MDTF in CUPiD externals. BM: We should do the same thing with CVDP. DC: Ideally long term we move both of these up to cupid control level. That said, conda environments are a bit tricky to bring in and Dani's installation is set up to make this easier. So, we decided that we will make a flag to bring MDTF in through ADF.

CVDP: Plan is to get it all into python by end of this year. Temporarily we can add ncl into env_mach_specific.sh. This would be good to run out of externals instead of out of Adam's home directory.

Additional xml variables to set: control case, Cecile could come up with a more detailed list.

CH: Helpful to run CUPiD after every 20 years as a sanity check.

Action Items:

    1. Fix bug: ADF config file specified in case.cupid command for generating ADF config file is using incorrect path.
    1. Make a flag to bring MDTF in through ADF.
    1. Run CVDP out of externals.
    1. Check in on additional XML variables with Cecile (include control case, etc)
Clone this wiki locally