On supporting science journalism
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
After five years of searching, researchers using data from NASA's exoplanet-hunting Kepler spacecraft have discovered what look to be two of the most Earth-like worlds yet. Dubbed Kepler 438 b and Kepler 442 b, both planets appear to be rocky and orbit in the not-too-hot, not-too-cold habitable zones of their stars where liquid water can exist in abundance. Astronomers announced the planets along with six other newfound small, temperate worlds today at a meeting of the American Astronomical Society in Seattle. Their findings will be published in The Astrophysical Journal. The discoveries double the number of known potentially habitable exoplanets. They also push Kepler's tally of vetted, confirmed worlds to just over 1,000, marking a milestone in the mission's epochal search for alien Earths.
Both planets are many hundreds of light-years away and orbit stars smaller and dimmer than our sun. Like most of Kepler's finds, they were discovered via transits—the shadows they cast toward our solar system as they cross the blazing faces of their stars. Transits allow astronomers to measure a planet's size, orbit and exposure to starlight. Kepler 438 b is only about 12 percent larger than Earth, and basks in 40 percent more starlight; Kepler 442 b is 30 percent larger and receives about 30 percent less light. Both spheres may be somewhat warmer than Kepler's two previous premier rocky worlds, Kepler 186 f and Kepler 62 f, each of which gets significantly less starlight—similar to that received by Mars. “We can't say for sure whether these planets are truly habitable—only that they are promising candidates for habitability," says study co-author David Kipping, an astronomer at the Harvard–Smithsonian Center for Astrophysics (CfA) in Cambridge, Mass.
When Kepler launched into orbit in 2009 to survey a patch of sky containing some 150,000 stars, one of its primary goals was to find mirror Earths, worlds about the same size as our own in approximately 365-day orbits around sunlike stars. The task was expected to take just over three years because many things can cause stars to dim besides transiting planets, and astronomers would need to glimpse the periodic recurrence of any mirror Earth's transit not once or twice but three times to be convinced that any particular dimming was due to a planet.
The sunlike stars had other plans, however, proving to be more variable in brightness than mission planners had anticipated, muddying Kepler's search for the faint shadows. Kepler's scientists consequently focused on smaller, dimmer, quieter stars and asked for more time to gather additional data. The mission was extended in 2012 but was struck by stabilizer equipment failures in spring of 2013 that sent Kepler's once-steady gaze drifting askew, bringing its survey to a premature close. Last year the spacecraft was resurrected as the "K2" mission after researchers devised a new method to aim the telescope. K2 is a more limited transit survey and has scant hope of finding a mirror Earth. The last, best option for the mission to succeed in discovering Earth twins was to sift through Kepler's archival data from 2013 and earlier, which is filled with thousands of unconfirmed candidate planets. "I think these latest planets are about as good as we're going to get from the Kepler data," Kipping says. "I would like to be surprised—and I'm hopeful—but I'm not sure we'll find planets closer to Earth twins than the objects we present in our paper."
Other researchers might disagree. At the same meeting, Fergal Mullally, a staff scientist and Kepler team member at NASA's Ames Research Center at Moffett Field, Calif., announced an additional 554 yet-to-be-confirmed candidate planets added to Kepler's catalogue, bringing its total candidate count to 4,175 possible worlds. Eight of those newly announced worlds are less than twice the size of Earth and orbit in their stars' habitable zones. Six of those eight orbit sunlike stars, and could potentially fulfill Kepler's elusive goal of finding an Earth 2.0—but only if follow-up studies can reveal more about them.
Like many of Kepler's most tantalizing finds, Kepler 438 b and 442 b were too small and far away to be confirmed in the traditional way, which involves independently measuring the gravitational wobble an orbiting planet induces on its star. Rather than being "confirmed," they were "validated" through a series of simpler follow-up observations and complex statistical tests designed to rule out false positives. According to lead author Guillermo Torres, another astronomer at the CfA, this validation process took more than a year and a half and relied on some of the world's largest telescopes as well as intensive simulation runs on a NASA supercomputer. As impressive as this validation process is, it cannot provide the crucial information available when a planet is confirmed by measuring its associated wobble. Such a measurement allows astronomers to estimate a world's mass, which, paired with the size estimate provided by a transit, yields a density and composition estimate. Planets revealed to be small and heavy must be made mostly of rock or metal whereas worlds seen to be large and light must possess much more gas.
Without confirmatory wobbles, determining that Kepler 438 b and 442 b are Earth-like worlds of rock and metal relies on statistics for now. "Lots of hard work over the past few years suggests the cutoff—the transition between rocky and gaseous planets—occurs at about 1.5 times the size of Earth," Kipping says. "Everyone is now converging on that number. Anything smaller than that is most likely rocky"—and therefore more likely to be habitable.
The latest study to bolster this argument was presented earlier in the meeting by lead author Courtney Dressing, another CfA astronomer, who measured the masses and sizes of a handful of small transiting planets to estimate the rocky-to-gaseous transition zone. Dressing pegs the general upper limit for rocky planets as 1.6 times Earth size. "We now know most planets in the galaxy are intermediate in size between Earth and Neptune but we don't have any examples of these worlds in our own solar system," Dressing says. "In order to figure out whether these planets, the most common in our galaxy, are mostly rocky and potentially habitable or mostly gaseous and probably not very habitable, we have to perform these measurements to learn where exactly this transition occurs and how broad it is."
Most planet hunters now pin their immediate hopes for discovering Earth-like planets on NASA's upcoming Transiting Exoplanet Survey Satellite, or TESS, which is slated to launch in 2017. TESS is expected to perform an all-sky survey focused on finding transiting rocky planets around nearby stars, planets that could then be studied in further detail by NASA's James Webb Space Telescope, which would launch no sooner than 2018. Even so, the quest for finding truly habitable—maybe even inhabited—planets beyond the solar system may require more ambitious, next-generation observatories.
"The impact of this work is to prove that these sorts of worlds are out there," Kipping says. "This is a journey where the end point is investigating planetary atmospheres to look for signs of life, and this result shows it's justifiable to build a space telescope to do exactly that."