An Efficient Automated Validation Procedure for Exoplanet Transit Candidates

Surveys searching for transiting exoplanets have found many more candidates than they have been able to confirm as true planets. This situation is especially acute with the Kepler survey, which has found over 2300 candidates but has confirmed only 77 planets to date. I present here a general procedure that can quickly be applied to any planet candidate to calculate its false positive probability. This procedure takes into account the period, depth, duration, and shape of the signal; the colors of the target star; arbitrary spectroscopic or imaging follow-up observations; and informed assumptions about the populations and distributions of field stars and multiple-star properties. I also introduce the concept of the "specific occurrence rate," which allows for the calculation of the FPP without relying on an assumed planet radius function. Applying these methods to a sample of known Kepler planets, I demonstrate that many signals can be validated with very limited follow-up observations: in most cases with only a spectrum and an AO image. Additionally, I demonstrate that this procedure can reliably identify false positive signals. Because of the computational efficiency of this analysis, it is feasible to apply it to all Kepler planet candidates in the near future, and it will streamline the follow-up efforts for Kepler and other current and future transit surveys.