[MITgcm-support] sponge limited to one process?

[Martin Losch]

You can use the rbcs package for sponge-layers that are independent of tile size.

M.

> On 18 Aug 2016, at 08:48, Sherry <schou at hawaii.edu> wrote:
> 
> Hi all,
> 
> I am setting up a sponge layer on the eastern boundary of a 2D configuration through making changes to data.obcs (following instructions of the manual). 
> The problem I have is when the number of points in one tile (on one process) is less than the number of points in the sponge (i.e. sNx < spongeThickness). In this case the sponge will start suddenly in the last tile, instead of farther away from the eastern boundary according to spongeThickness. This changes the extent of the sponge, as well as the value of Urelaxobcsbound at the beginning of the sponge.
> 
> I've been getting around this problem by using a smaller number of parallel-processes, or by making the xgrid coarser in the sponge so that it can fit into one tile. But I was wondering if this is a real problem or maybe I'm doing something wrong in my set-up.
> 
> Thanks for reading!
> 
> Sherry Chou
> University of Hawaii
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support