Planets around cool stars: a spectroscopic and photometric study of M dwarfs and their planets

Abstract

M dwarfs have become attractive targets for exoplanet searches. Smaller planets can be detected around stars with smaller radii (via the transit technique) or lower mass (via the radial velocity technique). Planets in the habitable zone of an M dwarf are closer in, and thus easier to detect, than those orbiting in the habitable zone of Solar-type stars. However, studying M dwarfs is difficult due to their intrinsic faintness and poorly established stellar parameters (radius, metallicity, etc.). For this dissertation I have worked to mitigate these problems with the goals of finding new transiting planets and establishing the role of stellar metallicity on the size and occurrence of transiting planets around M dwarfs. I describe a new approach to search for planetary transits around M dwarfs by slewing between targets several degrees apart on the sky. This method improves the efficiency of transit searches by enabling one to monitor multiple, separated, M dwarfs simultaneously from the ground. I show that the bright, late-type stars targeted by the Kepler spacecraft are predominately misclassified or unclassified giant stars. The level of giant star contamination has a significant impact on the metallicity distribution of the Kepler M star sample as well as the calculated planet occurrence. I significantly improve on techniques to measure M dwarf metallicities. Using wide binaries with a Solar-type primary and an M dwarf companion as calibrator stars, I perform an unbiased search for metal-sensitive features in visible and near-infrared spectra. This yields a list of spectral lines (or regions) that I then use to develop a series empirical calibrations (valid over different wavelength regimes) of M dwarf metallicities accurate to better than 0.1 dex. Lastly, I apply this technique to Kepler late K and M dwarf targets and planet hosts. I find that late-type stars hosting Neptune-sized and smaller planets have a metallicity distribution indistinguishable from that of the sample with no detected planets. This suggests that planet formation must be efficient in collecting material from the disk or that disk masses for M dwarfs are more massive than basic scaling would suggest.