The planet-hunting Kepler Space Telescope is deservedly celebrated for some of its more dramatic discoveries: a world with two suns, for example, a planet half-shrouded with clouds, a planet whose size and composition are a near match to Earth’s. But the ship is in some ways a pretty prosaic machine — a robot census taker, no more, no less, hanging in space and counting heads, to determine how many stars other than our own are home to planets. Like a census taker too, however, it has to make some pretty elaborate inferences.
It’s not possible to detect and count every single planet in the Milky Way, any more than it’s possible to shake the hand and take the name of every single person living in the U.S. In both cases, a sort of statistical sampling is often involved. When you know enough about most of the people living in any one town, county or state, you can make accurate inferences about the rest. The same goes in space, where Kepler is trying to determine the frequency of certain types of exoplanets. What percentage of stars host Jupiter-size worlds? How many have Neptunes? And most crucially for the search for extraterrestrial life, how many Earth-size planets orbit in their stars’ habitable zones, the regions in which temperatures are hospitable for living creatures?
Kepler has been chipping away at this question since the probe was launched in 2009, but a new report in Proceedings of the National Academy of Sciences has now moved scientists a giant step closer to the ultimate answer. According to the new analysis, a whopping 22% of sunlike stars have planets more or less the size of Earth in their habitable zones. That adds up to about 20 billion Earths in our galaxy alone, says lead author Erik Petigura, of the University of California, Berkeley. That in turn means that an Earth-like world is likely to be just 12 light-years away, and that its parent star is visible to the naked eye. “It’s really amazing when you think about it,” Petigura says.
It’s not just amazing, though: it’s crucial to the search for life. If the fraction of stars with life-friendly Earths was much smaller (a few years ago, says Petigura, some estimates put it as low as .0001%), the nearest one might easily be too distant to study for signs of biological activity, even with the most advanced telescopes currently on the drawing boards. If there’s one practically next door, as the new study suggests, that kind of remote analysis should be comparatively easy.
The new estimate of plausibly habitable planets is based on an intensive search through Kepler’s massive database — but not a comprehensive one. The satellite has been monitoring more than 150,000 stars, looking for the almost imperceptible dimming caused when a planet moves in front of its parent star. Many of these targets are anything but sunlike: they’re the smaller, redder, dimmer stars known as M-dwarfs. Astronomers already know that some M-dwarfs harbor Earth-size planets. A study earlier this year, in fact, suggests that the nearest of these is about 13 light-years away.
But it’s by no means certain that Earths around M-dwarfs could be habitable. Among other things, the planets always keep the same face toward their stars, the way the moon does to Earth, meaning one side could be a lot hotter than the other. So Petigura and his colleagues limited their search not to Kepler’s full target list, but to a subset of about 40,000 stars that range from .7 to 1.2 times the size of the sun. They’re not all perfect copies of our home star, but, says co-author Geoff Marcy, also at Berkeley, “if you flew past them in a spaceship you wouldn’t be able to tell the difference.”
In the end, the researchers turned up ten approximately Earth-size planets circling their stars at a distance at which water could exist in liquid form, which biologists think is essential for life—at least the kind of life we can imagine. To make the leap from ten to 20 billion, the astronomers had to extrapolate. For example, they had to take into account that many Earth-size planets got missed because their orbits didn’t happen take them directly in front of their stars.
They also realized that some of the planets that did pass in front of their stars would be missed as well, because their signals were too weak to emerge from the visual static that bedevils such high-precision observations. So Petigura created fake planets with weak signals, and by noting how many the software missed, he could tell how many real planets had probably escaped detection as well. And, of course, the researchers had to factor up from the 150,000 stars in Kepler’s field of view to the 300 billion or so that make up the Milky Way.
Taken together, these adjustments yielded the new 20 billion Earth-like planet estimate. “I do have to add a note of caution,” says Petigura. “These planets are only Earth-like in terms of their size and the amount of energy they receive from their stars. We don’t know if they have rocky compositions, oceans of water, plate tectonics or life.”
To answer the ultimate question of whether life exists beyond Earth, astronomers will have to go beyond statistics to identify an actual mirror Earth in our cosmic neighborhood, then scrutinize its atmosphere for gases like oxygen and methane, which could be telltale signs of biological activity. Those observations won’t be possible for years.
But at least they know they’re not looking for something so vanishingly rare, we’d never be likely to stumble on it. The Milky Way is evidently teeming with balmy, Earth-size planets. And unless nature is far more perverse than anyone expects, there’s got to be life on at least some of them.