ATLAST: The Gargantuan Telescope Designed to Find Life on Other Planets

We finally know what kind of telescope we need to see other earth-like planets. And we're getting ready to build it.

An artist's rendering of ATLAST (courtesy of the Space Telescope Science Institute)

Twenty years ago, in the year Bill Clinton was elected president, scientists first confirmed the existence of a planet outside our solar system. Now, we know there are thousands of other planets just in our galaxy, even if we've only detected them indirectly. We also finally know what it's going to take to glimpse an exoplanet, to actually see the places that might harbor life like ourselves (or otherwise). And the telescope that will eventually do so is on the drawing board. It has a profound name: ATLAST.

During the last three years, we have learned that our galaxy is teeming with planets. Since its launch in 2009, NASA's Kepler Mission has discovered more than 2,200 planet candidates orbiting distant stars in the Milky Way. Every year that goes by brings new exoplanet data, and new reasons to think that planets are a commonplace phenomenon in our universe. And yet, pressing questions about these planets remain. We aren't yet sure how many of them are capable of supporting life. The early data from Kepler indicates that as many as one in ten stars has a planet around it that can host liquid water on its surface. If that number holds up, then our galaxy could be home to more than ten billion watery planets, each a potential home for microbes, plants, or even intelligent beings like us. Some may be so close that we could use telescopes to detect signs of life in their atmospheres. The possibility that undiscovered Earths are hiding in every corner of our galaxy is completely reorienting the future of space science. Astronomers sense that they are on the brink of an epochal discovery, and they are keen to build the telescopes that will enable it.

The Space Telescope Science Institute in Baltimore, Maryland is at the leading edge of this effort. The Institute runs science operations for the Hubble Space Telescope, the most far-seeing instrument ever deployed by humans. The Hubble has had quite a run over its twenty-two years of service, but it is beginning to show its age. In 2009, NASA astronauts serviced the iconic telescope in orbit for the fifth and final time, outfitting it with a new camera and fresh batteries. Still, it's unclear if the Hubble's sensitive instruments can weather another decade of exposure to cosmic rays. Like the Voyager space probes, the Hubble is drifting slowly toward retirement.

The Institute is currently preparing for the launch of Hubble 2.0 -- the James Webb Space Telescope -- a massive infrared instrument that will be one hundred times more powerful than its predecessor. Unlike the Hubble, the James Webb will be difficult, if not impossible, to service. The delicacy of the Webb's infrared sensors require that it be positioned one million miles from Earth, which is too far for tune-ups. Without the benefit of regular maintenance, it is only expected to last five to ten years.

Because these machines take so long to build, the Space Telescope Science Institute is already planning for Hubble 3.0. A small working group at the Institute is starting to sketch the conceptual outlines of Webb's successor, a still larger space observatory called the Advanced Technology Large-Aperture Space Telescope (ATLAST). This telescope is being designed with a very special purpose in mind: to discover life on planets that orbit other stars.

Last week, I visited the Space Telescope Science Institute to meet with Matt Mountain, who has served as the Institute's Director for the last seven years. In an extended and wide-ranging conversation, Mountain told me about his vision for the future of astronomy, a vision built around ATLAST and the search for life elsewhere in our galaxy. "The discovery of life on another planet will be as important to the 21st century as Neil Armstrong stepping onto the Moon was to the 20th," Mountain said. "It will be bigger than Copernicus and Darwin rolled into one."

You've been the Director here at the Space Telescope Science Institute for 7 years now. How has astronomy changed in that short time?

Mountain: There are two really important dynamics that are changing the field and the community is still sort of struggling with them. The particle physicists struggled with these issues in the 70's and 80's.

First, to do cutting edge astrophysics it takes larger, more complex facilities than it once did. It's a matter of simple physics. The power of a telescope, its ability to detect a very faint signal against a noisy background, is directly proportional to the telescope diameter divided by the size of the object you're looking at. It's a very simple ratio. So, if you want to look for planets around other stars or very distant galaxies, those objects are going to be extremely small.

Our detectors today are almost perfect, so it's hard to gain anything by building better detectors. The only way we can get more information about planets around other stars, or distant galaxies, is to make larger telescopes. That's why we have to build these big observatories in Hawaii, and it's why we have to build the James Webb Space Telescope. It isn't because we want to spend billions of dollars, it's because we've been doing space science for four hundred years, and the low-hanging fruits have been picked.

To answer some of the more profound questions -- Is there life around another star? How did the first galaxies form? -- requires us to look at some very faint things, and we need large, complex facilities to do that.

That unfortunately moves you away from a traditional academic model of the solitary scientist writing a solitary paper to one where you need a complex machine and a complex organization like this one. And so the other thing you're seeing is a move towards teams; increasingly, it's large teams that are doing the really high impact research and that's because you need a multidisciplinary skill-set to do this stuff. This institution is an expression of that, and in some ways was slightly ahead of its time. We have scientists here, yes, but we also have engineers and software people---we've created a layer of interdisciplinary skills, and that layer allows astronomers to interface with very complicated machines like the Hubble Space Telescope in a very straightforward way. We've hidden the complexity.

It's a totally different paradigm, and it can be tough for some astronomers to wrap their heads around it, because they're wedded to the ideal of the lone astronomer going up to the mountaintop with his lab book and his worshipful post docs following behind. That's a model that has huge romance and pull, but it's actually not very effective anymore.

What are some of the most notable successes of the team model?

Mountain: Well take Adam Reiss and his team, who, together with two other teams, won the Nobel Prize last year for discovering dark energy. An individual couldn't have made this discovery. To do what they did, you needed people who understood the theory of supernova explosions, you needed people to figure out how to run these complicated telescopes, both on the ground and in space, and you needed people worrying about data and sophisticated statistics. And this is all very complicated stuff; the person who's an expert in Bayesian statistics and sampling methodologies isn't quite the same person who's an expert in getting the maximum signal from a really faint supernova. But in the end, there's a pay off: Reiss and his team spent most of their Nobel money getting the whole crew to the Nobel ceremony.

Now let's think about where the team model might take us next. Think about the big question right now: Are we alone? What would it take to answer that question? We've got the Kepler Space Telescope telling us that there are probably planets around every star, so we know that. But now we have another problem: these planets are really, really faint. So faint, in fact, that you need a big telescope to see them, and it has to be quite sophisticated because the planets are next to a very bright star. This is right at the limits of optical technology, which means you need experts in optics and telescopes. So let's say you get a spectrum of the planet's atmosphere, which will allow you to see its chemical makeup. Even then you're still not in the clear, because you've got to understand atmospheric circulation and ecosystems, not to mention how planets form.

Suddenly you realize that to understand whether there's life around another star, you need a huge, multidisciplinary team. An astronomer might be able to get a spectrum, but they wouldn't know what to do with it, because they weren't built to interpret it. That's why you see someone like Sara Seager -- who is very interested in the question of whether we're alone -- go to M.I.T. where she can do planetary science, but also astrophysics and remote sensing.

If you want to work at the frontier, with the very best technology possible, you need a huge team. If you want to be a solo theoretician and scribble in your notebook, then maybe you can still make breakthroughs, I don't know. But I do know that these big questions are going to take multidisciplinary teams and that's going to take a culture shift. People in this profession are going to ask themselves some tough questions. Like how do you give tenure to an astronomer who worked in a team of two hundred people? How do you measure their individual contribution? Who do you give the Nobel Prize to?

The Hubble Space Telescope has now been in operation for over 22 years, during which it has made more than a million astronomical observations. When is it due to be retired?

Mountain: The truth is we don't know. We have a probabilistic assessment of how long the instruments might last. We think the gyroscopes will last until at least 2020. We know the batteries will last at least that long, because the last ones lasted through more than a hundred thousand charge cycles. We know the solar cells work. The real question is how long the instruments will hold up. In principle, we think the instruments will last until at least 2016 or 2018, but it's a crapshoot; it depends upon cosmic rays and how well they built the electronics.

I want to talk about ATLAST today, but there is one question I want to ask you about the James Webb Space Telescope, the successor to the Hubble, which is due to launch in 2018. When I was first researching the Webb, it really stressed me out to think about how complex its deployment is going to be. In a strange way, the successful landing of Curiosity alleviated a lot of that stress. And maybe these things aren't comparable, but I thought I'd ask you: do you think the Webb deployment will be tougher to pull off than Curiosity's landing?

Mountain: We actually quantified this. We looked at the number of mechanisms in Curiosity's 7 minutes of terror video, and the number of mechanisms involved in deploying the Webb, and unfortunately the number for the Webb was a bit higher. I mean for one it's going to take a lot more time. It's going to take months, and so the experience of it is going to be slow, like water torture. You can't just shut your eyes for seven minutes and open them when you hear the first beep like you did with Curiosity. But Curiosity was encouraging; it showed we could do these very complicated things, and that's good, because if the Webb doesn't work we're screwed. There's no fallback option.

Tell me about ATLAST, the proposed successor to the James Webb Space Telescope. I know that at this point ATLAST is just a concept, but in some ways I sort of see it as a wish list for the next flagship telescope, and I'm curious what that wish list looks like?

Mountain: Let me give you a completely different way of thinking about it. The question is what is the future of space science? What questions do we want to answer? I'm a pragmatist when it comes to science, and I think the big question that everyone wants answered, and that we can answer, is whether or not we're alone. And we already know what kind of telescope we need to look for life around another star. This is not a difficult problem; it's a Physics 101 problem. We know where all the closest stars are and we know precisely the right distance a habitable planet will be from its star, and we know how bright a planet is. So now assume you can look at a star, but suppress its light, in order to get a spectrum from a planet that's orbiting it. That spectrum will tell us something about that planet's atmosphere; it might even tell us if there's life there. So how big a telescope do I need to do this?

If I assume that every single star has a planet around and that it's in exactly the right place, then I won't need to look at very many to have a good chance of finding life. With a 4-meter telescope, you can look at 10 systems, the 10 closest systems, in this way (the Hubble has a 2.4-meter mirror and the James Webb Space Telescope will have a 6-meter mirror). If I have an 8-meter mirror, I can observe hundreds of star systems in this way, and if I have a 16-meter mirror I can observe thousands. Those may sound like pretty big numbers, but remember in this scenario I'm assuming that there's an Earth in just the right place around every one of these stars. But we're not sure how many stars have a planet in just the right place, and we obviously don't know how many of those planets have an atmosphere with life in it. Kepler is giving us a handle on the first unknown, and it's looking like the answer is 0.1. It's looking like one in ten stars might have a planet in the habitable zone.

So now let's say I build my 4-meter telescope. That 1 in 10 chance that only gives me one or two habitable planets to look at, which isn't a very big sample size. Certainly not big enough to tell is whether we're alone or not. And so the question becomes how big do you need your sample size to be? From our perspective, the answer is about a thousand. If there's no life in the closest thousand stars, there's a good probability that we're pretty much alone. And that means I need a 16-meter telescope:

What sort of leap in the technology do we need to get there? 

Mountain: Well, at first glance, a 16-meter telescope sounds absolutely impossible. But let's think through this. Let's put space aside for a moment and ask ourselves how we got our big ground telescopes. What happened is these astronomers had telescopes with 4-meter mirrors and they realized they weren't big enough to see the faint objects they wanted to look at, objects like distant galaxies. So they tried to build bigger ones, but the problem was that simply scaling up the 4-meter technology didn't work, at least not without spending massive amounts of money. So people like Roger Angel and Jerry Nelson and Ray Wilson came up with brand new technologies; they used active systems and adaptive optics, which eventually earned them the Kavli Prize. These advances allowed them to produce a whole host of new telescopes, facilities like the Gemini Observatory, the Keck Observatory, and the Subaru Telescope.

So I called up my friends at Lockheed Martin and I said "let's figure out what technologies we need to make this work in space" and we both came to the same conclusion. We need the same technologies we've got on the ground: adaptive optics, which make mirrors very lightweight and floppy. We need adaptive optics in space. And I get silence on the other end of the phone. And then they say, "actually, we just bought two adaptive optics companies." And I say to myself, "of course these guys are interested in adaptive optics." Because if you want your spy satellites to have the same resolution as the Hubble, the resolution that allows you see the guy getting out of the car you've been following, then you need 16-meter telescopes out there. That's why you see Lockheed going after technologies like adaptive optics. That tells me that we need to build our partnership with industry.

That's how you get a new 16-meter telescope, a telescope like ATLAST. And that's a very different paradigm from just trying to do it alone. We don't want to become like the particle physicists, who have slowed way, way down, because the technologies they need are technologies that only they need. With these massive accelerators, they have to invent every new technology themselves. Our project of looking for life on other planets uses the same technologies (lightweight mirrors, good detectors, big rockets) that other people want. And to me that project is essential, because the discovery of life on other planets will be as important to the 21st century as Neil Armstrong stepping onto the Moon was to the 20th.

More important, I would think.

Mountain: Well, right, but in the framework of people who love the Space Age, the people who wish there was another Kennedy saying "we do this because it's hard, not because it's easy." This would be as radical as Copernicus and Darwin rolled into one. And you'll have kids asking how we can get there. You'd likely see a huge leap in rocket technology. I mean all we've been doing so far is building a better and better steam engine; SpaceX is just the James Watt of the steam engine. We need something beyond steam engines to get out to the stars, and finding life around another planet may be just the thing. It may motivate kids to ask "how do I get to somewhere that's ten parsecs away, because there's a living thing out there and it would be damn cool if we could get to it."

It sounds like the astrobiology is more of a driving force for you than cosmology. Is that right? Is looking for life more important than looking back further and further towards the Big Bang.

Mountain: Well, yes, it is definitely the preeminent goal. But then again, we have Adam Reiss on our staff. I mean it's my view that you have to be able to walk and chew gum at the same time. Cosmology has gotten to a really interesting place, but it's not clear how you make the next breakthroughs in cosmology. That's the problem. Maybe through gravitational waves, that may be the way to go. But, I do know how to look for life around other stars, and I'm a bit like the drunk -- I want to look for my keys underneath the lamppost. Because of Kepler, we know there are other Earthlike planets out there, we know we can get spectra from these places, and we think this is a pretty important question. Not that cosmology isn't full of important questions. But this looking for life is something I can explain and describe in very concrete terms. There's no mystery here.

So what does the timeline look like for ATLAST? If Webb goes up in 2018, how soon will ATLAST follow?

Mountain: Well we want to put it up before 2032, because if NASA is going to send humans to Mars it will likely do so between 2032 to 2035, when Mars is close to the Earth. It won't be that close again for another twenty-five years. We figure that if NASA decides to do a Mars mission, it's going to be spending most of its money there, because sending humans to Mars will be the most expensive thing we've ever done. There might not be much money left for anything else. So if we're going to get ATLAST up in space, we figure we'd better do it before 2032. But if it were up to me, we'd do it in 2025.