CAM-CLM tsi and other forcing inquiries

[wvsi3w]

Hello,
I am working with CESM2.1.3, and I have done several simulations with CLM5.
I have asked a question related to using different TSI and the related process that I have asked here:

Guidance on Running with LME datasets with CESM2.1.3 (CLM5)

Unfortunately, we can't provide any specific answers since we don't have any experience with this type of workflow. Our suggestion is to experiment with a short sequence of simulations, e.g., just a few years for each segment, to see how the workflow should be implemented.

bb.cgd.ucar.edu

And I have been referred to this CAM forum.

My questions are these:
1- How to use a different TSI file for this path inputdata/atm/cam/solar/ ? Is there a way to assign it in the namelist options, and how? I found that path in my IHIST case, which is weird because I can't find it in I1850 case! And as Sam said, this shouldn't be there if it is for land compsets. So, if I want to run a fully coupled or CAM-CLM case, how should I assign my tsi file (should I replace the existing file with mine by "overwriting" it? or is there a way to refer to it in the user-defined namelists?)
2- And how to use a different volcanic file
I have found these paths in here, but I dont know how to replace them:
(inputdata/atm/cam/physprops/
inputdata/atm/cam/chem/emis/CMIP6_emissions_2000climo/)
Should I just overwrite them with mine?

Thank you for your support

 

[peverley]

Hi,

A couple thoughts that might help you.

1. If you want a fully coupled CAM-CLM case, you'll want to use an F compset rather than an I compset (e.g. FHIST or F1850). "I" compsets are going to give you use a data atmosphere that you won't be able to configure the way you want.

2. Once you create a new case with an F compset, you can override the default tsi file with the following namelist in user_nl_cam:

solar_irrad_data_file = '<full path to file>'

3. What specific volcano file are you looking to replace? If you let me know, I can point you to more namelist options. Alternatively, if you create an F compset case and then run:

./case.setup
./preview_namelists

And then inspect the atm_in file in your run directory, you should be able to find the file that you are wanting to replace and use that namelist option in user_nl_cam.

4. FYI: You will find that F compsets run significantly slower than I compsets.

Hope that helps,
Courtney

 

[wvsi3w]

Hi,

A couple thoughts that might help you.

1. If you want a fully coupled CAM-CLM case, you'll want to use an F compset rather than an I compset (e.g. FHIST or F1850). "I" compsets are going to give you use a data atmosphere that you won't be able to configure the way you want.

2. Once you create a new case with an F compset, you can override the default tsi file with the following namelist in user_nl_cam:



3. What specific volcano file are you looking to replace? If you let me know, I can point you to more namelist options. Alternatively, if you create an F compset case and then run:

And then inspect the atm_in file in your run directory, you should be able to find the file that you are wanting to replace and use that namelist option in user_nl_cam.

4. FYI: You will find that F compsets run significantly slower than I compsets.

Hope that helps,
Courtney

Dear Courtney,
Thank you so much for your response, and I am sorry if I replied this late I am constantly dealing with multiple ideas from my supervisors so I couldn't answer you earlier.

I did try this:
create_newcase --case /MYCASES/CAM_CLM_test --res f09_f09_mg17 --compset F1850 --run-unsupported
./case.setup
./preview_namelists

And I see what you mean, thank you. Another good thing is that I have access to all of the files from my source.
This is the list of the files:

• fracdata_1.9x2.5_tnx1v1_c120120.nc
• ghg_meinshausen_0-2014_annual_c210324.nc
• griddata_1.9x2.5_060404.nc
• ozone_1.9x2.5_L26_1850clim_c091112.nc
• pft-physiology_c110425.nc
• rdirc_0.5x0.5_simyr2000_c101124.nc
• snicar_drdt_bst_fit_60_c070416.nc
• snicar_optics_5bnd_c090915.nc
• surfdata_1.9x2.5_simyr1850_c091108.nc
• USGS-gtopo30_1.9x2.5_remap_c050602.nc
• CCSM4_volcanic_forcing_0001-2000_ens01_eruption_month04.nc
• SOLAR_TSI_SATIRE_0-2015_annual_c080424.nc
• CCSM4_volcanic_forcing_0001-2000_ens02_eruption_month04.nc
• SOLAR_TSI_Shapiro_0-2016_annual_c080424.nc

I know that if we chose F1850 compset, we would probably have CO2 and aerosol being constant to the year 1850, right? if so, then I dont need the ghg_meinshausen file here (second file in the list). And I am not sure if I need to assign the USGS topography (last file in the list above), as I think all compsets have their own topo file ready (?)

A brief of what my objective is:
I have two sets of forcings, each of which has different solar variability. I already ran a land-only simulation using these datasets, but since in the land-only CLM5 simulation, you only need 7 variables to run (TPHWL, Solar, Precip) then I didn't need to go through all of these files (list above) in my land-only simulation. The forcings were two-degree resolution; I managed to make them one degree, and I made my own domain file for the land-only that I used successfully. The period of simulation is from year 500 to year 2000, with a monthly temporal resolution, and in my results from the land-only simulation that I have finished I was looking at several things like permafrost extent, subsurface thermal state, heat content, methane and soil carbon, etc, and I want to see what would be the results if I have actually used coupled simulation that includes the TSI at the top of atmosphere, and other forcings listed above, because land-only didn't produce significant change at the end and thats because the solar forcing that land-only simulations need is at the bottom layer and it has less variation compared to the actual TOA file. I have also uploaded the cam and clm namelist that my source has used in their own simulation for your reference.

So, based on what I explained, what compset do you think best suits my simulation? I guess it would be F1850 for 500 to 1849, then FHIST for 1850 to 2000? What if we chose fully coupled (B) compsets?

Which one of the files from the list above is needed? I am certain of the last four (green colors).

 

[peverley]

Hi,

I consulted with some folks in the paleo group and this is basically what I've got in terms of advice:

1. We are not aware of any Last Millennium simulation that has been done with CESM2(CAM6). Only CCSM4, CESM1, or CESM2(WACCM)*. So we'd recommend using CESM1, and then you can compare results with the CESM1 Last Millenium Ensemble, and forcing files are all available. If you have to use CESM2, then you have to develop a new volcanic forcing dataset for CAM6, which would take some time. If you have access to derecho, you can find some runs/resources for CESM3 Last Millennium here: /glade/campaign/cesm/development/palwg/LastMillennium/setup/runs

2. Unfortunately, CESM1 is not fully supported so I can't help you too much further. If you have further questions, you can try the Paleo forum

Sorry I couldn't be of more help.
Courtney

* For the CESM2 historical simulations, WACCM simulations with volcanic emissions were run first and then used to derive a volcanic aerosol dataset for CESM2 (see Section 5 in Gokhan's paper (Home Page10.1029/2019MS001916). Mike Mills, ACOM, produced this dataset. He retired several years ago.