Bayesian Statistical Inference of Giant Planet Physics
- Author(s): Thorngren, Daniel Peter
- Advisor(s): Fortney, Jonathan J
- et al.
The many exoplanet discoveries of recent years have opened new avenues for studying giant planets and their formation. The giant planets of our solar system have been studied up close and in great detail, and exoplanets can complement this with a rich population to examine statistically. More than just studying their occurrence rates, it is possible to combine physical and statistical models to uncover aspects of their physical processes. I apply this strategy here on a series of related topics. First, I study a set of cool giant exoplanets, infer their bulk compositions, and demonstrate that there is a relationship between a planet's mass and its composition. I further discuss the implications to their formation, and how a planet's bulk composition can usefully complement its observed atmospheric abundances. I also consider hot Jupiters, inferring the amount of internal heating required to explain their anomalously large radii, the cause of which is one of the longest standing open questions in exoplanet science. I show through a careful examination of their radii and parent star evolution that these objects appear to reinflate quickly when their equilibrium temperature is increased. This strongly constrains the physical mechanisms that are causing their inflation. Finally, I outline several immediately relevant areas for future work to better understand these objects.