[Mitgcm-support] Re:

[mitgcm-support at dev.mitgcm.org]

Kirill,

  Glad to hear it is more stable.

  Docs. will be avilable on Firday at http://mitgcm.org
  They are in final stages of editing.

  The transformations are more complex
  for velocity on the cube than for temperature, because the components
  of velocity get transposed as you move around
  the cube. There is a second script called
  "uvcube2latlon.m" which should work for velocity and
  other vector quantities.
  One day we will have the code writing quick look output
  that is already interpolated back to lat-lon. We don't have
  that at the moment I am afraid. However,
  I believe Jean-Michel has some additional Matlab tools for
  cube plotting.

Chris



Kirill Semeniuk wrote:

> Hello,
> 
> The "cubic sphere" version of the MIT GCM works for me.  It appears
> to be much more robust than the grid + Shapiro filter model (c39).
> 
> Is there documentation for the model?  It is producing odd behaviour
> in the temperature in the uppper levels (i.e. T is rapidly increasing
> and I am confident it is not the radiative equilibrium distribution
> or initial distribution I am suppying the model).
> 
> One final question, is there someone who is responsible for the
> matlab scripts supplied with the model.  The cube2latlon.m script
> butchers the fields for wind but not for temperature.  Similarly,
> plotcube.m is not combining the grid patches correctly except for
> the temperature field.
> 
> Thanks,
> 
> Kirill
> ---------------------
> kirill at rossby.mit.edu
> 
> 
> 
>