Specify the boundary condition at the inner edge of the polar mesh
Author: Frederic Masset
This parameter can be set either to
NON-REFLECTING. It specifies the boundary condition at the inner boundary.
RIGID sets the radial velocity to zero at the boundary. No material can flow from or to the mesh : the disk total gaseous mass is conserved (the outer boundary has a
WALL condition by default as well).
OPEN implies that the material can flow outside of the mesh, on its way to the central object : the total gaseous mass decreases with time.
NON-REFLECTING applies a transmitted wave boundary condition, as follows: the pitch angle of the wake at the boundary is evaluated using a WKB approximation. The content of the border ring is then copied into the ghost ring, properly azimuthally shifted by the amount dictated by the pitch angle. This technique is very efficient at removing any reflected wave, but it evaluates only one pitch angle, assuming the planet to have a semi-major axis a=1, so it efficiently only removes the reflected wake of this planet. Note that the non-reflecting boundary condition is also applied at the outer boundary, and that this condition does not strictly conserve the disk mass and angular momentum (the transmitted wave carries away some angular momentum from the mesh).
The figures below show the disk response to an embedded Saturn-mass planet after a few tens of orbits, at the same date and with strictly the same parameters, but with different boundary conditions.
Note that it is not possible to independently select the outer boundary and inner boundary conditions. There is no “OuterBoundary” parameter. This is a shortcoming of FARGO, but in practice it has never been needed. If however it was needed, it would be straightforward to implement (see
src/SideEuler.c). See also
OuterSourceMass. Also note that whatever is specified by the
InnerBoundaryparameter is ignored if fargo is run with the ’-e’ flag, which imposes the wave damping boundary conditions of the EU comparison problem.