• DOI:10.1088/2041-8205/713/2/L103
• Corpus ID: 119236240

Pre-Spectroscopic False Positive Elimination of Kepler Planet Candidates

@article{Batalha2010PreSpectroscopicFP,
  title={Pre-Spectroscopic False Positive Elimination of Kepler Planet Candidates},
  author={Natalie M. Batalha and Jason F. Rowe and Ronald L. Gilliland and Jon M. Jenkins and Douglas A. Caldwell and William J. Borucki and David G. Koch and Jack J. Lissauer and Edward Wood Dunham and Thomas Gautier and Steve B. Howell and David W. Latham and Geoffrey W. Marcy and Andrej Pr{\vs}a},
  journal={The Astrophysical Journal},
  year={2010},
  volume={713}
}

• N. Batalha, J. Rowe, +11 authors A. Prša
• Published 3 January 2010
• Physics, Geology
• The Astrophysical Journal

Ten days of commissioning data (Quarter 0) and 33 days of science data (Quarter 1) yield instrumental flux time series of ~150,000 stars that were combed for transit events, termed threshold crossing events(TCE), each having a total detection statistic above 7.1?. TCE light curves are modeled as star+planet systems. Those returning a companion radius smaller than 2RJ are assigned a Kepler Object of Interest (KOI) number. The raw flux, pixel flux, and flux-weighted centroids of every KOI are scrutinized to assess the likelihood of being an astrophysical false positive versus the likelihood of being a planetary companion. This vetting using Kepler data is referred to as data validation (DV). Herein, we describe the DV metrics and graphics used to identify viable planet candidates amongst the KOIs. Light curve modeling tests for (1) the difference in depth of the odd- versus even-numbered transits, (2) evidence of ellipsoidal variations, and (3) evidence of a secondary eclipse event at phase = 0.5. Flux-weighted centroids are used to test for signals correlated with transit events with a magnitude and direction indicative of a background eclipsing binary. Centroid time series are complimented by analysis of images taken in-transit versus out-of-transit, the difference often revealing the pixel contributing the most to the flux change during transit. Examples are shown to illustrate each test. Candidates passing DV are submitted to ground-based observers for further false-positive elimination or confirmation/characterization.