[MITgcm-devel] switching seaice_obcs from legacy to devel

[Menemenlis, Dimitris (3248)]

Gael, thank you for explanation.

One more thing: seaice_obcs/input/data.seaice contains a lot of parameters,
which were accumulated in order to preserve backward compatibility with results/output.txt.
Since we will be breaking output.txt in any case, I will remove these parameters.
The original seaice_obcs was carved out of lab_sea/input.salt_plume, which has evolved
quite a lot since.  So I am planning to:
(1) using the same data.seaice parameters as lab_sea/input.salt_plume and
(2) regenerating initial and lateral boundary conditions from lab_sea/input.salt_plume

Dimitris Menemenlis

On Dec 15, 2012, at 5:06 PM, Gael Forget wrote:

Dimitris,

I concur with Martin on both points. I had already taken the time
to search for the cause. The different is due to the second term in
          tmpscal2 = MAX(-HEFF(I,J,bi,bj),r_QbyATM_cover(I,J)+
     &        AREApreTH(I,J) * r_QbyOCN(I,J))
that "limit ice growth by potential melt by ocean". That term helps getting
rid of such a film of ice. It is one of Ian's improvements to the model.

In case you want to also resolve the issue at the initiation of the run,
I attach modified initial conditions with 0s where h<0.01m. The
open boundaries may also feed very thin ice, but may be that is fine.
Up to you.

Gael

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mitgcm.org/pipermail/mitgcm-devel/attachments/20121216/0ec2364b/attachment.htm>