Long-term orbital evolution of planetary systems: stability, migration, and other fates.

Abstract

This thesis develops two main topics related to the study of the long-term orbital evolution of planetary systems.The first one is the long-term dynamical stability of

planetary systems. By running a large number of orbit integrations I find a new and easy-to-use criterion for the long-term stability of two-planet systems. I discuss the fates of the dynamically unstable systems and show that at small orbital separations these systems generally lead to planet-planet collisions with small eccentricity and inclination excitation---a mechanism that is unable to account for the population of short-period planets in eccentric and/or large-obliquity orbits. This result poses a challenge to the theory of disk migration, motivating alternative models of planet migration.

The second main topic is the migration of gas giant planets into short-period orbits by

the gravitational interactions with a distant third body and tidal friction. First, I show that when the third body is a planet in an eccentric and low-inclination orbit, the migration occurs in nearly the same plane as the planets formed, producing low-obliquity hot Jupiters. Based on the observed periods, stellar obliquities, and occurrence rates, I show that this migration channel can account for most of the hot Jupiters.

Second, I show that if the third body is a wide stellar binary companion in a highly inclined orbit, the so-called Kozai-Lidov mechanism produces hot Jupiters with periods that are shorter than in the observations and it does not produce warm Jupiters. Based on the observed periods and occurrence rates, I show that the Kozai-Lidov mechanism can produce at most $\sim20\%$ of the hot Jupiters. Third, I show that when the third body is a planet in an eccentric and/or inclined orbit, it produces a population of warm Jupiters with a flat eccentricity distribution, possibly accounting for the population of eccentric warm Jupiters with outer planetary companions, otherwise difficult to account for by either disk-migration or other high-eccentricity migration channels.