[Mitgcm-support] Open boundary conditions

[mitgcm-support at dev.mitgcm.org]

Hi,

I've recently started working with the MIT GCM. Ultimately I want to do
high resolution (~1/12 degree) adjoint data assimilation of SST, SSH,
and in-situ obs in the East Greenland seas.

At the moment I am getting familar with just running forward
integrations of the model. I can get the model going well with closed
boundary conditions. The examples and the manual were very useful in
that respect -thanks. However, I'm stuck on the use of open boundary
conditions.

I think someone may have asked about the use of open boundaries, but I
can't find the message - I apologise for any repetition. What I'd like
to do, initially is have open boundaries which are fixed (in time). Is
this possible? The subroutine obcs_readparms refers to files OBNuFile
etc. do I need to set these? If so where/how? (I 'd like this to work on
a parallel machine). I've found the example "internal_wave" which seems
to be relevant, but I'm not quite sure what to do to adapt it to my
needs. Some of the files in there don't seem to be used e.g.
input/data.obcs.

Cheers

Daniel Lea