Mirror Earth Read Free

start that if the spacecraft worked as it was supposed to, it would blow the competition out of the water. This was true for several reasons. To start with, Kepler orbits high above Earth’s atmosphere, like the Hubble Space Telescope, but even farther out in deep space. As a result, the atmosphere’s blurring effect—the same thing that makes the stars seem to twinkle—isn’t a problem. Another huge advantage is that Kepler doesn’t look at one star at a time. It looks at more than 156,000 of them all at once, in a patch of the northern sky that lies between the constellations Lyra and Cygnus. And it keeps looking at those same 156,000 stars continuously, around the clock, day after day, month after month, year after year. That’s impossible from the ground: When the Sun rises, the stars vanish. Since one night’s observing isn’t nearly enough to find a planet around a distant star, you have to keep returning to a given star many times, over many nights, to get any information worth using. Kepler doesn’t return to any of its 156,000 stars because it never leaves them in the first place.
    If Kepler were a general-purpose telescope—even one like the Hubble—it wouldn’t be permitted to linger on a single patch of sky indefinitely. Most telescopes are used to study all sorts of cosmic phenomena, from distant galaxies to exploding stars to black holes. If you’re looking for planets with a general-purpose telescope, you might get to use it for a few nights at most before the next astronomer in line gets her turn. If you’re using the Hubble, which is vastly oversubscribed, you get morelike a few hours. Kepler, by contrast, was built to look at only one tiny patch of sky for its entire working lifetime. It will never avert its gaze from these 156,000 stars.
    By the time Borenstein, the AP reporter, had cornered Borucki, Kepler had proven itself to be technologically perfect, or pretty close to it. It had been staring at its target stars for nearly two years. Only the first six months’ worth of observations had been fully processed, though. That’s how long it takes for the Kepler team’s computers to pore through the terabytes of electronic data beamed down from the spacecraft, letting custom-written algorithms flag the tiny changes in starlight that might (or might not) betray the presence of a planet, weeding out false positives—things that look like planets but aren’t. If a potential planet passes all these tests, that’s still not good enough. The software has to pick up the planet’s signal not once, not twice, but at least three separate times for it to make the cut. Usually, though, since the signal is often very faint, it takes a lot more than three sightings: some run into the hundreds. And even then, a dozen or so mission scientists look at each of what they call KOIs—Kepler Objects of Interest. These are something like the persons of interest law-enforcement types talk about in criminal investigations. They’re not being charged … yet. But you shouldn’t be at all surprised if they end up being indicted.
    The reason Borucki hadn’t announced any new results at the Washington press conference was that he’d already presented everything he had just two weeks earlier, at a press conference at NASA headquarters, in Washington. There was plenty to say: In just the first six months’ worth of observations,Kepler had come up with no fewer than 1,235 possible planets, about 90 percent of which were almost certainly real. Kepler had barely warmed up, and it had identified at least twice as many planets as all the astronomers in the world had found in the previous sixteen years. “Astronomers have cracked the Milky Way like a piñata,” Dennis Overbye wrote in the
New York Times
, “and planets are now pouring out so fast that they do not know what to do with them all.”
    After he gave