This issue relates to a conversation I just had with @BWHindman about the turbulent kinetic energy budget. There, for example, the transport_pressure_pKE code 2704 (turbulent pressure work term) does not have the constant c_3 in front, making it only useful in the dimensional anelastic formulation of the code (reference_type = 2). (Also @BWHindman I just realized that the regular KE equation DOES have the c_3 implicitly there via the statement "pfactor = ref%dpdr_w_term/ref%density", so perhaps we should actually fix the turbulent KE diagnostics to make them consistent with the rest of the diagnostics).
But more generally, I wonder if we should try to make the diagnostics that are only applicable to one reference_type, applicable to as many other reference_types as possible? @feathern and I talked a bit about this a few years ago.
The other place I know this occurs is in the enthalpy fluxes, where pressure_specific_heat is hard-coded into the outputs (so making them only appropriate to reference_type = 2). We could make the enthalpy fluxes work at least for the for other anelastic types (reference_type = 3, 5).
This issue relates to a conversation I just had with @BWHindman about the turbulent kinetic energy budget. There, for example, the transport_pressure_pKE code 2704 (turbulent pressure work term) does not have the constant c_3 in front, making it only useful in the dimensional anelastic formulation of the code (reference_type = 2). (Also @BWHindman I just realized that the regular KE equation DOES have the c_3 implicitly there via the statement "pfactor = ref%dpdr_w_term/ref%density", so perhaps we should actually fix the turbulent KE diagnostics to make them consistent with the rest of the diagnostics).
But more generally, I wonder if we should try to make the diagnostics that are only applicable to one reference_type, applicable to as many other reference_types as possible? @feathern and I talked a bit about this a few years ago.
The other place I know this occurs is in the enthalpy fluxes, where pressure_specific_heat is hard-coded into the outputs (so making them only appropriate to reference_type = 2). We could make the enthalpy fluxes work at least for the for other anelastic types (reference_type = 3, 5).