[MITgcm-support] OBCs balance and sponge

[Matthew Mazloff]

Hi Julie,

The sponge layer slowly brings the interior to the prescribed values...
here is a sample data.obcs with sponge....
http://ecco2.jpl.nasa.gov/data1/arctic/run_template2/data.obcs
spongeThickness is the number of cells in sponge
(UVTS)relaxobcsinner is the time scale of relaxation at edge of sponge
(UVTS)relaxobcsbound is the time scale of relaxation at wall

obcs balance will balance the volume flux into the basin...i think  
this is done by adding a constant to make sure net normal flow is  
zero...

i think your problem is a likely a large vertical velocity at edge of  
your "OBCS prescribe area" ...do you have the prescribed normal  
velocity moving smoothly to zero at this edge?

-Matt

On Oct 9, 2007, at 9:35 AM, Julie Deshayes wrote:

> Hi,
> I am trying to set up a configuration with MITgcm  where the basin  
> is semi-enclosed, with an open boundary on one side, which is split  
> in two portions. In the first part, velocities are precribed as  
> inflow (where there is a gradient in topography, so this drives a  
> boundary current), while the rest of the open boundary is "open".   
> I tried using Orlanski radiation conditions, but I keep having very  
> high velocities just at the limit between the OBCS prescribe area  
> and the OBCS Orlanski area, and the model finally blow up. I would  
> like to try OBCS-balance and OBCS-sponge routines, but the manual  
> lacks details on what these routines actually do and which  
> parameters have to be specified.
> Any help would be grately appreciated ...
> Julie
>
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mitgcm.org/pipermail/mitgcm-support/attachments/20071010/581cfab2/attachment.htm>