The Saturn-sized planet Kepler-16b (black) orbits two dwarf stars. It is not thought to be habitable. Credit: NASA/JPL-Caltech

Geoffrey Marcy finds it strange to turn people away from a conference about planets beyond our Solar System. "This is a field that had three of us in 1995," marvels Marcy, a pioneering exoplanet hunter at the University of California, Berkeley.

But as an organizer of last week's Extreme Solar Systems II conference, he had to decline applicants by the dozen. The Jackson Lake Lodge auditorium in Wyoming's Grand Teton National Park was packed to its 330-person capacity as speakers announced a flood of new detections, and the air was alive with talk of a 'golden age' of exoplanet astronomy.

Along with the discoveries came some sobering news. Rocky, Earth-like planets may be less common than many hoped, and unexpectedly 'noisy' stars are slowing the hunt. Moreover, astronomers cannot learn much beyond the basics — mass or size and orbit — of the planets they do find. "What we need is a telescope in space that can image and take spectra of truly Earth-like planets," Marcy says. "We still need that desperately."

For now, however, indirect methods are keeping astronomers busy. One trove of discoveries came from a European team that watches stars for the slight wobble that signals the gravitational pull of an unseen planet. Their instrument, the High Accuracy Radial velocity Planet Searcher (HARPS), which is attached to a 3.6-metre telescope at the European Southern Observatory in La Silla, Chile, yielded 41 new planets, including one of the most Earth-like yet. At 3.6 times the mass of Earth, it sits in the 'habitable zone' around its star, the Goldilocks range of distances at which a planet's surface would be neither too hot nor too cold for water to be liquid.

Click for larger image

Click here for larger image

There was also news from astronomers working with Kepler, a NASA space telescope that stares fixedly at a field of about 155,000 stars in search of transits: the very slight dip in the brightness of a star as a planet crosses in front. The Kepler team announced that they have now detected 1,781 candidate planets, including 123 that are Earth-sized (see 'Sizing up the sample'). Among the objects was a novelty: a circumbinary, or a planet orbiting a pair of stars.

Both groups are now confident enough to start making pronouncements about the statistics of planets in orbit close to a star — the kind that both Kepler and HARPS are most sensitive to. The HARPS team, for instance, estimates that about half of Sun-like stars have at least one planet with an orbital period of 100 days or fewer, and that many of these systems boast several such planets. Greg Laughlin, an astronomer at the University of California, Santa Cruz, points out that the Solar System — in which only Mercury has such a short period (88 days) — might end up looking like the oddball. "The big news is that there's this staggering population of planets that you'd never suspect from looking at our own Solar System," he says.

Extrapolating from Kepler data, Wesley Traub, chief scientist of NASA's exoplanet exploration programme, took a stab at a more fundamental question: what fraction of Sun-like stars have rocky, Earth-sized planets farther from the star, in a habitable zone? His optimistic answer: 34%.

But many observers dismiss these calculations as premature. Marcy points out that the Kepler statistics suggest a drop off in the frequency of planets smaller than 'super-Earths' — exoplanets with a radius 1.25–2 times that of Earth. "Is Kepler having trouble detecting truly Earth-sized planets?" asks Marcy. "Or are they rare? We don't know."

The stars themselves are making the hunt difficult. Instruments such as HARPS detect the tug of a planet from tiny shifts in the frequency of a star's spectral lines. Many researchers hoped that a new tool for calibrating the position of those lines, called a laser frequency comb, would enable HARPS to detect the minuscule signal from a planet as small as Earth. But it seems that all but the most quiescent stars have enough surface turbulence to drown out so small a signal. Similarly, the Kepler team has said that because of stellar noise, the mission will now need 8 years, rather than the originally planned 3.5 years, to detect all of the Earth-like planets around its target stars (see Nature 477, 142–143; 2011).

Even after true Earth analogues have been detected, neither HARPS nor Kepler will be able to determine whether they have an Earth-like atmosphere containing, for instance, oxygen or carbon dioxide. For a few giant planets that pass in front of their stars, instruments such as the Spitzer Space Telescope have been able to analyse the spectrum of starlight shining through the planet's thick atmosphere, gleaning clues about its composition. But Marcy says that doing transmission spectro­scopy on the thin ring of atmosphere surrounding an Earth-sized planet is beyond the reach of even the 6.5-metre James Webb Space Telescope, the orbital observatory that NASA hopes to launch by 2018 if it does not fall victim to cost overruns. "James Webb, yea or nay, is not the answer to our prayers," Marcy says.

As a result, researchers such as Marcy and Traub stress the need to go beyond the indirect techniques of HARPS and Kepler, and gather the faint light from the planet itself, which is normally invisible in the glare of the parent star. For that, astronomers need either a giant space telescope equipped with a device for blocking starlight, or an interferometer, consisting of several telescopes flying in formation. NASA did develop a proposal for such a space telescope, called Terrestrial Planet Finder, and the European Space Agency hoped to fly a similar mission called Darwin. But budgetary constraints have left both missions in limbo, unlikely to advance to the front of either agency's queue until well into the next decade. At the conference, Traub raised the issue. "People are not thinking deeply about the distant future. People are wrapped up with what they're doing right now," he says. "Clearly, I'm concerned."

But Laughlin isn't as worried. He says that the enthusiasm and momentum in planet-hunting may lead to an unexpected solution. "People have a history of being inventive when they need to be," he says. He points to French philosopher Auguste Comte, who in 1835 wrote that astronomers might be able to learn about the shapes, sizes and motions of stars, but that stellar densities, temperatures and chemical compositions would be "forever denied to us".

Within three decades, astronomical spectroscopy was starting to answer all of those questions.

Credit: Source: NASA Kepler team