[Paper Review] SOPHIE velocimetry of Kepler transit candidates VII. A false-positive rate of 35% for Kepler close-in giant exoplanet candidates

A. Santerne, R. F. Díaz+7 more|Jun 4, 2012

Stellar, planetary, and galactic studies12 references95 citations

TL;DR

This study uses high-precision radial velocity measurements from the SOPHIE spectrograph to assess the false-positive rate (FPP) of Kepler close-in giant planet candidates. Among 46 candidates with transit depth >0.4%, orbital period <25 days, and Kp <14.7, the FPP is found to be 34.8% ±6.5%, significantly higher than prior estimates of <5%, indicating a substantial contamination of false positives from eclipsing binaries and brown dwarfs.

ABSTRACT

The false-positive probability (FPP) of Kepler transiting candidates is a key value for statistical studies of candidate properties. A previous investigation of the stellar population in the Kepler field has provided an estimate for the FPP of less than 5% for most of the candidates. We report here the results of our radial velocity observations on a sample of 46 Kepler candidates with a transit depth greater than 0.4%, orbital period less than 25 days and host star brighter than Kepler magnitude 14.7. We used the SOPHIE spectrograph mounted on the 1.93-m telescope at the Observatoire de Haute-Provence to establish the nature of the transiting candidates. In this sample, we found five undiluted eclipsing binaries, two brown dwarfs, six diluted eclipsing binaries, and nine new transiting planets that complement the 11 already published planets. The remaining 13 candidates were not followed-up or remain unsolved due to photon noise limitation or lack of observations. From these results we computed the FPP for Kepler close-in giant candidates to be 34.8% \pm 6.5%. We aimed to investigate the variation of the FPP for giant candidates with the longer orbital periods and found that it should be constant for orbital periods between 10 and 200 days. This significant disagrees with the previous estimates. We discuss the reasons for this discrepancy and the possible extension of this work toward smaller planet candidates. Finally, taking the false-positive rate into account, we refined the occurrence rate of hot jupiters from the Kepler data.

Motivation & Objective

• To measure the true false-positive rate (FPP) for Kepler close-in giant planet candidates using radial velocity follow-up.
• To investigate the nature of transiting candidates in the Kepler field that exhibit deep transits and short orbital periods.
• To assess whether the FPP remains constant for longer-period giant planets (10–200 days), challenging previous statistical estimates.
• To refine the occurrence rate of hot Jupiters by accounting for the newly measured FPP.
• To identify and characterize false positives such as eclipsing binaries, brown dwarfs, and blended systems among the candidates.

Proposed method

• Selected 46 Kepler candidates with transit depth >0.4%, orbital period <25 days, and Kp <14.7 for radial velocity follow-up.
• Conducted radial velocity measurements using the SOPHIE spectrograph on the 1.93-m telescope at Observatoire de Haute-Provence.
• Analyzed radial velocity variations to distinguish between planetary signals and stellar binary systems.
• Used orbital solution fitting and spectral energy distribution modeling to validate binary systems and detect blended eclipsing binaries.
• Applied statistical analysis to compute the FPP based on the number of confirmed planets versus false positives.
• Extended analysis to assess FPP trends across orbital periods from 10 to 200 days.

Experimental results

Research questions

• RQ1 What is the true false-positive rate (FPP) for Kepler close-in giant planet candidates with deep transits and short periods?
• RQ2 How do the FPP estimates from radial velocity follow-up compare to prior statistical estimates based on stellar population modeling?
• RQ3 Is the FPP for giant planet candidates constant across orbital periods between 10 and 200 days?
• RQ4 What is the impact of the revised FPP on the inferred occurrence rate of hot Jupiters in the Kepler sample?
• RQ5 What types of false positives (e.g., eclipsing binaries, brown dwarfs, blends) dominate the contamination in this sample?

Key findings

• The false-positive rate (FPP) for Kepler close-in giant planet candidates is 34.8% ±6.5%, significantly higher than the previously estimated <5%.
• Among the 46 candidates, five were identified as undiluted eclipsing binaries, two as brown dwarfs, and six as diluted eclipsing binaries, accounting for the majority of false positives.
• Nine new transiting planets were confirmed, complementing 11 previously published planets from the same sample.
• The FPP remains approximately constant for orbital periods between 10 and 200 days, suggesting a stable false-positive contamination level across this range.
• The high FPP contradicts earlier statistical estimates based on stellar population modeling, indicating a need to revise assumptions about false-positive probabilities.
• Accounting for the revised FPP, the true occurrence rate of hot Jupiters in the Kepler data is lower than previously inferred, requiring downward adjustment in exoplanet population statistics.