Slight modification in orographic gravity wave drag (OGWD) scheme

[rhaesung]

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:

This modification is from @JongilHan66

Description:

1. The reference height (href) in the OGWD (Orographic Gravity Wave Drag) scheme is a critical parameter that significantly influences the strength of OGWD. Currently, href is defined as the maximum of the maximum subgrid orographic elevation (Omax) and twice the standard deviation of subgrid orography (sigma), i.e.,
   href = max(Omax, 2*sigma).

   In the proposed modification, this is changed to:
   href = max(Omax, 2*sigma, hpbl),
   where hpbl is the planetary boundary layer height.

   The inclusion of hpbl in determining href is informed by prior studies on the OGWD scheme:

   href = hpbl (Kim and Arakawa, 1995)
   href = 2* sigma (Kim and Doyle, 2005)
   href = max(2*sigma, hpbl) (Hong et al., 2005)

2. This updated formulation of href not only improves the 500 hPa height anomaly correlation, but also helps to reduce too strong wintertime polar vortex in the upper stratosphere, where hr4a11 is the control run and hr5c01 is same as the control run but with the modified href.

Commit Message:

* UFSWM - slight modification in orographic gravity wave drag (OGWD) scheme
  * FV3 - slight modification in orographic gravity wave drag (OGWD) scheme
    * ccpp-physics - slight modification in orographic gravity wave drag (OGWD) scheme

Priority:

• High: Reason: this is needed by gfsv17

Git Tracking

UFSWM:

• None

Sub component Pull Requests:

• FV3:Slight modification in orographic gravity wave drag (OGWD) scheme NOAA-EMC/ufsatm#972
  • ccpp-physics:Slight modification in orographic gravity wave drag (OGWD) scheme ccpp-physics#287

---

Changes

Regression Test Changes (Please commit test_changes.list):

• PR Updates/Changes Baselines.

Input data Changes:

• None.

Library Changes/Upgrades:

• No Updates

---

Testing Log:

• RDHPCS
  • Hera
  • Orion
  • Hercules
  • GaeaC6
  • Derecho
• WCOSS2
  • Dogwood/Cactus
  • Acorn
• CI
• opnReqTest (complete task if unnecessary)

[rhaesung]

@jkbk2004 @grantfirl @JessicaMeixner-NOAA @RuiyuSun This is another high-priority PR required for GFSv17.

[rhaesung]

The CCPP physics code has been modified based on @mdtoyNOAA's comment. I'm re-running the RTs.

[jkbk2004]

@rhaesung can you sync up branches?

[rhaesung]

@rhaesung can you sync up branches?

done!

[DeniseWorthen]

WW3 is not up to date? The PR branch is pointing to f0e0285

[jkbk2004]

cpld_control_p8_gnu fails on hercules: /work2/noaa/stmp/jongkim/stmp/jongkim/FV3_RT/failed_rt_154197/cpld_control_p8_gnu/err

 80: 11 0x0000000000056690 ompi_mpi_finalize()  ???:0
 80: 12 0x0000000000ca8317 ESMCI::VMK::finalize()  ???:0
 80: 13 0x0000000000761d62 ESMCI::VM::finalize()  ???:0
 80: 14 0x00000000007ca0ff c_esmc_vmfinalize_()  ???:0
 80: 15 0x0000000000941bd8 __esmf_vmmod_MOD_esmf_vmfinalize()  ???:0
 80: 16 0x00000000008bcd0d __esmf_initmod_MOD_esmf_finalize()  ???:0
 80: 17 0x000000000042890c MAIN__()  /work/noaa/epic/jongkim/UFS-RT/hercules/rt-2752/driver/UFS.F90:473
 80: 18 0x0000000000427c4d main()  /work/noaa/epic/jongkim/UFS-RT/hercules/rt-2752/driver/UFS.F90:37
 80: 19 0x000000000003feb0 __libc_start_call_main()  ???:0
 80: 20 0x000000000003ff60 __libc_start_main_alias_2()  :0
 80: 21 0x0000000000427c85 _start()  ???:0
 80: =================================
 80:
 80: Program received signal SIGBUS: Access to an undefined portion of a memory object.

Note that develop branch runs ok on hercules for that case.

[rhaesung]

cpld_control_p8_gnu fails on hercules: /work2/noaa/stmp/jongkim/stmp/jongkim/FV3_RT/failed_rt_154197/cpld_control_p8_gnu/err

 80: 11 0x0000000000056690 ompi_mpi_finalize()  ???:0
 80: 12 0x0000000000ca8317 ESMCI::VMK::finalize()  ???:0
 80: 13 0x0000000000761d62 ESMCI::VM::finalize()  ???:0
 80: 14 0x00000000007ca0ff c_esmc_vmfinalize_()  ???:0
 80: 15 0x0000000000941bd8 __esmf_vmmod_MOD_esmf_vmfinalize()  ???:0
 80: 16 0x00000000008bcd0d __esmf_initmod_MOD_esmf_finalize()  ???:0
 80: 17 0x000000000042890c MAIN__()  /work/noaa/epic/jongkim/UFS-RT/hercules/rt-2752/driver/UFS.F90:473
 80: 18 0x0000000000427c4d main()  /work/noaa/epic/jongkim/UFS-RT/hercules/rt-2752/driver/UFS.F90:37
 80: 19 0x000000000003feb0 __libc_start_call_main()  ???:0
 80: 20 0x000000000003ff60 __libc_start_main_alias_2()  :0
 80: 21 0x0000000000427c85 _start()  ???:0
 80: =================================
 80:
 80: Program received signal SIGBUS: Access to an undefined portion of a memory object.

Note that develop branch runs ok on hercules for that case.

@JongilHan66 Can you please take a look? This test ran fine on Hera.

[JessicaMeixner-NOAA]

Just wanted to comment here that this looks like the same error I got here: #2737 (comment)

And it was not reproducible by others as it was tested by others and their test passed: #2737 (comment)

[jkbk2004]

@kbooker79 Is there a way to update jenkins file to capture the failed messages: 8699df6 ? it just reports as job completed in commit message. BTW, my manual run is ok. I am pushing derecho log.

[jkbk2004]

All tests are done ok across all machines. This pr is set for merging process.

[BrianCurtis-NOAA]

Hercules RT logs are missing two tests, and there are no rt.conf changes in this PR. @gspetro-NOAA were you still running some tests? What is the status of those?

[kbooker79]

@kbooker79 Is there a way to update jenkins file to capture the failed messages: 8699df6 ? it just reports as job completed in commit message. BTW, my manual run is ok. I am pushing derecho log.

@jkbk2004, it depends on what you mean by failed messages. The failures are usually captured in the logs. I

[gspetro-NOAA]

Hercules RT logs are missing two tests, and there are no rt.conf changes in this PR. @gspetro-NOAA were you still running some tests? What is the status of those?

Yes, I ran those two tests, and they passed. I believe @jkbk2004 updated the Hercules RT log with those results, basically combining logs? I'm also running the full RT suite again, but that will take a while.

[jkbk2004]

@kbooker79 8699df6 brings the commit message "Derecho Job Completed." even if the log contains failed cases. It will be great if the commit message clearly tells the test failed.

[BrianCurtis-NOAA]

Thanks. I missed the end of the logs.

[jkbk2004]

@rhaesung new fv3 hash is NOAA-EMC/ufsatm@921278b

[gspetro-NOAA]

I ran the full RT suite on Hercules after baseline update, and all tests passed. Happy to push the complete log (located at /work/noaa/epic/gpetro/hercules/RTs/wm-rts/2752-retest3/tests/logs/RegressionTests_hercules.log) if desired, but I know @jkbk2004 has already pushed a log showing the full suite passing in two parts.