Mirror Earth (10 page)

Dave Charbonneau
(Stephanie Mitchell/Harvard staff photographer)

The world of astrophysics was so consumed with cosmology in the mid-1990s that even though the first planets found by Mayor and Marcy and the rest were announced during his senior year in college, Charbonneau doesn't remember hearing a word about it. “It was just so new,” he said, “and it wasn't
percolating down to the undergraduates. I'm sure the graduate students knew about it, but what we teach undergrads is at least five years out of date in terms of excitement.” In fact, he recalled, “I was doing my fourth-year course in mathematical physics at the time—beautiful stuff, but the content hasn't changed in a hundred years.”

When he arrived in Cambridge as a graduate student, Charbonneau fully intended to stay with cosmology. The faculty had put together a series of afternoon seminars, though, in which professors were talking about new exciting things that they thought the grad students might want to get involved in. One of those new exciting things was exoplanets. The Mayor and Marcy discoveries had quickly caught the attention of astronomers around the world. Now that everyone knew you could actually find planets, it seemed that everyone wanted to get in on the action. “There was a guy named Bob Noyes,” said Charbonneau, “and he got up at one of these afternoon seminars and gave this talk about how there was this big debate as to whether these really were planets.”

In part, the doubt centered on the same problem that had faced Dave Latham and his collaborators back in 1989. If an object's orbit around a distant star lies edge-on from the perspective of Earth, then you're seeing 100 percent of the wobble it causes in the star. If it's not edge-on, then some of that wobble is going in a direction you can't detect. You'll see only 90 percent, or 80 percent, or even less, which means you'll underestimate the mass of the orbiting object. Marcy and Mayor were always careful to talk about their exoplanets'
minimum
masses—the mass the planets would have if they were truly
orbiting edge-on. If the handful of exoplanets known at the time had a lot more than the minimum mass, which was certainly possible, they were too massive to be true planets. Another possibility, as Marcy had anticipated, was that the stars themselves were pulsing; if so, maybe it was just the surface, not the entire star, that was moving toward and away from the telescope.

Marcy had good arguments for why both of those scenarios were unlikely, but scientists can be awfully clever at coming up with good arguments. If someone could confirm the existence of his planets and Mayor's in an independent way, it would all be a lot more convincing. Noyes, a senior faculty member whose office was right across the hall from Charbonneau's, thought he had a way to do it. If the planets really were orbiting edge-on, they should show phases, like the Moon—or, more aptly, like the phases of Venus, which Galileo discovered. When Venus is between the Sun and the Earth, we see its unlit side. When it's on the other side of the Sun, we see its fully illuminated side (every so often, Venus is
exactly
on the other side of the Sun, so we don't see it at all, but this happens rarely). The same should be true for an exoplanet: It should be dark when it's on the near side of its orbit and bright when it's on the far side. You can't actually see the planet; it's much too small, and too close to the glare of the star. But you can see the total amount of light coming from the star and planet together. So when the planet is bright, the sum of star plus planet should be a little bit brighter; when the planet is dark, the sum should be a little dimmer.

This is what Noyes proposed to look for, and when he
explained it, Charbonneau was hooked. “One thing that appealed to me,” he recalled, “is that the cosmology stuff is so amorphous. I mean, if you have to give a public lecture about why people should care about the CMB, you're crippled by the fact that the main thing you want to talk about is just not accessible to someone who doesn't understand spherical harmonics.” Exoplanets, he realized, were completely different. “The basic idea is something that I can explain to anyone in just a few sentences,” he said, “and it doesn't require any mathematics. People can visualize what you're talking about.”

Beyond that, as a grad student in cosmology he would be working on problems that some of the most accomplished scientists in the world had already been wrestling with for decades. “I had this feeling that to get going in cosmology I had to read ten or twenty years' worth of papers before I could begin work,” he said. That wasn't true for exoplanets. “Literally, I had one folder and as each paper came out on exoplanets, it would go in that folder. There was a period of time when there were basically five or ten papers that you had to have read.”

In short, he realized, it was one of those rare moments in scientific history when a brand-new field opens up and graduate students can identify and answer very simple questions and make a real contribution. You could also get research grants relatively easily for exoplanet research. “Normally, when you write a proposal for money or for telescope time, you spend most of the time explaining why what you're doing is interesting. But we never had to explain why it was interesting to go and look for new planets. It was obvious.”

Charbonneau was excited by the idea of working on exo-planets for a more fundamental reason as well. “It's life—the idea of searching for life in the universe. It used to be the case that professors were very hesitant to say that, so they would always say, ‘It's the physics of planet formation,' and that's as far as they would go. But I'm not afraid to say that it is absolutely this question of ‘Are there examples of life that arose independently from the life on the Earth?'” He was even willing to say it back in the mid-1990s, when all the known exoplanets were much too large to support life, or much too hot, or both. By the time we spoke in 2010, it was a more realistic question to ask.

“So,” he explained, “you basically have a mix, you have a ball of silicate rock, you put an ocean down, not too thick, and you have roughly the right temperature and atmosphere, then is it just a matter of time when life will arise?” Or would it come out differently? “Are we going to find, say ten or twenty examples of those”—a proposition that was looking better all the time, with Kepler already in orbit for more than a year and with at least one planet plausibly made of rock, not gas, discovered by Michel Mayor's group a year or two earlier—“study them carefully, find that, although they look just like the Earth in terms of their properties, they just don't have life for whatever reason?”

For all his excitement at the prospect of life, Charbonneau thinks either outcome is equally possible. “People push me on this,” he said, “and I really, honest to God, think it could go either way. When I teach, most of my students can't imagine the latter case. They think that of course there will be life.”
That doesn't mean for a moment that he's indifferent to the answer. He hopes life exists beyond Earth. “Even if they were completely foreign, I think we would feel less lonely, I think there would be this true loneliness if we found out that this was really it. And I think that does affect how people view how precious our planet is. Even though we could never go to those other places, I think that we would still view what we have differently if we knew that it was truly unique.”

These insights weren't quite so fully formed, of course, when Charbonneau initially switched from cosmology to planet-hunting. But Bob Noyes's scheme of looking for reflected light was instantly appealing. “I worked very hard on that,” said Charbonneau, “but we never did make a detection. It turns out that reflected light is a really hard problem. I always felt we were almost there, but it took Kepler to pull it off, six hundred million dollars and fifteen years later.” After this disappointing finish to the reflected-light project he went back to Noyes, who had become his adviser, to get advice on a research topic for his thesis, the grand finale to a graduate student's career. “Bob told me it probably wasn't the best bet to go into radial-velocity searches [like Marcy and Mayor were doing] because so many teams were so far ahead,” recalled Charbonneau. “Maybe I could try to confirm planets by looking for transits.” This made a lot of sense: It's hard to imagine something other than a planet that could make a star wobble
and
make it dim, with exactly the same timing. Even so, while it was now 1999, at least a decade and a half after Bill Borucki had begun working seriously on what would become the Kepler
mission, planetary transits were still something very few other astronomers were thinking about.

Confirming that a planet really existed was one reason to look for transits, but there was an even better one, which wasn't fully appreciated at the time. When Marcy and Mayor found the radial-velocity signature of an orbiting planet, they could tell you how much time the planet took to complete one orbit, and they could tell you its mass—or its minimum mass, anyway. If you could see that a planet made a transit as well, that would confirm you were seeing the orbit precisely edge-on, so you would know that the minimum mass was also the actual mass. But you would also know the planet's physical size. If it blocked, say, 2 percent of the star's light, that meant the planet's disk was 2 percent of the size of the star's disk. With a little bit of high school geometry, you could calculate the planet's volume. And once you had the mass and the volume, you knew the planet's density.

Planetary scientists already know from our own solar system that planets come in different densities. Mercury, for example, is denser than Earth because it has a higher proportion of iron to rock than Earth does. Pluto, whether you call it a planet or a dwarf planet, is less dense than Earth because it's made largely of ice. Saturn is even less dense, because most of its mass comes in the form of hydrogen and other gases, with only a little bit of rock down in its core. One of astronomy's longest-standing fun facts is that if you could find a big enough bucket of water to put it in, Saturn would float.

Whether you're interested in finding a place where life
might exist, or simply in understanding how an alien planet or solar system formed, it's crucial to know what the planet is made of. A planet with the mass of Earth bloated out to the size of Jupiter, to take an extreme and, in fact, physically impossible example, would be a wispy ball of dilute gas—not a good place to live. A planet the size of Earth but with a mass fifteen times as high would be so dense, and have such a crushing surface gravity, that probably nothing could live there either. For all of these reasons, the detection of planets where astronomers already had radial velocities would be incredibly valuable.

In 1999, when Dave Charbonneau was casting about for a thesis topic, Bill Borucki hadn't yet gotten approval for Kepler. But a handful of other astronomers had begun thinking about transits. Tim Brown was not only thinking about them; he was actively searching. Brown was originally interested in studying the Sun, and by the 1980s was working at the National Center for Atmospheric Research (NCAR), in Boulder, Colorado. NCAR, which mostly concerns itself with climate, isn't exactly a hotbed of astronomy, but since the Sun has a big effect on Earth's climate, understanding how it works is perfectly in keeping with the lab's mission. To give just one example, the Sun gets a little brighter overall when it has lots of sunspots, and dimmer when they almost disappear, in a regular boom-and-bust eleven-year cycle. During the 1600s and 1700s, though, sunspots pretty much disappeared entirely for an eighty-year stretch. At the same time, Europe experienced a period of unusual cold, known as the Little Ice
Age. There's good reason to think the cold spell was largely due to other factors, but even so, it's useful to try to figure out how this and other changes in the Sun might affect the Earth—especially since some solar physicists think the Sun may now be entering another prolonged sunspot drought.

While he was at NCAR, Brown built a spectrograph to measure subtle pulsations in the solar surface so he could figure out what was going on inside. It could measure pulsations in other stars as well, and Brown realized that he might be able to use spectrography to look for the wobbles caused by orbiting planets. “Planets were always sort of an afterthought,” he told me during a telephone conversation during the summer of 2011. But he was aware that Geoff Marcy was looking for planets, and he knew about Michel Mayor and the Swiss group, and about the group in Canada. By the mid-1990s Brown had teamed up with a few other astronomers including Bob Noyes at the Harvard-Smithsonian Center for Astrophysics, or CfA—the umbrella organization, located within a single building complex in Cambridge, that includes the Harvard University astronomy department and the Smithsonian Astrophysical Observatory.

The Brown-Noyes group set up shop at the Smithsonian's sixty-inch telescope on Mt. Hopkins, near Tucson, Arizona. They were mostly looking for pulsing stars, but they kept their eyes out for planets as well. “It's fair to say that we didn't get anywhere,” Brown told me in a phone conversation, “but it's a curiosity that I do have two observations of 51 Pegasi that predated Michel Mayor's discovery.” He made them in 1995, about six months before Mayor announced he'd found a planet.
Brown looked at the data, saw what turned out to be the signature of 51 Peg b, and thought, “Aha, the spectrograph is probably misbehaving again.” After that, the team dabbled a bit more in looking for planet-induced wobbles, but, said Brown, “we were never seriously in that game, although we tried to be.”