[MITgcm-devel] 3D diffusivity checked in

Jean-Michel Campin jmc at ocean.mit.edu
Thu Apr 5 11:45:40 EDT 2007


Hi Dimitris,

about your latest check-in, in short:
.cvsignore added : not sure that I will keep it.
 I would prefer to add a run dir + .cvsignore in
 all the experiments at the same time, just because it
 is easier when it's more uniform. We could decide
 on this later (next-week ?).
3d-vertical diffusivity:
- needs to be within #ifdef (anyone might want to be able
  to reduce the memory size to it's minimum).
  I would prefer to have it off (undef) by default (since
  many parametrizations are there to compute vertical diffusivity,
  providing a 3-D x Nb-tracers does not seems to go along the
  same spirit).
- not in PARAMS.h : I worked to remove 2-D arrays (coriolis),
  out of PARAMS.h, so I prefer not to have those ones there.
Why not in a dedicated common block, next to diffKr (used as a
control var for the adjoint), in DYNVARS.h ?

Cheers,
Jean-Michel




More information about the MITgcm-devel mailing list