boundary_forcing_heat_tendency needed in ocean_z

[StephenGriffies]

The diagnostic boundary_forcing_heat_tendency is 3d, as it includes penetrative shortwave. We should include this term as part of the ocean_z output in order to enable a closure of the heat budget in the ocean_z files.

[ashao]

As with #162, the column integrals of boundary_forcing_heat_tendency and boundary_forcing_salt_tendency aren't conserved under remapping. Looking into this now.

[adcroft]

I'm moving "boundary_forcing_heat_tendency" from "ocean_annual" to "ocean_annual_z".

What about "boundary_forcing_saln_tendency"? Why is it 3d and if it is 3d then it too needs to be in "ocean_annual_z"...

[adcroft]

@ashao Does NOAA-GFDL/MOM6#597 help fix this?

[ashao]

boundary_forcing_saln_tendency has to be 3D for the case where you have an outgoing freshwater flux (converted to meters) that is larger than the thickness of the top layer.

[adcroft]

Thanks - adding it to _z then...

[ashao]

@adcroft: Unfortunately, no. This case of nonconservative remapping for these salt/heat tendencies is partly due to the fact that the remapping thicknesses need to be updated, but there's something else that's going on too that I have yet to track down.

[ashao]

Also @adcroft, the equivalent diagnostic for boundary_forcing_heat_tendency is boundary_forcing_salt_tendency not boundary_forcing_saln_tendency