A new framework for classifying exoplanet systems
Recent work has revealed that exoplanetary systems possess markedly nonrandom architectures (Weiss et al., 2018, Fabrycky et al., 2014). Rather, they follow predictable patterns in the organization of their masses, orbital spacing, and inclinations of the planets. However, the origin of these patterns remains a mystery. Applying an information-theoretic approach (Lopez-Ruiz et al., 1995), we define new quantities to parameterize the complexity (C), monotonicity (M), and equipartitioning (Q) of exoplanetary systems. Our unique approach treats each system as a single coherent unit, in contrast to most prior work which has limited consideration only to pairs of planets. We demonstrate that a relatively small number of empirical parameters - C, M, and Q, plus multiplicity, planet-to-star mass ratio, and other related quantities - can effectively and efficiently group systems into several distinct categories. Moreover, we have begun building dynamical models linking these empirical classifications to distinct formation pathways. Our models build on the work done by Canup & Ward (2006) to reproduce the satellite systems of Jupiter, Saturn, and Uranus. Their models found that gas inflow and clearing rates during accretion are the dominant parameters that set the final properties of satellite systems, and I explore the hypothesis that gas inflow and dispersal play a similar role in protoplanetary disks. We propose a classification scheme for exoplanetary systems and thereby establish a dynamical-empirical framework for interpreting the observed diversity of known planetary systems.