Development of a new unified cumulus convection scheme

[lisa-bengtsson]

Description

A new unified convection scheme aimed to be used across scales (applications) in the UFS is being developed. As a starting point we merge samfdeepcnv, samfdshalcnv, cu_gf_deep and cu_gf_sh and add the following updates

1. As a starting point we copied all the GF related routines over from GF to “unified”.
2. We introduced new imfdeepcnv_unified and imfshalcnv_unified as flags for the new convection scheme, and updated GFS_typedefs and GFS_restart and radiation routines that took imf*conv flags as input.
3. We created a new suite definition file that can run GFSv17 suite with the unified convection scheme.
4. We added the Simpson and Wiggert 1969 wu2 and cloud average from saSAS in a new subroutine called subroutine calculate_updraft_velocity, called from both shallow and deep convection. I use the SAS formulation (Han et al. 2017) of drag and buoyancy terms. We also added output for omega_u in Pa/s needed for the prognostic closure.
5. We added the prognostic area fraction computation from Bengtsson et al. 2022 with a call to progsigma_calc.F90 from both shallow and deep convection.
6. We added a new closure in the calls cup_forcing_ens_3d and cup_output_ens_3d.
7. Added (1-sigmab)^2 scaling if prognostic closure is used
8. Generalized turbulence tendency input to Unified and saSAS schemes (removed specific satedmf dependency)
9. We saved vertically integrated rain evaporation from the unified scheme to be used for the cellular automata evolution.
10. We brought in the cellular automata field (Bengtsson et al. 2021) to be used (although not yet hooked up to the convection evolution)

This will be used as a starting point for the new development.