Matt Long (Jan 14 2021 at 14:37): Matt Long (Jan 14 2021 at 14:37):@Michael Levy, I think we should run a 1° companion integration identical in all init and forcing respects to the high-res.
Michael Levy (Jan 14 2021 at 14:39):that would be an interesting case to compare to the 0.1 degree run. I'll look into what would need to be done
Matt Long (Dec 02 2021 at 17:14):@Michael Levy, have you done anything on this 1° analog?
cc @Taka Ito
Michael Levy (Dec 02 2021 at 17:20):I haven't yet - looking back at my notes, I think I was aiming to have a year done by mid-February and even created a case (/glade/work/mlevy/hi-res_BGC_JRA/cases/g.e22.G1850ECOIAF_JRA_HR.TL319_g17.004_companion), but haven't run case.setup yet let alone tried to figure out comparable initial condition files and namelist settings
Michael Levy (Dec 02 2021 at 17:20):If this is a priority, I can definitely find some time in the next week or two
Matt Long (Dec 02 2021 at 18:09):Thanks @Michael Levy. I think this could actually be really helpful for some of the analysis @Kristen Krumhardt is working on. If the majority of the set up task are already done, it could be worth kicking off the run. Kristen is working on an analysis for a project meeting 14 Dec and I could see this being helpful in that context.
Kristen Krumhardt (Dec 02 2021 at 18:29):good idea. thanks!!
Michael Levy (Dec 09 2021 at 21:07):@Matt Long Kristen, Keith, and I just met and I think the companion run will require a little more work than I naively expected, but we should be able to get it up and running soon. I have a few questions about getting it all set up, though.
First off, the high-res run was a hybrid case that used physics from a run Alper did. He had run 4 61-year JRA cycles, and I don't think there is a comparable one-degree case... so as a first step I was going to run a 4-cycle case (without BGC). Some questions about that run:
There are also three CICE namelist changes (ndtd=2, r_snw=1.00, f_blkmask=true) that I want to ask Dave B about because we aren't sure what they do, and the high res run uses a specific coupler sequence (RASM_OPTION1) that might actually be the 1 degree default but I need to double-check.
While that run is going, I still need to generate the BGC initial condition file... but I think that'll be easy since the high res run just interpolated the tracer state from Kristen's run after 2 JRA cycles - for this run, I'll use the same output but the interpolation won't be necessary.
Also, I promised @Keith Lindsay I would point him to /glade/work/mlevy/hi-res_BGC_JRA/setup_1deg_case.sh, which is the script I'll run to actually configure the companion run after the physics-only JRA cycles are complete.
Matt Long (Dec 09 2021 at 22:58):Thanks @Michael Levy.
I was thinking that you'd simply continue the 1° spin-up integration that you used to generate the BGC initial conditions for the 0.1° case. This would be an imperfect comparison case, for sure.
The configuration details you mention are worth consideration. I am not sure, however, whether it's worth the effort at this time. Perhaps we should hold off on this step and decide once we progress with the 0.1° analyses whether to pursue running this more complete analog case.
Michael Levy (Dec 09 2021 at 23:03):I'm happy to run g.e22b05.G1850ECOIAF_JRA.TL319_g17.cocco.001 for another cycle or two (or three). @Kristen Krumhardt, do you want me to get that run going or would it be easier for you to just continue with a case out of your sandbox?
Kristen Krumhardt (Dec 10 2021 at 14:10):Hi @Michael Levy , I can continue the run. I found the case directory, but I don't have any of the restarts anymore... do you have copies of them somewhere on glade?
Michael Levy (Dec 10 2021 at 15:08):Kristen Krumhardt said:
Hi Michael Levy , I can continue the run. I found the case directory, but I don't have any of the restarts anymore... do you have copies of them somewhere on glade?
The restarts are on campaign -- I think we want to use /glade/campaign/cesm/development/bgcwg/projects/1deg_cocco_JRA/cases/g.e22b05.G1850ECOIAF_JRA.TL319_g17.cocco.001/rest/0123-01-01-00000 (that's the latest year available)
Kristen Krumhardt (Dec 10 2021 at 15:37):Thanks!
Michael Levy (Dec 14 2021 at 18:51):so as a first step I was going to run a 4-cycle case (without BGC)
@Fred Castruccio just stopped by my desk and pointed out that there is a low-res equivalent to the high-res run we branched off of. It's in /glade/campaign/collections/cmip/CMIP6/iHESP/FOSI/LR/, though you to be in the ihesp group to access the data and I'm not in the group yet... so it's not immediately clear to me what is in that directory.
I know for now we're just going to continue with Kristen's run, but Fred is working to get me access so if we do decide to set up a companion case that is more similar to the high-res run, we can take advantage of this companion spinup case existing already.
Michael Levy (Nov 28 2022 at 21:25):@Kristen Krumhardt -- Fred asked that we add TAUX and TAUY to the daily stream, so I created a new SourceMod for forcing.F90 defining TAUX_2 and TAUY_2; I've also updated tx0.1v3_tavg_contents_4p2z to include these in stream 2. You'll need to copy /work2/08815/mlevy/frontera/codes/frontera_highres_scripts/SourceMods/forcing.F90 into /work2/08815/kristenk/frontera/cesm_cases/g.e22.TL319_t13.G1850ECOIAF_JRA_HR.4p2z.001/SourceMods/src.pop and rebuild before submitting the next run
Kristen Krumhardt (Nov 28 2022 at 21:26):Sounds good!
Last updated: May 16 2025 at 17:14 UTC