Scheduled Downtime
On Tuesday 24 October 2023 @ 5pm MT the forums will be in read only mode in preparation for the downtime. On Wednesday 25 October 2023 @ 5am MT, this website will be down for maintenance and expected to return online later in the morning.
Normal Operations
The forums are back online with normal operations. If you notice any issues or errors related to the forums, please reach out to help@ucar.edu

No problem with hybrid, but a problem with branch

marcinkupilas

Marcin Kupilas
New Member
Hello

I would be very grateful to get some help with a branching issue I'm having. I have been managing to successfully "branch" simulations and make modifications via the user namelists, but only when running a hybrid run. I essentially follow the exact same procedure as for a branch (also making changes to the dates), but set the RUN_TYPE=hybrid.

For the equivalent branch run, when I set RUN_TYPE=branch, the model builds fine but fails on execution.

I describe the procedure in more detail below. I'm not sure what attachments are helpful, so I have attached my cesm.log, cpl.log and atm.log files, env files, diff output between the env files of the branch run that fails and the hybrid run that works, bashrc config, and submission script. Please let me know if any information is missing.

I am running cesm 2.2.0 on ARC4 at the University of Leeds

1. Create a new case using compset FWmaHIST and resolution f19_f19_mg17
2. Copy user_nl_cam, source mods, metal chemistry mods over from the REFCASE to the new case
3. Make changes to history output in user_nl_cam, and to CAM_CONFIG_OPTS to include metal chemistry in env_build
4. Copy entirety of relevant archived restart directory of REFACSE to run directory of new case
5. ./xmlchange RUN_REFCASE=f.e22.FWmaHIST.f19_f19.arc4.003,RUN_REFDATE=2006-01-01,RUN_STARTDATE=2006-01-01,RUN_TYPE=branch
6. ./case.setup, ./case.build
7. I submit the case using the attached script CASE.run (as I have trouble with using ./case.submit directly)


The above procedure works for hybrid runs, and I have been able to reproduce simulations bit-for-bit with others ran in the group. I'm confused however as to why doing a plain branch run does not work.

Best wishes
Marcin
 

Attachments

  • files.gz
    649.6 KB · Views: 7
Top