Add soil temperature and moisture IAU

[tsga]

Commit Queue Requirements:

• Fill out all sections of this template.
• All sub component pull requests have been reviewed by their code managers.
• Run the full Intel+GNU RT suite (compared to current baselines) on either Hera/Derecho/Hercules
• Commit 'test_changes.list' from previous step

---

Description:

Add Incremental Analysis Update (IAU) to soil temperature and moisture in the NoahMP land model of the UWM's CCPP.

Commit Message:

* UFSWM 
  * FV3
    * ccpp-physics - Add Incremental Analysis Update (IAU) capability to update soil temperature and moisture for CCPP's NoahMP land model

Priority:

• Normal

Git Tracking

UFSWM:

• Closes Add Land IAU Capability to NoahMP Driver in CCPP #2414

Sub component Pull Requests:

• FV3: Add soil temperature and moisture IAU NOAA-EMC/fv3atm#866
  • ccpp-physics: Add soil temperature and moisture IAU ccpp-physics#222
• noahmp: update for land IAU to ccpp-physics noahmp NOAA-EMC/noahmp#12

UFSWM Blocking Dependencies:

• None

---

Changes

Regression Test Changes (Please commit test_changes.list):

• PR Adds New Tests/Baselines.
  Additional land-IAU specific test to be added: control_c48_intel_land_iau (based on the "control_c48_intel" test, with the only difference being the turning on of land IAU and soil temperature increment files inside INPUT directory). The test inputs have been staged in hera platform. The baseline outputs created for the control_c48_lnd_iau_intel regression tests is at:
  /scratch1/NCEPDEV/stmp4/Tseganeh.Gichamo/REGRESSION_TEST_LAND_IAU/control_c48_lnd_iau_intel

Input data Changes:

• New input data.
  soil temperature increment files inside INPUT directory

Library Changes/Upgrades:

• No Updates

---

Testing Log:

• RDHPCS
  • Hera
  • Orion
  • Hercules
  • Jet
  • Gaea
  • Derecho
• WCOSS2
  • Dogwood/Cactus
  • Acorn
• CI
• opnReqTest (complete task if unnecessary)

[tsga]

Tagging relevant people here. Please add (tag) anyone you think should be aware.
@CoryMartin-NOAA, @barlage, @ClaraDraper-NOAA, @CatherineThomas-NOAA, @junwang-noaa

[grantfirl]

@tsga There should be an associated PR for the NoahMP changes in https://github.com/NOAA-EMC/noahmp, right? This component repo needs to have its internal NoahMP code updated to correspond to the changes in ufs-community/ccpp-physics#222. Please let me know if you need help with this or if I am mistaken.

[grantfirl]

@tsga Since there is a new namelist to read in (according to https://github.com/ufs-community/ccpp-physics/pull/222/files#diff-13b994b587cb8a841246d6db7df55ef9ff6fc1269777b9fe1b2e704ce9da629bR138), will an example of this be added to this repository for testing purposes?

[tsga]

@tsga Since there is a new namelist to read in (according to https://github.com/ufs-community/ccpp-physics/pull/222/files#diff-13b994b587cb8a841246d6db7df55ef9ff6fc1269777b9fe1b2e704ce9da629bR138), will an example of this be added to this repository for testing purposes?

@grantfirl yes there is an example namelist with the new Land IAU specific test at:
/scratch1/NCEPDEV/da/Tseganeh.Gichamo/ccpp-physics/test/control_c48_intel_land_iau/input.nml

Should I also add it as part of this repo (as one of the files)?

[grantfirl]

@tsga Since there is a new namelist to read in (according to https://github.com/ufs-community/ccpp-physics/pull/222/files#diff-13b994b587cb8a841246d6db7df55ef9ff6fc1269777b9fe1b2e704ce9da629bR138), will an example of this be added to this repository for testing purposes?

@grantfirl yes there is an example namelist with the new Land IAU specific test at: /scratch1/NCEPDEV/da/Tseganeh.Gichamo/ccpp-physics/test/control_c48_intel_land_iau/input.nml

Should I also add it as part of this repo (as one of the files)?

@tsga @jkbk2004 As of now, if we merged your work, there would be no tests that are utilizing it, so it could be broken with subsequent code changes. You mention in this PR description that a new RT should be added, but the files haven't been modified to add it yet. You'd need to edit tests/rt.conf, commit the new namelist and refer to it in the new test file to be added in tests/tests, and have the new data staged on the testing platforms in order for the test to be set up to be exercised.

[grantfirl]

@tsga Please see tsga#1,
tsga/fv3atm#1, and tsga/ccpp-physics#1

With these changes, I can use the new RT via: ./rt.sh -e -k -n "control_c48_lnd_iau intel" > rt_single_control_c48_lnd_iau.log 2>&1 &

It compiles successfully, although the run fails because the data is not staged. When the data is staged, we can complete testing.

[tsga]

@grantfirl Thank you! I added the increment files to the INPUT dir of the test you set up and it runs successfully (I am working on the other issues of gracefully exiting when there are no files, etc, you mentioned earlier).
In the mean time, can you copy these files (sfc_inc*.nc) /scratch2/NCEPDEV/land/data/DA/soil_moisture/LandIAU/INC/C48/INC09/*
to /scratch2/NAGAPE/epic/UFS-WM_RT/NEMSfv3gfs/input-data-20240501/FV3_input_data/INPUT_L127_gfsv17 (I don't have permission). These six files are the only ones needed to make the control_c48_lnd_iau intel test that you set up complete.

[grantfirl]

@grantfirl Thank you! I added the increment files to the INPUT dir of the test you set up and it runs successfully (I am working on the other issues of gracefully exiting when there are no files, etc, you mentioned earlier). In the mean time, can you copy these files (sfc_inc*.nc) /scratch2/NCEPDEV/land/data/DA/soil_moisture/LandIAU/INC/C48/INC09/* to /scratch2/NAGAPE/epic/UFS-WM_RT/NEMSfv3gfs/input-data-20240501/FV3_input_data/INPUT_L127_gfsv17 (I don't have permission). These six files are the only ones needed to make the control_c48_lnd_iau intel test that you set up complete.

@tsga I don't have permissions in that directory either. @jkbk2004 and/or other members of the UFS code management team can do this for us.

[jkbk2004]

@grantfirl Thank you! I added the increment files to the INPUT dir of the test you set up and it runs successfully (I am working on the other issues of gracefully exiting when there are no files, etc, you mentioned earlier). In the mean time, can you copy these files (sfc_inc*.nc) /scratch2/NCEPDEV/land/data/DA/soil_moisture/LandIAU/INC/C48/INC09/* to /scratch2/NAGAPE/epic/UFS-WM_RT/NEMSfv3gfs/input-data-20240501/FV3_input_data/INPUT_L127_gfsv17 (I don't have permission). These six files are the only ones needed to make the control_c48_lnd_iau intel test that you set up complete.

@tsga I don't have permissions in that directory either. @jkbk2004 and/or other members of the UFS code management team can do this for us.

@tsga @grantfirl I will follow up today. will keep you posted.

[jkbk2004]

@tsga /scratch2/NAGAPE/epic/UFS-WM_RT/NEMSfv3gfs/input-data-20240501/FV3_input_data/INPUT_L127_gfsv17 is for C96 resolution. Do you want me to copy over /scratch2/NCEPDEV/land/data/DA/soil_moisture/LandIAU/INC/C96/sfc_inc.tile*.nc ?

[jkbk2004]

@tsga Or if you want to stay with with C48, we have C48 test case at /scratch2/NAGAPE/epic/UFS-WM_RT/NEMSfv3gfs/input-data-20240501/FV3_input_data48/INPUT_L127_gfsv17. We can copy over your C48 sfc increment files there. Maybe both C48 and C96. Let me know.

[tsga]

@tsga Or if you want to stay with with C48, we have C48 test case at /scratch2/NAGAPE/epic/UFS-WM_RT/NEMSfv3gfs/input-data-20240501/FV3_input_data48/INPUT_L127_gfsv17. We can copy over your C48 sfc increment files there. Maybe both C48 and C96. Let me know.

@jkbk2004 yes, let's stay with C48. And disregard the earlier directory. The C48 increments are at: /scratch2/NCEPDEV/land/data/DA/soil_moisture/LandIAU/INC/C48/INC3

[jkbk2004]

@tsga /scratch2/NAGAPE/epic/UFS-WM_RT/NEMSfv3gfs/input-data-20240501/FV3_input_data48/INPUT_L127_gfsv17/sfc_inc.tile*.nc is available. Once test is ok on hera, then we can rsync up on other machines.

[barlage]

@tsga There should be an associated PR for the NoahMP changes in https://github.com/NOAA-EMC/noahmp, right? This component repo needs to have its internal NoahMP code updated to correspond to the changes in ufs-community/ccpp-physics#222. Please let me know if you need help with this or if I am mistaken.

@barlage I would like to hear your opinion here. My impulse is to finalize the current pull request for ccpp-physics/fv3/ufs-wm and then start work for the component model separately.
@grantfirl I am assuming there is no repo dependence here (like that exists between ccpp-physics->fv3->ufswm)? Whatever will be merged for ccpp-physics/noahmp will be ported to the NOAA-EMC/noahmp.

Would the component model require changes to build with the changes to CCPP, or just require changes to use the land IAU? If the former, then the PRs should probably be concurrent, but if the latter, then I support @tsga 's impulse to finish this first, and then move on to the component model.

I believe there will be an issue here due to change in the noahmpdrv_init subroutine. I set up a meeting with @uturuncoglu to discuss.

[ClaraDraper-NOAA]

looks good to me, thanks @tsga

[jkbk2004]

@tsga can you sync up branch? We can start working on this pr.

[tsga]

@jkbk2004 Thanks for fixing the git submodules and syncing up.
I have re-run the tests with the latest changes. Everything looks good and the sub-repos are cleaned up.
The baseline created for the control_c48_lnd_iau_intel regression tests is at:
/scratch1/NCEPDEV/stmp4/Tseganeh.Gichamo/REGRESSION_TEST_LAND_IAU/control_c48_lnd_iau_intel