Campus, Conference room 101X - Paul Allen Building
Direct imaging is a method to detect and study extra solar planets, which consists of spatially resolving the light of the planet from the light of the star around which it is orbiting. The method is currently sensitive to young gas giant exoplanets that still radiate the heat from their formation, at orbital separation similar to the gas giants in our solar system. A recurring theme in this work is the application of maximum likelihood and Bayesian inference to the detection and characterization of directly imaged planets. In particular, I have designed and implemented the planet detection scheme for the Gemini Planet Imager Exoplanet Survey, which is a 600-star direct-imaging survey spanning 2014-2018. In addition to improving the statistical standards of the field, the detection limits set by this work have allowed the derivation of planet occurrence rates for giant planets and informed their formation history. Additionally, I have demonstrated how Bayesian inference can be used to constrain the mass of undetected exoplanets, in particular those potentially carving the gaps of proto-planetary disks, which is the disk of orbiting gas a dust from which planets form. Finally, I developed a statistical framework for the analysis of medium resolution spectroscopic data, which I have used to characterize the planets orbiting the star HR 8799. In 2008, HR 8799 b,c, and d were the first exoplanets, and is still the only multi-planet system, that have been directly imaged. From these data, I derived the first radial velocity measurement of HR 8799 b and c and measured water and carbon monoxide in the atmosphere of HR 8799 d for the first time.