[MITgcm-devel] seaice*_if

[Martin Losch]

Hi Ian,

independently of the development of the seaice-thermodynamics, I'd like to understand what I need to do with the latent heat of sublimation in a coupled simulation. I thought that I had it right, because my simulations look OK, but I now I think that this is just an accident. In seaice_budget_ice_if.F, you nicely separate the different contributions to the heat flux, so that there is also a term for latent heat:
F_lh(I,J)    = D1I * UG(I,J) * (qhice_mm(I,J)-AQH(I,J,bi,bj))
For a closed fresh water budget, this heat should be used to sublimate/resublimate snow and ice, and the associated mass change then needs to be put into the atmosphere. In the current version of seaice_growth.F (and seaice_growth_if.F, maybe you can confirm) the heat is added to the net heat that is then used to _melt_ snow and ice, the resulting fresh water going into the ocean. I guess the resulting system still conserves mass but distributes the mass incorrectly. 
In a coupled system this F_lh should be taken out of the net F_ia, and it should be treated separately. Do you agree with this? Do you think that such a separation will destroy the adjoint?

Martin