[MITgcm-support] Orlanski and mass conservation

[Stefano Querin]

Hi,
I am making some tests using the Orlanski OBCs. I recently observed that when a non-negligible velocity field reaches the OB, the mass conservation in the domain is no longer satisfied.
For example, in the Plume on slope verification experiment, when the plume reaches the boundary, there is a volume increase in the basin (eta reaches 180 m in few days!). In this case, I simply made a longer run to see the evolution of the plume for some more days.
I observed similar problems even when I used the flag: exactConserv=.TRUE.
To be more clear I attach three pictures showing the problem:
the first two show the evolution of eta and u-velocity (varying with depth) in time, the third is a sketch of the temperature profile when the plume "reaches" the OB (after 2.5 days; the horizontal lenghtscale is of tens of meters).
My questions are:
- (first of all!) did I make some mistakes?
- is there a way to control the mass fluxes in the domain (in the orlanski_east(north, south, west).F routines there is no mention to eta)?
- if I impose the Orlanski condition on the eastern boundary (as prescribed in data.obcs), do I (implicitly) impose the same condition on the western OB (u-velocity is non zero on that OB)?
Thank you very much for any suggestion.

Sincerely,

Stefano Querin

P.S.: (for Patrick) any news about the active OBCs?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mitgcm.org/pipermail/mitgcm-support/attachments/20040802/e157db9d/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: eta(t).JPG
Type: image/jpeg
Size: 41689 bytes
Desc: not available
URL: <http://mitgcm.org/pipermail/mitgcm-support/attachments/20040802/e157db9d/attachment.jpeg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: u(z;t).JPG
Type: image/jpeg
Size: 49005 bytes
Desc: not available
URL: <http://mitgcm.org/pipermail/mitgcm-support/attachments/20040802/e157db9d/attachment-0001.jpeg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: T(z;x).JPG
Type: image/jpeg
Size: 64549 bytes
Desc: not available
URL: <http://mitgcm.org/pipermail/mitgcm-support/attachments/20040802/e157db9d/attachment.jfif>