[MITgcm-support] Open Boundary Conditions in MITgcm

[Martin Losch]

Hi Jody,

as Dimitris already pointed out the OBNu0/1 are two discrete time levels, read from a file. This is completely analogous to the way surface forcing is handled in the model (e.g. external_fields_load.F or, much more sophisticated and but less transparent, pkg/exf, in particular exf_getffields.F/exf_set_gen.F

I have updated the obcs description a little to describe the OBCS_PRESCRIBE business a little (<http://mitgcm.org/public/r2_manual/latest/online_documents/node236.html>), but have to admit that the package description is far from complete. E.g. I do not know the details of the orlanski code, so I cannot write documentation for that (and, as you point out, documentation requires that people need to be so busy that they are trying to run away by doing something important, that is not immediately urgent). If you do have text for the orlanksi part, I'd be happy to include it into the documentation.

I let others answer the wiki question. I remember that there used to be something like this, ...

Martin

On Feb 26, 2011, at 6:20 PM, Jody Klymak wrote:

> Hi Martin,
> 
> On Feb 25, 2011, at  0:23 AM, Martin Losch wrote:
> 
>>> I keep thinking that it would be worthwhile for someone with coding chops to add to obcs package the option to specify lateral boundary conditions, perhaps even reading them from a file.  I'd be interested to hear how other people do it.
>> There is code to read obcs-values from files (it's not really documented, as usual), you turn it on with
>> #define ALLOW_OBCS_PRESCRIBE
>> and (in data.obcs)
>> useOBCSprescribe = .true.,
>> The example exp4 uses this code (and it also works with exf).
> 
> That looks like it could work.  I'd have to play with it.  I'm mystified as to what OBNu0 and OBNu1 are (for instance).  I assume they are some halo, but they aren't used elsewhere in pkg/obcs.
> 
> WRT the documentation, has there been any thought to making a wiki where folks can contribute and write down how they do these things?  Of course it would be nice if the manual were kept up to date, but everyone is busy...
> 
> Thanks,  Jody
> 
> 
> --
> Jody Klymak
> http://web.uvic.ca/~jklymak/
> 
> 
> 
> 
> 
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support