>> From the Library of Congress, in Washington, DC. ^M00:00:20 >> Stephanie Marcus: Good morning. I'm Stephanie Marcus from the Science Technology and Business division here at the Library. I'd like to welcome you to today's program, which is the transiting exoplanet survey satellite, lovingly known as TESS. The program is the fourth in our 2015 series of lectures, presented through a partnership with our division and NASA Goddard. And this is actually the ninth year that we've had this wonderful relationship, and we've been able to bring these very stimulating talks to the Library. And we are definitely looking forward to our tenth season and starting to gather topics and get the speakers for next year. This year we still have a couple of programs. One in November and one in December. And in November, we're going to hear about asteroids, meteorites, comets, yeah, popular topic. And then in December, we get to hear about the Pluto flyby that everybody's been talking about. So please check our website and come to those talks. Today, our speaker is Dr. Stephen Rinehart, who's is an Astrophysicist and the Associate Chief of the Laboratory for Observational Cosmology at NASA Goddard. Dr. Rinehart knew as a little kid that he wanted to be a scientist, and he also decided as a child that he was going to go to MIT, which he did, and that's where he earned his BS in physics. And then he went to Cornell for his MS and PhD in physics. He was a postdoc research associate in London, and we won't talk about that. He seems to think that it was very difficult to live in London as a graduate student. And then he was able to get a National Research Council postdoc fellowship at NASA Goddard, and he's been there ever since. He became a full-time employee there in 2004 as an Astrophysicist. And his wife is also at NASA Goddard. So they have one child who's named Joshe, which means star in Japanese. And he said, of course, what else would a couple of astrophysicists name their child. And his mother-in-law is here, she's right there. All I have to say beyond that is exoplanets, the Goldilocks zone, alien life, these are things that humans have been interested in for a long, long time, even going back to the ancient Greeks who wondered if there were other worlds like ours. So I think that Dr. Rinehart can let us in on some of those secrets that TESS might tell us. Please welcome Dr. Rinehart to the Library of Congress. ^M00:03:25 [ Applause ] ^M00:03:31 >> Stephen Rinehart: Thank you. So that's a great introduction. My daughter Joshe is also at Goddard, they have the child development center there [laughter]. She also occasionally comes by some great pictures of her working in my lab. But don't tell labor people. I'm very happy to be here today to talk about TESS, the transiting exoplanet survey satellite. Before I really talk about it, I'm going to give credit where credit is due. It's being built in partnership by MIT, NASA Goddard, Orbital ATK, and a number of other partners. The principle investigator is Dr. George Ricker at MIT. Now, this is going to be the next NASA explorer mission, and it's really focused on finding new planets around other stars, new exoplanets, particularly those around nearby bright stars. So let me really start out with a little bit of context, and, in fact, history. Humans have known that there were planets besides the Earth for a very, very long time. In fact, the ancient Babylonians, over 4000 years ago, around the turn of the second millennium, recorded observations of Mercury, Venus, Mars, Jupiter, and Saturn. So as long as 4000 years ago, we've known that there were other planets in our solar system. We only missed 2, Uranus and Neptune. That's pretty amazing. So we've known for a very, very long time, longer than the Greeks, that there were other planets out there. Nobody knows this, this is like an obscure fact, by the way. If we skip forward in time from 4000 years ago to about 400 years ago, there was a guy that some of you may have heard of, Giordano Bruno. Bruno was a famous astronomer of his era. A very, very well-known, very well-respected astronomer. And in 1584, he laid out a postulate, and his postulate was, that if you look at the sky and see all these stars, all of those stars are suns like our own. Around each one of those suns there could be planets, and some of those planets could be like Earth. And on those planets, there could be life forms, something like us. And, fourth, and most importantly, from his perspective, those life forms could have souls. Now, that last bit is what got him burned at the stake, and I'm going to not talk about that all because that's theology and philosophy, not science. But we know that the other 3 parts, we know that the first one is essentially correct. While stars are, in fact, very, very diverse, our sun is a star like many others. We also know that the second part, that there are planets around all those stars, that's pretty close to true too. There are lots and lots of planets around a lot of stars. As for the third, that maybe there's life on some of those planets, we have a little way to go. And hopefully, within my lifetime, or my daughter's lifetime, we'll actually be able to talk about the possibility of life on other planets. And at the end of this talk, I'll say a little bit more about how we might get there. So let's just jump forward again in time, to now to about 40 years ago. I don't know if any of you are science-fiction fans, I am. But if you looked at science-fiction works that were made in this century, whether it be in TV, movies, or literature, the idea that there were planets around stars, in fact, planets around a lot of stars, is ubiquitous. It just assumes to be true. Now, at the time of all of these works, there was no evidence, direct or indirect, of a single exoplanet, a single planet around any other star. We had not yet found any evidence whatsoever. And yet the human imagination said, they simply have to be there. Now, all this changed in around 1989, when someone made the first measurement that showed it was possible for us to find exoplanets. And this kicked off the great exoplanet search. In fact, this week, ironically, because we didn't really think about this when we scheduled this talk, this week marks the twentieth anniversary of the first discovery of an exoplanet around a main sequence star. Since we realized we could do it, people started looking, and things have taken off. In fact, you'll notice in one second, I have changed the scale bar, in 2014, because there were over 900 planets confirmed last year alone. So to date, we have, and I have to look at my slide because the number keeps changing, as of Monday, we had 1892 confirmed exoplanets, planets around other stars that we've actually been able to observe and measure, mostly indirectly, but we know that they are there. Now, you'll notice on this slide I've got all these different colors. That's to indicate that there are a lot of different methods that people have used to find exoplanets. And they all have advantages and disadvantages. But the 2 that I want to talk about most today are the 2 that have been most successful. And also because they have the most relevance for TESS. The first of these is the transit method itself, and I'll describe that method in more detail in a minute. But all of these, all the objects that are represented by the red part of the bar, are transiting exoplanets. And over half of those planets are transiting planets. And in fact, over 1000 of the confirmed planets were discovered with one mission, the Kepler Mission, which has looked for transiting exoplanets. I'll say more about Kepler in a bit too. The other topic, the other method I want to say something about is the radial velocity method. Those of the objects in blue. That's the oldest and most venerable method for finding exoplanets. And it has a very important role in how we follow up on observations from TESS. And I'll say more about that in a minute. So what are all these objects? I'm a scientist, I like plotting things, so bear with me. If you look at the vertical axis here is the mass of a planet, in Jupiter masses. The horizontal axis is the duration of that planet's year, the orbital period. And what you can see is that if we plot all of these planets that have been confirmed so far, they're really pretty much all over the place. They're largely up in this corner. There's not a lot down here. But there are even a handful down here. Although those yellow ones are awfully hard to spot. To give you a little bit of context, let me add in our solar system planets. Now, there should be a couple of things that strike you right away about this. The first is that there are only a handful of exoplanets down in this regime where our own planets lie, in this plot. Now, there's 2 lessons that I take away from that. The first is that finding planets like the ones in our solar system is really, really hard. In fact, it is easier to find planets the bigger they are, pretty much no matter what method you are using. Which means we're bias towards finding things up in here. It is also, for the most successful methods, easier to find planets the closer they are to their host star. So things in here. So it's no surprise, the first planets we found were all up in this range. And as we've gotten better, as we've gotten better at the techniques and we've explored new techniques, we've pushed that boundary down, until now we're starting to just now see some of the planets that are like those in our own solar system. The second thing I want to take away from this is, to me, actually more interesting. And that's that while there aren't a lot of planets like those in our solar system, but gosh there are a lot of planets here. So while our solar system planets are down here, we don't have anything like these objects. These are completely new. These are things that we never saw until we started looking for exoplanets. And for theorists, this has been a big problem. Now, most of us, probably not the guy who wrote this particular theory, but most of us like new observations like this because they overturn theory. They make us say that the theory is no good, we need a new one. And that's always pretty exciting. We've found a lot of things in this data that we never expected to find. Let me give you a couple of quick examples. There's the hot Jupiters, also known as roasters. These are Jupiter -sized planets, but they're in orbits of less than about 10 days. They're closer to their host star than Mercury is to our sun. They're boiling hot. In some cases, they're so hot that you can see evidence that there are comet-like tails, as the atmosphere of the planet is being boiled off and blown away by its host star. Again, if you talked to a theorist 25 years ago, and you told him this object existed, they would tell you you were nuts. It makes no sense, it shouldn't be there. And yet now we know that they are. And theory has evolved to explain why we actually see them. Another class of objects we didn't expect were what we call the super-Earths. Again, if I went back in time, and I talked to theorists, what the one thing they would've told me is that, the way that planets form is that it's easy to form a rock about Earth size. Once you get beyond Earth size, if you have enough material, you go into sort of a runaway accretion mode. You just grab everything that's nearby. So you race straight from Earth sized up to Uranus sized with nothing in between. So if I were to look at a thousand exoplanets, I'll see a bunch down here, and I'll see a bunch up here, and I won't see anything in that box at all. Obviously the theorists were right -- I mean wrong. Because that box is very well populated with lots and lots of planets. And these are the super-Earths. Now, again, we have no examples of these planets in our solar system, so we don't actually know what they're like. Now, on one end of the large end, things like a 2 Earth radius planet, those are probably more like Uranus than Earth. They're probably big icy balls. But on the low end, something that's say 10% or 20% larger than Earth, well, that's probably a rocky planet. It might be Earthlike. It might even be habitable. So that leads me to the topic of habitability. The habitable zone and the goldilocks zone. Before I talk about that though, let me give you a couple of statistics. We have found, like I said, almost 1900 exoplanets. And from those studies, we've been able to draw some very serious and very important inferences about the population of exoplanets. So, for instance, we know that in our galaxy there is approximately 1 planet per star. That's not to say all stars have planets. Some stars like our own have many. Some have none. But on average, there are as many planets in this galaxy as there are stars. That's amazing at just the sheer number of planets that are out there. We also know that about 1/6, 17% of all stars have Earth sized planets around them. Again, this is, you know, shocking, 1/6 of a very, very large number is still a very large number. That's a lot of Earth-sized planets. That's a lot of potential. And perhaps most interestingly, 22% of stars that are like our sun have planets in their habitable zones. These are potential places where we might find life. Now, what do I mean by habitable zone. This also known as the goldilocks zone, for reasons that will become apparent very quickly. It is the range of distances from a star where you can have liquid water on the surface of the planet. If you get too close to the star, then the water gets too hot, it boils off, it then evaporates, and you're left with a dry dusty rock. No water. Not really a great place for life. If it gets too far away from the star, then all that water precipitates out, it freezes, it turns into ice, and now you have a rocky ball that's covered with a thick layer of ice on top of it. Again, not really a great place for life. But if you're not too hot, and you're not too cold, if you're just right, then you can have all the conditions -- one of the key conditions for formation of life. Now, you'll note, of course, that where that habitable zone is is different for different kinds of stars. On this plot, these are 4 different kinds of stars. And you can see that for a small cool star, the habitable zone is closer than it is for a star like our own, which, it's not hot, but it's hotter, where it gets much farther out. And you can see, these are some of the planets that we found that are in the habitable zones of their host stars. These kinds of objects are going to be really, really exciting things to study over the next few decades. In addition to all of that, there's a couple of really wacky things I always like to mention because I think they're weird. We have found planets around binary stars. Think tattooing, for you science-fiction fans. We have found planets in highly eccentric orbits, orbits that go really, really close to their host star but then come out past the orbit of Earth. No particular reason to believe that such an object should ever exist, in yet it does. We've even found planets around a wide range of host stars, including dead stars. In fact, in 1992, we found the first small exoplanets around neutron stars. These are the remnants, the husks, left over after supernova of normal stars. And it's hard to believe, but the planets, either formed later or survived, a supernova. Again, not something we would have expected. So all this now sort of leads me to the topic of TESS itself. And when I talk about TESS. ^M00:17:08 [ Inaudible ] ^M00:18:30 Shows you what we would get in real time. See that as the planet goes across, the brightest the star gets. These 3 plots on the right are actual transit data as measured by the Kepler mission. And these are detections of actual planets. Obviously, this bottom one here, that's a big planet. It's pretty easy to see that in that data, by eye, you can observe immediately that there's a transit there. This top one, on the other hand, that is -- well, I refer to that as a statistical tour de force. That was just an amazing bit of analysis by Tom Barclay and his collaborators to pull that out of the data. That planet was, and I believe it still is, the smallest exoplanet ever found. It's smaller than Mercury. And it really just goes to show that this technique can be very, very powerful in finding even the smallest of planets. It's not easy to do, because what you're trying to do, remember, is you're trying to measure a difference. And looking at a star, a star is already pretty faint. Now you're looking at a, in the case of planets, less than a percent difference in the brightness of the star. You're trying to measure a small change in something that's already pretty faint. So I've mentioned Kepler a couple of times now, so let me say a few words about what Kepler is. Kepler was a mission, or is a mission, it's now the K2 mission, looking for exoplanets by looking for transits, much like TESS. So why do we need TESS, what's the difference between the two? Well, the difference at its core is largely a philosophical one. Kepler was about finding lots of exoplanets. It was about statistics. Ultimately, the goal was to measure a quantity that we call eta-Earth, which is the frequency of Earth surround stars, how common are Earth-sized planets around all of the stars that we see. It was not designed to get detailed characteristics of all these exoplanets. It was just there to do a census. It was the mailer that we get every 10 years, that's all. TESS, on the other hand, is looking around the neighborhood to see what things are actually like. It's looking only within a well, very nearby distance, to look for bright nearby stars that have transit exoplanets. It's about finding the planets that are really well-suited for follow-up observations, for finding those objects that we're going to want to be studying for the next 10, 20, maybe even 30 years. Now, this philosophy is actually very clear, if you look at the way the different surveys are done. For Kepler, it looked in a relatively small portion of the sky. But it looked out a long, long way, over about 3000 light-years. TESS, by contrast, only looks about a few hundred light-years away, but it covers almost the entire sky, more than 400 times more area of sky thank Kepler covered. So it really is focused on trying to find those exoplanets that are around bright nearby stars. Now, you're going to hear me say bright nearby stars a lot, because it's really important. How does TESS do this? Let's go to the second part of the name. The transiting exoplanet survey satellite, the survey part of TESS. Tess has 4 cameras, each one of which has a 24 by 24 degree field of view. So at any one time, it's looking at a 24 by 96 degree patch of sky. This is a really big field of view. And in that field of view, it's looking at a whole bunch of individual targets with small sets of data, to try to find those transiting events. Each one of those observing sectors is observed for 27 days. And over the course of the year, we tile 13 of those observing sectors together to cover almost the entirety of one hemisphere. Then we flip around and we repeat that, and do the same thing for the other hemisphere. So at the end of 2 years, we've covered the entire sky with a coverage map that looks something like this. So you can see that most of the sky has been observed for at least 27 days. And where those observing sectors overlap, or those tiles intersect, you get much longer observing times. And in fact, up here, at the polls, you get almost 1 full year of time coverage. That's really important because it's at the polls that that longer time coverage that you're able to find planets with longer periods, and you're actually more sensitive to small planets. So this leads me to this question of, what will TESS actually find? Well, this is just a simulation, it's only a model. But all those red dots are planets that we expect to find with TESS in the core data. The blue dots, which are impossible to see for you, are other planets that we expect to find in ancillary data. But the key ones, the core ones, the ones that are most important to me, to us, are those red dots. Now, you'll see, they're all over the sky. Now, they are concentrated at the polls, as you would expect, because that's where our long baseline time coverage is. But there are a lot more around the entire sky. Let me convert those dots into numbers, because I'm a numbers guy. You see that with TESS we're expecting to find over 500 new small planets. That's just in the core data. That excludes the extra ancillary data that we're going to getting. And this is going to be a huge increase. We have only a few hundred small planets that we know of right now. So we're going to more than double the size of that set of targets. But that's not the best part. The best part is the fact that they're going to be close by. So let's take a look, if I look at how far away these planets are, and I compare planets that are known in blue, the black dots are Kepler detected planets, and the red dots are TESS expected TESS planets, you'll see so far we've got all red dots, a few blue. There's the first black dot, over 100 light-years away. Here's the Kepler targets. They go out a heck of a long way. In fact, they go out beyond the simulation. But the TESS targets are almost all within a few hundred light-years. So they really are the bright nearby stars. Why is this so important? Well, one of the things that we've learned from all the exoplanets that have been done so far, and this is a really exciting field didn't exist 20 years ago, but now it is like perhaps the single most vibrant part of the astronomical community. One of the things we've learned is that there's a lot more planets out there than we expected, and that they're very different. They're not homogenous bodies. They're not Earths and Venuses and Mars and Jupiters. There's planets that are sort of like this and sort of like this and like this, but they're all different. What we'd really like to do is to start doing what I call comparative planetology, actually making measurements of other planets that are out there, exoplanets, and say, these 2 planets look similar in some ways but they're different in other ways, can we understand why. Can we start to understand how planets turn into the objects that they are. And to do that we need to actually do the characterization measurements to actually measure the properties of individual exoplanets. Now, this all starts with the TESS data itself. You learn 2 important things just from looking at the transit data that TESS acquires. The first thing you learn is how big the planet is. This may be kind of obvious to some of you, but just in case. The size of that dip in brightness is determined entirely by how big the planet is relative to the star. So if you have Jupiter transiting in front of our sun, Jupiter will block about 1% of the light from the sun. If, by contrast, I have Earth orbiting the sun, Earth is this little dot right there, it only blocks about 1/100 of 1% of light from the sun. So, you know, that dip in brightness is just the ratio of the area, projected area, of the planet to the area of the star. Pretty easy, straightforward. You can also note that you can get the orbital period of a planet, you can tell how long a year is. If I'm looking at the Earth transiting the sun, then that transit event will occur exactly every 365 and 1/4 days. And I would go low and behold, that planet has a year that's 365 and 1/4 days long. Furthermore, if I look carefully at how long this transit lasts and that the shape of the entry and exit from that transit event, I can start talking about things like the eccentricity of the orbit, the inclination angle of the orbit, and a number of other parameters as well. I can learn a lot about the orbital parameters of a planet just from the transit data. Unfortunately, this is really not enough. If I were to take a bunch of planets, as I have done, and plot their radius on one axes, so this is their radius relative to Earth, and plot their mass on the other axes, that's the mass relative to Earth, they appear all sort of in this band on that plot. Now, you'll see the green above dots are super-Earth planets, blue dots are other exoplanets, particularly giants up here. And these pink squares are planets in our own solar system. These 4 lines, hydrogen, water, rock, and iron, are what we call phase curves. They are the how big a planet would be if it had this mass and was made out of this material. This is kind of obvious, right. If I have a bowling ball and a soccer ball, they have the same radius, but the bowling ball weighs a lot more. If I have a shock put and a bowling ball, they weigh about the same, but the shock put's a lot smaller in radius. The same thing is true for planets. If I just measure the radius of a planet, and let's say measure something here at 2 Earth radii, well, all those of those things are about 2 Earth radii and they vary from water-like planets to rocky planets to planets that are heavier and may even be iron-like. Vice versa, if I just had a mass measurement, then, well, I've got something here that looks heavier than iron, that's weird. I've got several things that are rocky, and one thing that's watery, and then some things that are kind of lying above that curve. So if I have just radius, it's not enough. If I had just mass, it's not enough. Both of those things give me some idea of what that planet might be like, but they don't really tell me nearly enough to start understanding what it really is like. I need both. If I have both, I have a density and I can tell you what the gross composition of that planet is. I can tell you whether it's a lump of rock, a big bowl of ice and water, a gas giant, or maybe something bizarre like an enormous chunk of iron in the middle of nowhere. ^M00:29:50 So how do I get that second measurement? With the transit data, I've gotten the radius. Now I want to get the mass. So this brings me to the radial velocity technique. The radial velocity technique is actually the oldest technique having been used for finding exoplanets. And it works like this. If I look at the light from a star and I break that into the spectrum, into very, very small chunks of light, then what I see are a whole bunch of very narrow absorption bands, absorption lines, even. If I watch those lines carefully, then I can learn a lot about the star. If that star has a planet orbiting it, then the planet doesn't just orbit the star, the star orbits the planet as well. In fact, they both orbit the center of mass of the system. So for every orbit this planet makes, the star is doing a tiny little pirouette. So if you're watching that spectrum, then what you will see is that as the star comes towards you in that pirouette, the lines all shift a little bit bluer, they get a little bit bluer than they would be otherwise, from the Doppler effect. As the star goes away from you, they get a little bit redder from the Doppler effect. So over time, I see these lines doing this, moving back and forth, and back and forth. And if I measure the frequency at which they go back and forth, I now know the orbital period, I know how long the year is for that planet. If I also measure the amplitude, how far those things move, I can tell you how massive that planet is. So if I have that measurement, in combination with the transit measurement, I can tell you what that planet is made out of, roughly speaking. So this means that follow-up observations of the TESS targets are going to be really, really critical. And in fact, part of the TESS program is to have people with ground-based telescopes following up on these targets, to try to make the measurements to get the masses. This is not an easy measurement to make. And well, in fact, it's extremely challenging. So let me give you an example. Kepler, with all those planets that go out very, very far away, while it has been responsible for the detection of over 1000 exoplanets, 1000 confirmed exoplanets, there almost 4700 additional exoplanet candidates that Kepler has found that have not yet been confirmed. And for many of those, they haven't been confirmed because those stars are down here. They are simply too faint for us to make that radial velocity measurement in order to measure the mass. So what would you like? You'd like planets around brighter stars, maybe bright nearby stars, in fact. And you see, that with TESS, of course, the stars are about, on average, 100 times brighter than those for Kepler. So for Kepler to get a mass of some of these planets down in here, I need weeks or months or many months of telescope time in order to make that measurement. With Kepler -- I'm sorry, with TESS, it will take hours. And in fact, some of these stars are visible to the naked eye. An amateur astronomer, somebody with 4 inch telescope, could go in their backyard and if they have a decent camera behind it, could actually measure that transit. Maybe not from DC, it's too hazy and polluted. But they go out to West Virginia somewhere and make that measurement. I think that's amazing. So a big key difference between Kepler and TESS is that with Kepler, you find a new exoplanet candidate, and you go, can we make that follow-up measurement? Is it even possible? With TESS, it's going to be, we've got all these exoplanet candidates, which one do we do next? It's a problem of prioritization, not one of possibility. So as I mentioned before, TESS actually has partners lined up to make a bunch of these measurements. The Las Cumbres Observatory -- actually it's the Las Cumbres Global Observatory Telescope, I believe. Pardon, if any of you speak French, I'm about to do something horrible. The Observatory Haute-Provence. My friend's pronunciation, I keep being told is awful. And the HARPS Instrument, which is the high accuracy radial velocity planetary search project. So these are all partners who are going to be helping us in doing that follow-up work. And there's a lot of other partners here, there's just not enough room to put them on a slide. Unfortunately, even once we have that measurement, we still don't really know that much about the planets. Now we know roughly what -- we can tell it's a rock, that's about it. Venus and Earth are both rocks. They have about the same mass. They have about the same radius. They're both in the habitable zone of the sun. And yet, it's a gorgeous day outside here in Washington, DC, on Earth. There is no place on Venus where we can stand on the surface and be alive for more than about a millisecond. I like it here, I'm not going to go there. We were talking outside about would I go to Mars, the answer is maybe. I wouldn't go to Venus. But if I just had the mass and radius, they look the same. I'd like to be able to tell the difference. I'd like to know which planets out there are Earthlike, which ones are Venous like, which ones are Mars like. How do I do that? Well, you need yet another measurement. What you really want to see is the spectrum of the atmosphere of the planet. Now, for Earth, this is this lime and green here. You see all these big deep features. Those are almost all due to water. Some of them are due to carbon dioxide, but they're almost mostly due to water. And that's a very good sign. We like water on planets. Mars, by contrast, here in blue, really doesn't have many features at all because it's atmosphere is thin and tenuous and doesn't have much in it. Venus has a lot of features, but they are not water features. They are things like carbon dioxide and sulfur-based chemicals. It's a very different atmosphere, so its spectrum is very, very different. If I had a spectrum of an exoplanet like this, I could tell you immediately if it was one of -- like one of these 3, or if it was something completely different, which I really think is likely. So how do you make that measurement? Now we're getting really into the weeds, now things get really hard to do. We do that with a technique we call transfer spectroscopy. Now, so imagine for a moment that you've got a planet. And let's imagine that this planet has an atmosphere, a big thick atmosphere, and that atmosphere absorbs all green light. All red light goes right through it. All blue light goes right through it. But all green light gets completely absorbed by the atmosphere. Then if I look at that planet, and I watch the transit occur in red light, what I see is that just the rock. I just see the rock passing from the star. I see the transit. Everything is great. If I look in blue light, I see the same thing. But if I look in green light, suddenly the atmosphere is absorbing light from the star too, so now the planet doesn't look like it's the rock, it looks like it's a bigger rock, because the atmosphere is also blocking out light from the star. So when I observe that transit, I see that the transit starts a little earlier, ends a little later, and it's a little bit deeper. If I do this at a lot of different wavelengths, then what I'm measuring is the apparent size of the planet as a function of wavelength. That tells me what's in the atmosphere. That is a spectrum of the atmosphere and tells me what its composition is. Now, this is a very, very, very challenging measurement to make. Remember, I talked about it before, a transit measurement itself is hard, because you're measuring the difference. You're measuring this change in brightness of something that's already pretty faint. To do transit spectroscopy, you're trying to measure the change in the change of something that's already pretty faint. So this is an extremely challenging measurement to make. And if you want to be able to make it, you need, let me guess, bright nearby stars. In fact, you want really bright stars. You want stars that you could see through a pair of binoculars, if not brighter than that. These are all the exoplanets that are currently known, as of last night, around very bright stars, and which transit, and there are 30 of them. Most of them, of course, are up here, they're the giant planets. In fact, they're the roasters. But there are a few down in here. Unfortunately, those are all -- there, A, only a handful of them. And, B, when I say that they are bright, they're right at the cutoff of what I consider bright enough. They're not really bright. TESS changes that. With TESS, suddenly we have hundreds of these planets that are bright enough for us to make that transit spectroscopy measurement. Hundreds of planets where we will be able to go and actually tell you what the atmosphere of that plan is made out of. And in fact, with future facilities, like the James Webb space telescope, which should launch in about 2018, and with future ground-based telescopes, they are people working on 20 and 30 meter ground-based telescopes, that eventually will actually get built. With facilities like those, we'll actually be able to measure the atmospheres of planets in this range, all of these we'll be able to do, but we'll also be able to do some of these small ones as well. We won't get down to the Earth-sized planets. That's going to take something even more powerful than what's in the drawing boards. But we will be able to do these super-Earth planets. And we may well find some actual signs of habitable worlds. And that's really, really exciting. Now, ultimately, we want to do better than even these transit spectroscopy. Ultimately, we would like to actually observe a planet directly and see a spectrum that looks like this. We would like to see a really high quality spectrum directly measure the planet. This is actually the spectrum of the Earth, as measured reflected off the moon. A very elegant little experiment. And you can see in this spectrum, there's a lot of stuff going on in our atmosphere. Our atmosphere is very complex. But there's a couple of key features that are, to me, as an astronomer, really interesting, and there are things, to me, that individually indicate that there might be a nice place for people. I see Rayleigh scattering, which tells me that there's a blue sky. I see very strong oxygen features. Well, where's the oxygen come from? It comes from a lot of places, but when they're this strong, there's something on that planet that's producing oxygen. In our case, it's plant life. Okay, it's not definitive, but there's a strong indication here that there might be plant life on this planet. Then I see water features. I see a lot of water features. That tells me that water is not only present but abundant on this planet. That there might even be big bodies of liquid water on the surface of this planet. Again, this is one of the things that we consider critical, necessary for the formation of life as we know it. Then we also see my favorite feature, the methane feature, out here. On Earth, methane is almost entirely produced by biological life. Whether it is bacteria on the ground or in jungles, breaking down organic matter, or flatulent livestock, almost all of the methane in our atmosphere is actually produced by biological life. So if we saw the combination of all of these features in the spectrum of an exoplanet, we would have, not a definitive, but a very, very, very strong case that that planet is habitable and maybe even habited. Again, we're not going to be able to make that measurement in the near future, that's something that will happen down the line. And maybe it will be a target that's been found by TESS. I think it is likely that the first time we see a spectrum like that, it will be something that TESS finds. But that spectrum will have to wait for the next generation of space telescopes. Which brings me to my final actual slide, which is to point out that TESS is part of a long line. Exoplanets is a new field, as I said before, it's really only about 20 years old. And it's really only been in the past 10 years that it's really taken off in a powerful way. And it's built on the legacy of a lot of other facilities and missions. It started with ground-based observatories. But then the Hubble space telescope has made great discoveries. Spitzer has been used to study exoplanets. And the Kepler mission that I talked about is responsible for more than half the exoplanets that we know about, that we've confirmed. TESS is the next logical step. Because we're no longer interested in just finding exoplanets, we're interested now in finding exoplanets that we can actually study. We're actually interested in learning what exoplanets are like. And this is a very big departure, this is a new direction, this is opening the door for comparative planetology. Beyond TESS, JWST is going to be a critical facility for following up these objects. We're going to have the target list that exoplanet scientists are going to be using for the next 20 years. And JWST is going to be the best way to study those objects over the next decade or maybe longer. Beyond that, the WFirst NASA mission, which is going to launch sometime in the next decade, I think it's 2023, but please don't quote me on that. That is going to also be used to study exoplanets. And beyond that, astronomers are looking at possible missions in the long-term, something like what's shown here is the New World telescope, even larger telescopes that are optimized solely for the purpose of getting a spectrum of an exoplanet, something that looks like that. And ultimately, maybe before I die, we will actually have not just a whole range of exoplanets, but we'll have a wide variety of exo-Earths. We will be able to talk about the fact that our Earth is a unique object, but there are other Earth's similar, similar enough that they could support life. And we can start understanding the differences between planets that are habitable, not just the differences between planets. Thank you. ^M00:44:07 [ Applause ] ^M00:44:13 >> That is exciting, and I hope we all live a really long time so we can learn more about this. We can have questions now. And would you please repeat the question so everyone can hear it. >> I can do that. ^M00:44:29 [ Inaudible Question ] ^M00:44:54 So let me see if I can pull this up. So the question was, the first exoplanets that were found were large planets, and they were close to their own sun; they were close in, these were the hot Jupiters. Have we gotten better or closer to finding things that are more like our own solar system? Well, the answer is, the short answer is, yes, we've gotten better. Originally we found things that were all up in this regime, but now you can see that there's a lot of planets that are in the same general area as Jupiter. There's a few that get close to Saturn. There's a few scattered in here with the rocky planets. So we are doing better, but we have a ways to go. One of the fundamental limitations is the fact that stars themselves are not well behaved. Stars are intrinsically noisy, if you've looked at sun spots or solar flares. All stars do that at some level, in fact, our star is, in the grand scheme of things, relatively quiet. So we try to make measurements of these planets around other stars, we have to take into account the fact that there's all this activity of the star that interferes with out measurement. But we are getting better. It is unclear how far we can go from the ground. It's unclear how far we can go with these techniques. But we're going to keep pushing that until we get as far as we can go. ^M00:46:14 [ Inaudible Questions ] ^M00:46:37 Let me see if I can repeat the question. So the question is, what about giant stars? Do we see smaller stars eclipsing or transiting other larger stars? Do we see -- okay, I think that's about it, right? The short answer is yes. In fact, we find a lot of what we call eclipsing binaries. And in many cases, those are 2 stars that are around the same size, that can be very different in size. But, in fact, one of the biggest sources of noise that we've got are eclipsing binaries. Because if you have a big star and a little star, it produces a signal that looks a lot like a transiting exoplanet. You know, if you look at it in more detail, you realize that's not a planet, but it takes some extra work. Now, as for giant stars, typically we don't observe giant stars on purpose. The reason for that is that, if I have a planet around a giant star, remember, that signal, that drop in brightness, is the ratio of the area of the planet to the area of the star. If the area of the star is this, then my signal drops like a rock, and now I've got no depth to measure, because it's so small it's not miserable. So we don't on purpose look at giant stars. Now, for TESS, this ancillary data, when I mentioned briefly, is we're going to be getting these full frame images. We're going to be taking pictures of the entire observation sector, that entire 24 by 96 degree field of view, every 30 minutes. And anything that's in that data, if there's a transiting something around a giant star, we'll have the data, it's possible we'll find it. That will be a very difficult bit of analysis to do. I'll ask a grad student to cover it. Other questions? ^M00:48:32 [ Inaudible Question ] ^M00:48:39 Earthlike planets, the estimate is there are 16% of stars have an Earth sized planet around them. ^M00:48:45 [ Inaudible Questions ] ^M00:49:02 The question is that, I quoted a statistic that 16% of stars have Earthlike planets around them. I also commented that some stars have no planets around them. Do we know that for a fact? And no, it's hard to know that for a fact. If a star has a planet in a very, very distant orbit, it would be very difficult to detect the methods that are really effective, like transfer rate of velocity. ^M00:49:33 [ Inaudible Question ] ^M00:49:38 What we can say is that there are some stars out there that don't have any planets within a certain distance, but that's about it. And there's other stars where we can make theoretical arguments that there shouldn't be any planets, right. If I see a supergiant, a star that has just recently gone into the supergiant phase, if it had any planets, they should've been sucked up and eaten. They shouldn't be there. But I don't know for a fact that there aren't any there. One of the limitations of both the rate of velocity technique and the transit technique is that it requires a given orientation. If I have, you know, if it's orbiting this way, then it transits, the planet passes in front of the sun. But if the orbit is inclined this way, it never transits, there is no transit. Likewise, the radial velocity measurement, if it's aligned this way, then that star is moving back and forth relative to your line of sight. But if that orbit is inclined this direction, then it's moving this way in your line of sight, and you get no Doppler effect, so you can't see it. So there's a lot of limitations, but it's essentially a statistical argument that what we're seeing is consistent with these numbers. If the number were significantly higher, we would expect to find more things. But if there's some unexpected unknown population of planets out at 10 times Pluto's distance, that we've never seen before, well, we can't count those because they shouldn't be there and we haven't seen them. >> Who's your favorite sci-fi writer? >> Who's my favorite science fiction writer. Well, that's going to depend on whether I talk about modern era or classic era. In the classic era, I think it has to be Frank Herbert. The Dune books were always -- I have a first edition of Dune, and that's the only first edition I own of anything. So that's going to be the classic era. In the modern era, the modern era, I'm a huge fan of John Schultz. But there's a lot of good work coming out now. It's very different than a lot of the classic science fiction but it's a lot of fun to read. >> Are we the only country studying exoplanets, or are there other countries doing similar things and we share information? >> Are we the only country studying exoplanets and do we share information? Everybody and their brother is studying exoplanets. In fact, one of the premier groups that has -- let me pull up the right plot. One of the premier groups that has made radial velocity measurements is a group out of Geneva. In fact, in these early days, it was a competition between a group at Geneva and a Carnegie California Hawaii collaboration, where the 2 who were competing and seeing who could get that one out first, and they were neck and neck. So it's a very international effort. In fact, even on TESS, you saw the OHP in France is collaborating with us to make follow-up observations. We also have the E-ELT telescope in Europe, the Japanese. We just had the science team meeting last week, and the Japanese are going to be collaborating on some of these follow-up measurements. Everybody wants to be involved. In the missions, so far, the only -- there have been a few dedicated exoplanet missions, Kepler is the big one. But the Europeans have flown something called Chiaps [phonetic]. They are now in development of a mission called Plato, which is a lot like TESS. It's a bigger version than TESS and it will observe more of the sky for longer, but it won't launch for about another 8 or 9 years, so it's a ways away. And they call it Plato, not Plato, I don't know why. And of course, they're collaborating. The Europeans are involved and the Canadians are involved with JWST. And we expect that for these kinds of missions where the science -- I hate to say this because it sounds so trite, but science doesn't have national borders. When it comes to this data, we share data openly and publicly, and we don't hide anything. Because we try to get to the truth. So I don't care who it is that helps me out. It sounds so -- anyway. Other questions? >> That's a lovely way to end. >> So cheesy. >> Thank you so much, Dr. Rinehart. ^M00:54:21 [ Applause ] ^M00:54:24 >> This has been a presentation of the Library of Congress. Visit us at LOC.gov.