[MITgcm-devel] status of OBCS changes

[Jean-Michel Campin]

Hi Martin,

OK. I might change the default then, but only after making
checkpoint62y.
Thanks,
Jean-Michel

On Thu, May 26, 2011 at 09:58:17AM +0200, Martin Losch wrote:
> Hi Jean-Michel,
> 
> I am not particularly attached to this OBCSfixTopo code, I just added it for convenience because I was tired of explaining to people that they have to adjust their topography along the OBCS. If this is no longer required, all the better, and yes, OBCSfixTopo=.false. is a sensible default in this case.
> 
> My little obcs-experiment does not seem to be affected by your changes either.
> 
> Martin
> On May 26, 2011, at 2:46 AM, Jean-Michel Campin wrote:
> 
> > Hi,
> > 
> > I have done most of the basic changes/cleaning I had in mind for OBCS.
> > (the last one was to bring the seaice advection/diffusion code 
> > at the same level as T,S & pTracers).
> > I did not change anything in the seaice dynamics part with OBCS
> > (but I don't think wider-stencil schemes are used there,
> > and don't know much about the solver/evp part).
> > 
> > Things that should now be possible is to move the OB further 
> > inside the domain (I tried with exp4) and to have concave
> > OB interior (not tried).
> > 
> > Going to make a checkpoint (62y) at some point (may be tomorrow ?)
> > 
> > A question to Matt: did you have a chance to update your 
> > code to check that it's still working with OBCS_CONTROL ?
> > (your simple test-exp is working, but not sure if I did not
> > break things when splitting OBCS.h)
> > 
> > And to Martin: I can now turn off (or on) OBCSfixTopo, it 
> > does not change anymore the results of the verification tests.
> > We could think at some point to change the default to False,
> > but keeping the code as it is, in case we need it for a 
> > particular set-up.
> > 
> > Cheers,
> > Jean-Michel
> > 
> > _______________________________________________
> > MITgcm-devel mailing list
> > MITgcm-devel at mitgcm.org
> > http://mitgcm.org/mailman/listinfo/mitgcm-devel
> 
> 
> _______________________________________________
> MITgcm-devel mailing list
> MITgcm-devel at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-devel