Another approach to achieve planetary gap

[miaoL1]

Example in documentation has demonstrated how to construct planetary gap by changing the alpha viscosity.

Here's another way to achieve planetary gap, which is directly adding a torch in evolution equation(Lin & Papaloizou(1986)):
$\frac{\partial \Sigma}{\partial t} = \frac{1}{R}\frac{\partial}{\partial R}[3R^{1/2} \frac{\partial }{\partial R}(\nu \Sigma R^{1/2}) - \frac{2 A\Sigma R^{3/2}}{(G M_\star )^{1/2}}]$
where A is torque density distribution.

The core issue is that Dustpy does not seem to allow changing the form of the evolution equation (torque cannot be treated as an external source as well).

Is it possible to implement a torque driven planetary gap without changing the fundamental code logic?

[matgarate]

@miaoL1 It is possible to include a planet torque as an advection term in the equation (for gas and dust) by modifying the gas Jacobian, (and the dust radial velocity for self-consistency).
If you are interested I could send you the current version of our planet torque module. Just send an email with your name, affiliation, and scope of your project to [email protected].

The module is now under development, so we would appreciate the opportunity to test it and receive feedback before making it public in the coming months.

Best regards,
Matías

[stammler]

That is a good point. Thank you!

We will try to add a field for the torque into the next release, such that this can be easily done.

I have to think about this in more detail, but I think in the current logic of the code it could be realized via the back reaction term.
Back reaction is implemented by adding a velocity term to the gas with "back reaction velocity"

$v = 2\ B\ \eta\ r\ \Omega_\mathrm{K}$

I think it could be possible by setting s.dust.backreaction.B accordingly to achieve the torque equation.
But maybe @matgarate has more information about this.

[matgarate]

The backreaction ´B´ term could be used to mimick the effect in the gas, but I do not recommend doing so, as it might lead to numerical errors near pressure maximum, and might lead to inconsistencies in the dust velocity which is also affected by the backreaction.
Adding and additional "advection velocity" in the gas jacobian has been my best solution to these type of problems.