A revised handling of cflx as a stealth feature

[huiwanpnnl]

This PR replaces PR #4

This PR includes the same essential changes in the old PR, but

• The implementation is based off Wuyin's branch (which has updated test capabilities).
• Comments in the code have been improved to clarify which tracers are affected and which are not.
• Initialization of cam_in%cflx is revised to ensure BFB restart.

---

This PR adds an optional revision to the numerical process coupling at the tphysbc/tphysac level to couple the surface emissions of tracers more tightly with the turbulent transport. This is done by revising where in the time integration loop the surface fluxes in cam_in%cflx(:,2:) are applied to update the tracer mixing ratios.

The revision was motivated by the known issues of overly short dust lifetime in EAMv1 and the very strong sensitivity of the dust lifetime to vertical resolution, see Feng et al., 2022 for the v1 results.

The implementation of the revised coupling is described below in the "code modifications" section.

Impacts of the revised coupling was initially evaluated in v1. More recently, evaluation was done in v2 using a 5-year F2010 simulation and a 30-year AMIP (F20TR) simulation using the huiwanpnnl/atm/aerosol_process_coupling branch in the E3SM repo. Highlights of the results can be found on this Confluence page and in this slide deck.

Code modifications

• The subroutine clubb_surface in clubb_intr.F90 was split into two parts:

  • The calculations of ustar and the Obukhov length remain in clubb_surface.
  • The calculation of tracer mixing ratio tendencies using cam_in%clfx was moved to a new subroutine cflx_tend in a new module in cflx.F90

• A new namelist variable cflx_cpl_opt was added.

  • _cflx_cpl_opt = 1 (default) gives the original coupling in v2._

  There, the subroutine cflx_tend is called immediately after clubb_surface followed by a call physics_update(...). This gives the original process coupling in EAMv2 and the results are BFB.

  • _cflx_cpl_opt = 2 activates the revised coupling._

  The call of cflx_tend and the corresponding call physics_update(...) are moved to right before the cloud macro-microphysics sub-cycles in tphysbc, so that the tracers emitted to the lowest model layer can be transported to upper layers by CLUBB or dropmixnuc.

  Note that cflx_cpl_opt = 2 does not affect water vapor. Neither does it affect tracers whose portion of cam_in%cflx is zero when clubb_surface is called.

• The initialization of cam_in%cflx(:,2:) with zeros was moved from the atm_import subroutine in components/eam/src/cpl/atm_import_export.F90 to the read_restart_physics subroutine in components/eam/src/physics/cam/restart_physics.F90 to avoid incorrectly zeroing out the values read in from a restart file. This ensures BFB restart for both the default model and for cflx_cpl_opt = 2.

[rljacob]

Are you starting development branches from the NGD_v3atm branch? I didn't think that was in the workflow.

[wlin7]

Are you starting development branches from the NGD_v3atm branch? I didn't think that was in the workflow.

@rljacob . I might not understand your comments correctly. Git log does show this branch started from the NGD_v3atm branch of Sept. 26.

[wlin7]

Merged to NGD_v3atm.