Update Prototype test suite to initial P8 configuration; turn on RRTMGP

[DeniseWorthen]

PR Checklist

• Ths PR is up-to-date with the top of all sub-component repositories except for those sub-components which are the subject of this PR. Please consult the ufs-weather-model wiki if you are unsure how to do this.

• This PR has been tested using a branch which is up-to-date with the top of all sub-component repositories except for those sub-components which are the subject of this PR

• An Issue describing the work contained in this PR has been created either in the subcomponent(s) or in the ufs-weather-model. The Issue should be created in the repository that is most relevant to the changes in contained in the PR. The Issue and the dependent sub-component PR
  are specified below.

• Results for one or more of the regression tests change and the reasons for the changes are understood and explained below.

• New or updated input data is required by this PR. If checked, please work with the code managers to update input data sets on all platforms.

Instructions: All subsequent sections of text should be filled in as appropriate.

The information provided below allows the code managers to understand the changes relevant to this PR, whether those changes are in the ufs-weather-model repository or in a subcomponent repository. Ufs-weather-model code managers will use the information provided to add any applicable labels, assign reviewers and place it in the Commit Queue. Once the PR is in the Commit Queue, it is the PR owner's responsiblity to keep the PR up-to-date with the develop branch of ufs-weather-model.

Description

• adds new default values of DO_RRTMGP and CCPP_SUITE=FV3_GFS_v16_coupled_p8 for the coupled model in default_vars
• adds same two settings to all standalone (control_[]_p7) tests since these are not default settings for the standalone
• adds namelist variables required for the DO_RRTMGP option to cpld_control_run.IN
• updates test names from p7 to p8 and makes required changes in rt.conf
• retains cpld_bmark_p7 test by setting DO_RRTMGP to false and CCPP_SUITE=FV3_GFS_v16_coupled_nsstNoahmpUGWPv1
• adds 35d versions of cpld_bmark_p8 and update rt_35d.conf.

Issue(s) addressed

Link the issues to be closed with this PR, whether in this repository, or in another repository.
(Remember, issues must always be created before starting work on a PR branch!)

• fixes Update Prototype Test suites P7->P8 #940
• fixes Indexing bug in SW cloud-sampling routine. (dustinswales/ccpp-physics@0a545c6#diff-ab383f7b24008085d38216e27611d9db32967aa782caf00525886f1c46407665L102)
• Reverts bug introduced by Move sfc emis alb calculation to its own scheme and z0 composite consistent NCAR/ccpp-physics#627. A little background. In this PR, the surface albedo calculation was moved out of the SW scheme and to the start of the radiation loop. In RRTMGP, the calculation of the solar-zenith angle was not moved out of the scheme along with the surface-albedo calculation. The SZA is an input to the subroutine to compute the surface-albedo.

Testing

• New baselines are required for all P7 tests (standalone and coupled) in order to update to the initial P8 feature set with RRTMGP turned on
• An initial baseline was created for hera.intel and all tests passed against that baseline.
• Baselines were run on hera.intel against develop-20111130. The following tests failed as expected:

Missing baselines:

• all cpld p8 tests
• all control p8 tests

Answer changes:

• regional_RRTMGP
• control_rrtmgp
• control_rrtmgp_c192
• control_rrtmgp_debug

However, the cpld_bmark_p7 test also failed. This was not expected since RRTMGP is not active in this test and the SDF remains FV3_GFS_v16_coupled_nsstNoahmpUGWPv1.

At 8f53730, hera.gnu baselines all pass except for the following:

Missing baselines:

• control_p8
• control_debug_p8
• cpld_control_c96_p8
• cpld_debug_p8

Answer changes:

• control_rrtmgp
• control_rrtmgp_debug

At 8f53730, hera.intel baselines showed the same results as previously but the cpld_bmark_p7 now passed, as expected.

How were these changes tested? What compilers / HPCs was it tested with? Are the changes covered by regression tests? (If not, why? Do new tests need to be added?) Have regression tests and unit tests (utests) been run? On which platforms and with which compilers? (Note that unit tests can only be run on tier-1 platforms)

• hera.intel
• hera.gnu
• orion.intel
• cheyenne.intel
• cheyenne.gnu
• gaea.intel
• jet.intel
• wcoss_cray
• wcoss_dell_p3
• opnReqTest for newly added/changed feature
• CI

Dependencies

If testing this branch requires non-default branches in other repositories, list them. Those branches should have matching names (ideally).

• Changes to FV3 for ccpp-physics #782 NOAA-EMC/ufsatm#428
• RRTMGP cleanup from ORTs NCAR/ccpp-physics#782

[DeniseWorthen]

@dustinswales Please update your FV3 branch for further testing. I have found that the cpld_bmark_p7 test does not reproduce against the current baseline, which is not expected (see the PR description for all the testing). Would you expect that any of your changes would change answers if RRTMGP is not active?

[dustinswales]

@DeniseWorthen
The answers should not change if GP is off.
I'm updating my code now and re-running opnReqTest script.

[MinsukJi-NOAA]

@DeniseWorthen The answers should not change if GP is off. I'm updating my code now and re-running opnReqTest script.

@dustinswales opnReqTest does not compare against the saved baselines (I believe @DeniseWorthen is referring to the baseline from 20211130: https://github.com/DeniseWorthen/ufs-weather-model/blob/745cebde8cb4ad950ed0de37f15f09c911f83af9/tests/rt.sh#L463). You will need to use ./rt.sh -n cpld_bmark_p7 -k

[DeniseWorthen]

@dustinswales I got a seg fault when I tried to run the cpld_debug_p8 test in rt_gnu.conf. Note I have not yet updated FV3 in my PR branch. The error is:

  0:  in fcst comp init, fcst_ntasks=         144
  0:  num_atmos_calls=          30 time_init=        2021           3          22           6           0           0 time_atmos=        2021           3          22           6           0           0 time_end=        2021           3          22          12           0           0 dt_atmos=         720 Run_length=       21600
 78: At line 3675 of file /scratch2/NCEPDEV/climate/Denise.Worthen/WORK/ufs_p8_gnu/FV3/ccpp/data/GFS_typedefs.F90
 78: Fortran runtime error: Cannot match namelist object name h2o_co2_o3_n2o_ch4_o2
 78:

My run directory on hera is /scratch1/NCEPDEV/stmp2/Denise.Worthen/FV3_RT/rt_21968/cpld_debug_p8. Could you check if there seems anything wrong w/ the input.nml?

(edit: the control_p8, control_debug_p8 and cpld_control_c96_p8 test also failed w/ the same error).

[DeniseWorthen]

I think the issue is the missing quotes around the file names in the input.nml. That seems to have resolved the gnu debug issue. Note I had no problem w/ the intel debug tests, even though they also lacked the quotes.

[dustinswales]

@DeniseWorthen
Yesterday we found an issue using gnu, dustinswales/fv3atm@e34a7d4, which is the same line you are having an error at.
Did you update the fv3atm you are using?

[DeniseWorthen]

@dustinswales I thought that replacing the quotes in the input.nml (see my last commit) resolved the GNU problem but now I'm not sure. I will update my FV3.

[DeniseWorthen]

I'm having trouble updating FV3. It is trying to checkout out ccpp/physics hash 4031e8f but can't find it.

[dustinswales]

@DeniseWorthen
Sorry, I forgot to push my physics. Should be all good now.

[dustinswales]

@DeniseWorthen
Once small thing that got lost along the way that you need to add in:
diff --git a/tests/parm/cpld_control.nml.IN b/tests/parm/cpld_control.nml.IN
index 339f3050..020183bd 100644
--- a/tests/parm/cpld_control.nml.IN
+++ b/tests/parm/cpld_control.nml.IN
@@ -225,10 +225,7 @@ deflate_level=1

-  use_LW_jacobian    = .true.
-  damp_LW_fluxadj    = .true.
-  lfnc_k             = 3125.
-  lfnc_p0            = 60000.
+  use_LW_jacobian    = .false.

[DeniseWorthen]

The cpld_bmark_p7 test is now reproducing after fixing the namelist and updating FV3.

[dustinswales]

The cpld_bmark_p7 test is now reproducing after fixing the namelist and updating FV3.
Excellent!

[DeniseWorthen]

@dustinswales I will be closing this PR since the feature is not being included in the P8 feature set at this time. Please open a new PR with your FV3 feature branch and add just your single test cpld_control_p7_rrtmgp. We'll then be able to at least commit your current RRTMGP fixes.

[dustinswales]

@DeniseWorthen
Can we just reopen my original PR...

[DeniseWorthen]

I can do that, but it seemed that PR would require quite a bit of cleanup to remove all the additional tests. Opening a new PR won't push you later in the Q if that is what you are concerned about.

[dustinswales]

@DeniseWorthen
I need to do some cleanup first, then will open a new PR

[DeniseWorthen]

OK, we're discussing the Q right now, and it looks like 12/7 would be the target date.

[DeniseWorthen]

Closing; The updates and fixes to RRTMGP will come in a new PR, currently scheduled for commit on 12/7.