Revising Properties of Planet-host Binary Systems I.: Methods and Pilot Study

TL;DR

This study introduces a new MCMC method for analyzing binary star systems with unresolved data, improving the accuracy of stellar and planetary property measurements crucial for understanding planet formation in binaries.

Contribution

The paper presents a novel MCMC fitting algorithm for binary systems that utilizes unresolved spectra and photometry, providing more accurate stellar and planetary parameters.

Findings

Primary star temperatures revised upward by ~200 K.

Planetary radii increased by 20-80% depending on the host star.

Binary contrast in Kepler band averages 0.75 mag, affecting host star identification.

Abstract

To fully leverage the statistical strength of the large number of planets found by projects such as the Kepler survey, the properties of planets and their host stars must be measured as accurately as possible. One key population for planet demographic studies is circumstellar planets in close binaries ($\rho < 50$au), where the complex dynamical environment of the binary inhibits most planet formation, but some planets nonetheless survive. Accurately characterizing the stars and planets in these complex systems is a key factor in better understanding the formation and survival of planets in binaries. Toward that goal, we have developed a new Markov Chain Monte Carlo fitting algorithm to retrieve the properties of binary systems using unresolved spectra, unresolved photometry, and resolved contrasts. We have analyzed 8 Kepler Objects of Interest in M star binary systems using literature data, and have found that the temperatures of the primary stars (and presumed planet hosts) are revised upward by an average of 200 K. The planetary radii should be revised upward by an average of 20% if the primary star is the host, and 80% if the secondary star is the planet host. The average contrast between stellar components in the Kepler band is 0.75 mag, which is small enough that neither star in any of the binaries can be conclusively ruled out as a potential planet host. Our results emphasize the importance of accounting for multiplicity when measuring stellar parameters, especially in the context of exoplanet characterization.