Save the Planet, Feed the Star: How Super-Earths Survive Migration and Drive Disk Accretion

Two longstanding problems in planet formation include (1) understanding how planets survive migration, and (2) articulating the process by which protoplanetary disks disperse---and in particular how they accrete onto their central stars. We can go a long way toward solving both problems if the disk gas surrounding planets has no intrinsic diffusivity ("viscosity"). In inviscid, laminar disks, a planet readily repels gas away from its orbit. On short timescales, zero viscosity gas accumulates inside a planet's orbit to slow Type I migration by orders of magnitude. On longer timescales, multiple super-Earths (distributed between, say, $\sim$0.1--10 AU) can torque inviscid gas out of interplanetary space, either inward to feed their stars, or outward to be blown away in a wind. We explore this picture with 2D hydrodynamics simulations of Earths and super-Earths embedded in inviscid disks, confirming their slow/stalled migration even under gas-rich conditions, and showing that disk transport rates range up to $\sim$$10^{-7} M_\odot~{\rm yr^{-1}}$ and scale as $\dot{M} \propto \Sigma M_{\rm p}^{3/2}$, where $\Sigma$ is the disk surface density and $M_{\rm p}$ is the planet mass. Gas initially sandwiched between two planets is torqued past both into the inner and outer disks. In sum, sufficiently compact systems of super-Earths can clear their natal disk gas, in a dispersal history that may be complicated and non-steady, but which conceivably leads over Myr timescales to large gas depletions similar to those characterizing transition disks.