Merged global namelist

[dpsarmie]

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 PR addresses issue #1248. The changes are meant to allow compatibility for all global regression tests with GFS_v17_p8. Changes have been made to the regression tests for this update and changes have also been made to the default_vars.sh file to add needed variables and change some variable values under the export_fv3 section.

global_control.nml.IN is the new global template namelist for the RTs. Variable values in this namelist have been converted to pointers and default values have been aggregated to default_vars.sh.

A new section has been made in default_vars.sh. The export_tiled section will take the ATMRES variable (set to C96 as default) and set the appropriate file names for the required parameter files.

Global regression tests have been updated to reflect these changes and are now all using the GFSv17_p8 core. New baselines will need to be created with the update to GFSv17 for some of the regression tests. There are a few RTs that crash, mainly those that use stochastic physics, and those will need to be sent to the physics groups to work on a compatibility fix. The groups have been notified and these tests will be temporarily turned off while the issues are addressed. (See rt.conf for the affected tests; discussion with the groups are below on this PR)

Commit Message:

* UFSWM - global_control.nml_IN has been added as the new regression test namelist template for all global regression tests. The namelist now uses pointers (i.e. @[abc]) for variables and default values have been added to the default_vars.sh script. A new section in default_vars.sh has been added (export_tiled) to account for tiled RTs that pulls the correct parameter files using the ATMRES variable.
Regression tests have been modified to account for these changes. Tests that were not compatible with the GFSv17_p8 core have been disabled for now. They will be turned on as they are updated from GFSv16 to GFSv17.

Priority:

• Normal

Git Tracking

UFSWM:

• Closes UFS WM regression test namelist template updates #1248

Sub component Pull Requests:

• None

UFSWM Blocking Dependencies:

• None

---

Changes

Regression Test Changes (Please commit test_changes.list):

• PR Updates/Changes Baselines:

• control_c192 intel
• control_c384 intel
• control_c48 intel
• control_CubedSphere_debug intel
• control_CubedSphereGrid intel
• control_CubedSphereGrid_parallel intel
• control_diag_debug intel
• control_latlon intel
• control_wrtGauss_netcdf_parallel_debug intel
• control_wrtGauss_netcdf_parallel intel
• merra2_thompson intel
• atmaero_control_p8_rad_micro intel
• control_c48 gnu
• control_diag_debug gnu

Input data Changes:

• None.

Library Changes/Upgrades:

• No Updates

---

Testing Log:

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

[dpsarmie]

This is a link to a spreadsheet showing the results of the regression tests using this PR.
Tests in magenta are runs that caused FV3 to crash due to incompatibilities with GFSv17_p8 (ex. any run with stochastic physics will crash).
Tests in yellow are runs that were able to produce output but need a new baseline for GFSv17_p8.
Merged nml results spreadsheet

[junwang-noaa]

@yangfanglin, @Qingfu-Liu, @pjpegion, @MatthewMasarik-NOAA, @dpsarmie is updating global and coupled RT tests to use GFSv17 namelist. We need your help to update following tests (listed in purple in the google sheet above), we are thinking of turning off those tests at this moment until the namelist of these tests are updated to be consistent with GFSv17, currently these tests are still based on GFSv16 or prototype P8 configuration. The tests failed when using GFSv17 namelist. Thanks

control_flake
control_stochy
control_iovr4
control_iovr5
control_p8_mynn
control_csawmg
control_csawmgt
control_ras
control_wam
control_atmwav
control_c384gdaswav

and the debug tests of the test above.

[MatthewMasarik-NOAA]

@yangfanglin, @Qingfu-Liu, @pjpegion, @MatthewMasarik-NOAA, @dpsarmie is updating global and coupled RT tests to use GFSv17 namelist. We need your help to update following tests (listed in purple in the google sheet above), we are thinking of turning off those tests at this moment until the namelist of these tests are updated to be consistent with GFSv17, currently these tests are still based on GFSv16 or prototype P8 configuration. The tests failed when using GFSv17 namelist. Thanks

control_flake control_stochy control_iovr4 control_iovr5 control_p8_mynn control_csawmg control_csawmgt control_ras control_wam control_atmwav control_c384gdaswav

and the debug tests of the test above.

Thank you for bringing this to my attention, @junwang-noaa. I've discussed this with @JessicaMeixner-NOAA offline and we will be discussing it further with our full group at our GFSv17-waves meeting on Monday. I'll be able to reply back at that point with how we can proceed. Thank you

[yangfanglin]

Besides Wave, I am brining the appropriate POC for the following RTs and provide my recommendation where it applies

control_flake @barlage Do we still need to maintain this ?

control_stochy @pjpegion

@ZhanZhang-NOAA should the two be removed ? I don't think they are needed anymore.
control_iovr4
control_iovr5

@joeolson42 ?? or this belongs to someone else ?
control_p8_mynn

@AnningCheng-NOAA Do you need both ?
control_csawmg
control_csawmgt

@Qingfu-Liu Qingfu, can you take care of this one ? Please ask Anning to help if needed.
control_ras

@[email protected] @[email protected]
control_wam

[AnningCheng-NOAA]

@yangfanglin and @junwang-noaa, control_csawmgt can be removed.

[Qingfu-Liu]

@fanglin Yang - NOAA Federal ***@***.***> I can take a look at the suite: control_ras

…

On Thu, Mar 14, 2024 at 9:22 AM AnningCheng-NOAA ***@***.***> wrote: @yangfanglin <https://github.com/yangfanglin> and @junwang-noaa <https://github.com/junwang-noaa>, control_csawmgt can be removed. — Reply to this email directly, view it on GitHub <#2173 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AGTS6UVTLJAGFWQZBRD6NS3YYGQCHAVCNFSM6AAAAABEJU3U3CVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSOJXGQ2TCNBTGM> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[joeolson42]

@yangfanglin I don't know the origin of control_p8_mynn.

[barlage]

@yangfanglin my suggestion would be to remove all flake tests until we resurrect that work: control_flake, control_flake_debug, rap_flake_debug

[uturuncoglu]

@DeniseWorthen @dpsarmie If there is an issue, we could address it. So, I don't think it is good idea to deactivate those tests.

[DusanJovic-NOAA]

There are many variable in the new input namelist template file (global_control.nml.IN) whose value is replaced with placeholders named FALSE_VAL:

$ grep -r FALSE_VAL .
./parm/global_control.nml.IN:  fv_debug = @[FALSE_VAL]
./parm/global_control.nml.IN:  reset_eta = @[FALSE_VAL]
./parm/global_control.nml.IN:  nudge_dz = @[FALSE_VAL]
./parm/global_control.nml.IN:  hydrostatic = @[FALSE_VAL]
./parm/global_control.nml.IN:  phys_hydrostatic = @[FALSE_VAL]
./parm/global_control.nml.IN:  use_hydro_pressure = @[FALSE_VAL]
./parm/global_control.nml.IN:  gfs_phil = @[FALSE_VAL]
./parm/global_control.nml.IN:  ncep_ic = @[FALSE_VAL]
./parm/global_control.nml.IN:  adjust_dry_mass = @[FALSE_VAL]
./parm/global_control.nml.IN:  consv_am = @[FALSE_VAL]
./parm/global_control.nml.IN:  dwind_2d = @[FALSE_VAL]
./parm/global_control.nml.IN:  no_dycore = @[FALSE_VAL]
./parm/global_control.nml.IN:  checker_tr = @[FALSE_VAL]
./parm/global_control.nml.IN:  pre_rad      = @[FALSE_VAL]
./parm/global_control.nml.IN:  ltaerosol    = @[FALSE_VAL]
./parm/global_control.nml.IN:  lradar       = @[FALSE_VAL]
./parm/global_control.nml.IN:  oz_phys      = @[FALSE_VAL]
./parm/global_control.nml.IN:  do_mynnsfclay = @[FALSE_VAL]
./parm/global_control.nml.IN:  do_ugwp      = @[FALSE_VAL]
./parm/global_control.nml.IN:  do_tofd      = @[FALSE_VAL]
./parm/global_control.nml.IN:  do_ugwp_v0_orog_only = @[FALSE_VAL]
./parm/global_control.nml.IN:  pdfcld       = @[FALSE_VAL]
./parm/global_control.nml.IN:  cal_pre      = @[FALSE_VAL]
./parm/global_control.nml.IN:  hybedmf      = @[FALSE_VAL]
./parm/global_control.nml.IN:  debug        = @[FALSE_VAL]
./parm/global_control.nml.IN:  iau_drymassfixer = @[FALSE_VAL]
./parm/global_control.nml.IN:  do_sedi_heat = @[FALSE_VAL]
./parm/global_control.nml.IN:  const_vi = @[FALSE_VAL]
./parm/global_control.nml.IN:  const_vs = @[FALSE_VAL]
./parm/global_control.nml.IN:  const_vg = @[FALSE_VAL]
./parm/global_control.nml.IN:  const_vr = @[FALSE_VAL]
./parm/global_control.nml.IN:  prog_ccn = @[FALSE_VAL]
./parm/global_control.nml.IN:  use_ppm = @[FALSE_VAL]
./parm/global_control.nml.IN:  de_ice = @[FALSE_VAL]
./parm/global_control.nml.IN:  LDEBUG   = @[FALSE_VAL]
./default_vars.sh:export FALSE_VAL=.false.

or TRUE_VAL:

$ grep -r TRUE_VAL .
./parm/global_control.nml.IN:  nudge_qv = @[TRUE_VAL]
./parm/global_control.nml.IN:  do_vort_damp = @[TRUE_VAL]
./parm/global_control.nml.IN:  external_eta = @[TRUE_VAL]
./parm/global_control.nml.IN:  fill = @[TRUE_VAL]
./parm/global_control.nml.IN:  z_tracer = @[TRUE_VAL]
./parm/global_control.nml.IN:  agrid_vel_rst = @[TRUE_VAL]
./parm/global_control.nml.IN:  filtered_terrain = @[TRUE_VAL]
./parm/global_control.nml.IN:  gfs_dwinds = @[TRUE_VAL]
./parm/global_control.nml.IN:  h2o_phys     = @[TRUE_VAL]
./parm/global_control.nml.IN:  use_ufo      = @[TRUE_VAL]
./parm/global_control.nml.IN:  oz_phys_2015 = @[TRUE_VAL]
./parm/global_control.nml.IN:  bl_mynn_tkeadvect = @[TRUE_VAL]
./parm/global_control.nml.IN:  lwhtr        = @[TRUE_VAL]
./parm/global_control.nml.IN:  swhtr        = @[TRUE_VAL]
./parm/global_control.nml.IN:  cnvgwd       = @[TRUE_VAL]
./parm/global_control.nml.IN:  redrag       = @[TRUE_VAL]
./parm/global_control.nml.IN:  dspheat      = @[TRUE_VAL]
./parm/global_control.nml.IN:  trans_trac   = @[TRUE_VAL]
./parm/global_control.nml.IN:  nst_anl      = @[TRUE_VAL]
./parm/global_control.nml.IN:  effr_in      = @[TRUE_VAL]
./parm/global_control.nml.IN:  sedi_transport = @[TRUE_VAL]
./parm/global_control.nml.IN:  rad_snow = @[TRUE_VAL]
./parm/global_control.nml.IN:  rad_graupel = @[TRUE_VAL]
./parm/global_control.nml.IN:  rad_rain = @[TRUE_VAL]
./parm/global_control.nml.IN:  do_qa = @[TRUE_VAL]
./parm/global_control.nml.IN:  fast_sat_adj = @[TRUE_VAL]
./parm/global_control.nml.IN:  use_ccn = @[TRUE_VAL]
./parm/global_control.nml.IN:  mono_prof = @[TRUE_VAL]
./parm/global_control.nml.IN:  z_slope_liq  = @[TRUE_VAL]
./parm/global_control.nml.IN:  z_slope_ice  = @[TRUE_VAL]
./parm/global_control.nml.IN:  fix_negative = @[TRUE_VAL]
./default_vars.sh:export TRUE_VAL=.true.

What's the purpose of this? I do not understand it. How is this supposed to work? What do we get by replacing string .false., with placeholder @[FALSE_VAL]? We can never change the value of FALSE_VAL variable in any individual test configuration file to be anything other than .false.. Ever. Otherwise all of the above variables will get different and probably incorrect values. This should not be how we use template placeholders.

[BrianCurtis-NOAA]

@dpsarmie looks like the linter probably wants you to export WAV_TASKS after line 1091 in default_vars.sh

[dpsarmie]

@dpsarmie looks like the linter probably wants you to export WAV_TASKS after line 1091 in default_vars.sh

Yep, got it. Thanks for the heads up on the BASH subsection that details all the conflicts.

Also, I noticed that the current default_vars.sh has the following line (661):
export SECS=$(( SHOUR*600 ))

Should this be 3600 and not 600? I currently have it as 3600 in this commit. But we should resolve this before the final commit.

Also, we should run the RTs again since there were changes to the default_vars.sh. Don't know if EPIC wants to take the lead on that or if I should run the tests on my end. Let me know either way.

[BrianCurtis-NOAA]

@dpsarmie looks like the linter probably wants you to export WAV_TASKS after line 1091 in default_vars.sh

Yep, got it. Thanks for the heads up on the BASH subsection that details all the conflicts.

Also, I noticed that the current default_vars.sh has the following line (661): export SECS=$(( SHOUR*600 ))

Should this be 3600 and not 600? I currently have it as 3600 in this commit. But we should resolve this before the final commit.

Also, we should run the RTs again since there were changes to the default_vars.sh. Don't know if EPIC wants to take the lead on that or if I should run the tests on my end. Let me know either way.

Yes please adjust that to 3600, looks like a typo in my editing. All the other SECS are OK.

[zach1221]

Ok, looks like we're finished testing. We can proceed with the merging process.

[DeniseWorthen]

@dpsarmie What is the meaning of the "stack build" label?