The Kepler Follow-Up Observation Program. II. Stellar Parameters from Medium- and High-Resolution Spectroscopy

Abstract

We present results from spectroscopic follow-up observations of stars identified in the Kepler field and carried out by teams of the Kepler Follow-Up Observation Program. Two samples of stars were observed over six years (2009-2015): 614 standard stars (divided into "platinum" and "gold" categories) selected based on their asteroseismic detections and 2667 host stars of Kepler Objects of Interest (KOIs), most of them planet candidates. Four data analysis pipelines were used to derive stellar parameters for the observed stars. We compare the $T_{\mathrm{eff}}$, $\log$(g), and [Fe/H] values derived for the same stars by different pipelines; from the average of the standard deviations of the differences in these parameter values, we derive error floors of $\sim$ 100 K, 0.2 dex, and 0.1 dex for $T_{\mathrm{eff}}$, $\log$(g), and [Fe/H], respectively. Noticeable disagreements are seen mostly at the largest and smallest parameter values (e.g., in the giant star regime). Most of the $\log$(g) values derived from spectra for the platinum stars agree on average within 0.025 dex (but with a spread of 0.1-0.2 dex) with the asteroseismic $\log$(g) values. Compared to the Kepler Input Catalog (KIC), the spectroscopically derived stellar parameters agree within the uncertainties of the KIC, but are more precise and are thus an important contribution towards deriving more reliable planetary radii.