J/ApJ/767/64 Benchmark light curves for exoplanet eclipses (Rogers+, 2013)
Benchmark tests for Markov Chain Monte Carlo fitting of exoplanet eclipse
observations.
Rogers J., Lopez-Morales M., Apai D., Adams E.
<Astrophys. J., 767, 64 (2013)>
=2013ApJ...767...64R 2013ApJ...767...64R
ADC_Keywords: Planets ; Stars, double and multiple ; Models ; Photometry
Keywords: binaries: eclipsing; methods: analytical; methods: statistical;
planetary systems; stars: individual: CoRoT-1; techniques: photometric
Abstract:
Ground-based observations of exoplanet eclipses provide important
clues to the planets' atmospheric physics, yet systematics in light
curve analyses are not fully understood. It is unknown if measurements
suggesting near-infrared flux densities brighter than models predict
are real, or artifacts of the analysis processes. We created a large
suite of model light curves, using both synthetic and real noise, and
tested the common process of light curve modeling and parameter
optimization with a Markov Chain Monte Carlo algorithm. With synthetic
white noise models, we find that input eclipse signals are generally
recovered within 10% accuracy for eclipse depths greater than the
noise amplitude, and to smaller depths for higher sampling rates and
longer baselines. Red noise models see greater discrepancies between
input and measured eclipse signals, often biased in one direction.
Finally, we find that in real data, systematic biases result even with
a complex model to account for trends, and significant false eclipse
signals may appear in a non-Gaussian distribution. To quantify the
bias and validate an eclipse measurement, we compare both the
planet-hosting star and several of its neighbors to a separately
chosen control sample of field stars. Re-examining the Rogers et al.
(2009, J/ApJ/707/1707) Ks-band measurement of CoRoT-1b finds an
eclipse 3190-440+370ppm deep centered at
φme=0.50418-0.00203+0.00197. Finally, we provide and
recommend the use of selected data sets we generated as a benchmark
test for eclipse modeling and analysis routines, and propose criteria
to verify eclipse detections.
Description:
We selected a number of representative light curves from our studies
that can be used as benchmark testing for any routines designed to fit
and measure eclipses: 10 data sets that span the full breadth of our
tests, both the synthetic noise models (S1-S7; white and red noise)
and the real data systematics (R1-R3; from the CoRoT-1 Ks-band
photometry). Various baselines from 160 to 506 minutes and sampling
rates from 25 to 250 points per hour are featured. The depth and
central phase of the eclipses that are input are provided, so that any
team can check the results from their analysis routine against them.
Objects:
-----------------------------------------------------------------
RA (ICRS) DE Designation(s) (Period)
-----------------------------------------------------------------
06 48 19.17 -03 06 07.8 CoRoT-1b = CoRoT-1b (P=1.508956)
-----------------------------------------------------------------
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
table10.dat 79 2865 Data sets S1-S7: synthetic noise
table11.dat 116 2067 Data sets R1, R2 and R3: real noise
--------------------------------------------------------------------------------
See also:
B/corot : CoRoT observation log Release 13 (CoRoT, 2009-2014)
J/AJ/143/39 : Analysis of hot Jupiters in Kepler Q2 (Coughlin+, 2012)
J/A+A/530/A5 : WASP-4b Ks-band detection of thermal emission (Caceres+, 2011)
J/A+A/528/A49 : HAT-P-1b Ks-band secondary eclipse (de Mooij +, 2011)
J/A+A/513/L3 : Exoplanet WASP-19b H-band thermal emission (Anderson+, 2010)
J/ApJ/707/1707 : Ks-band light curve of CoRoT-1b (Rogers+, 2009)
J/A+A/506/359 : FORS2 and HAWKI photometry of CoRoT-1 (Gillon+, 2009)
J/A+A/482/L17 : CoRoT space mission. I. (Barge+, 2008)
Byte-by-byte Description of file: table10.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1 I1 --- Type Dataset description code (1)
3- 11 F9.7 --- Phase [0.33/0.67] Orbital phase
13- 22 F10.8 --- OFlux [0.995/1.009] Observed flux
24- 33 F10.8 --- e_OFlux [0/0.002] Error in OFlux
35- 44 F10.8 --- EShape Noiseless eclipse shape (G1)
46- 55 F10.8 --- NBase Combined noise base
57- 67 F11.8 --- WNoise White noise components
69- 79 F11.8 --- RNoise Red noise components
--------------------------------------------------------------------------------
Note (1): Benchmark data set attributes as in table 9:
-------------------------------------------------------------------------
Type σw σr Obs. Win Samp. Rate De φme
(ppm) (ppm) (min) (ppm)
-------------------------------------------------------------------------
1 = Synth 1000 0 240 75 4000 0.500
2 = Synth 1000 0 240 75 3300 0.475
3 = Synth 1000 0 160 75 1000 0.500
4 = Synth 1000 0 506 75 1600 0.500
5 = Synth 1000 200 240 75 2000 0.500
6 = Synth 1000 1000 240 250 4000 0.500
7 = Synth 1000 200 318 25 2600 0.500
--------------------------------------------------------------------------
σw and σr = white and red noise amplitude (input values
that we started with)
Samp. Rate = sampling rate (in points per hour)
De = Input eclipse signal's depth
φme = mid-eclipse phase
--------------------------------------------------------------------------
--------------------------------------------------------------------------------
Byte-by-byte Description of file: table11.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1 I1 --- Type Dataset description code (1)
3- 12 F10.8 --- Phase Orbital phase
14- 23 F10.8 --- OFlux [0.984/1.019] Observed flux
25- 34 F10.8 --- e_OFlux [0.003/0.007] Error in OFlux
36- 45 F10.8 --- EShape [1/1.004] Noiseless eclipse shape (G1)
47- 56 F10.8 --- NBase [0.98/1.02] Combined noise base
58 I1 --- OPos [0/1]? Offset position (2)
60- 69 F10.8 --- AirM [1.2/1.8] Airmass
71- 81 F11.8 pix FWHM [10.4/12.1] Full-width at Half-Maximum of
stellar image
83- 93 F11.8 pix xDist [-6.1/6.9] X-position on the chip
95-105 F11.8 pix yDist [-4.4/6.8] Y-position on the chip
107-116 F10.6 ct/pix Sky [-63/58] Sky brightness
--------------------------------------------------------------------------------
Note (1): Benchmark data set attributes as in table 9:
--------------------------------------------------------------------------
Type σw σr Obs. Win Samp. Rate De φme
(min)
--------------------------------------------------------------------------
1 = Star 2 6188 2239 293 141 4000 0.500
2 = Star 0 7318 1225 293 141 1500 0.500
3 = Star 0 7318 1225 293 141 0 0.500
--------------------------------------------------------------------------
σw and σr = white and red noise amplitude (measured by
fitting to Equation (1), in ppm; see section 2.2).
Samp. Rate = sampling rate (in points per hour)
De = Input eclipse signal's depth (in ppm)
φme = mid-eclipse phase
--------------------------------------------------------------------------
* Set R1 treats comparison star 2 as the target and adds a 4000ppm
eclipse, modeled using the characteristics of the model "Exoplanet X".
Refer to table 1 and Section 4.1 for all the parameters that are needed
to reconstruct the eclipse signal in light curve.
* Sets R2 and R3 use the light curve from the actual target star,
with set R2 having an extra 1500 ppm eclipse added on top of the
true signal. See the Appendix for further explanations.
Note (2): Offset position as follows:
1 = initial field position;
0 = second offset position, 16arcsec East of initial position.
--------------------------------------------------------------------------------
Global note:
Note (G1): For our purposes, we considered the planet to be seen as a
uniformly bright disk, allowing us to model the eclipse shape as:
Fecl(φ)=Fb*{1+De*[1-Pecl(φ,φme_)]} (Eq.6) where
Fb is a normalizing baseline flux parameter.
Pecl(φ,φme) was found using Equations (3)-(5),
the mid-eclipse phase φ_me, and the other fixed parameters, and
then used with De to model the expected flux from the star and
planet throughout the eclipse (flux of 1 in full eclipse, 1+De when
the planet is fully visible, normalized to the brightness of the star
alone). Thus, with all the other star and planet parameters held
constant, the model for the eclipse shape depends only on Fb, De,
and φme. See section 3.1.
--------------------------------------------------------------------------------
History:
From electronic version of the journal
(End) Greg Schwarz [AAS], Emmanuelle Perret [CDS] 19-Nov-2014