[Mitgcm-support] Re: SEAICE_GRID.h

[mitgcm-support at dev.mitgcm.org]

Great.
Alistair and I should have the exchange stuff sorted as soon as
our Christmas deliveries of German beer or Greek ooooooozoo (?)
arrive from Pasadena.

Chris
On Thu, 12 Dec 2002, Jinlun Zhang wrote:

> Chris,
> 
> You are exactly right. It is indeed north-east B-grid. Hmm... it IS a problem and
> you just confirm it. I'll definitely need your new exch_XXX and other help. Let me
> know once you get it done. Thanks. Seems Dimitris' German beer got him into the
> wrong place at the right time.
> 
> Jinlun
> 
> Chris Hill wrote:
> 
> > Hi Jinlun,
> >
> >  Great. I took a brief look yesterday and thought that
> > it looked like you were on a B-grid, and that that
> > was why maybe tiles edges were bad (in addition to Dimitri's
> > in ability to code, drink German beer and watch soccer
> > at the same time).
> >
> >  It looked to me like a north-east origin B-grid too?
> > That is definitely incompatible with the current
> > EXCH routines (south-west C-grid). I am getting close
> > to done with a revised EXCH() library that can be
> > extended fairly easily to for example a
> > north-east B-grid. There may still need to be some
> > custom coding but not as much.
> >
> > Chris
> >
> > On Thu, 12 Dec 2002, Jinlun Zhang wrote:
> >
> > > Chris,
> > >
> > > The ice model is based on B grid. Yesterday I found that, on an ice model
> > > generated B grid, the LSR ice dynamics solver works nicely on parallel machines
> > > - velocity difference at tile edges decreases when the tolerance value
> > > decreases. This is really encouraging. However, Dimitris found that's not the
> > > case with MITgcm. Since MITgcm uses C grid and the coupled ice model derives B
> > > grid from the C grid,  I suspect that there may be a bug with ice-ocean grid
> > > conversion. I'll try to fix the problem (if it is a problem) and then get
> > > correct grid info in SEAICE_GRID.h. But before that let me know if you have any
> > > questions about the grid.
> > >
> > > Cheers, Jinlun
> > >
> > >
> > > Dimitris Menemenlis wrote:
> > >
> > > > >> Also SEAICE_GRID.h has absolutely no explanation of what the grid
> > > > >> variables are! That's bad :-).....
> > > >
> > > > Jinlun, could you add a description for the variable in SEAICE_GRID.h
> > > > You can base your description on what the MIT folks do in model/inc/GRID.h
> > > >
> > > > Thanks, Dimitris
> > > >
> > > > --
> > > > Dimitris Menemenlis                      menemenlis at jpl.nasa.gov
> > > > Jet Propulsion Lab, MS 300-323           tel: 818-354-1656
> > > > 4800 Oak Grove Dr, Pasadena CA 91109     fax: 818-393-6720
> > >
>