Mirror Earth Read Online Free PDF Page A

spotting even a single transit, therefore, you’d have to monitor hundreds, or even thousands, of stars at once. The easiest planets to see would obviously be the biggest ones, since they block the most light. But big planets like Jupiter orbit pretty far from their stars, as far as anyone knew at the time. Jupiter’s year is about eleven Earth years long. So if your telescope was powerful enough to see only the dimming of a big planet, you’d have to wait a long time to see anything at all, even in an edge-on solar system. If you’re looking at many hundreds of stars at once, a few score should be edge-on, and in some fraction of those, a Jupiter should be just about to make its transit. “Based on the stated assumptions,” wrote Borucki and Summers, “a detection rate of one planet per year of observation appears possible.”
    This “simple way” did, however, require light detectors of unprecedented precision. If a Jupiter-size planet transited in front of a Sun-like star, the star’s light should dim byabout 1 percent, or one part in a hundred. At the time, when astronomers talked about high-precision photometry—that is to say, brightness measurements—they were talking about a sensitivity of one part in ten, or ten times less sensitive. “The first thing we had to do,” said Borucki, “was to show we could build photometers with the necessary precision.”
    He knew it couldn’t be done with photomultipliers, the detector technology astronomers were using at the time. Borucki thought it might be possible to use CCDs, or charge-coupled devices, a sort of detector-on-a-chip that had been invented in 1969 at Bell Laboratories. CCDs are the detectors that have replaced film in modern digital cameras. At the time, though, they were rare and expensive. Astronomers were just beginning to adopt them. “Younger astronomers were familiar with CCDs,” he said, “but for the older ones, the attitude was, ‘If God wanted you to have CCDs you would have been … born with a CCD in your mouth instead of a silver spoon or … something like that.”
    So Borucki and a couple of colleagues set up an experiment in the basement. “We built this thing with bricks and aluminum,” he said, “and we had a light shining under an aluminum plate with a bunch of holes in it.” Little holes were dim stars and big holes were bright stars. At first, the CCDs didn’t seem to be sensitive enough to make the measurements he needed, but Borucki wrote a series of equations that corrected for the inaccuracies. “You can correct it out to ten parts, even one part per million,” he said, “even with a poor detector.” They’d shown, in other words, that they could find planets, and not just Jupiters but smaller planets as well—if someone would letthem try. But when Borucki tried to sell the project, he met with a brick wall of resistance. “You go to all sorts of meetings,” he said, “and you tell other astronomers, ‘You know, we can find other planets. We can find small planets. We can find them with a CCD.’” The other astronomers would say no, that’s impossible. “They would get up and show why it couldn’t be done. I would say, ‘We’ve done it. It can be done.’ They would go to my boss and see if I could get taken off the project, because obviously we were wasting money, but he had enough faith in us that he let us continue.”
    His boss evidently had faith to spare. The CCD experiments happened in the late 1980s, and in the early 1990s Borucki took his sales pitch on the road. He went to NASA with an official proposal for a space-based planet-hunting telescope. It was rejected. He addressed the agency’s criticisms, and reproposed the mission. He was rejected. In all, Kepler was rejected four or five different times before the satellite was finally approved