Usage of flux_param in g_tracer_add

[jessluo]

@tanyamarshall and I are wondering about flux_param in g_tracer_add:

It looks like in some cases this refers to the molecular weight of a tracer with a unit conversion:

ocean_BGC/generic_tracers/generic_COBALT.F90

Line 1495 in c10390d

flux_param = (/ 14.0067e-03 /), &

but in other cases the molecular weight is given as flux_gas_molwt:

ocean_BGC/generic_tracers/generic_COBALT.F90

Line 1462 in c10390d

flux_gas_molwt = WTMN, &

@tanyamarshall is trying to add in 15N isotopes so would you recommend adding in its molecular weight directly in the FMS / constants.F90 and referencing it here, or just adding it in here directly?

Also, where can we find aof_set_coupler_flux() which is referenced in generic_tracer_utils.F90? In it, the comments say refer to documentation for subroutine aof_set_coupler_flux() but we are unable to locate this subroutine.

ocean_BGC/generic_tracers/generic_tracer_utils.F90

Line 817 in c10390d

! Aray of parameters for non-gas flux (refer to documentation for subroutine aof_set_coupler_flux() ).

[andrew-c-ross]

I believe this is the aof_set_coupler_flux() being referred to: https://github.com/NOAA-GFDL/FMS/blob/d13c1298f15ba3c409dec70a57a02a0e29069bcb/coupler/atmos_ocean_fluxes.F90#L114

[andrew-c-ross]

Also, my understanding is that you would use specify flux_gas_molwt for gases that are exchanged with the atmosphere, and flux_param for tracers with sources from runoff, deposition, etc:

ocean_BGC/generic_tracers/generic_tracer_utils.F90

Lines 1137 to 1167 in c10390d

	if(g_tracer%flux_gas) then
	g_tracer%flux_gas_ind = aof_set_coupler_flux(g_tracer%flux_gas_name, &
	flux_type = g_tracer%flux_gas_type, &
	implementation = trim(g_tracer%implementation), &
	mol_wt = g_tracer%flux_gas_molwt, &
	param = g_tracer%flux_gas_param, &
	ice_restart_file = g_tracer%ice_restart_file, &
	ocean_restart_file= g_tracer%flux_gas_restart_file, &
	verbosity = verbosity &
	)
	endif
	if(g_tracer%flux_runoff) then
	g_tracer%flux_runoff_ind = aof_set_coupler_flux(g_tracer%flux_runoff_name, &
	flux_type = 'land_sea_runoff', &
	implementation = 'river', &
	param = g_tracer%flux_param, &
	ice_restart_file = g_tracer%ice_restart_file, &
	verbosity = verbosity &
	)
	endif
	if(g_tracer%flux_wetdep) then
	g_tracer%flux_wetdep_ind = aof_set_coupler_flux(g_tracer%flux_wetdep_name, &
	flux_type = 'air_sea_deposition', &
	implementation = 'wet', &
	param = g_tracer%flux_param, &
	ice_restart_file = g_tracer%ice_restart_file, &
	verbosity = verbosity &
	)
	endif

See for example oxygen which has both flux_gas_molwt and flux_param and uses flux_param for units conversion:

ocean_BGC/generic_tracers/generic_COBALT.F90

Lines 1500 to 1513 in c10390d

	call g_tracer_add(tracer_list,package_name, &
	name = 'o2', &
	longname = 'Oxygen', &
	units = 'mol/kg', &
	prog = .true., &
	flux_runoff= .true., &
	flux_param = (/ 1.0e-3 /), &
	flux_gas = .true., &
	flux_gas_name = 'o2_flux', &
	flux_gas_type = 'air_sea_gas_flux_generic', &
	flux_gas_molwt = WTMO2, &
	flux_gas_param = (/ as_coeff_cobalt, 9.7561e-06 /), &
	flux_gas_restart_file = 'ocean_cobalt_airsea_flux.res.nc', &
	flux_bottom= .true. )

[tanyamarshall]

Hi @andrew-c-ross, thank you for the explanation!