Presentation #405.06 in the session Extrasolar Planets: Populations — iPoster Session.
Orbital eccentricities of directly imaged substellar companions such as giant planets and brown dwarfs are a powerful probe of these objects’ formational and dynamical histories. Recently, hierarchical Bayesian modeling (HBM) was used to find evidence that imaged giant planets and brown dwarfs have different population eccentricity distributions, pointing to distinct formation mechanisms for these objects. Here, we study the effect of hyperpriors on the population-level eccentricity distributions inferred from HBM, and find that the choice of hyperpriors can impact the recovered population-level eccentricity distributions, an effect that grows stronger for decreasing sample size and orbital coverage. We conduct forward modeling experiments to establish constraints on the family of minimally biased hyperprior distributions, sample size, and orbital coverage required to accurately recover an observational sample’s underlying eccentricity distribution using HBM. Applying our method to an observational sample of imaged giant planets at 5-100 AU, we find that the population eccentricity distribution is broadly consistent with those of radial velocity ”Warm Jupiters’, similar to previous results. We package our HBM and forward modeling code into a Python package, ePop!, and make it freely available for the community to use.