How does one set up a SSP585-ESM simulation with CESM2

[UllaH]

Hi forum,

I'm interested in running experiments to explore how altered ocean chemistry can influence atmospheric CO2.

I want to do this using ssp585 scenarios.

I have the following compset: BSSP585
SSP585_CAM60_CLM50%BGC-CROP_CICE_POP2%ECO%ABIO-DIC_MOSART_CISM2%NOEVOLVE_WW3_BGC%BDRD

and in env_run.xml, the following settings:
CCSM_BGC=CO2C
CLM-CO2_type=diagnostic
OCEAN_CO2_type=diagnostic


Would this give me an SSP585-ESM run? Or would there be additional changes required?

I find the CO2C option a bit confusing as the description says "this adds prognostic CO2 and diagnostic CO2 at the lowest model level to be sent from the atmosphere to the land and ocean and the surface upward flux of CO2 to be sent from the land and the open ocean back to the atmosphere".

What does it mean that it is both prognostic CO2 and diagnostic CO2? Is it possible to have an experiment without any prognostic CO2? (i.e. everything is calculated by the model based on emissions and fluxes from the ocean/land)?

Thanks in advance,

Ulla


Ulla

 

[Eric]

Hi Ulla,

I am also interested in using ESM-SSP585 compset to do future climate projections. May I ask where did you get this compset?
SSP585_CAM60_CLM50%BGC-CROP_CICE_POP2%ECO%ABIO-DIC_MOSART_CISM2%NOEVOLVE_WW3_BGC%BDRD
I searched it on the compset website while did not see it.
Thank you!

Best
Eric

 

[UllaH]

Hi Eric,

I think it may depend on the model version you are using, but for CESM2.1.3, there is a BSSP585 compset, which is what I used.

CESM2.1.3 Component Sets Definitions

Component Sets Definitions for CESM2.1.3

www.cesm.ucar.edu

Still very much hoping for more insight into my original question if anyone knows!

Thanks again,
Ulla

 

[erik]

I'm not quite sure what you are asking. The compset you list is a SSP5-8.5 future scenario.

Note, that CESM2 experiments are listed here:

CESM2 Experiments, Data & Diagnostic Output

Multi-decadal model runs of given component sets at the target resolution, followed by scientific review of the model output diagnostics

csegweb.cgd.ucar.edu

And the official CMIP6 SSP5-8.5 experiment was b.e21.BSSP585_BPRPcmip6.f09_g17.CMIP6-esm-ssp585.001 which uses the compset:

SSP585_CAM60_CLM50%BGC-CROP-CMIP6DECK_CICE%CMIP6_POP2%ECO%ABIO-DIC_MOSART_CISM2%NOEVOLVE_WW3_BGC%BPR

The main difference between it and the compset you list is that CMIP6 user-mod directories are used to get extra CMIP6 output. It also labels CO2 as prognostic and radiatively prognostic rather than diagnostic. I'm not clear on what settings that changes though, it doesn't change CCSM_BGC, so must impact other control flags such as for CAM.

 

[Eric]

Hi Erik,

I do not find this BPRP compset on the "available compsets" website, and only BDRD (the compset Ulla listed) is provided for users. But I noticed that on the CESM2 experiments website, the caseroot of the official CMIP6 SSP5-8.5 experiment was given as: https://svn-cesm2-expdb.cgd.ucar.edu/public/b.e21.BSSP585_BPRPcmip6.f09_g17.CMIP6-esm-ssp585.001 However, this website is not accessible. Do you know where to find the case under NCAR's storage system? I guess it is under /glade but have no idea which subdirectory I should look into. Thank you!!

I was also wondering, why CESM2 does not make SSP585-BPRP compset publicly available if CESM2 has already used this compset for CMIP6 ScenarioMIP?

Best
Eric

 

[erik]

It looks like that specific case wasn't checked into the svn experiment database. You can see other cases under https://svn-cesm2-expdb.cgd.ucar.edu/public/. I'm not sure why it isn't under svn, but it is available as you point out on NCAR's glade storage system here:

/glade/work/cmip6/cases/C4MIP/b.e21.BSSP585_BPRPcmip6.f09_g17.CMIP6-esm-ssp585.001

You can run with any compset using the long-name format to create_newcase. The ones with short names are the ones that are given the most use and testing, but long-names can be used to get other combinations. I can't say if there was a specific reason this particular case was left off the list, even though there was a simulation done with it.

 

[Eric]

Hi Erik,

Thank you very much! This is super helpful!!

I created a BSSP585-BPRP case by using the long-name format, and then compared it with the one in the glade storage system (I will refer to it as the official case). I noticed two differences:

1. The official case setted "do_circulation_diags = .true." in the user_nl_cam.
2. In the official case's env_run.xml: <entry id="CLM_NAMELIST_OPTS" value="use_init_interp=.true.">
But in my case's env_run.xml, "CLM_NAMELIST_OPTS" does not have a value.

Do you have any clue on why they did these settings? Do I have to also make these settings?

Thank you very much!

Best
Eric

 

[erik]

The cllm namelist item is just saying to interpolate initial conditions from the input file. That's something needed if the input file isn't an exact match to the case being run.

I'd lookup the CAM namelist item from the namelist definition file in the component/cam/bld/namelist_files directory.

I'd also suggest doing a comparison of the namelists in the CaseDoc directories for each. That might point out other namelist items that are different. There will likely be a ton that have to do with history fields, so it might be difficult to swim through all the differences, but it might also show something important. I expect that it should somehow show that CO2 is prognostic rather than diagnostic and that must show up in the namelists in some fashion. I know what that would be for CLM, but POP and CAM must also have something that triggers it as well.

 

[Eric]

Thank you very much for your help, Erik! :)

Best
Eric

 

[Eric]

Hi Erik,

Do you know how I can find the codes to calculate the surface downward solar radiation at each time step at each grid cell in CAM6.0? I posted this question in the Atmosphere module while nobody answers. I would appreciate it if you could give me some hints.

Best
Eric

 

[UllaH]

Hi again,

Returning to my original question. Ive now run a few years with this compset:
BSSP585
SSP585_CAM60_CLM50%BGC-CROP_CICE_POP2%ECO%ABIO-DIC_MOSART_CISM2%NOEVOLVE_WW3_BGC%BDRD

I've further set up a modified run where I've changed the CO2 emission file:
/Users/ullaklintheede/Downloads/emissions-cmip6_CO2_anthro_surface_ScenarioMIP_IAMC-REMIND-MAGPIE-ssp585_201401-210112_fv_0.9x1.25_c20190207.nc

However, when I look at the output ATM_CO2 in pop, the two model runs have identical CO2 concentrations.

This makes me think that the model is using a prescribed CO2 concentration rather than relying on the emission files?

How can I change this?

 

[katec]

Did you make a new case that is a branch or hybrid run referencing your first case? CESM will not update forcing or history streams during a regular CONTINUE_RUN.

 

[UllaH]

Hi katec,

My first run is a hybrid run

Ulla

 

[katec]

Your first run can be a hybrid run, but you need to make a new run for the model to use a different CO2 forcing file. The case with the changed CO2 emission file should be a different case than the first one, and the new RUN_TYPE should be a branch or hybrid run. The new case should set the REF_CASE to be the old case with the constant CO2. For more information on branching and changing settings during a run, read this: 5. Running a Case — CIME cime5.6 documentation

 

[Eric]

Hi Kate,

I met a similar problem with Ulla and have posted it in the CAM section. Ulla, you are supposed to use a BPRP case rather than BDRD. A BDRD case uses prescribed CO2 concentration while a BPRP case calculates CO2 concentration.
I used a BPRP case to carry out experiments to change land carbon uptake, however, the CO2 concentration is still identical to the pre-experiment condition, as Ulla said. That means even a BPRP case still uses prescribed CO2 concentration. So I was wondering where we can find the codes calculating atmospheric CO2 concentration by emission, and land and ocean carbon fluxes in the CESM2 model.

Thanks.

 

[UllaH]

Hi Eric -

Thanks for this message, so if I understand correctly, you are using a BPRP case rather than a BDRD case - but the results are the same (i.e. the concentration is still prescribed rather than calculated?) So you still have the same problem as me?

@katec, thanks! Just to clarify - each model run I've set up with different CO2 emission files are separate model runs. They each use the same ref case as the standard SSP585 run.

Ulla

 

[Eric]

Yes the CO2 concentration is prescribed rather than calculated under BPRP, although it is not supposed to.

 

[UllaH]

Just bumping this - in case anyone has any answers for how the model can be set up to calculate CO2 rather than have a prescribed concentration? Seems that both Eric and I are stuck with this.

Ulla

 

[klindsay]

As UllaH pointed out, the longname for the BSSP585 compset is
SSP585_CAM60_CLM50%BGC-CROP_CICE_POP2%ECO%ABIO-DIC_MOSART_CISM2%NOEVOLVE_WW3_BGC%BDRD
The BDRD at the end of this longname describes how CO2 is used in the experiment.
The BD indicates that the biogeochemistry modules in the surface components use diagnostic (i.e. prescribed) CO2.
The RD indicates that atmospheric radiation computations use diagnostic (i.e. prescribed) CO2.

If you replace BDRD in the compset longname with BPRP, then the biogeochemistry modules in the surface components and the atmospheric radiation computations will both use prognostic (i.e. computed) CO2. It sounds like this is what you want.

I've attached some slides from the BGC practical session of the CESM tutorial that (somewhat tersely) explains this. In particular, please see slide 3.

Note that I advise against initializing a BPRP experiment from a BDRD experiment. The atmospheric CO2 tracers in a BDRD compset are usually out of balance with the carbon cycle. So you would get large transients in carbon fluxes at the beginning of your experiment. You should initialize using a reference case that was BPRP. The inputdata repository has restart files for 2015-01-01 from b.e21.BHIST_BPRP.f09_g17.CMIP6-esm-hist.002 that suffices for this.

At one point we had submitted BPRP SSP585 results from CESM2.1 to CMIP6. However, a bug was found in the forcing and the results were retracted. The experiment has not yet been rerun.

I think the absence of this compset in the CESM 2.1 release series is just an oversight. I don't recall having a reason to intentionally omit this compset.

 

[UllaH]

Hi Keith,

This is exactly the information I need!

I will set up experiments with BPRP and let you know how I get on-

Thanks again, this is very helpful

Ulla