J/ApJS/234/9 A spectral approach to transit timing variations (Ofir+, 2018)
A spectral approach to transit timing variations.
Ofir A., Xie J.-W., Jiang C.-F., Sari R., Aharonson O.
<Astrophys. J. Suppl. Ser., 234, 9 (2018)>
=2018ApJS..234....9O 2018ApJS..234....9O
ADC_Keywords: Stars, double and multiple ; Exoplanets
Keywords: catalogs ; methods: numerical ; planetary systems
Abstract:
The high planetary multiplicity revealed by Kepler implies that
transit timing variations (TTVs) are intrinsically common. The usual
procedure for detecting these TTVs is biased to long-period, deep
transit planets, whereas most transiting planets have short periods
and shallow transits. Here we introduce the Spectral Approach
technique to TTVs that allows expanding the TTV catalog toward lower
TTV amplitude, shorter orbital period, and shallower transit depth. In
the spectral approach, we assume that a sinusoidal TTV exists in the
data and then calculate the improvement to χ2 that this model
allows over that of the linear-ephemeris model. This enables detection
of TTVs even in cases where the transits are too shallow, so that
individual transits cannot be timed. The spectral approach is more
sensitive because it has fewer free parameters in its model. Using the
spectral approach, we (a) detect 129 new periodic TTVs in Kepler data
(an increase of ∼2/3 over a previous TTV catalog); (b) constrain the
TTV periods of 34 long-period TTVs and reduce amplitude errors of
known TTVs; and (c) identify cases of multi-periodic TTVs, for which
absolute planetary mass determination may be possible. We further
extend our analysis by using perturbation theory assuming a small TTV
amplitude at the detection stage, which greatly speeds up our
detection (to a level of few seconds per star). Our extended TTV
sample shows no deficit of short-period or low-amplitude transits, in
contrast to previous surveys, in which the detection schemes were
significantly biased against such systems.
Description:
We used Kepler Data Release 24 as the source data and the Kepler
Objects of Interest (KOIs) table downloaded from the NExSci archive on
2015 December 25 as the source of list of candidate signals, and
processed 4706 object not dispositioned as "false positive". We remove
eclipsing binaries from the candidates list (see section 4.1 for
further details).
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
table1.dat 264 437 Results of the combined perturbative approximation
(PA) and the full MCMC spectral approach fit
--------------------------------------------------------------------------------
See also:
V/133 : Kepler Input Catalog (Kepler Mission Team, 2009)
J/ApJS/197/2 : Transit timing observations from Kepler. I. (Ford+, 2011)
J/MNRAS/421/2342 : 4 Kepler systems transit timing obs. (Steffen+, 2012)
J/ApJ/756/185 : Kepler TTVs. V. Metrics catalog (Ford+, 2012)
J/ApJS/208/16 : Kepler transit timing observations. VIII. (Mazeh+, 2013)
J/A+A/555/A58 : New Kepler planetary candidates (Ofir+, 2013)
J/ApJS/208/22 : Transit timing variation for 12 planetary pairs (Xie, 2013)
J/ApJ/787/80 : 139 Kepler planets transit time variations (Hadden+, 2014)
J/ApJS/210/20 : Small Kepler planets radial velocities (Marcy+, 2014)
J/ApJ/784/45 : Kepler's multiple planet candidates. III. (Rowe+, 2014)
J/ApJS/210/25 : Transit timing var. for 15 planetary pairs. II. (Xie, 2014)
J/A+A/573/A124 : Kepler-117 transit-timing variations (Bruno+, 2015)
J/MNRAS/455/4136 : Kepler triples (Borkovits+, 2016)
J/ApJS/225/9 : Kepler TTVs. IX. Full long-cadence data set (Holczer+, 2016)
J/AJ/151/68 : Kepler Mission. VII. Eclipsing binaries in DR3 (Kirk+, 2016)
J/ApJ/822/86 : False positive prob. for Q1-Q17 DR24 KOIs (Morton+, 2016)
J/A+A/587/A64 : Physical properties of giant exoplanets (Santerne+, 2016)
J/AJ/154/5 : Transit timing variations of Kepler planets (Hadden+, 2017)
http://exoplanetarchive.ipac.caltech.edu/ : NASA Exoplanet Archive home page
Byte-by-byte Description of file: table1.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 7 F7.2 --- KOI [12.01/7594.01] Kepler Object of Interest
identifier
9 I1 --- f_KOI [0/2]? Flag on KOI (1)
11- 19 F9.4 10-4/d fTTV [1.7/7487] Frequency of TTV signal
21- 27 F7.4 10-4/d E_fTTV [0.008/12] Upper uncertainty in fTTV
29- 34 F6.4 10-4/d e_fTTV [0.007/8] Lower uncertainty in fTTV
36- 39 I4 d PTTV [1/5666] Period of TTV signal
41- 46 I6 --- dchi2 [10/376607] The Δχ2 of TTV signal
over linear model
48- 52 F5.3 --- Area [0.02/2.5] Cumulative Δχ2 area test
54- 61 F8.3 --- Single [-929/-0.7] Cumulative Δχ2
single-point test
63- 71 F9.3 --- RMS [1/24580] Cumulative Δχ2 RMS test
73- 77 F5.3 --- Corr [0.6/1] Cumulative Δχ2 correlation
test
79- 87 F9.4 min Amp [0.1/1080] TTV signal amplitude
89- 96 F8.4 min E_Amp [0.007/197] Upper uncertainty in Amp
98-105 F8.4 min e_Amp [0.005/306] Lower uncertainty in Amp
107-115 F9.4 d T0 [10/1940] TTV signal reference time; KBJD
117-124 F8.4 d E_T0 [0.005/722] Upper uncertainty in T0
126-133 F8.4 d e_T0 [0.006/713] Lower uncertainty in T0
135-139 F5.3 --- Conf [0.9/1] Bootstrap test confidence estimation
from 1000 runs
141-145 F5.2 --- Ratio [1/15.5] Ratio of scatter (2)
147-179 A33 10-4/d OFreq ? Other PA-detected independent TTV
frequencies (3)
181-184 F4.2 10-4/d e_OFreq [0/7]? Uncertainty on OFreq
186-190 F5.1 10-4/d SFreq [1.7/985] Expected stroboscopic frequency
192-204 A13 --- Ref Previous reference code(s) (4)
206-264 A59 --- Comm Comments
--------------------------------------------------------------------------------
Note (1): Flag as follows:
0 = not a new detection.
1 = new detection.
2 = new determination of the TTV period to a previously-polynomial TTV.
Note (2): Around the linear model to the median error.
Note (3): Sub-list of other significant frequencies. If no other significant
frequencies were found this row is blank. If up to five other
significant frequencies were identified they are provided in a comma
separated list. If more than five other significant frequencies were
identified then the row is ">5".
Note (4): This is a comma separated list. Code as follows:
1 = Holczer et al. (2016, J/ApJS/225/9);
2 = Xie (2013, J/ApJS/208/22) + Xie (2014, J/ApJS/210/25);
3 = Hadden & Lithwick (2014, J/ApJ/787/80) (high significance);
4 = Hadden & Lithwick (2016ApJ...828...44H 2016ApJ...828...44H);
5 = Jontof-Hutter et al. (2016ApJ...820...39J 2016ApJ...820...39J);
6 = Van Eylen & Albrecht (2015ApJ...808..126V 2015ApJ...808..126V)
See Section 5.1 for details.
--------------------------------------------------------------------------------
History:
From electronic version of the journal
(End) Prepared by [AAS], Emmanuelle Perret [CDS] 13-Mar-2018