CESM3 Land Spinup

[wwieder]

Description:
Land spinup from CPL_HIST files for use in CESM3 contributions for CMIP7 simulations.

Let's use this issue to track everything needed for our spinup case(s)

• CLM cases from:
  • master (up to 5.4.055) &
  • wetlands branch (needs to be updated to master, above)
  • 2026/9/1 @slevis-lmwg merging ctsm5.4.054 to the wetlands branch: Tag wetlands.n04.ctsm5.4.054.
• CPL_HIST DATM files from CESM_dev 377 or 392 @dlawrenncar will advise:
  • 2026/9/1 CESM Proj. Meeting: Start from the end of 377.
  • see /glade/derecho/scratch/hannay/archive/b.e30_alpha09d_m.B1850C_MTso_Gris_Marbl.ne30_t233_wgx3.377/cpl/hist
  • datm_yr_start and _end 0132 to 0151 would avoid a jump in Nino34 SSTs.
• NDEP, aerosols and ozone from a MTt4s run, @tilmes please advise.
• glc coupler history files are set to be written @gunterl
  • user_nl_cpl histaux_l2x1yrg = .true.
• Get @slevis-lmwg or @wwieder added to special queue? @cecilehannay, can help?
  Sam L. and Will have been added to CESM0023.

We'll hopefully aim to start these simulations soon after the machine comes back on line.

---

Case directory:
Locally:
/glade/work/$USER/$TAG/cime/scripts/$CASE

---

Sandbox:
Locally:
/glade/work/$USER/$TAG

---

user_nl_ changes:

---

SourceMods:

---

Diagnostics:
Diags (if available)

https://webext.cgd.ucar.edu/I1850/$CASE/lnd/

---

Output:
Output (if still available):
/glade/scratch/$USER/archive/$CASE/

---

Contacts:
@slevis-lmwg
@wwieder
@olyson

---

Extra details:
Planned lengths of simulations
AD 400 yrs: X% of the land sfc not in eqm
SASU 240 yrs based on my past records, which originated from Daniel's guidance: Y% of land not in eqm
pSASU 160 yrs: Z% of the land sfc not in eqm

[wwieder]

@gunterl, can you clarify what you mean with the request that "coupler history files are set to be written"

[cecilehannay]

Get @slevis-lmwg or @wwieder added to special queue? @cecilehannay, can help?

=> @dlawrenncar can add you to the special queue. I will add you to the chat where we discuss which runs get a spot in the special queue.

[gunterl]

@wwieder - I mean to set histaux_l2x1yrg = .true. in user_nl_cpl when you set up your case.

[wwieder]

I'm moving this to a discussion, as that's that it really is. Also noting that:

• @klindsay28 will look at ENSO characteristics to identify a period from 377 that looks appropriate for us to loop over for CPL_HIST forced runs
• I'll run LDF to check for stability in DATM fields over the 169 years that have been run.

[wwieder]

See also longer land timeseries here

[klindsay28]

CPL_HIST files are not available in 377 in years 0152-0169. (It's a long story why.)
The SSTs in the Nino3.4 region on dates 01-01-0132 and 01-01-0152 are very close.
So running with CPL_HIST over the 20-year span 0132 to 0151 would avoid a jump in Nino34 SSTs.

[wwieder]

I'm still a little confused on checking to make sure we're getting NDEP and aerosols from the right place in our CLP_HIST case.

Big question: Can we get NDep, aerosols and CO2 from the files here?
/glade/derecho/scratch/hannay/archive/b.e30_alpha09d_m.B1850C_MTso_Gris_Marbl.ne30_t233_wgx3.377/cpl/hist/*.cpl.hx.atm.24h.avrg*

Detailed questions:
I checked the atm_in file in @cecilehannay's 377 MTso run, which look like it's reading in NDep from this climatology:
/glade/derecho/scratch/shawnh/Forcings_tmp/20260706/b.e30_alpha09b.B1850C_MTt4s.ne30_t233_wgx3.360/atm/proc/NDepositionCyclic1850/fndep_clm_MTt4s_3Dmonthly_1850climo330_c20260706.nc.

@tilmes I'm assuming this is the most recent MTt4s run we should be getting data from? Should we be using NDEP from the CPL_HIST files or from Shawn's Forcing_tmp directory direcly?

It also look like aerosols (black carbon, organic carbon, dust) and CO2 are coming in the daily *.cpl.hx.atm.24h.avrg* files.
I'm also assuming that the MTso model calculated the aerosol fluxes, or does this require inputs different from the MTt4s run? If we need information from MTt4s, what files should we be using?

[tilmes]

For NDEP, yes, that is correct, it is from the latest MTt4s model run.

[wwieder]

Thanks. Do you have any input on the aerosol question, @tilmes?

[cecilehannay]

Note that we were originally planning to include the NDep files on the trunk (see ESCOMP/CAM#1665). However, since these are not yet the final versions of the files, @tilmes recommended keeping the NDep changes in user_nl_cam for now. I am fine either way.

[wwieder]

A quick look at global DATM input suggests shows a bit of dip in TBOT and FLDS between years 110-140.

This may correspond to the slushy period seen in the Lab Sea plots (but somewhat difficult to estimate)?

Talking with @olyson and @slevis-lmwg we're inclined to use roughly the last 20 years from 377 (150-169) for the land spinup, contingent on suggestions regarding ENSO state and SSTs during this time from @klindsay28's ocean analysis.

[klindsay28]

Note that CPL_HIST files are not available in 377 in years 0152-0169. (It's a long story why.)

[ekluzek]

The test case I setup to test our setup is here:

/glade/work/erik/ctsm_worktrees/post5.4/cime/scripts/test_spinup_AD

[ekluzek]

The wetlands branch tag to use for spinup is: branch_tags/wetlands.n05.ctsm5.4.054
The master branch tag to use for spinup is: branch_tags/cdeps_updates_for_cplhist.n01.ctsm5.4.054

[wwieder]

Most of the fields we're reading into our CPL_HIST I-case in #178 look as expected (Diags here). There's a slight difference in the winds, but not enough that I'm overly concerned.

Note, previously we'd been worried that NDEP wasn't identical between the two cases, but on further investigation the global fluxes are the same for both cases ~9.3 Tg/y the the apparent difference was basically just roundoff diffs between the B-case and the CPH_HIST forced run.

[wwieder]

Maybe one of the bigger differences is in 1850 Ndeposition rates. This link shows diags from the same CPL_HIST case, #178 vs. the CTSM5.4 release tag, #123. This reflects Ndep differences in CESM3 (9.3 Tg/y) vs. CESM2 (18.8 Tg/y) . Is this expected @tilmes?

[gunterl]

The Resulting Land Ice spin-up given the spun-up Land SMB on top of Greenland yields on acceptable results within the range of regional models MAR and RACMO typically used for comparison. The aggregated SMB in this new spin-up is slightly lower compared to the one used initially in the long 377 run. It results in a slight decrease in caving flux by ~10 Gt/yr.

The new spin-up file is located here:
/glade/campaign/cesm/development/liwg/leguy/CESM3/CESM3_input/GrIS/input_for_BGrun377_spinup/gris_init_377.cesm.comp_spinup.nc

[ekluzek]

@gunterl to be clear above, do you mean "Greenland yields an acceptable result" or "Greenland yields unacceptable results"? I was worried, because at first I thought it was the second and we'd need to make changes. But, I know realize it may be the former, and we are fine.

[gunterl]

Sorry for the typo, we are fine for now.

…

On Thu, Sep 10, 2026 at 2:24 PM Erik Kluzek ***@***.***> wrote: @gunterl <https://github.com/gunterl> to be clear above, do you mean "Greenland yields an acceptable result" or "Greenland yields unacceptable results"? I was worried, because at first I thought it was the second and we'd need to make changes. But, I know realize it may be the former, and we are fine. — Reply to this email directly, view it on GitHub <#176?email_source=notifications&email_token=AB6QV6A2HWH2LX6OEQ3GWSD5OMEXBA5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCOBTHEYTMNZUUZZGKYLTN5XKO3LFNZ2GS33OUVSXMZLOOSWGM33PORSXEX3DNRUWG2Y#discussioncomment-18391674>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AB6QV6DXS5C4YJ7ASBZ7KF35OMEXBAVCNFSNUABIKJSXA33TNF2G64TZHM3DQMRWG4ZDSMRZHNCGS43DOVZXG2LPNY5TCMBXGM3DANRZUF3AE> . You are receiving this because you were mentioned.Message ID: ***@***.***>

-- Gunter Leguy, Ph.D Scientist NSF National Center for Atmospheric Research cell: (575) 418 1021 desk: (303) 497 1790