Characterizing the Variability of Stars with Early-Release Kepler Data

Abstract

We present a variability analysis of the early-release first quarter of data publicly released by the Kepler project. Using the stellar parameters from the Kepler Input Catalog, we have separated the sample into 129,000 dwarfs and 17,000 giants, and further sub-divided the luminosity classes into temperature bins corresponding approximately to the spectral classes A, F, G, K, and M. Utilizing the inherent sampling and time baseline of the public dataset (30 minute sampling and 33.5 day baseline), we have explored the variability of the stellar sample. The overall variability rate of the dwarfs is 25% for the entire sample, but can reach 100% for the brightest groups of stars in the sample. G-dwarfs are found to be the most stable with a dispersion floor of $\sigma \sim 0.04$ mmag. At the precision of Kepler, $>95$% of the giant stars are variable with a noise floor of $\sim 0.1$ mmag, 0.3 mmag, and 10 mmag for the G-giants, K-giants, and M-giants, respectively. The photometric dispersion of the giants is consistent with acoustic variations of the photosphere; the photometrically-derived predicted radial velocity distribution for the K-giants is in agreement with the measured radial velocity distribution. We have also briefly explored the variability fraction as a function of dataset baseline (1 - 33 days), at the native 30-minute sampling of the public Kepler data. To within the limitations of the data, we find that the overall variability fractions increase as the dataset baseline is increased from 1 day to 33 days, in particular for the most variable stars. The lower mass M-dwarf, K-dwarf, G-dwarf stars increase their variability more significantly than the higher mass F-dwarf and A-dwarf stars as the time-baseline is increased, indicating that the variability of the lower mass stars is mostly characterized by timescales of weeks whi...astroph will not allow longer abstract!