Light of the Stars

by Adam Frank

  “People came away from that [second] meeting with a sense of what was possible,” Tarter told me. “The reflex motion approach was seen as particularly promising if the technology could be hammered out. I think a lot of folks were really excited.”9

  Astronomer and SETI research leader Jill Tarter.

  Not everyone was so happy, however. While the transit method was raised at the Maryland meeting, its prospects were deemed to be dim. The final report concluded, “The Workshop considered the role of photometric [transit-based] studies in an effort to detect other planetary systems and upheld the conclusions of earlier studies, namely, that photometric studies are not practical.”10

  That conclusion didn’t go down well with one tenacious scientist. “There was a young NASA researcher named Bill Borucki,” Tarter said. “He felt the transit method had a lot of promise, even if everyone else thought it was hopeless. I think he was determined to prove them wrong.”

  THE FALL OF A THREE-THOUSAND-YEAR-OLD QUESTION

  In 1995, at an astronomy conference in Florence, Swiss scientist Michel Mayor walked up through the audience and took his place at the podium. The other astronomers present looked around the room and wondered why a film crew had just appeared. That’s when Mayor dropped his epoch-making bombshell. He and his partner Didier Queloz had firm evidence for the existence of another planet orbiting another star.11 When it came to solar systems, at least, we were not alone.

  In the decade and a half following the Ames and Maryland meetings, the hurdles blocking the way to reflex motion–based planet searches had been overcome. In the US, astronomers Geoff Marcy and Paul Butler had built a series of ever more sensitive instruments to monitor a long list of nearby stars. Theirs was the world’s most complete planet-hunting program.

  But Marcy and Butler were expecting other solar systems to look like ours. They thought they’d need years of tracking before the signal of a Jupiter-sized planet in a Jupiter-sized orbit would appear in their data (Jupiter takes twelve years to make one swing around the Sun). The European researchers Mayor and Queloz had an observational program oriented toward finding close binary stars. They’d gotten lucky with their planet detection, but also had the insight to recognize what they’d found.12

  Mayor and Queloz discovered their planet orbiting the star 51 Pegasi, which is fifty light-years from Earth (one light-year equals six trillion miles). The planet, called 51 Pegasi b, was Jupiter-sized, but swung around its star once every four days. That meant it was almost ten times closer to its star than our innermost planet, Mercury, is to the Sun.13 A giant planet on a tiny orbit was not what astronomers were expecting when it came to solar systems.

  Back in the US, Marcy and Butler quickly began looking for planets on such short orbits. It didn’t take long for results to appear. At a press conference just a few months after Mayor’s talk in Florence, Marcy and Butler announced their own discovery of two more Jupiter-sized worlds.14

  New planets began to pile up after the discovery of 51 Pegasi b. As the shock of discovering exoplanets wore off, astronomers got down to the work of building a census of the new worlds.

  But the real prize still lay in the search for Earth-sized worlds living in the star’s habitable zone, where water and perhaps life could exist on the surface. The Earth’s mass is one three-hundredth that of the Sun, a fact that meant even more precision was needed to detect Earth-sized worlds. That need for greater precision was also coupled with the problem that reflex motions only worked on one star at a time. What astronomers desperate for data needed was a precise way to discover planets wholesale. That threshold would be broken by the stubborn resilience of a man who simply refused to accept rejection.

  Bill Borucki was a longtime NASA scientist who had cut his teeth on the physics of spacecraft heat shields. In the late 1970s, he decided to switch fields. The problem of planet detection offered the kind of technical challenges he loved, and after the Maryland meeting where transit-based planet-hunting methods had been dismissed, Borucki became determined to show these methods could work. In a now-famous 1984 paper, he and a coauthor laid out the basic framework for how to build a precise device to detect tiny changes in a star’s light output. Then, in 1992, he proposed a space-based telescope using the same technology for planet hunting.15

  While NASA thought the idea was interesting, it didn’t believe Borucki’s detectors would work. Unperturbed by the proposal’s failure, Borucki began systematically addressing NASA’s concerns. He built prototypes on the cheap to demonstrate that his system could hit the needed goals. After months of exhaustive work, Borucki’s designs worked exactly as he said they would. In 1994, he spent months putting together the documentation needed to propose his transit-based telescope again. The proposal was rejected a second time.

  A different set of concerns was raised in the second rejection. The new questions focused on Borucki’s claim that he could do transit-detection on many stars at once. Once again, Borucki cobbled funds together and carried forward the extraordinary efforts needed to address each and every objection. Four years later, he and his team sent in their new version of the proposal. The proposal was shot down a third time.16

  A reasonable person might have given up at that point. But in this regard, Borucki was not reasonable. He knew he was right. He knew the transit method would be a game changer. The only direction he could allow himself to move was forward.

  Eventually, Borucki prevailed. After more than two decades of working on the same idea and having that idea rejected as scientifically unsound, Borucki’s proposal was finally accepted. What would come to be known as the Kepler Mission was given the green light.17

  Kepler was designed to stare at a single portion of the sky. In that small patch of cosmic real estate, about 156,000 individual stars had been identified as worthy of attention.18 The satellite would patiently watch the same stars week after week, year after year. The patience was needed to accumulate enough transits—enough dips in light output—to provide an unambiguous signal of an orbiting exoplanet.

  On March 6, 2009, the Kepler telescope rode into space on a Delta II rocket.19 The launch was flawless. After so many years of rejection, Borucki and his team were staring across the frontier, ready to see how well his decade-spanning vision would work. They didn’t have to wait long.

  “As soon as the data started coming in from the spacecraft, we could see transits,” recalls Natalie Batalha, a NASA astronomer who joined Borucki ten years earlier. “You could see the dips as clear as day. We were literally just sitting there in our office, watching as new planets were discovered with each transit.”20

  The first confirmed detections of exoplanets by Kepler came in January 2010, but they weren’t the real news. Along with these detections, thousands of Kepler “candidates” were identified. These were stars showing dips in light that hadn’t yet been confirmed as real planet detections. With so many exoplanet candidates, the Kepler team was sitting on the equivalent of a cosmic piñata. By 2014, that piñata had been busted wide open. That year, the Kepler team announced the discovery of 715 exoplanets in a single news release.21 Wholesale planet hunting was the new reality. By 2015, the combination of Kepler and other methods had given astronomers 1,800 new worlds that were ready for detailed investigation.22

  As the list of exoplanets grew, one of the first and most impor­tant conclusions was how different the architectures of other solar systems could be from our own.

  Here on Earth, we all grew up learning about our solar system’s tidy arrangement of small, rocky worlds tucked close to the Sun and larger gas giants splayed out at ever-greater orbital distances. The very first exoplanet discovered, 51 Pegasi b, showed that this arrangement was anything but universal. It’s an example of what is called a “hot Jupiter”—a gas giant that somehow ended up on an outrageously tight orbit. Big planets on small orbits are easy to find in reflex-motion searchers, so many more of these hot Jupiters were quickly added to the exoplanet tally. Lots of stars were al
so found to have Jupiter-sized worlds on orbits the size of Earth’s, rather than out at the farther reaches of their solar systems.

  Eventually, other kinds of planets living close to their parent stars would be found—“hot Neptunes” and even “hot Earths.” Inner rocky worlds and outer gas giants were clearly not the only way nature laid out her planetary families. Systems with hot Jupiters were the most dramatic examples of “weird” solar systems, but there were many other surprises. Systems consisting of only smaller rocky worlds were found, and even they looked weird by our standards.

  “One of the big surprises was our discovery of what we call ‘compact multis,’ ” says Batalha. “These are planetary systems with a bunch of small planets clustered very close to each other.”23 In our solar system, Earth and Venus are the nearest neighbors, coming as close to each other as twenty-five million miles. That’s why it takes many months for us to reach these worlds. But in the compact multiplanet system Kepler 42, for example, there are three planets stuffed into remarkably tight orbits. These worlds get one hundred times as close to each other as Venus ever gets to Earth.24 If you lived on one of Kepler 42’s worlds, you could travel to your neighbor planet in just a week or so, using the kind of spacecraft that got us to the Moon back in 1969.

  The architecture of planetary systems wasn’t the only surprise. “We found a whole class of planet out there that don’t even occur in our solar system,” says Batalha. There are no planets orbiting our Sun with a mass between those of Earth and Neptune. That represents a considerable gap since Neptune is a big mix of gas and ice and weighs in at fourteen times the mass of Earth. Earth and Neptune are, in other words, very different kinds of planets. But as the exoplanet revolution matured, astronomers soon found worlds—a lot of them—with masses right in that gap between one and fourteen Earth masses. They called these “super-Earths,” and it soon became clear that this new kind of planet, which doesn’t even occur in our solar system, might be the most common in the universe.25

  “We don’t even understand what these worlds will look like,” says Batalha. “Some of them could be rocky. But some could be water worlds with deep oceans surrounded by thick water-vapor atmospheres. Others could be a mix of rock and ice and gas. The possibilities are pretty broad.”

  Beyond the general findings, there were the incredibly weird specific cases. For example, there’s J1407B, the “super-Saturn” located 434 light-years from Earth. The rings orbiting this gas giant stretch two hundred times farther than the gossamer disk surrounding Saturn.26 Then there’s 55 Cancri e, which is forty light-years away. Its diameter is only twice as great as Earth’s, but it has a mass almost eight times higher, resulting in a density so great that it may be a planet made of diamond.27 And not to be missed is the ominously named WASP-12b. It’s a hot Jupiter with a temperature of nearly 4,100 degrees Fahrenheit, making it one of the hottest exoplanets ever discovered. Astronomers can see a trail of debris surrounding the planet as WASP-12b boils away in a torrent of evaporating gas.28

  In the end, though, what matters most are not hot Jupiters, super-Saturns, or super-Earths. The numbers as a whole are what make the exoplanet revolution so important for us. At the beginning of the second decade of the second millennium of the Common Era, humanity finally learned that, in one very real sense, we were not alone. There were other worlds out there. Just as important, with a full census of planets being built, the first three terms in the Drake equation were now fully known. With that advance, questions not only about planets, but even about civilizations other than our own, could be seen in an entirely new light.

  DRAKE AND THE EXOPLANET REVOLUTION

  The first term in Drake’s equation describes the rate of making stars (called N*). It has been known with some accuracy since the late 1950s, and subsequent studies have only refined that value (about one star per year).29 But when Drake first wrote his equation in 1961, the second term, describing the fraction of stars with planets (called fp), and the third term, describing the number of planets in a star’s habitable zone (called np), were anyone’s guess. By 2014, in the wake of Kepler and other exoplanet studies, there was enough data in hand to give scientists meaningful—that is, statistically significant—values for those numbers.

  The implications of this advance are stunning enough to change our experience of the night sky. Let’s consider the fraction of stars with planets first. Remember that, during the early part of the twentieth century, astronomers believed planet formation was a rare event, meaning the fraction of stars with planets would be very low. But by 2014, the agreed-upon value for fp was about 1.30 In other words, pretty much every star you see in the night sky hosts at least one planet.

  The next time you find yourself outside at night, take a moment to stop and consider the implications of this result as you gaze at all those pinpricks of light. Every one of them hosts at least one world, and most stars will have more than one planet. Solar systems are the rule and not the exception. They’re everywhere.

  The advent of Kepler also allowed astronomers to reach a firm conclusion about the average number of habitable-zone planets orbiting each star. Remember that the habitable or Goldilocks zone is a band of orbits around a star where liquid water can exist on a planet’s surface. That means any planet in a star’s habitable zone might be a world of rain and rivers and oceans—a world potentially capable of supporting life. There are currently two planets in the Sun’s habitable zone—Earth and Mars—and both have had water running in torrents across their surfaces.

  From the exoplanet data, astronomers can now say with confidence that one out of every five stars hosts a world where life as we know it could form.31 So, when you’re standing out there under the night sky, choose five random stars. Chances are, one of them has a world in its Goldilocks zone where liquid water could be flowing across its surface and life might already exist.

  The importance of the achievement represented by nailing these two numbers cannot be overstated. Through the hard-won efforts of a generation of astronomers, we increased the number of known terms in Drake’s equation by 200 percent. Where there was darkness, there now is light. Where there was ignorance, there now is knowledge.

  YES, THERE PROBABLY HAVE BEEN ALIENS

  But what, if anything, could the trove of data leading us to these numbers reveal about the possibility of other worlds inhabited by technology-deploying, civilization-building species? We still have zero evidence that such civilizations exist. Is there any way to leverage the achievement of the exoplanet revolution to say something—anything—about exo-civilizations? Addressing exactly that question was the task Woody Sullivan and I took on at the beginning of 2015.

  I first met Woody Sullivan in the late 1980s, when I was a physics graduate student at the University of Washington. He’s tall and slender with a wry sense of humor and a passion for sundials and baseball (the Seattle Mariners, in particular). Most importantly, Woody is a radio astronomer with an unwavering interest in SETI. When I was a graduate student, he was the only person on the faculty at the University of Washington who worked on the question of exo-civilizations. This was well before NASA began serious funding for astrobiology. The exoplanet revolution was still a decade from its inception. In the 1980s, SETI and its astrobiological surroundings were still considered a bit “out there” for many folks. But Woody didn’t care. He was interested, and he thought there was science to be done. So he pressed on and wrote a number of important papers on the subject.

  I once helped Woody teach a course called “Life in the Universe.” He set the class up to deal with everything from the nature of physical law to the prospects for life on other worlds. His perspective was broad and imaginative. I loved being involved with that course, and its perspectives shaped my thinking for decades. It was also the first time Woody and I started talking about exo-civilizations. Those conversations have been going on ever since, even before I did any direct work in astrobiology.

  In 2014, Woody and I found ourselv
es asking if all the new exoplanet data could be used to infer a definite conclusion about technological civilizations on other worlds. The astonishing progress made since the first exoplanet discovery had to be good for something. Wasn’t there some way to use it with an eye toward answering Drake’s original question about our uniqueness in the universe? We soon saw there was a path forward, but to take it, we’d have to turn Drake on his head.

  Drake built his famous equation on a simple question: How many exo-civilizations exist now? He chose that focus because his real interest was in finding signals from alien civilizations. For his equation to make sense, those aliens had to be out there, emitting radio signals right now (relatively speaking). But to make the kind of progress Woody and I were interested in, we realized we had to change the focus. We had to ask a different question—one that could be answered by the exoplanet data. Our new question was only slightly different, but the small change we made would mean everything in terms of results. Our question was this: How many exo-civilizations have there ever been across the entire history of the universe?

  Taking this approach gave us a strategy for getting an empirically based number concerning the existence of exo-civilizations. First, we combined all the astronomical terms in Drake’s equation into one. This was easy, since they were all known. Then we began thinking differently about three unknown probabilities involving life in Drake’s equation (fl, fi, and ft). Rather than dealing with them separately, our approach lumped them all together, too. We were interested in the process as a whole, going from the origin of life all the way up to an advanced civilization. We called our new term the “bio-technical probability,” fbt, and it is the product of multiplying all the usual life-centric terms in the Drake equation together. In the language of math: