Updates and bug fix for MYNN surface layer scheme

[joeolson42]

1. Bug fix for rare situation when denominator of flux equation becomes <= 0.
2. Added brute-force iterative method
3. Added GFS psi (stability) functions as an option (set psi_opt in module_MYNNSFC_wrapper.F90)
4. Removed 2D surface variables (relics from WRF days), making them 1D.

[DomHeinzeller]

@joeolson42 are these changes ready for testing?

[DomHeinzeller]

Regression testing on hera.intel using rt.conf against official EMC baseline. Tests pass except those expected to fail (because of previous changes and changes in this commit; note that all these tests do run to completion):

fv3_ccpp_hrrr 040 failed in check_result
fv3_ccpp_rap 039 failed in check_result
fv3_ccpp_rrfs_v1beta 043 failed in check_result
fv3_ccpp_gsd 038 failed in check_result
fv3_ccpp_gsd_debug 052 failed in check_result
fv3_ccpp_gsd_diag3d_debug 053 failed in check_result

rt_hera_intel.log
rt_hera_intel_fail_test.log

[DomHeinzeller]

Regression testing on hera.intel using rt_ccpp_gsd.conf: first create new baseline, then verify against it: all tests pass.

rt_ccpp_gsd_hera_intel_create.log
rt_ccpp_gsd_hera_intel_verify.log

[DomHeinzeller]

Regression testing on hera.gnu using rt_gnu.conf against official EMC baseline. Tests pass except those expected to fail (because of previous changes and changes in this commit; note that all these tests do run to completion):

fv3_ccpp_thompson_no_aero 009 failed in check_result
fv3_ccpp_thompson_no_aero 009 failed in run_test
fv3_ccpp_rrfs_v1beta 010 failed in check_result
fv3_ccpp_rrfs_v1beta 010 failed in run_test
fv3_ccpp_gsd 007 failed in check_result
fv3_ccpp_gsd 007 failed in run_test

rt_hera_gnu.log
rt_hera_gnu_fail_test.log

[joeolson42]

Thanks Dom!

…

On Fri, Aug 28, 2020 at 5:23 PM DomHeinzeller ***@***.***> wrote: Merged #45 <#45> into gsd/develop. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#45 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADLRR3SONJNX6AP7UBNDJUDSDA4APANCNFSM4PQQDWGA> .

-- Joseph Olson Meteorologist Environmental Prediction Advancement Division Global Systems Laboratory NOAA-Earth System Research Labs Boulder, Colorado