Population-level eccentricity distributions of imaged exoplanets and brown dwarf companions: dynamical evidence for distinct formation channels.
BOWLER B.P., BLUNT S.C. and NIELSEN E.L.
Abstract (from CDS):
The orbital eccentricities of directly imaged exoplanets and brown dwarf companions provide clues about their formation and dynamical histories. We combine new high-contrast imaging observations of substellar companions obtained primarily with Keck/NIRC2 together with astrometry from the literature to test for differences in the population-level eccentricity distributions of 27 long-period giant planets and brown dwarf companions between 5 and 100 au using hierarchical Bayesian modeling. Orbit fits are performed in a uniform manner for companions with short orbital arcs; this typically results in broad constraints for individual eccentricity distributions, but together as an ensemble, these systems provide valuable insight into their collective underlying orbital patterns. The shape of the eccentricity distribution function for our full sample of substellar companions is approximately flat from e = 0-1. When subdivided by companion mass and mass ratio, the underlying distributions for giant planets and brown dwarfs show significant differences. Low mass ratio companions preferentially have low eccentricities, similar to the orbital properties of warm Jupiters found with radial velocities and transits. We interpret this as evidence for in situ formation on largely undisturbed orbits within massive extended disks. Brown dwarf companions exhibit a broad peak at e ≃ 0.6-0.9 with evidence for a dependence on orbital period. This closely resembles the orbital properties and period-eccentricity trends of wide (1-200 au) stellar binaries, suggesting that brown dwarfs in this separation range predominantly form in a similar fashion. We also report evidence that the "eccentricity dichotomy" observed at small separations extends to planets on wide orbits: the mean eccentricity for the multi-planet system HR 8799 is lower than for systems with single planets. In the future, larger samples and continued astrometric orbit monitoring will help establish whether these eccentricity distributions correlate with other parameters such as stellar host mass, multiplicity, and age.