A Toolbox for Exoplanet Exploration

Abstract

In this thesis, I develop a new suite of tools to address two questions in exoplanet science: how common are Earth-mass planets in the habitable zones of Solar-type stars, and can we detect signs of life on other worlds? Answering the first question requires a method for detecting Earth-Sun analogs. Currently, the radial velocity (RV) method of exoplanet detection is one of the most successful tools for probing inner planetary systems. However, degeneracy between a spectrometer's wavelength calibration and the astrophysical RV shift has limited the sensitivity of today's instruments. In my thesis, I address a method for breaking this degeneracy: by combining a traditional spectrometer design with a dynamic interferometer, a fringe pattern is generated at the image plane that is highly sensitive to changes in the radial velocity of the target star. I augmented previous theoretical studies of the method, creating an end-to-end simulation to 1) introduce and recover wavelength calibration errors, and 2) investigate the effects of interferometer position errors on the RV precision. My simulation showed that using this kind of interferometric system, a 5-m class telescope could detect an Earth-Sun analog. Addressing the occurrence rate of Earth twins also requires an understanding of planet formation in multiple star systems, which encompass half of all Solar-type stars. Gravitational interactions between binary components separated by 10-100 astronomical units are predicted to truncate the outer edges of their respective disks, possibly reducing the disks' lifetimes. Consequently, the pool of material and the amount of time available for planet formation may be smaller than in single star systems. The stars' rotational periods provide a fossil record of these events: star-disk magnetic interactions initially prevent a contracting pre-main sequence star from spinning up, and hence a star with a shorter-lived disk is expected to be spinning more quickly when it reaches the zero age main sequence. In order to conduct a large-scale multiplicity survey to investigate the relationship between stellar rotation and binary system properties (e.g. their separations and mass ratios), I contributed to the commissioning of Robo-AO, a robotic laser guide star adaptive optics system, at the Kitt Peak 2.1-m. After the instrument's installation, I wrote a data pipeline to optimize the system's sensitivity to close stellar companions via reference star differential imaging. I then characterized Robo-AO's performance during its first year of operations. Finally, I used Robo-AO to search for binaries among the 759 stars in the Pleiades with rotational periods measured using the photometric data of the re-purposed Kepler telescope, K2. Detecting signs of life on other worlds will require detailed characterization of rocky exoplanet atmospheres. Polarimetry has long been proposed as a means of probing these atmospheres, but current instruments lack the sensitivity to detect the starlight reflected and polarized by such small, close-in planets. However, the latest generation of high contrast imaging instruments (e.g. GPI and SPHERE) may be able to detect the polarization of thermal emission by young gas giants due to scattering by aerosols in their atmospheres. Observational constraints on the details of clouds physics imposed by polarized emission will improve our understanding of the planets' compositions, and hence their formation histories. For the case of the brown dwarf HD19467 B orbiting a nearby Sun-like star, I demonstrated that the Gemini Planet Imager can detect linear polarizations on the order predicted for these cloudy exoplanets. My current pilot programs can produce the first detections of polarized exoplanet emission, while also building expertise for reflected starlight polarimetry with future observatories.