Add F2010 equivalent compset and supporting files for v3atm

[wlin7]

The compset still starts wth F2010-chemUCI-Linozv3 to save it for separate purpose of using chemUCI. The additional settings to complete the fourth full smoke config are done via run script.

A template script for running F2010 equivalent is provided.

[wlin7]

The 2010 specific input files as of now in this PR can be considered placeholders, as they are all single year 2010 data extracted from the 20TR equivalent. Once the input files for 2010 compset are finalized, a new compset (e.g., F2010-V3-4F) will be created, and user_nl settings in the run-script will be mostly for output, plus maybe additional tunings.

[rljacob]

Recall we decided not to put version numbers in compset aliases.

[wlin7]

Recall we decided not to put version numbers in compset aliases.

Thanks for the timely reminder, @rljacob . We will not do that then.

[wlin7]

Thanks for the review, @mingxuanwupnnl . No changes needed for ocean and ice as existing settings applicable for all 2010 compsets. For lnd, if there is a preferred IC, we can change it. For now, the elm.r.2010-01-01 from the fourth smoke test is used. It should be fine in general.

BTW, Mingxuan, if the new data you are preparing for 2010 compset can be available soon, we can update this PR first before merging.

[mingxuanwupnnl]

@wlin7 The new data are almost ready. I updated on the F2010 confluence page (https://acme-climate.atlassian.net/wiki/spaces/ATMOS/pages/3665070019/Creation+of+F2010+Compsets). C2H4, C2H6, C3H8, CH2O, CH3CHO, CH3COCH3, CO, ISOP, C10H16, NO, E90, NO2 files should be under /inputdata/atm/cam/chem/trop_mozart_aero/emis/chem_gases/2degrees/

SOAG0 surface and elevated files should be under /inputdata/atm/cam/chem/trop_mozart_aero/emis/DECK_ne30/

What is left is the so2 elevated emission file. Not sure how we treat volcano eruption. Current F2010 so2 elevated emission file might be good to use and we don't need to change it.

[wlin7]

This is great, thanks, @mingxuanwupnnl . I think the SO2 elevated emission file needs to be updated. It turns out containing high frequency data, even sub-daily for some years/months, including during 2010.

[keziming]

This is great, thanks, @mingxuanwupnnl . I think the SO2 elevated emission file needs to be updated. It turns out containing high frequency data, even sub-daily for some years/months, including during 2010.

@wlin7 the high frequency data is due to eruptive volcanic emission. For PD-PI simulations, A quick fix can be exclude all eruptive volcanic emissions, then this natural factor has no impact. Or keep PD and PI as same as 2010 volcanic emission.

[wlin7]

@keziming , for elevated SO2 emission file from volcano eruption, between the one you created and that E3SMv2 used, only variables 'altitude',, 'altitude_int' and 'contvolc' are common. Are these the only variables used in the model, other than time variables? There are a few other fields in different names.

[wlin7]

@keziming , @mingxuanwupnnl , @tangq , @hwangacme , please take another look of the changes for 2010 input files, just the commit ec4df1f.

For the elevated SO2 emission, the same '2010_clim' file used in V2 is used here.

During the v3atm meeting, I think we intended to use 1850 condition for 2010 compset. Such a file is also ready (not in this PR yet): atm/cam/chem/trop_mozart_aero/emis/DECK_ne30/cmip6_mam4_so2_elev_1850-as-2010_c230218.nc, created by extracting 1850 data from the file for transient AMIP then changed the date to 2010 to as use common cycling year for the forcing.

[wlin7]

Thanks for the careful review, @mingxuanwupnnl . After the commit ec4df1f, only the run.NGD_v3atm.F2010.P10.sh template is applicable. I am going to remove the other one (that was before incorporating your 2010_clim data).

I must have also confused you a little. The use_case file is for the base compset F2010-ChemUCI-Linozv3, and it is not using SOAG0. All the additional settings for 4th full smoke equivalent still go via the run script. We will later put in a new compset to cover all.

[wlin7]

Merged to NGD_v3atm.

See verification simulations and the differences of using 2010 vs 1850 volcano SO2 emissions. (Note: the pair of 'New 10-year simulations')

[oksanaguba]

@wlin7 i am trying to use this capability and i ran into this error

 0:  ERROR: open_trc_datafile: cycle year not found
 4:  open_trc_datafile: cycle year not found :         2010

with res ne4pg2_oQU480 . Could i use ne4 grid here or does it have to be ne30 grid as you listed in tun script (submittied with this PR)? thanks!

Update: this is exactly what i tried

replay.sh:/home/ac.onguba/acme-v3/cime/scripts/create_newcase --case "${CASEDIR}" --res ne4pg2_oQU480 --mach chrysalis --compiler intel --compset F2010_chemUCI-Linozv3

[wlin7]

Hi @oksanaguba , indeed the compset was created having only ne30pg2 in mind. No test was done for ne4pg2. The surprise is that it was complaining about forcing file. I thought it would hit initialization problem first. I feel there is an easy fix for the forcing 'cycle year'. Can you point me to your casedir?

[wlin7]

It appears ext_frc_specifier for the following species pom_a4, so4_a1 and so4_a4 uses 2000 files (ar5_mam3_pom_elev_2000_c130422.nc, ar5_mam3_so4_a1_elev_2000_c120315.nc, ar5_mam3_so4_a2_elev_2000_c120315.nc) that are not what used in the use_case file for F2010-chemUCI-Linozv3. The causes to be investigated as these specifiers are currently grid independent. Thank you @oksanaguba for exposing this problem.

[wlin7]

This problem also occurs to case created for ne30pg2 grid. Currently the runs rely on user_nl_eam in the runscript specify the forcing files correctly -- thus no problem for ongoing tests with the 4th full smoke config.

Similar problem likely also happens to case with standalone F20TR-chemUCI-Linozv[23] if not overidden via user_nl_eam in runscript.

[mahf708]

@wlin7 and @oksanaguba what would it take for me to run this compset with the highest resolution available? Simply changing the resolution (what would it be?) and then finding the corresponding files (or making them)?