Running a Restart/Hybrid simulation

[pgoddard08]

Hello,

I am attempting to run a hybrid simulation using restart files that I previously created. Below is the relevant code in the run script:

# Run options
readonly MODEL_START_TYPE="hybrid"  # 'initial', 'continue', 'branch', 'hybrid'
readonly START_DATE="0069-02-01"

# Additional options for 'branch' and 'hybrid'
readonly GET_REFCASE=TRUE
readonly RUN_REFDIR="/N/project/MSB_Project/E3SMv2.0_data/0054-0073_Restarts_v24_code_1xCO2/v2.0.LR.F2010_restart_0069-01-01-00000_1yr_v24_code_1xCO2_20250529/run"
readonly RUN_REFCASE="v2.0.LR.F2010_restart_0069-01-01-00000_1yr_v24_code_1xCO2_20250529"
readonly RUN_REFDATE="0069-02-01"   # same as MODEL_START_DATE for 'branch', can be different for 'hybrid'

The RUN_REFDIR contains monthly restart files for the year 0069 (ie, contains restart files for 0069-02, among others).

The simulation compiles, but only runs for 15 secs and no log files are created so I'm not sure where to begin to troubleshoot.

Please let me know if I can provide more information. Thank you!

Paul

[pgoddard08]

The relevant restart files in the RUN_REFDIR are:

rpointer.atm
rpointer.drv
rpointer.ice
rpointer.lnd
rpointer.ocn
rpointer.rof
v2.0.LR.F2010_restart_0064-01-01-00000_1yr_v24_code_1xCO2_20250529.cpl.r.0064-02-01-00000.nc
v2.0.LR.F2010_restart_0064-01-01-00000_1yr_v24_code_1xCO2_20250529.docn.rs1.0064-02-01-00000.bin
v2.0.LR.F2010_restart_0064-01-01-00000_1yr_v24_code_1xCO2_20250529.eam.r.0064-02-01-00000.nc
v2.0.LR.F2010_restart_0064-01-01-00000_1yr_v24_code_1xCO2_20250529.eam.rs.0064-02-01-00000.nc
v2.0.LR.F2010_restart_0064-01-01-00000_1yr_v24_code_1xCO2_20250529.elm.r.0064-02-01-00000.nc
v2.0.LR.F2010_restart_0064-01-01-00000_1yr_v24_code_1xCO2_20250529.elm.rh0.0064-02-01-00000.nc
v2.0.LR.F2010_restart_0064-01-01-00000_1yr_v24_code_1xCO2_20250529.mosart.r.0064-02-01-00000.nc
v2.0.LR.F2010_restart_0064-01-01-00000_1yr_v24_code_1xCO2_20250529.mosart.rh0.0064-02-01-00000.nc
v2.0.LR.F2010_restart_0064-01-01-00000_1yr_v24_code_1xCO2_20250529.mpassi.rst.0064-02-01_00000.nc

[pgoddard08]

I was able to get an E3SM log on my latest try and received this error:
GETFIL: FAILED to get v2.0.LR.F2010_restart_0064-01-01-00000_1yr_v24_code_1xCO2_20250529.eam.i.0064-02-01-00000.nc

I guess I didn't think hybrid runs needed a eam.i file?

[rljacob]

I think hybrid cases do start the atmosphere from an initial file. If you have any initial file from you model version, try just renaming it to what is expects.

[pgoddard08]

Hi Robert,

I do have an initial file from the start of the year --- I guess I just want to make sure that using a eam.i file from 0064-01-01 for hybrid runs starting on the first of each month for 0064 won't cause an issue with the output. (And if I'm reading your comment correctly, I'll rename eam.i 0064-01-01 to eam.i 0064-02-01 through eam.i 0064-12 for my simulations).

Thanks!
Paul

[mt5555]

chiming in to say that I thought branch runs do not use the .i file. branch runs are restart runs that allow the user to change things like start date and I/O, but they restart from restart files, not .i files.

[rljacob]

But a hybrid run is different from a branch run. I think you can only change the start date on a hybrid run.

Paul, this is all from fuzzy memory and some docs (https://ncar.github.io/CESM-Tutorial/notebooks/modifications/xml/modify_run_type/hybrid_branch_restart.html) . The original reason for a hybrid start was to take restart files from long ocean-only and land-only spinup runs and combine them with an atmosphere initial condition to start a new simulations. Because those are usually long simulations, the atmosphere initial condition doesn't really matter.

[pgoddard08]

Great - thank you all for your help and information. The link to the run types on the CESM-Tutorial was very helpful. I think my question has been answered. Cheers