Mirror Earth (12 page)

Sara Seager
(Justin Knight)

Some of the faculty on her thesis committee thought that the planets didn't even exist and that nobody would be able to observe them and test her theoretical predictions in any case. “But Bob Noyes was also on my thesis committee,” Seager told me, “and he really believed in these planets. He was like, ‘Don't listen to these people, this is going to be good.'” For the next two years, Seager and Sasselov worked together on the atmospheres of hot Jupiters and on the early universe, both at the same time. “I'm sure nobody does this anymore,” Sasselov told me, “and from the perspective of ten years or more of time, it feels like we jumped from one field to another in a sort of instantaneous split. But if you look at it with the resolution of a month, you can see that it didn't really happen instantaneously. You're in one field, and then, at some point, you cross into the other field. I don't work on the early universe anymore.”

As she went ahead and worked out her thesis, the same insight came to Seager that had come to Tim Brown. If these hot Jupiters really existed, they were hugging their stars so tightly that at least some of them should produce transits. “In retrospect,” she said, “one of the things I was really good at was having hunches about things that were really going to happen.
So now I try to take action on those hunches because I am older and wiser.” At the time, Dave Latham was working in an office right nearby, and, recalled Seager, “every time I saw him I was like, ‘Dave, you've got to look for transits, transits are about to happen.' I knew he had a few radial-velocity candidates and I kept telling him over and over again, because it was fresh on my mind, ‘You've got to do this. I'm sure you must have a transit in there.' ”

Latham remembers things the same way, except he remembers Dave Charbonneau bugging him as well. And Charbonneau remembers getting the idea from Bob Noyes. Trial lawyers know that eyewitnesses to an event often remember vivid details—but that the details don't match up with the memories of other witnesses. Whatever the case, it's clear that the idea of looking for transits was in the air at Harvard in 1998 and 1999. Maybe Seager thought of it first, or maybe it was Bob Noyes, or maybe it just emerged out of multiple conversations between different pairs of astronomers, without anyone deserving full credit. What happened next is that Seager, her thesis finished and approved, headed off to the Institute for Advanced Study, in Princeton, New Jersey, to take a five-year postdoctoral fellowship. She was still a theorist, not an observer, but she couldn't stop thinking about transits.

“When I went to Princeton,” she said, “I tried to figure out if I could make an observation myself. But do you know how many photometric nights they get a year in New Jersey?” she asked, referring to the kind of clear, dry, Moonless nights when you can make high-precision measurements of an object's brightness. “They get one per year. I mean, come on, it's
hazy all the time there.” Besides, Princeton's only good-size telescope didn't have the light detectors she needed to make that kind of observation. She checked back at the University of Toronto, but they didn't have the right equipment either. “So after that, I dropped it. It wasn't really my thing anyway.”

This was in the fall of 1999. The following January, she was at a conference where she ran into Dave Latham. He gave her a huge hug. “I mean, people don't usually give hugs in astrophysics. It's not protocol. And I was like, ‘What's going on?' And he told me he was just so happy I'd bugged him about transits, because Dave Charbonneau, who is one of the luckiest people on Earth, had just found the first transiting planet. So I was like, ‘Oh, my God, I could have found the first transiting planet, but I just didn't pursue it enough.' “

Seager was hired at the Institute for Advanced Study mostly because of her work on cosmology. The institute, founded by New Jersey department-store magnate Louis Bamberger, is completely independent of Princeton University, which lies a mile or so to the east, across a golf course. The scientists at both institutions work closely together, however. Albert Einstein, for example, the institute's first hire, would frequently shuffle over to the university to give or attend talks and consult with colleagues. Combined, the university and the institute allow Princeton to rival Cambridge or Pasadena or the other Cambridge as one of the world's leading centers of cosmology—which was tantamount, when Seager arrived, to astrophysics. When Geoff Marcy's partner Paul Butler had come through in 1996 to give an early talk about exoplanets, the local astronomers listened politely and asked a couple of
questions, but they didn't quite understand what the fuss was about.

The fact that Seager had moved on from cosmology didn't bother John Bahcall, though. Bahcall was head of the institute's School of Natural Sciences, and the man who hired Seager. He was a cosmologist like everyone else, an expert on, among many other things, neutrinos, a type of elementary particle so ephemeral that it could speed through a wall of lead a trillion miles thick without noticing there was anything in its way. Bahcall was also a key figure in getting NASA (and by extension, Congress) to build and launch the Hubble Space Telescope. Bahcall's policy was to find the best people he could and bring them to the institute, whether they were in the mainstream or not. The institute's other natural science faculty members were less certain, however. “I learned after he died [in 2005] that they weren't so sure about bringing me on,” said Seager. “They knew I'd be working on exoplanets, but who would I work with?”

The answer came about a year later. Although it had taken them a little longer than their counterparts at Harvard, several of the Princeton cosmologists had begun to take an interest in exoplanets. One was David Spergel, a MacArthur “genius” grant winner for his work on cosmic structure and other esoteric ideas. In the early 2000s, Spergel was still deeply involved in a satellite called the Wilkinson Microwave Anisotropy Probe (WMAP), which would dramatically improve on the data beamed down from the Cosmic Background Explorer a decade earlier. Another was Edwin Turner, whose specialty was gravitational lenses—a phenomenon, predicted by Einstein in the
1930s and finally found in the 1970s, in which a massive foreground object, such as a galaxy, warps and distorts the light from a more distant background object.

Turner, Spergel, an aerospace engineer named Jeremy Kasdin, and a few other interested parties, including a couple of students, began meeting informally once a week to talk about exoplanets in general. Among other topics, they also talked about one of the projects NASA had been dreaming about for twenty years, at least, but which was now, thanks to the discovery of actual exoplanets, getting some research funding. This was the Terrestrial Planet Finder (TPF), a space telescope powerful enough to image exoplanets directly, and maybe even search their atmospheres for telltale signs of life. “So I was sort of the local resident expert on exoplanets,” said Seager, “and David Spergel said, ‘Do you want to join this team?' And what followed was one of the most enlightening and invigorating series of discussions—I won't say that I've ever had, because this happens a lot in exoplanets, but it was amazing. The word
no
was not really allowed. It was like, ‘Can we detect lightning on an exoplanet?' That kind of thing. It didn't get documented, because we weren't fast enough to get it on paper. And Spergel came up with the pupil idea that summer.”

The pupil idea is something people at Princeton still shake their heads over. The biggest problem with imaging a planet directly is that planets and their stars are very close together, and stars are vastly brighter. It's like trying to pick out a candle sitting next to a searchlight. So you need to blot out the star's image somehow, and the simplest answer—just paint a
black dot on the telescope, or the equivalent—is really hard to do for complicated reasons involving the laws of optics. David Spergel became fascinated with the problem, but he didn't know much about optics. So he took home a textbook one weekend, read it through, and showed up the following week with a radical proposal. Instead of blotting out the starlight directly, you could achieve the same effect by putting a mask over the telescope's opening with a cutout in a shape resembling a cat's eye.

Because of the way light diffracts, or bends, at the edge of the cutout, the opening redirects starlight so that it more or less disappears in big areas of the image. It's a bit like the way ocean waves bend around promontories and other features on a shoreline to leave some areas calm and others choppy. If there's a planet in the place where the starlight has been removed, you can see the planet, faint though it is. “This is a completely new idea,” Michael Littman, an optical engineer at Princeton and a member of the discussion group told me at the time. “But once you see it, it's obvious. I kick myself that I didn't see it myself.” Ed Turner was a little more colorful: “My jaw dropped when I heard about it,” he said.

Seager found the conversations exhilarating, and they led to collaborations with Turner and a Princeton grad student named Eric Ford on how TPF might detect the presence of life on an Earth-like exoplanet if and when the planet were found—a difficult job, given that all you'd see would be a single dot of light, but they thought of a couple ways to do it. But TPF wasn't going to happen for years, and Seager had
plenty of other research ideas. She thought she'd given up on searching for transits herself, but then she started up another search, working with a postdoc named Gabriela Mallén-Ornelas. Mallén-Ornelas was from Chile, where several of the world's biggest, most powerful telescopes are located, in the high desert in the north of the country. In return for providing the sites, Chile gets 15 percent of the time on the telescopes. “She and I sat down and said, ‘What can we do with all that telescope time?'” They decided to search for transits. “We did it all by ourselves, starting from scratch,” Seager said. “Unfortunately,” she continued, “we failed in the end because if you do a complicated experiment you need to get every single step right, and we made a small mistake. We didn't appreciate how hard it was going to be to work with very faint stars.”

In retrospect, Seager thinks John Bahcall suspected she might fail. “He did say to me afterward, ‘Now you really understand data. That's maybe all I wanted you to get out of this project.'” What Bahcall was getting at was the fact that Seager was a theorist and a builder of computer models. She had spent most of her career in the realm of the abstract, but it was important to go up against the real world of observations as well. Bahcall also thought it was important to take scientific risks. “At that time,” she said, “I was traveling all around the country giving talks. There was always one person in the audience who would come up to me afterward, usually some middle-aged man, some professor, who would say, ‘This is so exciting, if I was a young man I would be doing this.' I don't know if they would though. It was a big risk to invest so much in a field that a lot of people thought would go nowhere. But
I had an advantage. At the institute, I was supported to do anything I wanted, and I had John's guidance. He didn't say, ‘Do this, do that, I advise this, I advise that.' I'd say, ‘I am so excited about this paper I'm working on, because it's going to be so useful for observers.' And he would say, ‘It's not going to be useful if you don't get it published.'” Bahcall mentored her, she said, “in this really subtle but persuasive way.”

For plenty of young scientists, Bahcall could also be a little bit terrifying. Once a week, for many years, he would host what was known by astrophysicists around the world as Tuesday Lunch, in a dining room off the main institute cafeteria. Physics and astrophysics faculty from the institute and the university, along with postdocs, grad students, and eminent visiting scientists, would sit around a long, U-shaped table, with Bahcall at the head. First he'd call on the person sitting next to him, who was always the visiting scientist who had just finished giving a formal talk in the hour before lunch. Bahcall would ask him or her to speak briefly on an entirely different topic, off the cuff. Then he'd go around the room, sometimes in order, sometimes at random, asking whomever his eye landed on to speak for a few minutes about his or her research project. He was always perfectly friendly about it, but if he asked a follow-up question that exposed a lack of rigor or knowledge on the speaker's part, it was considered deeply embarrassing.

By Seager's lights, though, he was simply a great guy. “People told me a story about a time when child care at the institute was a real problem—mostly for the wives, because the postdocs were mostly male.” There was a day care center
on campus, but they wouldn't take infants, and they didn't seem interested in changing that policy. “One day the wives spoke to John,” she said, “and the next day they took infants. I mean John had real power and authority. And his idea was just to do whatever could help the postdocs become great scientists.”

With a Ph.D. from Harvard, an appointment to the Institute for Advanced Study, and John Bahcall's blessing, you'd think Seager would have had plenty of job offers. The transit project didn't work out, it's true, but shortly after she'd arrived, she and her Ph.D. adviser, Dimitar Sasselov, had written an important paper together. As a planet transited in front of a star, they realized, some of the starlight would shine through the planet's atmosphere. Whatever elements or compounds were present in the atmosphere would absorb the starlight at specific wavelengths—specific colors of light. They'd create the same kind of spectral lines that Mayor and Marcy were using, except that these absorption lines were coming from the planet's atmosphere, not the star's.

When the planet was in front of the star, those extra lines would be superimposed on the star's own lines; when it went behind the star, the extra lines should disappear. Seager and Sasselov figured out which of the lines should be most easily visible in a telescope; they should come, the co-authors said, from sodium compounds. A couple of years later, Dave Charbonneau and Tim Brown used the Hubble Space Telescope to look for this effect in their original transiting planet, HD 209458 b. Sure enough, there they were. It was the first detection ever of an exoplanet's atmosphere. “As a result,” Charbonneau
told me, “Sara is really credited with a lot of those first calculations—because they turned out to be right.”