Date of Award

1-1-2014

Language

English

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

College/School/Department

Department of Physics

Content Description

1 online resource (vi, 104 pages) : illustrations (some color)

Dissertation/Thesis Chair

Kevin H Knuth

Committee Members

Ariel Caticha, Daniel Angerhausen, Jesse Ernst

Keywords

Bayesian, Exoplanet, Photometric, Extrasolar planets, Astronomical photometry, Bayesian statistical decision theory

Subject Categories

Astrophysics and Astronomy

Abstract

Exoplanets are known to be responsible for a variety of photometric effects, which collectively can be used for both exoplanet detection and characterization. A portion of the observed flux variation originates directly from the exoplanet itself as both a reflected light component and thermal emission. Additional effects originate from the influence of the exoplanet on its host star. These include Doppler boosting, or beaming, caused by the radial velocity variations due to the stellar wobble, as well as variations in flux caused by the ellipsoidal shape of the star, which is induced by the planetary tidal forces. The newly developed EXONEST algorithm uses Bayesian inference in order to estimate the values of the physical parameters on which these effects depend. In addition, EXONEST computes the Bayesian evidence, which can be used to test a variety of models, some of which may either allow for, or neglect, these effects. Presented here is a comprehensive study of model-generated synthetic data demonstrating EXONEST's ability to perform parameter estimation and model selection, two confirmed exoplanets KOI-13b and Kepler-2b further demonstrating EXONEST's ability to work with real data, another confirmed Earth-like planet Kepler-91b, where EXONEST is applied to determine if there is a trojan planet present, and finally KIC-5436161, which is most likely a hierarchical triple star system discovered using the EXONEST algorithm. By considering only the non-transiting portions of the light curve, we demonstrate that it is possible to estimate the photometrically-relevant model parameters of KOI-13b and Kepler-2b, and that the orbit of KOI-13b has a detectable eccentricity. Furthermore, Bayesian model selection shows that there is likely a trojan planet around Kepler-91b based on the photometric data alone.

Share

COinS