Protoplanetary disks and planet formation: A computational perspective

by Isaac Jonathan Backus

Institution: University of Washington
Year: 2018
Keywords: High performance computing; Numerical methods; Planet formation; Protoplanetary disks; Simulations; Smoothed Particle Hydrodynamics; Astrophysics; Computational physics; Astronomy; Physics
Posted: 02/01/2018
Record ID: 2219057
Full text PDF: http://hdl.handle.net/1773/40959


In this thesis I present my research on the early stages of planet formation. Using advanced computational modeling techniques, I study global gas and gravitational dynamics in proto- planetary disks (PPDs) on length scales from the radius of Jupiter to the size of the solar system. In that environment, I investigate the formation of gas giants and the migration, enhancement, and distribution of small solidsthe precursors to planetesimals and gas giant cores. I examine numerical techniques used in planet formation and PPD modeling, especially methods for generating initial conditions (ICs) in these unstable, chaotic systems. Disk simulation outcomes may depend strongly on ICs, which may explain results in the literature. I present the largest suite of high resolution PPD simulations to-date and argue that direct fragmentations of PPDs around M-Dwarfs is a plausible path to rapidly forming gas giants. I implement dust physics to track the migration of centimeter and smaller dust grains in very high resolution PPD simulations. While current dust methods are slow, with strict resolution and/or time-stepping requirements, and have some serious numerical issues, we can still demonstrate that dust does not concentrate at the pressure maxima of spiral arms, an indication that spiral features observed in the dust component are at least as well resolved in the gas. Additionally, coherent spiral arms do not limit dust settling. We suggest a novel mechanism for disk fragmentation at large radii driven by dust accretion from the surrounding nebula. We also investigate self induced dust traps, a mechanism which may help explain the growth of solids beyond meter sizes. We argue that current apparent demonstrations of this mechanism may be due to numerical artifacts and require further investigation.Advisors/Committee Members: Quinn, Thomas (advisor), den Nijs, Marcel (advisor).