There’s Something in the Air Up There

The same cosmic trick of the light that has allowed us to discover planets much too remote for our telescopes to see is also helping us determine which of their alien atmospheres could possibly support life

BY MICHAEL D. LEMONICK

SCORCHED PLANET HD 209458b is 150 light-years away in the constellation Pegasus. This rendering depicts its proximity to a sunlike star.

How scientists “see” exoplanets

GJ 1214 b, orbiting its star, is a super-Earth.

WHEN A TRAILBLAZING handful of astronomers first detected exoplanets orbiting distant stars back in the 1990s, the effect on their field was nothing short of game-changing. Before their discoveries, the hottest area in astrophysics was cosmology, the study of the origin and evolution of the universe. Afterward, senior stargazers in droves dropped what they were doing to turn their attention to alien planets, and graduate students by the score rethought their career paths in a hurry. “When I would talk to nonscientists about my work in cosmology,” says Harvard’s David Charbonneau, who was, in fact, one of those grad students, “they looked confused.” Once he switched interests, though, his explanations got far fewer blank stares. Everyone could understand the guy who held up two fists—one representing a star, the other its planet—then mimicked the one orbiting the other to demonstrate how worlds revealed their existence.

NASA boarded the bandwagon, too, developing its own long-range plan for exoplanet research. Step 1: Find as many worlds as possible by indirect means (that is, the real-life version of Charbonneau’s dual-fist demonstration, the lone means of detection at the time). Step 2: Build a Next Generation Space Telescope capable of collecting direct images of the biggest exoplanets. Step 3: Follow up with a gigantic Terrestrial Planet Finder telescope that could image smaller, more Earth-like planets.

The phases of WASP-43b, a hot Jupiter, as it orbits its star

In the end, harsh reality overtook the government agency’s ambitious to-do list. Budget and technical woes shrunk the Next Generation project—now known as the James Webb Space Telescope, or JWST—and delayed it to 2018, and the Terrestrial Planet Finder was delayed indefinitely.

Elsewhere, however, scientists didn’t receive the austerity memo. As far back as 2001, they were already beginning to figure out ways not only to detect exoplanets but to analyze their atmospheres, in the hope of finding chemicals—oxygen, say—that suggest the presence of life. The earliest subjects were giant Jupiter-size bodies, but more recently super-Earths, which fall between Earth and Neptune in size, have received similar treatment. And if signs of life remain elusive, evidence of water vapor, clouds and other life-teasing features speak to how far the field has advanced in a short while—and hint at even more significant exploration to come. “We’ll be able to use our techniques to get to the sensitivity required to really study Earth-size planets,” says Laura Kreidberg of the University of Chicago.

NASA’S James Webb Space Telescope, successor to—and twice as large as—Hubble, launches in 2018. Among other things, it will be more sensitive to water vapor.

To be clear, no one is getting any direct looks at exoplanets just yet; in fact, that remains a far-off goal. Instead we get our glimpses by observing changes in light. One way to do that is to focus on the parent star. When a planet passes directly between its star and Earth, its silhouette slightly dims the amount of starlight. The light’s character, from which we glean evidence of its chemistry, changes a bit too, as it passes through the planet’s atmosphere on the way to our telescopes.

The other way to see what we otherwise can’t is to consider the star and planet together. When a planet orbits behind a star, their combined light lessens as the planet’s glow is obstructed and we pick up the star’s alone. Similarly, the character of their light changes. The planet’s blocked glow contains spectral hints of its atmospheric makeup. By noting what is missing from the overall chemical mix, observers can deduce what compounds belong solely to the planet.

Kepler-7b, 1.5 times the radius of Jupiter(bottom), was the first exoplanet to have its clouds mapped

An artist’s conception of what clouds look like on other planets.

Another image of a planet with clear skies, resembling HAT-P-11b, which is the size of Neptune

An illustration of HD 189733b, a huge gas giant that has temperatures of more than 1,800° and fierce 4,300-mph winds. Scientists say it rains molten glass there

Kreidberg and her colleagues use both methods. In one instance they were able to calculate the amount of water vapor in the atmosphere of a world known as WASP-43b—a so-called hot Jupiter because although it is also a giant gas ball like our solar system’s biggest planet (albeit about twice as massive), its orbit is far tighter than Mercury’s, making it too hot for life. Still, that water vapor measurement is an important advance. Water, of course, is an essential nourisher of biology—at least, of the biology we know—and being able to determine its exact amount should someday tip off scientists to worlds where life might thrive.

Another University of Chicago astronomer, Jacob Bean, used the Hubble telescope to chart a precise temperature map of WASP-43b’s surface. He watched the planet through three consecutive 20-hour orbits, and as it went through its phases—a thin crescent as it neared the front, showing a dark backside to Earth; swelling to nearly full as it prepared to duck behind—Bean determined how much heat emanated from it during each one. Recording different wavelengths of light, he also saw that temperature varied with altitude and, coincidentally, water concentration did too. Considering we are nowhere close to seeing exoplanets, we sure can learn a lot about them. “This was an unprecedented use of the Hubble,” says Bean, “and really points the way to what we will do with JWST.”

Actually, all the research on WASP-43b is a dry run for the kinds of analyses astronomers hope to do on more Earth-like planets—those similar to ours not just in size and temperature but also atmosphere-to-planet ratio. One of the problems with planets like WASP-43b (and our own Jupiter, for that matter) is that they are shrouded in a thick envelope of gases. Whatever solid surface it may have sits far below, so even if it resided in a less scorching orbit, it wouldn’t be habitable. It’s why astronomers are re-aiming their atmospheric analyses to super-Earths—to date, the smallest exoplanets found in any significant number.

KREIDBERG, FOR ONE, HAS been sniffing around these worlds. She and her team used the Hubble to solve a long-standing question about the atmosphere of one known as GJ 1214b. Charbonneau and his group discovered the exoplanet in 2009, but they couldn’t tell if it was surrounded by an extended, gassy atmosphere like, say, Neptune’s or a more compact one like Earth’s. Kreidberg watched the starlight—specifically, how it was altered as it filtered through the planet’s atmosphere.

But what she saw was no alteration at all. It was both an unusual result and a common one. “Astronomers threw the kitchen sink at this planet for three years after its discovery,” says Kreidberg, implying that none of them had had any luck breaking its code. The reason for everyone’s trouble, she realized, was that the planet’s atmosphere is filled with clouds, maybe made of potassium chloride or zinc sulfide, but in any case dense enough to keep any starlight from escaping.

A team led by Caltech’s Heather Knutson encountered the same issue as they did an atmospheric analysis of another super-Earth, GJ 436b. “Clouds are cool in their own right,” Knutson says, “but they can be frustrating.”

In fact, it was getting so that planet hunters were beginning to fear that most, if not all, smallish planets were shrouded by clouds. Then, in September 2014, a University of Maryland team led by Jonathan Fraine detected water vapor in the atmosphere of HAT-P-11b. (Names of stars and exoplanets often refer to the survey that found them, which is why they can seem so different and random.) This time it wasn’t the water vapor that was the important discovery; rather, it was that a relatively cloud-free world close to Earth’s size existed. Once one was found, scientists were confident they would uncover more.

A series of observations carried out by MIT’s Brice-Olivier Demory and his team affirms Fraine’s discovery. Their target was Kepler-7b, one of the first planets discovered by the Kepler space telescope after it launched in 2009, and to see it they looked not at starlight passing through its atmosphere but at light coming off the planet itself. Surprisingly, the brightness of the planet just before it ducked behind the star was distinctly different than its brightness when it emerged on the other side. Demory’s conclusion: one side of 7b, and only one side, is shrouded in cloud cover. “It is becoming apparent that a number of these planets across the board, from hot Jupiters down to super-Earths, seem to be hosting clouds or hazes of some kind,” says Nikku Madhusudhan, an expert on planetary atmospheres at Yale. It’s just as apparent that those clouds differ in concentration and makeup from one world to the next.

Of course, there is much more work to do on exoplanet atmospheres. Madhusudhan, for example, recently measured water in three hot Jupiters and found them to be astonishingly dry. “This opens a huge can of worms,” he said in a statement released by the Hubble Space Telescope Science Institute. “We expected all of these planets to have lots of water in them.” The finding sent theorists back to reconsider their models.

As the pace of exoplanet atmospheric analysis continues to accelerate, what nobody needs to reconsider is its usefulness. It isn’t about to reveal alien life to us (not with our current telescopes, at any rate). But what it has taught us—and promises to continue to teach us—about the “air” way up there has surely pointed us in the right direction.