[CIG-LONG] Gale Velocity Boundary Condition

[Walter Landry]

feng lin <rimbong at gmail.com> wrote:
> Dear Walter,
> 
> In fact, I don't want the material in lower layer flow out of the bottom.
> I'm trying to model an extension process considering both Lithosphere and
> Asthenosphere.  What I need to observe is how lower layer( asthenosphere )
> affects the upper layer's behaviour ( Lithosphere ). The extension is
> simulated as only stretching the lithosphere( top 0.3 ), and the
> athenosphere( lower 0.2 ) flows passively.
> But when I set the side boundary as a step function ("vx" :
> "step(y-0.2)*1") than there will be convection in the lower layer and make
> material accumulate in the right side ( because the vx in lower layer is
> zero vx=0). so I think if, by any chance, I can mix Dirichlet and Neumann
> boundaries on a wall then it will be great.
> As they couldn't be mixed, I guess I should think of some other way to do
> this.

If you could make the box infinitely big, what would your conditions
on the right side look like?  Would you see a rigid lithosphere and a
linear gradient in the asthenosphere?  Or would the lithosphere and
asthenosphere move as a unit?

Cheers,
Walter Landry