For those seeking life-sustaining worlds far, far away, 2011 was a year of discoveries for NASA's Kepler telescope that culminated with images of the first two Earth-size planets observed outside the Solar System. Unfortunately, their orbits of a star much like our Sun are so close that they are hotter than a pizza oven.

Nearly twins, they are called Kepler-20e and -20f and are part of a solar system in the constellation Lyra 1,000 light-years away that is not playing by the rules. The host star, Kepler-20, is slightly cooler and smaller than the 5700K Sun, but otherwise quite similar. Nevertheless, scientists report that the five planets orbiting it are a strange lot, mixing three Neptune-size gaseous orbs with the rocky -20e and -20f in ways not seen before.

These mysteries are delighting Kepler's science team and offer proof of the parts-per-million measuring accuracy of the spacecraft. Built by Ball Aerospace with mission management by the Jet Propulsion Laboratory (JPL), the Kepler telescope has a 1.4-meter (4.6-ft.) primary mirror and a 105-deg. field of view. Most importantly, it carries the largest photometer ever put into orbit, 42 charge-coupled devices, each with 2,200 X 1,024 pixels, continuously staring at a field of 100,000 stars, detecting the slightest diminishment in an individual star's brightness that might signal the transit of a planet. Kepler was launched into a solar orbit trailing the Earth by 9 million mi. in March 2009 and is nearing the end of its funded three-year lifespan but with the promise that its instruments can serve twice that long, assuming Congress agrees to extend funding.

Even as the Kepler-20 results were being announced and published on Dec. 20, a University of Arizona astronomy team was reporting in Nature the discovery of two smaller planets. These Kepler Objects of Interest (KOI) have been “deep-fried” after being engulfed by a star similar to the Sun, except the star is dying and has gone through the swelling red-giant phase as its hydrogen burned away.

All of these reports are from data observations that may have been made soon after Kepler's observations began, when its “possible” planetary list started swelling. But a single “sighting” of what appears to be a planet might be something else, such as a sunspot or the wink of a binary star. So additional transits of the candidate planets must be taken and then the Kepler observations independently confirmed, usually by ground observatories.

Kepler is the most sensitive planet-finder in use, but other spacecraft and ground observatories are also at work. JPL reports there have been 709 confirmed planet sightings, including two that arose from a scrub of Hubble Space Telescope images taken in 1998. Kepler has made 33 confirmed finds and identified 2,326 candidate planets.

“We tried to study -20e a year ago” without success because of a lack of data, says Francois Fressin of the Harvard-Smithsonian Center for Astrophysics in Cambridge, Mass., lead author of the peer-reviewed scientific papers on Kepler's most recent finds. The “Eureka moment” of the first confirmation that it was truly an Earth-size planet came only recently when more Kepler data flowed in, he says. “This is the first time we have crossed the Earth-size threshold.”

Kepler's initial discoveries were of planets far larger than Earth, some many times Jupiter's size. It has only been more recently that smaller planet sightings have been confirmed. The first of these—Kepler-10b— came last January. It is 1.4 times larger than Earth but with the density of a dumbbell and orbits its host star in less than a day. As the year progressed, the Kepler revelations became more varied, including the first confirmation of Kepler-16b as a planet that circles two stars. It was promptly nicknamed Tatooine in honor of the fictional planet in the Star Wars movies.

Kepler-20e is a rocky orb of silicate and iron slightly smaller than Venus, 0.87 times the radius of Earth. It was immediately dismissed from the “possibly habitable” list because it orbits Kepler-20 in just 6.1 days and has a surface glowing at 1400F, hot enough to melt glass. Its slightly smaller fraternal twin, Kepler-20f, has a radius 3% larger than Earth's but orbits in 19.6 days. Its surface temperature is 800F, about the same as Mercury's.

The -20e/-20f findings followed another tantalizing report on Dec. 5 of Kepler-22b, the first planet found in a habitable zone, which in our own Solar System roughly equates to the range of Venus to Mars (AW&ST Dec. 12, 2011, p. 16). Although not much is known about Kepler-22b, it is in the so-called Goldilocks orbit of “not too hot, not too cold” for water to remain liquid and life to be possible.

But it is probably the wrong size to be habitable, says Linda Elkins-Tanton, director of the Carnegie Institution for Sciences' Department of Terrestrial Magnetism in Washington. She is an outside but “fascinated and admiring observer” of the latest Kepler findings. At 2.4 times the size of Earth, -22b presumably has a large gas envelope without the rocky surface that would make it suitable for habitation.

Elkins-Tanton and others emphasize that they have much to learn about all the near-Earth-size planets they are seeing. Given the right conditions, for instance, they may discover that these or other planets detected by Kepler have moons. But doing so will not come about by direct observation from the ground—the planets are too far away—or through the slight dimming technique that Kepler uses for the planets themselves. Instead, moons might be surmised if there is a slight wobble in the planet's transient passage, indicating it is experiencing a gravitational interplay with the moon.

All of this work fits fairly neatly into the known world of stars and planets. But the mystery of the orbits of Kepler-20 is large enough that Harvard astronomy Prof. David Charbonneau, part of the Kepler-20e/f team, says he intends to revise his lecture notes.

Unlike planets in our own Solar System—which can be neatly separated by an “ice line” between the smaller, rocky planets orbiting closest to the Sun and the larger, colder gaseous planets such as Jupiter and Neptune farther out—the Kepler-20 system mixes them up.

“The architecture of the system is crazy,” says Charbonneau. First there is a Neptune-like gas bag, followed by -20e, then another Neptune, followed by -20f, and finally another Neptune. “It's big, little, big, little, big . . . that we know of,” since more planets may be found, he says. Not only that, the orbits of all five are squeezed together. “All five orbit within a distance of Mercury to the Sun,” he says. Mercury has an elliptical orbit; but even at its aphelion of 44 million mi., it is more than twice as close to the Sun as is Earth.

Not surprisingly, the current model for a solar system is based on our own pattern of rocky planets inside and gaseous planets farther out. The Kepler-20 results confront this ice-line theory. “So we really are challenging scientists to figure out how they are formed,” says Elkins-Tanton.

The masses of astronomical bodies are typically calculated using a technique called radio velocity. But it is less applicable for smaller bodies like -20e. So the mass readings of Kepler's latest findings are still to be determined, although Charbonneau notes that there is a large radio-velocity effort being made in that regard.

In the meantime, astronomers such as Charbonneau are devising new instruments to measure the mass of these distant objects. Of the coming year, he says, “you can bet this is one of the main goals of astronomers—to measure them.”

Findings from Kepler are flowing in so quickly that Charbonneau is telling his Harvard freshmen that they can expect theory-startling discoveries during their undergraduate years.

A team at the University of Arizona's Steward Observatory identified two planets, KOI 55.01 and 55.02, while investigating pulsations in KOI 55, a 28000K red giant star 1,400 light-years away in the constellation Cygnus. Tiny brightness changes gave evidence of the planets, which are, respectively, 0.76 and 0.87 times the mass of Earth and reside in orbits of their star of just 5.76 and 8.23 hr.

“We spent about a year trying to get our discovery published, but since the technique we used has never been applied to a planet finding, there was healthy skepticism among our peer reviewers until we were able to answer all the questions they could think of,” says team member Elizabeth Green, an associate astronomer at the University of Arizona. UA's technique works only on extremely hot planets, and confirmation of the findings came through the collection of additional spectra from ground-based observatories and asteroseismological analysis of the brightness changes that led to the discoveries.

The effect of the budget impasse in Congress raises concerns about a continuation of funding for Kepler beyond its prescribed three-year mission lifespan. “Kepler light curves were essential for the discoveries in both KOI 55 and Kepler-20,” says Green. “Without the Kepler data we would not have known planets were there.”

Before Kepler's launch, astronomers were unaware of how “noisy” stars are or that Sun-like stars pulsated in so many frequencies. The filtering effects of Earth's atmosphere made most stars appear to have rather constant brightness levels. Kepler has shown this is not quite true.

“Because these small brightness changes are significant compared to the exceedingly small brightness changes caused by transits of Earth-size planets [in front of stars], it means that Kepler has to take data for a longer time period than the Kepler planners thought,” Green says. Only then is there enough signal from a planet's transit to overcome whatever small background 'noise'” of brightness changes might be present because of other things, she says.

“Without another couple of years of Kepler data, many, many Earth-size planets will not be detected, because there simply won't be enough data,” she says.