[Mitgcm-support] c48

[mitgcm-support at dev.mitgcm.org]

I've generated a new checkpoint48 on the main branch.
It's the latest, it's differentiable,
it has all GM fixes and the adjoint verif's use
all GMREDI_OPTIONS.
I've also fixed it to use exactConserv.with the adjoint,
needed for the Held-Suarez setup.

THe GM results are expected to change because
of slight numerical changes.
I haven't updated the corresponding verif's.
Somebody shoould double-check that the GM changes
didn't compromise the forward GM.

The gradients with GM turned on are, of course,
larger than without, i.e. sensitivities increase.
That means, long-integration sensitivities remain to be
analyzed.

The verification/carbon/ now also contains
a tangent linear (forward mode) experiment

c48 is a good candidate for updating the
ecco-branch at some point and update Detlef's
state estimation setup.

Cheers
-Patrick

-- 
_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
Patrick Heimbach     Massachusetts Institute of Technology
FON: +1/617/253-5259                    EAPS, Room 54-1518
FAX: +1/617/253-4464               77 Massachusetts Avenue
mailto:heimbach at mit.edu                 Cambridge MA 02139
http://www.mit.edu/~heimbach/                       U.S.A.