[MITgcm-support] Prescribe eta as open boundary condition

[Eivind Støylen]

Dear all,
This may be an old problem, I have seen it discussed in mails some years
back, but I would appreciate an updated answer on the following.

I am simulating a 3D stratified fjord with a sill and one open boundary.
On this boundary I want forcing in terms of barotropic tide, in order to
analyse the baroclinic tide response inside the fjord. The most
"realistic" forcing would then in my opinion be to apply open boundary
forcing in terms of oscillating surface, i.e. setting OBWeta=Acos(omega*t)
in obcs_calc.F. (Ive made some attempts with prescribed velocity+sponge,
but as the prescribed velocities are merely guesses, the result isnt very
realistic..).

How should I proceed to do this? Do I need to turn on NONLIN_FRSURF?
Seeing as eta is calculated from the divergence of the flow I fear I need
some extensive code manipulation here...

Regards,
Eivind Støylen
PhD student
University of Oslo,
Norway