[MITgcm-support] parameters in OBCS sponge layer in ocean modeling

[Jody Klymak]

Hi Yangxin

If the time constant is 0 then the sponge is not a sponge, its like setting the boundary value, and your non-compliant signal will reflect back into the domain from the edge of the sponge

If the time constant is infinite, then the sponge has no effect, and your non-compliant signal will reflect from the boundary. 

The best effect is to set the inner to something large and the outer to something smaller and to make the sponge region a good fraction of a mode-1 wavelength.

Cheers,   Jody


Sent from my iPad

> On Nov 10, 2017, at 9:39 AM, Yangxin He <y67he at uwaterloo.ca> wrote:
> 
> Hi,
> 
> 
> I am not sure if my questions are related with the coding stuff, but I guess I can give it a try. In the OBCS sponge layer, I need to set 4 time scales as the following,
> 
> "&OBCS_PARM03
>   Urelaxobcsinner=864000.E0,
>   Urelaxobcsbound=43200.E0,
>   Vrelaxobcsinner=864000.E0,
>   Vrelaxobcsbound=43200.E0,"
> I have read the "obcs_sponge.F" code, and see how these parameters fit in the sponge forcing. But I am still a bit confused.
> Then I looked through MITgcm archives and Matthew suggested these numbers in one correspondence back in 2011. My question is, how and why did he set these values? For example, if I set Urelaxobcsinner=5000.E0, why it is not good? Is there any numerical/theoretical reason behind it?
> By the way, I am modeling internal waves in the ocean, if its helpful.
> Thanks a lot.
> Yangxin
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.mitgcm.org/pipermail/mitgcm-support/attachments/20171110/71ef7c33/attachment.html>