net_sfc_heating does not include all ice-ocean heat fluxes for ACCESS

[rmholmes]

As discussed in an issue on the ACCESS-OM2 repository (COSIMA/access-om2#139), the net_sfc_heating diagnostic is missing the heat fluxes associated with the mass fluxes wfimelt, wfiform and licefw (noting that this last one has been added only recently).

These heat fluxes are included in sfc_hflux_pme through the pme variable:

MOM5/src/mom5/ocean_core/ocean_sbc.F90

Lines 3570 to 3581 in 4c3318e

	pme(ii,jj) = (Ice_ocean_boundary%lprec(i,j) + Ice_ocean_boundary%fprec(i,j) &
	#if defined(ACCESS_CM) || defined(ACCESS_OM)
	! PME is meant to include "melt", in a MOM+SIS configuration it
	! is added by the coupler. We add it here.
	+ Ice_ocean_boundary%wfimelt(i,j) &
	+ Ice_ocean_boundary%wfiform(i,j) &
	+ Ice_ocean_boundary%licefw(i,j) &
	#endif
	- Ice_ocean_boundary%q_flux(i,j))*Grd%tmask(ii,jj,1)
	liquid_precip(ii,jj) = Ice_ocean_boundary%lprec(i,j)*Grd%tmask(ii,jj,1)
	frozen_precip(ii,jj) = Ice_ocean_boundary%fprec(i,j)*Grd%tmask(ii,jj,1)
	evaporation(ii,jj) = -Ice_ocean_boundary%q_flux(i,j)*Grd%tmask(ii,jj,1)

But they are not included in net_sfc_heating:

MOM5/src/mom5/ocean_core/ocean_sbc.F90

Lines 5543 to 5555 in 4c3318e

	if(id_net_sfc_heating > 0) then
	wrk1_2d(:,:) = 0.0
	do j=jsc,jec
	do i=isc,iec
	wrk1_2d(i,j) = T_prog(index_temp)%conversion*( &
	T_prog(index_temp)%stf(i,j) &
	+ T_prog(index_temp)%runoff_tracer_flux(i,j) &
	+ T_prog(index_temp)%calving_tracer_flux(i,j) &
	+ (frozen_precip(i,j)+liquid_precip(i,j)+evaporation(i,j))*T_prog(index_temp)%tpme(i,j) )
	enddo
	enddo
	call diagnose_2d(Time, Grd, id_net_sfc_heating, wrk1_2d(:,:))
	endif

A second related issue is that there is no separate mh_flux diagnostic for the heat flux due to melting ice (although this term is correctly included in sfc_hflux_coupler).

I am about to submit a PR that addresses these issues.

[aekiss]

thanks @rmholmes, this has been merged - can the issue now be closed?

[rmholmes]

Yes I think so!