Scientists are on their way to discovering thousands of new planets, potentially including hundreds of worlds the size of Earth, in Earth-like orbits around sunlike stars. They expect to achieve that goal within three years or so. But they'll start with the weirdest worlds.
The most advanced planet-hunting probes — the European Space Agency's COROT satellite and NASA's Kepler spacecraft — are designed to spot close-in planets most easily. That means the first revelations will be about planets in orbits much smaller than Mercury's.
So when Kepler's scientists announce their first official results next month, expect to hear about "hot Jupiters" and "super-Earths" whirling so close to their stars that they sizzle. And you just might hear about phenomena so strange that the scientists can hardly believe their instruments.
"I was not prescient enough to anticipate something that we're seeing," David Latham, a mission co-investigator from the Harvard-Smithsonian Center for Astrophysics, told msnbc.com. "There are some good things coming."
Kepler's principal investigator, William Borucki of NASA's Ames Research Center, expects that his science team will present about 30 papers at next month's American Astronomical Society meeting in Washington.
"We have planets to announce, and we will have planets to announce next year — quite a few more, in fact," he said.
Borucki said some of the results being turned up by Kepler are so unusual that he's not sure when the team will be ready to go public with them. "The worst thing you could do with exoplanets is announce false positives," he said.
Latham said "the flow into the faucet" during the first 45 days of data collection amounted to about 200 targets of interest. But neither Borucki nor Latham would drop any hints about the nature of the findings that were being turned up.
Scientists are taking so much care with their data in part because the stakes are so high. Kepler and COROT could answer a question that has dogged humans for centuries: Are planets like Earth so rare that we're essentially marooned in the universe, or are they so common that life could find many other homes beyond the solar system?
"A 'dry hole' would be a very, very interesting result," Borucki said. "You will have dramatically affected mankind's future. ... There'd be no 'Star Trek' in that case, no place to go."
But Borucki and his colleagues expect to find many Earths — and that could help focus future observation and exploration for decades or even centuries to come.
"The biggest impact has to be to support the idea that we aren't alone, in the sense that there are other planets out there rather like the Earth," Latham said. "We're confident that they're out there, but we don't have any yet."
How planet-hunting is done
Astronomers have been detecting extrasolar planets for more than 15 years, but most of those finds have been made using a method known as radial velocity, which takes spectral measurements to figure out how much a star rocks back and forth due to a planet's gravitational influence.
That method works well for detecting planets as massive as Jupiter or Saturn — but not so well for planets the size of Neptune or Earth, which have a much weaker gravitational effect. Planets in the range of two to 10 times Earth mass fall into a category that astronomers call "super-Earths," and such worlds are the prime target for Kepler and COROT.
The two probes detect planets not by tracking radial velocity, but by checking planetary transits. Kepler and COROT stare at stars, one by one, and measure the slight dip in their brightness as planets periodically pass over the stars' disks. By factoring in some assumptions about the size of the star in question, astronomers can figure out how wide the planet is — but not how massive it is. For that measurement, scientists have to fall back on the radial velocity method.
They also have to make sure they're seeing a planet rather than some kind of eclipsing star or observational glitch. For that reason, Kepler's scientists want to record three periodic dips in starlight that fit the pattern for a planetary orbit.
If the planet is in a tight orbit, that might happen in a matter of weeks or months. But if the planet is in an Earth-like orbit, the job could take two and a half to three years of painstaking observations. The job is more complicated for Kepler because one of its data channels turns out to have more random "noise" than expected.
Borucki compared the problem to telling time with a clock that perennially runs fast. "Our 'clock' is fine, but we do have an error that we have to subtract off," he said. Improved software will eventually compensate for the error, he said.
Kepler studies just a sampling of the night sky: an area in the constellation Cygnus that takes in about as much area as your fist does when you hold it out at arm's length. "You're talking about a rather small portion of the galaxy," Borucki said. "It's a local neighborhood, in that sense."
Nevertheless, Kepler is scheduled to check 150,000 stars during its 3 1/2-year mission, which should indicate how common alien Earths are in a representative sample of the universe. "We have a pretty good idea when we get a result what it means," Borucki said.
Making sure of super-Earths
To confirm the detections, the Kepler team is working with astronomers at ground-based telescopes. "Telescope time at the big telescopes is the biggest bottleneck for these discoveries," Borucki said.
Currently, the best instrument for the job is the HARPS spectrometer installed on the European Southern Observatory's 12-foot (3.6-meter) La Silla telescope in Chile, which was used by the COROT team to confirm its first "super-Earth" detection. Another spectrometer, known as HARPS-N, is due to go into operation in 2011 at the 13.7-foot (4.2-meter) William Herschel Telescope on the Canary Islands — and that should accelerate the discovery rate for super-Earths.
Astronomers would have to use more sensitive techniques to confirm Kepler's detection of planets the size of Earth or smaller.
To be hospitable for life, a planet doesn't necessarily have to be a carbon copy of Earth in an Earth-like orbit around a sunlike star. "The excitement, at least in the scientific community, is the possibility that there will be super-Earths that might be even better for the evolution of life in the universe," Latham said.
Closer-in planets would be OK as well, if they were orbiting cooler stars with a closer-in habitable zone. Those planets would be easier for Kepler and COROT to detect. "You don't have to have the attention span of a Solomon," Latham said. "It'd be instant gratification."
In any case, astronomers should know by 2013 how rare or common habitable planets are in the universe. That is, if the space probes and the analysis programs work the way they expect.
"This is not a slam-dunk," Borucki said. "What we're doing is at the limits of what humankind can do right now."
Latham isn't disheartened by that degree of difficulty; on the contrary, the challenges of the planet quest suit him just fine. "If it was easy, it wouldn't be any fun," he said.
Science editor Alan Boyle offers daily doses of space and science at Cosmic Log and is the author of "The Case for Pluto: How a Little Planet Made a Big Difference."