[MITgcm-support] Problem with Bathymetry

[Martin Losch]

Abbas,

if you don't have enough grid points to resolve a topographic  
feature, then you need to increase the number of grid points. If you  
cannot do that, because of limited resources, then you cannot do that.

You need to dig out your canyon by hand, which will probably give you  
a too wide canyon, so resolution is still an issue, it might preserve  
the "topology" of your domain, though.
Alistair once wrote a script that could do such things, but it was  
never really published/documented as far as I remember.

I if see it right, most of your domain is either shallow, or even  
dry. If the latter is the case, then you could rotate your grid, so  
that it is aligned with the main direction of the lake, then you'd  
save many, many grid points and could use them to resolve your  
topography better.

Martin


On 10 Nov 2008, at 15:07, Abbas Dorostkar wrote:

> Hi Martin,
>
> Thanks for your prompt reply.
>
> I don't have any problem with my bathymetry file if I use large  
> number of grid points (e.g. nx=ny=1000) to generate it(see fig1).  
> The problem arose when I use small number of grids (e.g. nx=ny=240)  
> so that the canyon in my topography cannot be seen (see fig 2).  I  
> need to use small number of grids due to lack of enough  
> computational power. Do you have any idea how to fix that.
>
> Thanks,
>
> Abbas
>
> <fig1.jpg><fig2.jpg>
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support