Changes needed for full land model coupling with ClimaAtmos

[juliasloan25]

We need the following changes to ClimaLand source code so we can couple the full land model with the ClimaAtmos model:

Done

• functions to compute q, \rho from land, atmos inputs, and to return land thermo state
  • function signatures like: get_\rho(T_land, T_atmos, \rho_atmos), get_q(T_land, \rho_land): Add functions which compute the land surface fluxes #1062
  • call these functions inside land turbulent flux calculation function? already has access to atmos thermo state
• functions to return the total water and energy stored in the land model, for conservation checks
  • something like total_water, total_energy : Adds functions for models to compute total energy and water volume(mass) per area #1071
• move diffuse_fraction update out of update_aux! and into a function that dispatches on radiation driver type
  • in the CoupledAtmosphere case, this will be updated by the coupler so we don't need to do anything in ClimaLand
  • see Update the diffuse fraction in drivers; use cosine of zenith angle; unneeded h access #1046
• use cos(zenith angle) instead of zenith angle (\theta_s) so we don't have to convert atmos value using arccos
  • this isn't strictly required for coupling, but will make things cleaner :)
  • see Update the diffuse fraction in drivers; use cosine of zenith angle; unneeded h access #1046
• remove access to bc.atmos.h in Canopy update_aux! here Update the diffuse fraction in drivers; use cosine of zenith angle; unneeded h access #1046
• move snow_surface_specific_humidity call out of update_aux! and into turbulent_fluxes! (this might require a new turbulent_fluxes! method that dispatches on SnowModel)
  • relax SoilDrivers atmos type annotation; compute SnowModel humidity in model-specific method #1056
• relax the type restriction for SoilDrivers to allow a CoupledAtmosphere in addition to PrescribedAtmosphere (here)
  • relax SoilDrivers atmos type annotation; compute SnowModel humidity in model-specific method #1056

[kmdeck]

@juliasloan25 would we like this"function signatures like: get_\rho(T_land, T_atmos, \rho_atmos), get_q(T_land, \rho_land)" to return a tuples (one element of the tuple for each land component), and for the bucket to return a tuple of 1, or for bucket to return a scalar?

[juliasloan25]

Everything in this issue is now done :)