[MITgcm-devel] teos-10 temperature conversion

[Martin Losch]

Hi Jean-Michel,

I have now copied the relevant rountines (functions) to convert conservative temperature into potential temperature from the teos-10 toolbox into seawater.F (not yet checked-in because never called so far). We now should talk about the procedure to include the conversion. Without having checked every detail, the surface temperature is required in at least these routines:
- external_forcing_surf (for restoring boundary conditions, temperature of the rain, etc)
- exf_radiation, exf_bulkformulae (here you actually sometimes need ksrf and ksrf+1 for extrapolation from the cell centers to the surface)
- bulkf_forcing
- seaice_growth (basal melting) and equivalent for thsice.
- shelfice_thermodynamics/icefront_thermodynamics, where “surface” has a different meaning

I guess the best way is to have a global 2D surface temperature field thetaSurf or tSurf that is filled once per time step before or at the beginning of load_fields_driver. By default tSurf = theta(k=1) (or different for shelfice,icefront, not clear if such a field is useful for these packages), and if eosType = ‘TEOS10’ convert: tSurf = gsw_pt_from_ct(salt,theta(k=1)).

Since this change will probably affect all sorts of computations (adjoint!), we should think about this carefully before actually starting such a procedure. Let’s see if we can make progress before I leave.

Martin