Noah MP updates for Prototype 8 (GFSv17)

[HelinWei-NOAA]

1. Add some extra surface fields required to close the offline energy/water budget computation. These fields are listed below:

• LAND FRACTION
• INST PREC ADVEC HEAT
• WATER IN AQUIFER
• ACM EIWATER(CANOPY WATER EVAPORATION)
• ACM PLANTTR( PLANT TRANSPIRATION)
• ACM SOILSE (BARE SOIL EVAPORATION)
• AVE PREC ADVEC HEAT

2. The surface loop control/iteration removal when Noah MP is used
   Surface skin temperature is determined inside the Noah MP LSM with its own surface scheme/iteration.
   'guess_run'/iteration (for determination of skin temperature with weak winds) in surface loop control is
   set to 'false' when Noah MP LSM (lsm=2) is used.

3. New coupling interface
   The land model provides feedback to the atmospheric boundary layer. The required bulk Richard
   number input to the boundary layer scheme is computed using grid-aggregated momentum (from the
   roughness lengths over the tiles and vegetation fraction) and thermal roughness (from interpolation
   of kB^-1 from the tiles) lengths.

4. The addition of GFS surface scheme in Noah MP surface layer call
   To be consistent with the surface layer treatment for the other part of the surface in the model, the original
   GFS surface layer scheme is added to the Noah MP surface layer option.

5. Fix the inout issue of smc in Noah LSM (sflx.f)

[climbfuji]

@HelinWei-NOAA @barlage Now that we are ready to merge P8 features, can we take up this PR again? We will need to update the PR with the latest changes frorm the authoritative repository (same for fv3atm, ufs-weather-model) and then complete the review process. We are probably looking at the first days in December for a commit date (because of Thanksgiving fewer PRs are scheduled for next week). Does this work for you?

[HelinWei-NOAA]

Yes. I have ufs-weather-model branch ready on my fork for this change. I will create a ufs-weather-model PR once I have the other branch updated to current. Thanks.

@HelinWei-NOAA @barlage Now that we are ready to merge P8 features, can we take up this PR again? We will need to update the PR with the latest changes frorm the authoritative repository (same for fv3atm, ufs-weather-model) and then complete the review process. We are probably looking at the first days in December for a commit date (because of Thanksgiving fewer PRs are scheduled for next week). Does this work for you?

[grantfirl]

Do you really need to define your own gravity variable? I don't actually see it being used, but if it is, for continuity with other physics, you should pass in the variable with CCPP standard name of gravitational_acceleration.

[grantfirl]

This looks good to me, although the changes related to removing NoahMP from the "guess" iteration should be double-checked to make sure that there aren't unintended consequences for the sfc_nst scheme. The other comments relate to code cleanup.

[climbfuji]

One thing I forgot in the review so far. Since physics/module_sf_noahmp_glacier.f90 now depends on sfc_diff's stability subroutine, we should add a dependency on sfc_diff.f in physics/sfc_noahmp_drv.meta:

diff --git a/physics/sfc_noahmp_drv.meta b/physics/sfc_noahmp_drv.meta
index 52b35473..b4367bfd 100644
--- a/physics/sfc_noahmp_drv.meta
+++ b/physics/sfc_noahmp_drv.meta
@@ -1,7 +1,7 @@
 [ccpp-table-properties]
   name = noahmpdrv
   type = scheme
-  dependencies = funcphys.f90,machine.F,module_sf_noahmp_glacier.f90,module_sf_noahmplsm.f90,noahmp_tables.f90,set_soilveg.f
+  dependencies = funcphys.f90,machine.F,module_sf_noahmp_glacier.f90,module_sf_noahmplsm.f90,noahmp_tables.f90,set_soilveg.f,sfc_diff.f

 ########################################################################
 [ccpp-arg-table]

[HelinWei-NOAA]

done

[climbfuji]

@grantfirl can you confirm that the issues you raised have been addressed? Thanks!

[climbfuji]

PR looks good to me.

[climbfuji]

All requests for changes have been addressed, and regression testing is completed. Merging.