Analysis of transit signals measured with Corot prior to follow-up observations

Transcription

1 Analysis of transit signals measured with Corot prior to follow-up observations Pascal Bordé & Olivier Demangeon Institut d astrophysique spatiale Université Paris-Sud Planet validation workshop LAM, Marseille 14 May 2013 Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 1 / 29

2 Outline 1 Aperture photometry in the exoplanet channel 2 Modeling of exoplanet imagettes 3 Chasing exoplanet false positives Corot-1: a genuine planet : a contaminating eclipsing binary 4 Bayesian analysis for the ranking of transits 5 Conclusions Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 2 / 29

3 Aperture photometry in the exoplanet channel 1 Aperture photometry in the exoplanet channel 2 Modeling of exoplanet imagettes 3 Chasing exoplanet false positives Corot-1: a genuine planet : a contaminating eclipsing binary 4 Bayesian analysis for the ranking of transits 5 Conclusions Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 3 / 29

4 Aperture photometry in the exoplanet channel From photometric masks to (3-color) light curves Corot-1 as seen by Corot Corot-1 white and color LCs Because the PSF is large (10 pix 23 ), neighboring PSFs overlap. The fluxes of neighbor stars leak into the target mask. Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 4 / 29

5 Aperture photometry in the exoplanet channel Consequences for exoplanet search Flux contamination by neighboring stars: modifies transit relative depths, hence inferred companion radii; creates exoplanetary-like transits = false positives. Know your target environments + correct for flux contaminations. Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 5 / 29

6 Modeling of exoplanet imagettes 1 Aperture photometry in the exoplanet channel 2 Modeling of exoplanet imagettes 3 Chasing exoplanet false positives Corot-1: a genuine planet : a contaminating eclipsing binary 4 Bayesian analysis for the ranking of transits 5 Conclusions Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 6 / 29

7 Modeling of exoplanet imagettes Using higher resolution images Corot imagette Palomar + Exodat Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 7 / 29

8 Modeling of exoplanet imagettes The need for a good stellar catalog PPMXL catalog Exodat catalog Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 8 / 29

9 Corot PSFs Modeling of exoplanet imagettes Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 9 / 29

10 Modeling of exoplanet imagettes Contamination computation Original Computed Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 10 / 29

11 Modeling of exoplanet imagettes Contamination computation Original Target removed Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 11 / 29

12 Modeling of exoplanet imagettes Contamination results white = 3.1 %, blue = 0.6 %, green = 1.1 %, red = 5.0 % Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 12 / 29

13 Modeling of exoplanet imagettes What about accuracy/uncertainty? Complex problem should be Monte-Carlo computed (on-going work). Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 13 / 29

14 Chasing exoplanet false positives 1 Aperture photometry in the exoplanet channel 2 Modeling of exoplanet imagettes 3 Chasing exoplanet false positives Corot-1: a genuine planet : a contaminating eclipsing binary 4 Bayesian analysis for the ranking of transits 5 Conclusions Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 14 / 29

15 Chasing exoplanet false positives Corot-1: a genuine planet Transit detection by matched filtering Correlation map Period and phase correlograms Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 15 / 29

16 Chasing exoplanet false positives Corot-1: a genuine planet Transit depth, duration, and stellar density tests Classical trapezoid model ρ /ρ = ± T max 14 = 3.5 h assuming 2.9 M Round-bottom trapezoid model ρ /ρ = 0.80 ± 0.29 True density: 0.69 ± 0.12 Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 16 / 29

17 Chasing exoplanet false positives Individual transit check Corot-1: a genuine planet Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 17 / 29

18 Chasing exoplanet false positives Individual transit statistics Corot-1: a genuine planet Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 18 / 29

19 Chasing exoplanet false positives Corot-1: a genuine planet Odd/even transit depth difference and secondary transit tests Separate modeling of odd & even transits Search for a secondary transit at orbital phase 0.5 Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 19 / 29

20 Chasing exoplanet false positives : a contaminating eclipsing binary Colors as spatially separated photometric channels Folded white transit Folded color transits Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 20 / 29

21 Target neighborhood Chasing exoplanet false positives : a contaminating eclipsing binary Computed imagette Palomar + Exodat Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 21 / 29

22 Chasing exoplanet false positives Contamination calculations : a contaminating eclipsing binary # red (%) dr/r green (%) dg/g blue (%) db/b > > > > > > > > For a CEB, we expect: dr/r dg/g db/b. the culprit is contaminant #3! Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 22 / 29

23 Bayesian analysis for the ranking of transits 1 Aperture photometry in the exoplanet channel 2 Modeling of exoplanet imagettes 3 Chasing exoplanet false positives Corot-1: a genuine planet : a contaminating eclipsing binary 4 Bayesian analysis for the ranking of transits 5 Conclusions Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 23 / 29

24 Bayesian analysis for the ranking of transits BART: set of models Olivier Demangeon s PhD thesis Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 24 / 29

25 Bayesian analysis for the ranking of transits BART: parameter estimation Olivier Demangeon s PhD thesis Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 25 / 29

26 Bayesian analysis for the ranking of transits BART: Markov Chain Monte Carlo Olivier Demangeon s PhD thesis Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 26 / 29

27 Bayesian analysis for the ranking of transits BART: preliminary results Olivier Demangeon s PhD thesis Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 27 / 29

28 Conclusions 1 Aperture photometry in the exoplanet channel 2 Modeling of exoplanet imagettes 3 Chasing exoplanet false positives Corot-1: a genuine planet : a contaminating eclipsing binary 4 Bayesian analysis for the ranking of transits 5 Conclusions Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 28 / 29

29 Conclusions Transit analysis & sorting based on the light curve The target star neighborhood needs to be properly studied; Many sanity checks can be performed on the light curve; Some exoplanet false positives can be readily identified; Bayesian methods can: provide a nice way to integrate all these tests, sort exoplanet transit candidates using probabilities, save time by prioritizing follow-up observations; This kind of analysis could be the starting point for a more thorough (and computationaly intensive) analysis with e.g., Blender or Pastis; For a similar approach, see Morton, ApJ 761, 6 (2012). Pascal Bordé & Olivier Demangeon (IAS) Corot transit analysis Planet validation workshop 29 / 29