Discovery and atmospheric characterization of gian planet Kepler-12b : an inflated radius outlier

Abstract

We report the discovery of planet Kepler-12b (KOI-20), which at 1.695 ± 0.030 RJ is among the handful of planets with super-inﬂated radii above 1.65 RJ.

Orbiting its slightly evolved G0 host with a 4.438-day period, this 0.431 ± 0.041 MJ planet is the least-irradiated within this largest-planet-radius group, which has important implications for planetary physics.

The planet’s inﬂated radius and low mass lead to a very low density of 0.111 ± 0.010 g cm−3.

We detect the occultation of the planet at a signiﬁcance of 3.7σ in the Kepler bandpass.

This yields a geometric albedo of 0.14±0.04; the planetary ﬂux is due to a combination of scattered light and emitted thermal ﬂux.

We use multiple observations with Warm Spitzer to detect the occultation at 7σ and 4σ in the 3.6 and 4.5 µm bandpasses, respectively.

The occultation photometry timing is consistent with a circular orbit, at e < 0.01 (1σ), and e < 0.09 (3σ).

The occultation detections across the three bands favor an atmospheric model with no dayside temperature inversion.

The Kepler occultation detection provides signiﬁcant leverage, but conclusions regarding temperature structure are preliminary, given our ignorance of opacity sources at optical wavelengths in hot Jupiter atmospheres.

If Kepler-12b and HD 209458b, which intercept similar incident stellar ﬂuxes, have the same heavy element masses, the interior energy source needed to explain the large radius of Kepler-12b is three times larger than that of HD 209458b.

This may suggest that more than one radius-inﬂation mechanism is at work for Kepler-12b, or that it is less heavy-element rich than other transiting planets.

Subject headings : planetary systems; stars: individual: (Kepler-12, KOI-20, KIC 11804465), planets and satellites: atmospheres, techniques: spectroscopic.

Introduction :

Transiting planets represent an opportunity to understand the physics of diverse classes of planets, including massradius regimes not found in the solar system.

The knowledge of the mass and radius of an object immediately yields the bulk density, which can be compared to models to yield insight into the planet’s internal composition, temperature, and structure (e.g., Miller & Fortney 2011).

Subsequent observations, at the time of the planet’s occultation (secondary eclipse) allow for the detection of light emitted or scattered by the planet’s atmosphere, which can give clues to a planet’s dayside temperature structure and chemistry (Marley et al. 2007; Seager & Deming 2010). NASA’s Kepler Mission was launched on 7 March 2009 with the goal of ﬁnding Earth-sized planets in Earth-like orbits around Sun-like stars Borucki et al. 2010).

While working towards this multi-year goal, it is also ﬁnding an interesting menagerie of larger and hotter planets that are aiding our understanding of planetary physics.

Early on in the mission, followup radial velocity resources preferentially went to giant planets, for which it would be relatively easy to conﬁrm their planetary nature through ameasurement of planetary mass.

This is how the conﬁrmation of planet Kepler-12b was made, at ﬁrst glance a relatively standard “hot Jupiter” in a 4.438 day orbit. However, upon further inspection, the mass and radius of Kepler- 12b make it an interesting planet from the standpoint of the now-familiar “radius anomaly” of transiting giant planets (e.g.Charbonneau et al. 2007; Burrows et al. 2007; Laughlin et al. 2011).

Given our current understanding of strongly-irradiated giant planet thermal evolution, around 1/3 to 1/2 of known transiting planets are larger than models predict for severalGyr-old planets that cool and contract under intense stellar irradiation (Miller et al. 2009).

The observation that many Jupiter- and Saturn-mass planets are be larger than 1.0 Jupiter-radii can be readily understood.

It is the magnitude of the effect that still needs explanation.

The ﬁrst models of strongly irradiated planets yielded the prediction that these close-in planets would be inﬂated in radius compared to Jupiter and Saturn (Guillot et al. 1.996).

The high incident ﬂux drives the radiative convective boundary from less than a bar, as in Jupiter, to pressures near a kilobar.

The thick radiative zone transports less ﬂux than a fully convective atmosphere, thereby slowing interior cooling, which slows contraction.

A fairly uniform prediction of these strongly irradiated models is that 1.2 − 1.3 RJ is about the largest radii predicted for planets several gigayears-old (Bodenheimer et al.2003;

Burrows et al. 2007; Fortney et al. 2007; Baraffe et al. 2008). However, planets commonly exceed this value.

The mechanism that leads to the radius anomaly has not yet been deﬁnitively identiﬁed.

However, constraints are emerging.

One is planet radius vs. incident ﬂux, which could also be thought of as radius vs. equilibrium temperature, with an assumption regarding planetary Bond albedos.

Since low-mass planets are relatively easier to inﬂate to large radii than higher mass planets (e.g. Miller et al. 2009), we plot the planets in three mass bins.

The lowest mass bin is Saturn-like masses, while the middle mass bin is Jupiter-like masses.

The upper mass bin ends at 13 MJ , the deuterium burning limit. Kepler-12b is shown as a

black ﬁlled circle.

The largest radius planets are generally the most highly irradiated (Kovács et al. 2010; Laughlin et al.2011; Batygin et al. 2011).

The near-universality of the in- ﬂation, especially at high incident ﬂuxes, now clearly argues for a mechanism that affects all close-in planets (Fortney et al.2006), rather than one that affects only some planets.

The distribution of the radii could then be understood in terms of differing magnitudes of the inﬂation mechanism, together with different abundances of heavy elements within the planets (Fortney et al. 2006; Guillot et al. 2006; Burrows et al. 2007; Miller & Fortney 2011; Batygin et al. 2011).

Within this emerging picture, outlier points are particularly interesting: those that are especially large, given their incident ﬂux.

These are the super-inﬂated planets with radii of 1.7 RJ or larger.

These include WASP-12b (Hebb et al. 2009), TrES- 4b (Mandushev et al. 2007; Sozzetti et al.

2009), WASP-17b (Anderson et al. 2010), and now Kepler-12b, which is the least irradiated of the four. In the following we describe the discovery of Kepler-12b, along with the initial characterization of the planet’s atmosphere.

Transiting planets enable the characterization of exoplanet atmospheres.

The Spitzer Space Telescope has been especially useful for probing the dayside temperature structure of close-in planetary atmospheres, as thermal emission from the planets can readily be detected by Spitzer at wavelengths longer than 3 µm.

Data sets are becoming large enough that one can begin to search for correlations in the current detections (Knutson et al. 2010; Cowan & Agol 2011).

A powerful new constraint of the past two years is the possibility of joint constraints in the infrared, from Spitzer, and the optical, from space telescopes like CoRoT (e.g., Gillon et al.2010; Deming et al. 2011) and Kepler (Désert et al. 2011a).

The leverage from optical wavelengths comes from a measurement (or upper limit) of the geometric albedo of the planet’s atmosphere, although this is complicated by a mix of thermal emission and scattered light both contributing for these planets.

Detection of relatively low geometric albedos Ag < 0.15 is consistent with cloud-free models of hot Jupiter atmospheres (Sudarsky et al. 2003; Burrows et al. 2008), and can inform our understanding of what causes the temperature inversions in many hot Jupiter atmospheres(Spiegel & Burrows 2010).

In this paper we discuss all aspects of the detection, validation, conﬁrmation, and characterization of the planet.

Section 2 discusses the detection of the planet by Kepler, while §3 covers false-positive rejection and radial velocity conﬁrmation.

Section 4 gives the global ﬁt to all data sets to derive stellar and planetary parameters, while §5 concerns the observational and modeling aspects of atmospheric characterization.

Section 6 is a discussion of the planet’s inﬂated radius amongst its peers, while §7 gives our conclusions.

Discovery

The Kepler science data for the primary transit search mission are the long cadence data (Jenkins et al. 2010b).

These consist of sums close to 30 minutes of each pixel in the aperture containing the target star in question.

These data proceed through an analysis pipeline to produce corrected pixel data, then simple unweighted aperture photometry sums are formed to produce a photometric time series for each object (Jenkins et al. 2010c).

The many thousands of photometric time series are then processed by the transiting planet search (TPS) pipeline element (Jenkins et al. 2010c).

The candidate transit events identiﬁed by TPS are also vetted by visual inspection.