[MITgcm-support] grid nesting help

[Martin Losch]

Maxiaohui,

the code is not set-up to do two-way-nesting. For one-way nesting  
(that is you run a coarse resolution experiment and use the solution  
fields to drive a high-resolution experiment that is embedded in the  
coarse-resolution domain), you need to save the state variables  
U,V,T,S (and maybe Eta) in the coarse external model, interpolate  
them to the open boundaries of the high-resolution internal model and  
use the obcs package. An example of the usage of the obcs package is  
in exp4 (and there are others).

Alternatively you can try to use a generalized (quasi-) orthogonal  
curvilinear grid with high resolution in the center and lower  
resolution towards the edges (so a stretched grid), but I think there  
is little experience with this approach (with the MITgcm).

Martin

On 23 Jun 2008, at 16:05, Maxiaohui wrote:

> Hi,all
>
> I want to do a grid nesting experiment, from a large area to a  
> small area. Does our model have this kind of example or does  
> anyone  have done the similar experiment ? Could you give me a  
> similar example ? Thank you very much!
>
> Maxiaohui
>
> 雅虎邮箱，您的终生邮箱！
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support