Mirror Earth (20 page)

MEarth had been under way for several months now. It was essentially a scaled-up version of the telescope he and Tim Brown had used to find HD 209458 b from a parking lot in Colorado a decade before. Instead of one telescope, MEarth had eight automated telescopes, each with a sixteen-inch-wide mirror, located on Mount Hopkins, near Tucson, Arizona. “They're essentially high-end amateur telescopes,” he told me. Each one cost about $80,000, but, he said, “there are lots of amateurs who have invested eighty thousand dollars in their habit, often much to the chagrin of their partners.” The MEarth group was made up of Charbonneau himself, two grad students, two postdocs, and an extremely earnest and serious undergraduate named Lauren Weiss, who has since gone on to grad school at Berkeley and to working with Geoff Marcy. “It's small enough,” Zach Berta (he's one of the grad students) once said, “that we can all fit around a single table at a cheap Chinese restaurant.”

Most of the team's weekly meetings actually took place over pizza at the Center for Astrophysics in Cambridge (Dave Charbonneau would send out e-mail invitations that began “Dear
Exoplaneteers,” which is almost certainly where the term first arose). The meetings were largely devoted to going through possible candidates and weeding out false-positive detections. MEarth's biggest problem, just like Kepler's, was eclipsing binary stars that looked on first glance like transits. The MEarth pipeline was something like the Kepler pipeline, but, because MEarth was looking at so many fewer stars, and was looking at each one for a much shorter time, it was a lot simpler. If there were any actual transits in the data, these would be discussed here—or, if a team member was away from Cambridge, via e-mail. So before he tied his bow tie, Charbonneau logged in to see what might be new. “I'd been copied on an e-mail,” he said, “from Jonathan Irwin [one of the postdocs] to Zach Berta, saying, ‘Nice shooting, Zach.' There was a plot of a light curve attached, and as soon as you looked at it, you could see that there was this beautiful event.” MEarth had found its first planet. “It was really fun to go to the dinner after that,” said Charbonneau, “because I knew that this risky project we'd taken on had delivered after just six months of operation. And Dave Latham was there, so I got to confide in him that we'd found this really exciting planet.”

The discovery, which the always-quotable Geoff Marcy would publicly declare to be a “top-of-the-top discovery in the quest for Earth-size planets,” was a planet less than three times the diameter of Earth, according to the team's calculations. Its mass, determined by the European HARPS spectrometer, was about 6.5 times that of Earth. Nominally, it was a little bigger than CoRoT-7b, but observations of this new planet, called GJ 1214 b (the GJ prefix means the star appears in a version of the
Gliese catalog later expanded on by the German astronomer Hartmut Jahrheiss), weren't afflicted by the kind of uncertainties that clouded CoRoT-7b. When you do the calculation, the planet's density comes out at about 2 grams per cubic centimeter. If it were a blob of pure water, the density would come out at exactly 1. If it were perfectly Earth-like, it would be around 5.5. Neptune comes in at a little over 1.6. “So GJ 1214 b is in this funny place,” Berta told me later, “in this weird intermediate place between Earth and Neptune, where it's less dense than Earth, but more dense than Neptune. It's difficult to understand.”

Geoff Marcy described one possibility in an article he wrote for the journal
Nature
, in which GJ 1214 b was formally announced the following December: “It is likely,” he wrote, “that this new world has nearly 50% of its mass in water surrounding an Fe/Ni [iron-nickel] core and a silicate [that is, rocky] mantle.” That being the case, he wrote, “it probably has an extraordinarily deep ocean.” Because it's so close to its star, with a year only 1.6 days long, GJ 1214 b is hot. But since the star itself is a dim red dwarf, it isn't
that
hot. The surface temperature would be about 320° Fahrenheit, well above the boiling point of water. So, wrote Marcy, “a sauna-like steam atmosphere is possible.” The ocean would be more than a thousand miles deep, however, so the pressure at depth would be enormous, keeping most of GJ 1214 b's water in liquid form. It's pretty much impossible to imagine life existing on the molten-lava surface CoRoT-7b. It's not quite impossible with GJ 1214 b.

“What you want for life,” Charbonneau told me at the time, “is a nice toasty ocean with a little bit of atmosphere. That's
not going to be happening here. I think it would be foolish to say categorically that GJ 1214 b doesn't have life. But we have no basis for thinking it could.” Jack Lissauer, a theorist at Ames and a Kepler project scientist, agreed. “We sometimes think of Earth as a water world, but we're not. If you have oceans hundreds of kilometers deep then you don't have little warm ponds.” This was a reference to Charles Darwin's speculation that this is where life began on Earth. “I'd be very concerned,” continued Lissauer, “if we found that most of the planets the size of Earth are water worlds. Of course, if you found signatures of life on them—well, you change your view. But you've got to do what you can with what you know.”

Bill Borucki had said that the original Kepler team toasted with champagne when NASA finally approved the project in 2000 after so many rejections. By the time the satellite launched in 2009 it had taken nearly as long to go from approval to launch as it had from the first proposal to the last, successful one. That's how painstaking and careful the scientists had been in designing the mission, and how careful the engineers at Ball Aerospace, who put together the actual satellite, had been at building Kepler. The telescope was originally supposed to go up in October 2008, but the project took longer to finish than anyone expected for all sorts of reasons, so the launch was delayed until the following March—even more reason to celebrate when it finally happened.

Even up until the last minute, of course, Borucki was biting his nails, at least metaphorically. “Everyone,” said Borucki, “was thinking of CONTOUR.” This isn't exactly a household name, but in the space community, CONTOUR, the COmet Nucleus TOUR Mission, which was supposed to visit at least two and maybe three comets, was a well-known cautionary
tale: It failed shortly after launch back in 2002 (by far the best Google result you get when you search for
Contour
and
spacecraft
: “UFO Destroyed Contour Spacecraft”). “In the back of my mind,” Borucki said, “I was imagining this thing going up … and then going
plunk!
into the ocean.”

But the spacecraft didn't fail. “It was a night launch,” Borucki said. “It was beautiful. It was just so wonderful after all those years. To know we're now going to get data instead of just trouble. You can't imagine the elation we felt.” So was there a second champagne toast? “Not exactly,” he said. Kepler went into space atop a Delta II rocket from Launch Complex 17B at the Cape Canaveral Air Force Station. “It was a government thing,” he explained, “so there couldn't be any alcohol at the launch or on the bus. We went back to a hotel and that was where we had our party, and I think there's a picture somewhere of me and Dave Koch drinking champagne. That was exquisite. I think we even had Dave dancing up on a table or something.”

The launch was clearly a crucial milestone. Even so, Kepler wouldn't be announcing its first five planets until the following January. First, the spacecraft had to be checked out, during a two-month commissioning phase where engineers tested all of the electronics and other hardware to make sure it was working in space exactly as they'd planned. In practice, a spacecraft never does work exactly as planned, but knowing the difference between expectations and reality—that this light-detecting CCD, for example, is a few percent less efficient than the one next to it—lets the astronomers correct for any hardware or software errors. Real science observations
began in May. Even so, however, it would take time for the software pipeline to identify Kepler Objects of Interest, and even longer to weed through them for planet candidates, and even
more
time to go out and confirm at least some of those candidates with radial-velocity measurements.

Those first five, moreover, which would finally be announced at the American Astronomical Society's 2010 Winter Meeting in Washington, would all be as big as or bigger than Neptune. It would take even longer for Kepler to get down to the level of a Mirror Earth, or even a super-Earth. Extracting a good signal from a transiting planet in the face of a star's roiling surface and flares and sunspots and weeding out electronic noise from the detectors required many, many observations. So even after Dave Charbonneau formally announced the discovery of GJ 1214 b, eight months after the launch, there was time for other groups to try to find the very first potentially habitable planet. GJ 1214 b wasn't the first; it was too wet. CoRoT-7b wasn't the first; it was too hot.

The Gliese 581 system, however, was proving to be just as rich a planet-hunting ground as Michel Mayor's team had first begun to realize back in 2005. The Swiss team had found four planets orbiting the star—planets b, c, d, and e. Planet c had looked like a good place for life, but it turned out to be too hot, so it was ruled out. But early in 2011, d, which had seemed clearly too cold, was belatedly ruled in. The problem with 581 c had been that any atmosphere it might have would likely trap too much of the star's heat, just as Venus traps the Sun's heat. A runaway greenhouse effect would, according to climate modelers who simulated the planet with a plausible,
carbon-dioxide-rich atmosphere, make it too hot to be livable.

But then the modelers turned their attention to 581 d, a planet whose minimum mass was about seven times that of Earth. So 581 d was plausibly a super-Earth, even better than the real Earth at holding on to an atmosphere. If that atmosphere were rich in carbon dioxide—just as plausible for this planet as for its brother 581 c—the greenhouse effect would work in its favor. The modelers had more good news. Originally, the astronomers had realized that Gliese 581 d was almost certainly tidally locked to its star, showing just one face to the star at all times (just as the Moon shows only one face to the Earth). That implied that the perpetually dark side would be so cold that the atmosphere would freeze and collapse. The climate modelers showed, however, that this wouldn't happen: The atmosphere would circulate enough from the warm side to the cold side that the temperature differences should even out.

It was plausible, then, that Gliese 581 d was habitable. Without a transit to confirm its size and density, and maybe even to detect its atmosphere, there was no way of knowing. Maybe it was significantly more massive than its tug on the star would suggest. Even if it wasn't more than seven times as massive as Earth, Charbonneau's discovery of GJ 1214 b more than a year earlier had shown that a planet you might expect to be rocky and Earth-like based on size alone could instead be a water world, or even something more like Neptune than Earth—a rocky or icy core surrounded by a crushing, poisonous, hydrogen-rich atmosphere.

Mayor's team was not the only one looking at this star, however, and by the time Gliese 581 d had sneaked, barely, and only possibly, into the habitable zone, yet
another
planet had shown up in the Gliese 581 system. In September 2010, reporters got an e-mail alerting them to a press conference at the National Science Foundation where a major announcement about exoplanets would be made. Geoff Marcy's longtime collaborators Paul Butler and Steve Vogt were to be the speakers. When reporters saw the details, it was clear that the word
major
was justified. By combining their own observations with data from the Swiss team, Butler and Vogt had teased out the signal of two more planets, Gliese 581 f and Gliese 581 g. Planet f had about the same minimum mass as d, or about seven Earth masses, but was unambiguously outside the star's habitable zone. Planet g, however, had a minimum mass less than four times that of the Earth—and it was smack in the middle of the zone. “We're pretty excited about it,” Vogt told me when I reached him at Butler's house the night before the press conference. “I think this is what everyone's been after for the past fifteen years.” I contacted James Kasting, of Penn State, for an outside opinion, and Kasting, who is considered the world expert on habitable zones, agreed. “I think,” he said, “they've scooped the Kepler people.”

At the press conference the next day, Butler and Vogt acknowledged that without a transit to confirm Gliese 581 g's density, there was no way to guarantee that the planet had a rocky surface for life to walk or crawl around on. It could in principle be like Dave Charbonneau's water world, GJ 1214 b. There was no ambiguity at all about its surface temperature,
however; it would be somewhere between -22° and -10° Fahrenheit—without an atmosphere. With any sort of reasonable atmosphere, it would be warm enough, at least on some parts of the planet (like others in the system, it always turns one face to its star), for water to be liquid most of the time.

Reporters naturally wanted a better headline than that, however (recall how the AP's Seth Borenstein buttonholed Bill Borucki after another press conference to pry loose something headline-worthy), so they asked explicitly about the prospects for life on the planet. At first, the astronomers demurred. They weren't biologists, Vogt protested, and Butler said, “I like data,” making it clear that he didn't feel comfortable speculating. He wanted only to talk about things he could actually measure. But then, pressed repeatedly by reporters, Vogt gave in, and said something he would quickly regret. “The chances for life [on Gliese 581 g],” he said, “are 100 percent.”

The reporters loved this, naturally, and ran with it. But many of his colleagues were appalled, or even worse. Some talked privately about Vogt as if he were a clueless buffoon. I spoke with Vogt a couple of weeks after the event, and he was clearly embarrassed at having taken the bait. “I'm not good at these media things where you have to speak in sound bites,” he said. “I have no knowledge of whether there's life on the planet, obviously. I was expressing a personal opinion, or really, just a personal speculation.” It was informed speculation, he argued, based on the fact that life on Earth seems to thrive under a huge range of conditions. Bacteria, in particular, have been found in super-heated water inside geysers and near
volcanic vents at the bottom of the ocean. They've been found on pools of super-salty water, and pools tainted with radioactive mine tailings, and in pockets of water within Arctic sea ice, and encased in rock a mile underground. “Life seems to find a way, given even the smallest of chances,” he said. “I look at 581 g, if it's there, and it is a place where life has a lot of chances of having a foothold.”