[Mitgcm-support] lots of headaches...

[mitgcm-support at dev.mitgcm.org]

... with c46m_post:

- any good reason why to keep pload when
   #undef ATMOSPHERIC_LOADING?
   (I don't think, there is one)

- pressure is now part of the model "state"
   whether you want it or not.
   So why not put it in DYNVARS.h?
   I have to place the EOS.h in routines where
   it shouldn't really be to make it visible to TAF.

- kRef in convective_adjustment.F: TAF says "UFFF"

- something else weired provokes TAF wanting
   to store the whole state at the innermost chkpt.-loop.
   Didn't have this kind of problem before.
   Are you sure your computations produce a "clean" model state
   at the time stepping level (forward_step)?
   It seems to be related to dynamics.F
   which is always recomputed.

-p.

-- 
_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
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.