Mirror Earth (29 page)

The view wasn't lost on the several hundred astronomers who had come here in September 2011. They would frequently stop and gaze out the windows, or step outside to take it in without the intervening glass. Many of the scientists had tacked on extra days for hiking or rafting; several had brought their families. But while the view and the outdoor activities were the only reason most visitors were here, it was secondary to the astronomers, who spent most of their days in a shade-darkened meeting room watching one presentation after another on the latest discoveries in the search for planets around other stars. The conference was titled Extreme Solar Systems II; it was a sequel to the first Extreme Solar Systems conference held in Santorini, Greece, in June 2007. That first one was convened in honor of the discovery of pulsar planets fifteen years earlier—the first solar system ever found beyond our own, and an extreme set of worlds by any definition.

That was just the excuse for a conference, however; most of
the talks in Santorini, like most of the talks here, had been about the more conventional worlds discovered since then. Many of the world's leading exoplaneteers were here, mingling unobtrusively with the tourists. (I ran into someone I knew from home, unexpectedly. “What are you doing here?” she asked. “What's all this about ‘extreme solar systems'?”) Looking around during the morning or afternoon coffee breaks you could see Geoff Marcy, Michel Mayor, Dave Charbonneau, Natalie Batalha, Bill Borucki, Dave Latham, Eric Ford, Jack Lissauer, and many more eminent observers and theorists, deep in conversation with one another and with the newest generation of exoplaneteering graduate students and postdocs.

Even back in 2007, there had been plenty to talk about, of course. It had been twelve years, at that point, since Michel Mayor had announced the discovery of 51 Pegasi b. Between them, Marcy and Mayor had found hundreds of planets since then, and the astronomers who had flooded into the field in the aftermath of the great revolution of 1995 and 1996 had found dozens more with their own searches. They'd found exoplanets using radial-velocity wobbles and transits and gravitational microlensing. But in 2007, Kepler was still nearly two years away from launch.

Now, everyone agreed, Kepler had changed everything. The 1,235 candidates that had been released in February had completely overwhelmed the community's ability to keep up. Even now, more than a year and a half after Bill Borucki had announced Kepler's first five confirmed planets, only 20 or so of the 1,235 candidates had crossed over the line to qualify as
actual planets—although, as Geoff Marcy told me during one of the coffee breaks, “we know that most of them are going to be confirmed.” But the number 1,235 was about to go out of date. There were already more candidates in the pipeline, a fact that was evident in an exchange on Natalie Batalha's Facebook page, where she and Geoff Marcy got a little bit silly sometime in the small hours of the morning just before the conference got under way.

BATALHA: Savoring the moment just before calculating the number of new planet candidates by looking out at the Grand Teton wilderness. Best. Job. Ever … Let the history books show that numerous new planet candidates were vetted with the aid of Cocoa Crunchies in the middle of the night.

MARCY: Wait! Surely you mean “Cocoa Crunchy Candidates.” A Cocoa Crunchy ain't nothin' until vetted by difference munching and high resolution smelling. You should take the Fourier Transform of each and every Cocoa Crunchy. Admittedly, they never taste quite as good when you transform them back.

BATALHA: @geoff: severe scattering, tidal disruption, and accretion have annihilated all cocoa crunchies. No candidates left. Abort mission! ;)

The dozens of talks that would happen over the four days of Extreme Solar Systems II (with Wednesday afternoon left clear for the exoplaneteers to take hiking excursions into the
surrounding wilderness) covered every conceivable aspect of exoplanetology. “It's really quite wonderful,” Bill Borucki said during a coffee break after a session on planet formation. “I've just been taking a lot of notes. There's just so much here.”

I knew what he meant; I mentioned to Geoff Marcy during another break that I felt like I was falling farther behind every minute. “Welcome to the club,” he said. “What do you think we feel like?”

Borucki knew beforehand about some of the announcements that would be made, of course: No exoplanet conference could take place these days without a whole series of new results from Kepler, and Borucki was naturally up to date on all of these. One had actually come out just a week before the conference began: Sarah Ballard, a Harvard graduate student, had been the lead author on a paper for the
Astrophysical Journal
announcing the discovery of Kepler-19b and -19c, orbiting a star about 650 light-years from Earth. The first planet, 19b, was found straightforwardly: It transited across the face of its star.

It hadn't been confirmed, since the star, Kepler-19, was too far away and too faint to make it possible to pick up radial-velocity wobbles. It was validated, though, by Dave Charbonneau's Spitzer Space Telescope follow-up program, which confirmed the transit in infrared light. And thanks to variations in the timing of 19b's transit, Ballard's co-author Dan Fabrycky, the creator of the Kepler Orrery, was able to identify a second, unseen planet. It was something like Matt Holman's inference of a third, unseen planet in the Kepler-9 system a year earlier, but this one was more solid.

An even bigger announcement would come out toward the end of the meeting. The Kepler exoplaneteers were concentrating their attention mostly on single stars, even though there are plenty of double stars in the telescope's field of view. The reason is that for years, theorists had considered it something of a long shot that a planet could form and survive in a double-star system, where the ever-changing gravitational pull of two different suns would tend to make for an unstable environment. “When two elephants are waltzing,” I wrote in
Time
, “it can be difficult for mice to tiptoe safely under their feet.”

Still, Kepler sees all, and since its launch the satellite has discovered plenty of binary stars, including two thousand that eclipse each other. One such pair, which orbits each other every 41 days, showed just the sort of pattern of dimming you'd expect as the light of two stars becomes the light of one, over and over. A team led by the SETI Institute's Laurance Doyle noticed that there was a tiny bit of extra dimming, however, which showed up every 229 days, on average. There was clearly a planet here. It was the size of Saturn. Its transits varied by about 9 days in either direction, because the stars themselves were in different positions each time the planet came around. A transit would start earlier when the stars were side by side and later when they were one on top of the other—and in the former case, the transit would last longer as well, as it had to cross two stars rather than one.

Not only that: Because of the complex interplay between the gravitational tugs of the stars on each other, and on the planet, and of the planet on the stars, the transits and the eclipses all showed the sort of transit-timing variations Matt Holman and
Sarah Ballard had found in the other Kepler systems. By solving this cosmic three-body problem, Doyle and his co-authors could nail down the masses of the stars and the planet with terrific accuracy. Once they knew how massive the stars were, they could calculate their physical sizes, so the size of the planet was easy to determine with high accuracy as well.

All of this might have been a little technical to get the general public as excited as the astronomers clearly were, except for one obvious but brilliant marketing move by the NASA press office. In the first
Star Wars
movie back in the 1970s, Luke Skywalker's home planet, Tatooine, orbited a pair of suns. In one eerie scene, you see the two suns sinking slowly toward the horizon together at the end of the day. So the NASA people put
Star Wars
right in the first paragraph of their press release, and brought in John Knoll, a special-effects supervisor for George Lucas's Industrial Light & Magic, to sit on the expert panel at the press conference (they tried to get Lucas himself, evidently, but he was unavailable). The director had been well aware, Knoll revealed, that theorists didn't think such a planet was likely to exist. He decided to go with it anyway. NASA's strategy worked: The discovery generated headlines around the world. It also became clear just how many exoplaneteers are
Star Wars
fans: They threw around the name Tatooine as though it needed no explanation—which for me, it actually did.

While the Tatooine discovery was the most accessible to the public, the talk Natalie Batalha gave on the first day of the meeting was also important. In it, she announced the planet candidates she had calculated during her Cocoa Crunch–fueled all-nighter. There were now 123 Earths on the list, up
from the 68 Bill Borucki had announced the previous February; 412 super-Earths, up from 288; 988 Neptunes, up from 662; and 204 Jupiters, up from 165. “It's about a 45 percent increase in candidates overall from the February release,” she said during one of the lunch breaks. The statistics available to calculate eta-sub-Earth were even richer than they had been a few months earlier—partly due to the longer observing time, but partly due also to a new algorithm in the software pipeline that screened out variability in the stars themselves more effectively.

Kepler's wasn't the only exoplanet census that was featured on the first day of the conference. Michel Mayor also stood up before the audience to announce no fewer than fifty new planets of his own, including sixteen super-Earths—an extraordinary haul for any search other than Kepler. Among them was a world named HD 85512 b, just three and a half times as massive as Earth, orbiting in at the edge of its star's habitable zone (no word on whether it might be rocky, since the Swiss team didn't have a transit to calculate its density). The Swiss had also analyzed all the planets they'd found to date by size and orbital distance and come up with the claim that 40 percent of Sun-like stars have at least one planet smaller than Saturn.

Geoff Marcy, Andrew Howard, and the American Eta-Sub-Earth Survey team, meanwhile, had come up with their own figure of 15 percent, but that wasn't a direct comparison, since the Swiss, with their more sensitive HARPS spectrometer, included planets out to a larger orbital distance. And then there was Kepler, which didn't address mass at all (except in
a few cases) but rather size—which might or might not be directly translatable.

The result, said Marcy, was that “by the end of Monday the meeting was left in an unstable mode. Natalie got up to discuss five hundred new planets—or planet candidates, but we all know they're mostly real. Andrew Howard got up to discuss the statistics of the occurrence rates of planets from Kepler and from radial velocity, looking at a different set of masses and different orbital distances. And Michel got up and did similar things.” But without a good way to compare the three sets of data, it was hard to tell whether they agreed or disagreed. “One speaker,” said Marcy, “who I won't name, suggested that there was a factor of ten disagreement. But there actually isn't.”

He felt confident in saying so because, said Marcy, “in the intervening seventy-two hours we've been working behind the scenes, especially Andrew Howard. And it turns out the occurrence rates of the planets of different masses, different sizes, and different orbital distances agree. Kepler, Eta-Sub-Earth with Keck, and the Swiss team, they all agree.” This still wasn't about true Mirror Earths, or even Earth-size planets. The radial-velocity teams hadn't measured such subtle motions, so only Kepler had any data at all about worlds this small. But Kepler, the new kid on the block, was getting the same answer for hot Jupiters that Marcy and Mayor were getting.

(While nobody has yet found an Earth-mass planet with radial-velocity wobbles, Michel Mayor told me at the meeting that his team was now capable of doing so. “Yes, yes, yes, yes,” he said, when I asked him to confirm what I thought I'd heard
him say during his talk—that he could not only detect an Earth-mass planet, but could do so in an Earth-like orbit. He could find a Mirror Earth. “We can measure as small as 0.5 meters per second. So, the only thing I can say is it's extremely expensive in terms of telescope time.” The best use of this capability, he said, would be to follow up on Kepler candidates. Since Kepler's field of view is in the northern hemisphere, HARPS couldn't do that, but he had teamed up with Dimitar Sasselov, Dave Latham, and other exoplaneteers at Harvard to build a twin, HARPS North, which was nearing completion at the Italian Telescopio Nazionale Galileo, in the Canary Islands.)

The reconciliation of the different groups' statistics was hammered out in real time, as the meeting went on—something that doesn't happen often. In fact, said Marcy, “I've never seen anything like it. This is what happens when you run a meeting with three hundred smart people and deny them Internet in their rooms. They aren't holed up fiddling away; instead they're talking. Finding out that everybody's a human being and that we're much stronger when we work together. I was around here at about twelve thirty A.M. last night, and there were twenty-five people still there talking, with their laptops open. Yeah, okay, Kepler found five hundred planets, and the Swiss team also announced some planets, and there's a circumbinary planet [that was the Tatooine announcement]. Forget the circumbinary planet! Look at the sociological experiment that's been such an enormous success!” When Andrew Howard heard this sentiment, he smiled
wryly and said, “Yeah, it's easy for Geoff to be so enthusiastic.
He
has Internet in his room.”

For all the excitement about new planets and taking steps toward a reliable calculation of eta-sub-Earth, however, there was also an undercurrent of concern, especially for members of the Kepler team. Finding at least a handful of true Mirror Earths—the same size, the same density, the same temperature as the original—was always Kepler's goal. “That was the driver of the mission,” Jack Lissauer said over lunch one day. But it was based on the assumption that the mission would last at least four years, and that the Sun was about typical in how noisy it is—how much it varies in brightness due to sunspots and flares and other activity.