Flicker as a tool for characterizing planets through Asterodensity Profiling

Abstract: Variability in the time series brightness of a star on a timescale of 8 hours, known as 'flicker', has been previously demonstrated to serve as a proxy for the surface gravity of a star by Bastien et al. (2013). Although surface gravity is crucial for stellar classification, it is the mean stellar density which is most useful when studying transiting exoplanets, due to its direct impact on the transit light curve shape. Indeed, an accurate and independent measure of the stellar density can be leveraged to infer subtle properties of a transiting system, such as the companion's orbital eccentricity via asterodensity profiling. We here calibrate flicker to the mean stellar density of 439 Kepler targets with asteroseismology, allowing us to derive a new empirical relation given by $\log_{10}(\rho_{\star}\,[\mathrm{kg}\,\mathrm{m}^{-3}]) = 5.413 - 1.850 \log_{10}(F_8\,[\mathrm{ppm}])$. The calibration is valid for stars with $4500$K$<T_{\mathrm{eff}}<6500$K, $K_P<14$ and flicker estimates corresponding to stars with $3.25<\log g_{\star}<4.43$. Our relation has a model error in the stellar density of 31.7% and so has $\sim8$ times lower precision than that from asteroseismology but is applicable to a sample $\sim40$ times greater. Flicker therefore provides an empirical method to enable asterodensity profiling on hundreds of planetary candidates from present and future missions.