Implemented the Noah-MP Land Surface Model

[ldfowler58]

This PR adds the implementation of the Noah-MP land surface model (LSM):

• The sourcecode of the Noah-MP LSM was downloaded from the Noah-MP github repository and corresponds to the release-v5.0.1. It is located in the directory ./src/core_atmosphere/physics/physics_noahmp.

• ./src/core_atmosphere/physics/Registry_noahmp.xml contains all the run options and variables needed to link the Noah-MP sourcecode to the MPAS physics framework.

• ./src/core_atmosphere/physics/mpas_atmphys_lsm_noahmpinit.F includes the initialization of the Noah-MP LSM.

• ./src/core_atmosphere/physics/mpas_atmphys_driver_lsm_noahmp.F is the main driver for the Noah-MP LSM.

[mgduda]

I think we may have discussed what to do about this before (apologies for my poor memory!), but we get a compilation error with recent GNU compilers:

RunoffSurfaceDynamicVicMod.F90:205:4:

  151 |                           goto 1003
      |                                   2
......
  205 | 1003                if ( DepthYTmp <= 0.0 ) DepthYTmp = 0.0
      |    1
Error: Legacy Extension: Label at (1) is not in the same block as the GOTO statement at (2)
RunoffSurfaceDynamicVicMod.F90:205:4:

  175 |                              goto 1003
      |                                      2
......
  205 | 1003                if ( DepthYTmp <= 0.0 ) DepthYTmp = 0.0
      |    1
Error: Legacy Extension: Label at (1) is not in the same block as the GOTO statement at (2)

Should we try to fix this locally (since we have a copy of the Noah-MP code) and separately feed a fix upstream to the Noah-MP repository?

[ldfowler58]

I modified RunoffSurfaceDynamicVicMod.F90 so that the file can be compiled with gcc/12.2.0. I am not entirely sure if the modified file breaks the logic of the sourcecode or not.

[mgduda]

Right now, we get quite a lot of output from Noah-MP-related code in every timestep:

 --- enter subroutine driver_lsm_noahmp:
 --- enter subroutine lsm_noahmp_fromMPAS: itimestep = 46
 --- enter subroutine lsm_noahmp_sounding_fromMPAS: 1
 --- end subroutine lsm_noahmp_sounding_fromMPAS:
 --- end subroutine lsm_noahmp_fromMPAS.
 
 --- enter subroutine NoahmpDriverMain:
 
 --- enter subroutine noahmpdrivermain:
 --- NoahmpIO%itimestep         = 46
 --- NoahmpIO%soiltstep         = 0.00000
 --- NoahmpIO%dtbl              = 360.000
 --- NoahmpIO%soil_update_steps = 1
 --- NoahmpIO%calculate_soil    = T
 
 --- NoahmpIO%isurban_table     = 13
 --- NoahmpIO%urbtype_beg       = 50
 --- NoahmpIO%sf_urban_physics  = 0
 --- NoahmpIO%iri_urban         = 0
 
 --- NoahmpIO%yearlen           = 365
 --- NoahmpIO%yr                = 2010
 --- NoahmpIO%month             = 10
 --- NoahmpIO%day               = 23
 --- NoahmpIO%julian            = 295.188
 --- NoahmpIO%xice_threshold    = 0.200000E-01
 
 --- enter subroutine lsm_noahmp_toMPAS:
 --- end subroutine lsm_noahmp_toMPAS:
 --- end subroutine driver_lsm_noahmp:

Most of this looks like it might have been used during development but could be removed now; what do you think?

[ldfowler58]

modified sourcecode accordingly.

[ldfowler58]

I rewinded the sourcecode to get back to the original sourcecode for RunoffSurfaceDynamicVicMod.F90. Also removed print statements that are printed to the logfiles during the initialization of Noah-MP. As we talked about, Michael will modify the flags used to compile Noah-MP so that we can run MPAS-Model with the latest versions of the gnu compiler with optimization.

[mgduda]

I've just left a couple of additional small requests.

Otherwise, I'm working on some changes to deal with compilation failures with the GNU compilers.

[mgduda]

Although the model won't compile with the gnu build target at present, this can be addressed by a future PR.

[ldfowler58]

Thanks for your help.