I didn’t know there was such a thing as the Kepler Mission happening. The Kepler satellite is part of NASA Discovery Mission #10, which aims to collect systematic data on the prevalence and size of planets orbiting other stars in the Milky Way. Kepler is equipped with a powerful telescope that has a very wide range of vision, which is pointed constantly at a particular range of space for four years. In that time, Kepler can monitor the spectral signatures of 100,000 and measure the deviation in brightness to detect the transit of a planet around the star. With such a large sample, the handlers expect to find something on the order of 50 planets, if most are the size of earth and perhaps 640 if planets tend to be much larger. Put simply, Kepler doesn’t just look for planets; it’s looking for data on the prevalence of planets.
But even if planets are common, who such a big sample size? There are all sorts of really delicious details on the link above, but this one knocked my socks off—
For a planet to transit, as seen from our solar system, the orbit must be lined up edgewise to us. The probability for an orbit to be properly aligned is equal to the diameter of the star divided by the diameter of the orbit. This is 0.5% for a planet in an Earth-like orbit about a solar-like star. (For the giant planets discovered in four-day orbits, the alignment probability is more like 10%.) In order to detect many planets one can not just look at a few stars for transits or even a few hundred. One must look at thousands of stars, even if Earth-like planets are common. If they are rare, then one needs to look at many thousands to find even a few. Kepler looks at 100,000 stars so that if Earths are rare, a null or near null result would still be significant. If Earth-size planets are common then Kepler should detect hundreds of them.
The picture below, from here, shows how sensitive Kepler’s output is relative to ground-based detection systems. The dip is the drop in light intensity as a planet transits a distant star.

a.j.m.

