Enlarge

OBSERVATION LOUNGEENLARGE | The eight small telescopes that make up the MEarth project at the Whipple Observatory sit atop Mount Hopkins in Arizona. The telescopes seek habitable, rocky planets around M stars.Jonathan Irwin / Harvard

For years, planet hunters have been preoccupied with hot Jupiters—giant, gaseous planets that tightly hug their sunlike parent stars. These massive, close-in planets, not yet directly seen, are the easiest to find because they induce the largest wobble in the motion of the stars they orbit. But now astronomers are following a rockier road—seeking rocky, icy planets only a few times as massive as Earth. Soon, astronomers predict, they will discover an Earth-sized planet that orbits within the habitable zone of its parent star. And if David Charbonneau has any say about it, that historic find will come from eight tiny telescopes his team has just finished assembling at the Fred Lawrence Whipple Observatory on Mount Hopkins in Arizona.

The telescopes, each only 40 centimeters in diameter, are designed to scan the 2,000 closest small, low-mass stars in the northern skies. The telescopes will look for signs that an orbiting planet periodically passes between the star and Earth, blocking a tiny but detectable amount of starlight every pass, or transit.

Nine years ago, Charbonneau, now at the Harvard-Smithsonian Center for Astrophysics in Cambridge, Mass., and his then adviser, Tim Brown of the National Center for Atmospheric Research’s High Altitude Observatory in Boulder, Colo., pioneered the transit method of hunting planets. The technique offers a key advantage over the wobble method, which reveals both transiting and nontransiting planets but provides only their minimum mass and the time it takes to orbit.

By measuring the precise amount of light obscured by a passing planet, transit studies reveal the planet’s size. When combined with the wobble data, transits also indicate a planet’s true mass. In addition, the starlight filtering through the atmosphere of a transiting planet, like a flashlight shining through a fog, reveals the composition of gases clinging to these alien orbs—even though the planet itself would lie too close to the glare of its star to be imaged.

Before the discovery of the first transiting planet, HD 209458b, in 1999 “no one appreciated that these sorts of studies would be the ones” that would first provide information on a planet’s composition, notes Charbonneau. “I don’t think people understood at the time or really realized … this would allow you to directly detect light from the planet without having to take its picture.” Eventually, scientists from all over the world embraced the strategy.

For the first few years, Charbonneau and his collaborators, along with a slew of other astronomers, limited their search for transiting planets to relatively large, sunlike stars. An orbiting planet has a higher probability of making a transit, as viewed from Earth, when the body lies close to its parent star. And in order to block enough light from a big star to be detected, the transiting planet must also be relatively large. So the transiting objects researchers initially found were all giant, star-hugging planets—the hot Jupiters.

Enlarge

UP CLOSEENLARGE | A close-up of one of the MEarth telescopes. MEarth is the first project devoted to finding habitable, rocky planets. Each telescope will scan a few hundred M stars nightly for such orbs.Brad Ehrhorn

But around 2005, Charbonneau got another idea. If he could look for transits around an abundant group of dwarf stars, called M stars or M dwarfs, which are only about one-third the mass of the sun and smaller in size, he could find planets as small as twice Earth’s diameter. “M stars are so small that you really could detect something as small as a superEarth—a planet five to 10 Earth masses—orbiting them if the planet passed in front,” says Charbonneau.

Then he took another leap. Instead of just looking for a rocky planet not much bigger than Earth, he thought, “what about going for the big kahuna—habitability?” Because M stars emit much less heat and light than sunlike stars, a planet closely orbiting one of these dwarfs might still lie in the habitable zone, where water would remain liquid. So not only could Charbonneau hope to find small planets, but also ones that might be capable of supporting life.

Calculations by Charbonneau and Philip Nutzman of Harvard-Smithsonian revealed that such a search would require telescopes with mirrors four times bigger than the 10-centimeter instruments used to find transiting Jupiters. Luckily, Charbonneau says, 40-centimeter telescopes are large, but still within the limit of what’s commercially available, without the need for years of special design and ordering.

Armed with an unexpected windfall—funds from a Packard Fellowship for Science and Engineering—Charbonneau’s team put the plan into action. This fall, the team installed the last three telescopes in the eight-telescope facility that Charbonneau calls MEarth (for transiting Earthlike planets around M stars) and has now begun hunting for habitable planets, he reported in October in Ithaca, N.Y., at the annual meeting of the American Astronomical Society’s Division for Planetary Sciences.

“This is the single greatest opportunity” in extrasolar planet studies, Charbonneau says, “because you can detect these planets from the ground and you can study them with Hubble’s successor, the James Webb Space Telescope, which is already being built.”

Theorist Sara Seager of MIT says MEarth is a good idea. She likens it to hunting for a lost item under the first available streetlight. The item may not lie under the streetlight, but it’s the easiest place to start looking. Ultimately, she notes, “we want to find Earth analogs around a sunlike star, but it’s easier to find such planets around small stars than it is to find a small planet around a big star.”

Enlarge

IN TRANSITENLARGE | A planet (1–3) crosses in front of its parent star, creating a mini-eclipse that blocks a small amount of starlight from reaching Earth.ESA; Source: Charbonneau et al.

The hunt begins

Most planet transit surveys stare at a single patch of sky, using visible light to look for transits around stars in that patch. In contrast, MEarth will track 2,000 M stars all over the Northern Hemisphere at infrared wavelengths only slightly longer than those of visible light—ranging from 700 to 1,000 nanometers—and will eyeball a few hundred of the bodies every night as they drift across the sky. Each of the eight robotic telescopes will monitor an individual M star, which is brightest at infrared wavelengths, every 20 minutes, the time needed to catch a close-in planet’s transit, which might last only an hour. If one of the telescopes finds a star that undergoes periodic dimming—a possible transit—it will trigger the other seven to track the same star.

Once a transit is identified, the star must still be observed by telescopes other than MEarth. These would carefully monitor the star’s speed, confirming that the body is truly being pulled to and fro by an orbiting planet. Combining data from these observations could provide more details about the planet, including its mass, radius and whether it is dense enough to be rocky or puffy enough to be gaseous.

To look for possible signs of life—the presence, for instance, of ozone in the transiting planet’s atmosphere—astronomers will have to wait until 2013, when the James Webb Space Telescope is set for launch. Only JWST, with its large, 6.5-meter mirror flying above Earth’s turbulent atmosphere, will collect enough photons to record the faint spectra of an M star’s light filtering through the atmosphere of a small, rocky planet, Charbonneau says.

Even with the capabilities of the orbiting observatory, taking those measurements will still be extraordinarily difficult, notes theorist Tom Greene of NASA’s Ames Research Center in Moffett Field, Calif. That’s because the atmospheres of terrestrial planets are small and thin compared with the sizes of the stars they transit. Within those small atmospheres, the region in which a molecule like water would strongly absorb starlight is even tinier, only one-two-hundredth the area of the planet’s atmosphere. So even for a terrestrial planet transiting a very bright, nearby M star, obtaining spectra would require some 50 hours of observations with JWST, Greene says. Trying to study the composition of a more massive solid planet, a superEarth, may be even harder because the larger gravitational tug of these orbs shrinks the size and area of their atmosphere, he adds. 

Only a handful of the 2,000 stars surveyed by MEarth may harbor a transiting habitable planet. (The number of habitable planets around M stars that don’t happen to transit might be 10 times that number.) But the survey, he says, will also help answer other questions.

Enlarge

HABITABLE ZONESENLARGE | The habitable zone (gray) — the region where water stays liquid — lies much closer to tiny M stars (below left) than it does to brighter, more massive stars like the sun (right). Earth’s orbit lies beyond the sun’s habitable zone, but atmospheric gases warm the planet.ESA; Source: Charbonneau et al.

“In our solar system, the biggest rocky planet is Earth,” notes Charbonneau. “We want to find out why we don’t have a [rocky] friend that’s three times the mass of Earth, why the next heaviest planets are the ice giants,” Uranus and Neptune.

A decade ago, when researchers first examined the diversity of Jupiter-like planets, it was shocking, recalls Charbonneau. “We expected to find gas giants in circular orbits with 12-year periods, just like Jupiter.” Instead, most whipped about their stars in just a few days, orbiting within roasting distance.

“We now want to explore the diversity of rocky [and icy] planets just as we have explored the diversity of giant planets like Jupiter,” Charbonneau says.

Measuring the mass and radius of a terrestrial planet reveals a planet’s average density, and that in turn may indicate where the body formed. For instance, a close-in planet with a low density, suggesting a mixed composition of rock and ice, is likely to have migrated from a spot farther from the star, where a reservoir of icy material once resided.

While MEarth won’t find nearly enough rocky planets to do a complete census, “all we want to do is open the door” to exploring the nature of the first few terrestrial planets, he says.

Other teams, including a group of planet hunters led by Michel Mayor of Geneva Observatory in Sauverny, Switzerland, have adopted a different approach for seeking habitable planets. Rather than looking for transits among a large group of stars, many of which aren’t likely to have a planet in the first place, they first use the wobble method to winnow down the sample. Only stars whose back-and-forth motion indicates the presence of an orbiting planet are closely examined for a periodic dip in brightness—the telltale sign of a transiting planet. There’s a 50-50 chance that this approach may yield the first habitable, transiting planet around an M star, says MIT’s Seager.

Caveats for the hunt

While M stars offer some distinct advantages for finding terrestrial planets, some of the stars’ quirks might not bode well for finding those bodies that support life, theorists caution.

“We’re really excited about M stars, but there are a lot of reasons not to like them,” Seager says. For instance, M stars are active for most of their long lives, with strong magnetic fields that can trigger flares and hurl parcels of gas into space. These flares are similar to the solar outbursts known as coronal mass ejections, in which the sun explosively ejects billion-ton clouds of charged particles.

In addition, although M stars overall emit much less radiation than the sun, they radiate copious amounts of extreme ultraviolet light. Both the radiation and outbursts could erode the atmosphere of a close-in terrestrial planet, and without the protection of an atmosphere, life similar to that on Earth may not be possible.

Another factor, separate from the M star questions, is that MEarth can look only for planets twice Earth’s size or several times heavier. A terrestrial planet even a few times more massive than Earth is more likely to have an entirely solid core, rather than a liquid outer core surrounding a solid center, says Seager. The churning of material within a liquid outer core is believed to be a prerequisite for generating a planet’s magnetic field. Without a magnetic field acting as a shield, harmful cosmic rays and other charged particles are more likely to hit the planet.

Theorist Jack Lissauer of NASA-Ames has similar concerns. Because an M star’s habitable zone lies so close to the star, a planet residing there could be particularly vulnerable to the vicissitudes of its parent, such as a sudden outburst. In addition, a planet twice the size of Earth could be 12 to 15 times as massive, he says, and it’s unclear exactly how Earthlike such a planet could be. The interior of such a body could be extremely hot, he notes.

Charbonneau says he agrees that the strong magnetic fields of M dwarfs can lead to increased ultraviolet emission and activity. But, he adds, “While we can ponder what the effect of all this would be on the evolution of life on the planet’s surface—it could be harmful, but then again it could promote random mutations—we have absolutely no hard evidence as to what the actual effect will be.

“My goal is very much to learn about the robustness of life in different stellar environments. If we find planets in the habitable zones of low-mass stars, and determine that these planets have all the right building blocks for life—for example that they are rocky, are at room temperature and have liquid water—but find no life upon them, that would be a very interesting result indeed.”

Other searches, other transits

MEarth isn’t the only set of instruments scanning the skies for the transit of rocky planets around nearby stars. COROT, a French-European Space Agency mission launched in December 2006, has a single, 30-centimeter telescope that stares at two much smaller patches of sky but for a much longer time than MEarth. Astronomers calculate that by the end of its 60,000-star survey next year, COROT will find 10 to 40 transiting rocky planets. Each of these rocky orbs, dubbed superEarths, would have a diameter several times that of Earth. COROT will also detect dozens of previously unknown gas giants, ESA scientists predict.

Every 150 days, COROT rotates by 180 degrees and switches observations from one patch of sky to an oppositely located region. The observatory focuses on transiting planets of sunlike stars, rather than the low-mass M stars in the MEarth mission. Although these planets lie too close to their massive and hot parent stars to be habitable, COROT will provide a better census of how common superEarths are in the Milky Way galaxy.

NASA’s more comprehensive Kepler Mission, scheduled for launch next March, will examine 100,000 stars over three and a half years, finding planets as small as half the size of Earth. With its 0.95-meter mirror, the satellite “has the ability to find actual, honest-to-God analogs of the sun-Earth system,” says David Charbonneau of the Harvard-Smithsonian Center for Astrophysics. Follow-up studies will be tricky. Even the James Webb Space Telescope won’t have the collecting area to analyze the feeble amount of starlight filtering through the atmospheres of these planets. But Kepler will answer an enduring cosmic riddle—whether planet Earth is an oddball in the pantheon of planetary systems or just another face in the crowd.

A proposed space mission by MIT’s Sara Seager and her colleagues, the Transiting Exoplanet Survey Satellite, or TESS, would use six wide-field cameras to search for dips in brightness among as many as 100,000 M stars during its two-year mission. TESS would examine a much larger group of M stars for transiting, habitable planets and would have the potential to discover several hundred of them. Because these stars would be nearby, bright systems, it would be relatively easy for JWST to study the planets’ atmospheres, notes Charbonneau. If NASA decides to fund the mission, it could be ready for launch in 2013, Seager says.

• Print
• |
• Comment