[MITgcm-devel] ptracer conservation

[Bruno Deremble]

Sorry, this modification was not well implemented. I wasn't aware of the
adjoint requirements.

moreover, it appears that it doesn't entirely solves the issue. This
modification causes the model to blow up after 100 days in our
configuration (gulf of mexico regional model).

I *think* it is related to hfac in shallow area close to the coast. It
is actually hard to track the point where the model blows up.
I wonder if Oliver ever encountered this kind of issue (I think he was
the original author of this modification)

thank you
bruno