Science - An Earth-Sized Planet in the Habitable Zone of a Cool Star lyrics

An Earth-Sized Planet in the Habitable Zone of a Cool Star Elisa V. Quintana,1,2* Thomas Barclay,2,3 Sean N. Raymond,4,5 Jason F. Rowe,1,2 Emeline Bolmont,4,5 Douglas A. Caldwell,1,2 Steve B. Howell,2 Stephen R. Kane,6 Daniel Huber,1,2 Justin R. Crepp,7 Jack J. Lissauer,2,8 David R. Ciardi,9 Jeffrey L. Coughlin,1,2 Mark E. Everett,10 Christopher E. Henze,2 Elliott Horch,11 Howard Isaacson,12 Eric B. Ford,13,14 Fred C. Adams,15,16 Martin Still,3 Roger C. Hunter,2 Billy Quarles,2 Franck Selsis4,5 1SETI Institute, 189 Bernardo Avenue, Suite 100, Mountain View, CA 94043, USA. 2NASA Ames Research Center, Moffett Field, CA 94035, USA. 3Bay Area Environmental Research In- stitute, 596 1st Street, West Sonoma, CA 95476, USA. 4Univer- sity of Bordeaux, Laboratoire d'Astrophysique de Bordeaux, UMR 5804, F-33270, Floirac, France. 5CNRS, Laboratoire d'Astrophysique de Bordeaux, UMR 5804, F-33270, Floirac, France. 6San Francisco State University, 1600 Holloway Avenue, San Francisco, CA 94132, USA. 7University of Notre Dame, 225 Nieuwland Science Hall, Notre Dame, IN 46556, USA. 8Depart- ment of Geological and Environmental Sciences, Stanford University, 450 Serra Mall, Building 320, Stanford, CA 94305, USA. 9NASA Exoplanet Science Institute, California Institute of Technology, 770 South Wilson Avenue, Pasadena, CA 91125, USA. 10National Optical Astronomy Observatory, 950 North Cherry Avenue, Tucson, AZ 85719, USA. 11Southern Connecticut State University, New Haven, CT 06515, USA. 12University of California, Berkeley, CA 94720, USA. 13Center for Exoplanets and Habitable Worlds, 525 Davey Laboratory, The Pennsylvania State University, University Park, PA 16802, USA. 14Depart- ment of Astronomy and Astrophysics, The Pennsylvania State University, 525 Davey Laboratory, University Park, PA 16802, USA. 15Michigan Center for Theoretical Physics, Department of Physics, University of Michigan, Ann Arbor, MI 48109, USA. 16Department of Astronomy, University of Michigan, Ann Arbor, MI 48109, USA. *Corresponding author. E-mail: elisa.quintana@nasa.gov The quest for Earth-like planets is a major focus of current exoplanet research. Although planets that are Earth-sized and smaller have been detected, these planets reside in orbits that are too close to their host star to allow liquid water on their surfaces. We present the detection of Kepler-186f, a 1.11 T 0.14 Earth-radius planet that is the outermost of five planets, all roughly Earth-sized, that transit a 0.47 T 0.05 solar-radius star. The intensity and spectrum of the star's radiation place Kepler-186f in the stellar habitable zone, implying that if Kepler-186f has an Earth-like atmosphere and water at its surface, then some of this water is likely to be in liquid form. In recent years, we have seen great progress in the search for planets that, like our own, are capable of harboring life. Dozens of known planets orbit within the habitable zone (HZ), the region around a star within which a planet can sustain liquid water on its surface (1–4). Most of these HZ planets are gas giants, but a few, such as Kepler- 62f (5), are potentially rocky despite having a larger radius than Earth. Hitherto, the detection of an Earth-sized planet in the HZ of a main-sequence star has remained elusive. Low-ma** stars are good targets in the search for habitable worlds. They are less luminous than the Sun, so their HZs are located closer in (6). The shorter orbital period and larger planet-to- star size ratio of a planet in the HZ of a cool star relative to planets orbiting in the HZ of solar-type stars allow for easier transit detections. M dwarfs, stars with 0.1 to 0.5 times the ma** of the Sun (M⨀), are very abundant, constituting about three quarters of all the main-sequence stars in our galaxy (7). They also evolve very slowly in luminosity, thus their HZs remain nearly constant for billions of years. Kepler-186 (also known as KIC8120608 and KOI-571) is a main-sequence M1-type dwarf star with a temperature of 3788 T 54 K and an iron abundance half that of the Sun [(8) and supplementary materials (SM) section 2]. The star was observed by the Kepler spacecraft at near- continuous 29.4-min intervals. The presence of four planets, designated Kepler-186b to Kepler- 186e, all smaller than 1.5 Earth radius (R⊕), with orbital periods between 3.9 and 22.4 days, was confirmed with the first 2 years of data (9, 10). The fifth planet candidate, Kepler-186f, which we discuss here, was detected with an additional year of data. We compared the observed data to a five- planet model with limb-darkened transits (9, 11), allowing for eccentric orbits to estimate the phys- ical properties of Kepler-186f. We used an affine invariant Markov-chain Monte Carlo (MCMC) algorithm (12, 13) to efficiently sample the model parameter posterior distribution. Kepler-186f has an orbital period of 129.9 days and a planet-to- star radius ratio of 0.021. The additional constraint on stellar density from the transit model allowed us to refine the stellar radius that was previously derived by modeling spectroscopic data. Interior models of cool main-sequence stars such as Kepler- 186 show systematic differences from empirically measured stellar properties (14–16) (SM section 2). To account for discrepancies between the em- pirically measured radii and those derived from model isochrones at the measured temperature for Kepler-186, we have added a 10% uncertain- ty in quadrature to our stellar radius (R✭) and ma** estimate, yielding a final estimate of R✭ = 0.472 T 0.052 and a planet radius of 1.11 T 0.14 R⊕ (Fig. 1 and table S2). The Kepler-186 planets do not induce a de- tectable reflex motion on the host star or dynam- ically perturb each other so as to induce substantial non-Keplerian transit ephemerides, both of which can be used to help confirm the planetary nature of Kepler's planet candidates (17, 18). Instead, we used a statistical approach to measure the confidence in the planetary interpretation of each candidate planet (19, 20). We obtained follow-up high-contrast imaging observations using the Keck- II and Gemini-North telescopes (SM section 5) to restrict the parameter space of stellar magnitude/ separation where a false-positive–inducing star could reside and mimic a planetary transit. No nearby sources were observed in either the Keck- II or Gemini data; the 5s detection limit set the brightness of a false-positive star to be Kp = 21.9 at 0.5′′ from Kepler-186 and 19.5 at 0.2′′, where Kp is the apparent magnitude of a star in the Kepler bandpa**. The probability of finding a background eclipsing binary or planet-hosting star that could mimic a transit in the parameter space not ex- cluded by observations is very low: 0.5% chance relative to the probability that we observe a planet orbiting the target. However, this does not ac- count for the possibility that the planets orbit a fainter bound stellar companion to Kepler-186. Although we have no evidence of any binary com- panion to the target star, faint unresolved stellar companions to planet host stars do occur (21). We constrained the density of the host star from the transit model by a**uming that all five planets orbit the same star. The 3s upper bound of the marginalized probability density function of stellar density from our MCMC simulation is 11.2 g cm−3. If Kepler-186 and a hypothetical companion co- evolved, the lower limit on the stellar ma** and brightness of a companion would be 0.39 M⨀ and Kp = 15.1, respectively. Given the distance to Kepler-186 of 151 T 18 pc, a companion would have to be within a projected distance of 4.2 astronomical units (AU) from the target to avoid detection via our follow-up ob- servations. However, a star closer than 1.4 AU from the primary would cause planets around the fainter star to become unstable (22). The prob- ability of finding an interloping star with the spe- cific parameters needed to masquerade as a transiting planet is very small relative to the a priori probability that the planets orbit Kepler- 186 (<0.02%). Therefore, we are confident that all five planets orbit Kepler-186. Although photometry alone does not yield planet ma**es, we used planetary thermal evolu- tion models to constrain the composition of the Kepler-186 planets. These theories predict that the composition of planets with radii less than about 1.5 R⊕ is unlikely to be dominated by H/He gas envelopes (23). Although a thin H/He envelope around Kepler-186f cannot be entirely ruled out, the planet was probably vulnerable to photoevaporation early in the star's life, when extreme ultra- violet flux from the star was significantly higher. Hence, any H/He envelope that was accreted would probably have been stripped by hydro- dynamic ma** loss (23). Although Kepler-186f probably does not have a thick H2-rich atmo- sphere, a degeneracy remains between the rela- tive amounts of iron, silicate rock, and water, because the planet could hold on to all of these cosmically abundant constituents. Ma** estimates for Kepler-186f can therefore range from 0.32 M⊕ if composed of pure water/ice to 3.77 M⊕ if the planet is pure iron, and an Earth-like composition (about 1/3 iron and 2/3 silicate rock) would give an intermediate ma** of 1.44 M⊕ (table S3). For Kepler-186, the conservative estimate of the HZ (i.e., likely narrower than the actual annulus of habitable distances) extends from 0.22 to 0.40 AU (4). The four inner planets are too hot to ever enter the HZ. Kepler-186f receives 32þ6 % of −4 the intensity of stellar radiation (insolation) re- ceived by Earth from the Sun. Despite receiving less energy than Earth, Kepler-186f is within the HZ throughout its orbit (Fig. 2). It is difficult for an Earth-sized planet in the habitable zone of an M star to accrete and retain H2O (24, 25), but being in the outer portion of its star's HZ reduces these difficulties. The high coplanarity of the planets' orbits (given by the fact that they all transit the star) suggests that they formed from a protoplanetary disk. The leading theories about the growth of planets include in situ accretion of local material in a disk (26, 27), collisional growth of inward- migrating planetary embryos (28, 29), or some combination thereof. We performed a suite of n-body simulations of late-stage in situ accre- tion from a disk of planetary embryos around a star like Kepler-186 (SM section 9). We found that a ma**ive initial disk (>10 M⊕) of solid material with a very steep surface density profile is needed to form planets similar to those in the Kepler-186 system. Accretion disks with this much ma** so close to their star (<0.4 AU) or with such steep surface density profiles, however, are not com- monly observed (30), suggesting that the Kepler- 186 planets either formed from material that underwent an early phase of inward migration while gas was still present in the disk (31) or were somehow perturbed inward after they formed. Regardless, all simulations produced at least one stable planet in between the orbits of planets e and f, in the range from 0.15 to 0.35 AU (fig. S5). The presence of a sixth planet orbiting between e and f is not excluded by the observations; if such a planet were to have a modest inclination of a few degrees with respect to the common plane of the other planets, we would not observe a transit. Planets that orbit close to their star are sub- jected to tidal interactions that can drive the planets to an equilibrium rotational state, typically either a spin-orbit resonance or a pseudosynchronous- state in which the planet co-rotates with the star at its closest approach (32, 33). The proximity of the inner four planets to Kepler-186 suggests that they are probably tidally locked. Kepler-186f, however, is at a large enough distance from the star that un- certainties in the tidal dissipation function preclude any determination of its rotation rate (34). Regard- less, tidal locking (or pseudosynchronous rotation) does not preclude a planet from being habitable. The 5.6 M⊕ planet GJ 581d (35) probably rotates pseudosynchronously with its star and in addition receives a similar insolation (27%) as Kepler- 186f. Detailed climate models have shown GJ 581d to be capable of having liquid water on its surface (36, 37). Taken together, these consid- erations suggest that the newly discovered planet Kepler-186f is likely to have the properties re- quired to maintain reservoirs of liquid water.