Planetary Universe

- Good evening, everybody, and welcome to lecture five in our series on Cosmic Revolutions. For centuries, we only knew about the planets in our own solar system. And what we're going to learn about tonight, is the advances that the last few years have have brought in our revolution of our understanding of planets in outer space, beyond our own solar system. But we're going to start very much in our solar system. And I want to show you what I think is a really special photograph. It's the first ever a photograph of planet Earth taken from a spacecraft well away from Earth. This particular photograph was taken by Bill Anders in Apollo 8, the first ever spacecraft to orbit Earth's Moon. I think it's a really special photograph, and it's termed Earthrise because of the way planet Earth is rising above the horizon of the Moon. You can see some of the Moon's craters in the foreground. What happened next after this photograph was taken, was that Bill Anders rifled in his bag for a roll of color film to put in his camera. So this was the second photograph of planet Earth. It was taken on Christmas Eve in 1968, as I say, aboard Apollo 8. And this photograph is a reminder that Earth is floating through space, orbiting through space, revolving through space. Orbiting as it does with our Moon and orbiting as they do around our Sun. So this is a cartoon reminder of our solar system. Our solar system is rich in geology. The planets in our solar system are believed to have formed from the same spinning cloud of gas from which the Sun formed at the very center. This gas cloud called the solar Nebula was composed mainly of hydrogen and helium, but many other elements in smaller proportions. It is just as well for us that there were elements other than hydrogen and helium; otherwise, there would be no lecture theaters, no laptops, and no humans, but more of that, another time. This is a movie now, cartoon movie of Saturn and its moons in orbit around Saturn, and Saturn and its moons are, of course, in orbit quite a long way out in the solar system, but still around the exact same Sun that planet Earth moves. So just so that we are beginning with a basic sense of how planets are believed to be formed, I want to show you this cartoon movie. A gas cloud, mainly comprising hydrogen and helium, but lots of other elements from further down the periodic table as well. The gas cloud collapses in such a way that a plane is formed, and that is to do with the action of gravity pulling the matter together. But the fact that it's in a plane, which is still rotating and spinning closer and closer as you get to the center, that's to do with the conservation of angular momentum. Heating and cooling thermodynamics are also hugely important in determining what forms where. So just as a star can form right at the center, as in our solar system, so too can rocky planets quite close in, they can survive and withstand the great heat that comes from the central star about which everything is orbiting. Further out still, you see some planets that are surrounded by protoplanetary disks, as they are called, and the collapse of those under gravity, under the law of conservation of angular momentum, under the laws of thermodynamics and of chemistry, give rise to things that somewhat resemble Jupiter and Saturn, the gaseous giants in our solar system that have abundant moons orbiting around them. So with some idea then of our place in space, let's remind ourselves of the challenges of trying to find and study planets that are in orbit around stars. What orbits our Sun has less than 1000th of the mass of the Sun. And as I've already said, what is in orbit around our Sun is determined by gravity, angular momentum, thermodynamics and chemistry. We'll see a bit more how that plays out later on in this lecture. But let's begin with the start of the intellectual revolution that was the discovery of what we refer to as exoplanets. An exoplanet is a planet outside of our solar system. An exoplanet is a planet that is nothing to do with our Sun. The very first exoplanets to be discovered orbited a very exotic star indeed. Not a star on the main sequence at all, but in fact, orbiting around a pulsar. And that pulsar was discovered by the Arecibo radio telescope in Puerto Rico. That discovery was made before this telescope featured in the James Bond movie, "Golden Eye," And indeed before, sadly, a lot of the structure collapsed sadly under gravity, not in a good way, due to decay after many years of wonderful service in pursuit of finding exotic stars such as pulsars. So a signal was discovered from this particular pulsar that I'm going to talk about. Let me remind you that what pulsars are famous for is having an incredibly good clock. They release pulses that we can receive on Earth with radio telescopes with incredibly precise timing. But the pulse from this particular pulsar was anything but stable and steady in its periodicity. Although the approximate periodicity at about 1.935 millisecond, sorry, was moderately stable, at least on average, some very clear variations were seen, not random noisy fluctuations. This wasn't a question of low signal to noise, but fluctuations that varied in seemingly systematic ways following various different sinusoidal curves throughout time. These are some observations that began in 1990 and proceeded through 1991. Subsequent follow up confirmatory observations were made with the very large array radio telescope in New Mexico, in the USA. And those follow up observations were able to determine absolutely that the only way that you could explain that solitary yet somewhat stable pulse period from that pulsar was by the presence of a planetary system in orbit around that radio pulsar. The masses of the two planets in question were about three Earth masses in each case. Their respective distances from the pulsar that they were orbiting around were about half the distance between the Earth and the Sun. That's a distance that we refer to as an astronomical unit or an AU, and about a third of an astronomical unit. So distances and masses, somewhat comparable with Earth and somewhat comparable with the Earth-Sun separation. In fact, we now know that there's a third planet around this particular pulsar, PSR 1257+12. And that explains some of the finer details that were suspected in this paper, in this letter published in "Nature" in 1992. So that was the very first exoplanet to be discovered, but how about some of the others? How about all of the subsequent ones? How do you go about finding exoplanets and how many are they out there? Well, let me show you and describe to you one principle method for discovering exoplanets. And that's a method that we refer to as transits. Transits in the light curves of stars that have planets orbiting around them in front of them. Let me begin by talking about transits of planets in our solar system across face of our Sun.. So this is an image of our Sun taken in 2006 by my colleague, Steven Lee. In 2006, Mercury transited the Sun from the vantage point of Earth, and this image here shows on the periphery some sunspots. That's not what this lecture is about. Those were covered in my previous lecture entitled Magnetic Universe, but I want to talk about that tiny dot in the lower left quadrant. That dot is due to planet Mercury. Mercury is eclipsing a tiny amount of light from the Sun. And successively, in fact, at every five minutes, Steven Lee took successive exposures as Mercury transited across the Sun. Now it's a slightly extreme situation, but in fact, the brightness that we would get from the Sun would drop a teeny tiny amount as Mercury transited all the way across the face of our Sun. If you have a planet that subtends a larger solid angle, it will block more of its stars light. So now let's look at a transit of planet Venus, and this image was taken by John Gleason, and that rather more obvious black circle there was where Venus was. This was in June of 2012. I'm now going to show you a movie of Venus transiting across our Sun, not an image quite like this, but an image taken through a filter that permits only a very narrow range of wavelengths, those that are emitted normally by the hydrogen atom undergoing one of its particular electronic transitions. So I'm going to set this movie going. It's in this so-called hydrogen alpha filter, which is why you see some of the swirls and sunspots around. The shimmying around and jittering is because the seeing was dreadful on this particular day in June 2012. But nonetheless, apart from the clouds, which are taking some of the Sun's light away there, you can see very, very clearly the passage of Jupiter across, sorry, forgive me, the passage of Venus across the Sun. So if you measure the brightness of a star and then you have a planet transiting across it, the brightness of the star dips a little while, as long as the planet is completely in front of the star. So this cartoon shows exactly that phenomenon. When the planet is in position one, it doesn't coincide with the stellar disk at all, and light is received from both bodies. Although, of course, the light that you receive from the planet is almost certainly negligible. When it just slides into the projection of the stellar disk, then you see the first of the slanting red lines. Then when it's fully coincident with the stellar disk, the brightness has dropped completely. And then as it bids for freedom, you see a slanting line again. And then in position four, you see the light return to its full brightness again. That is known as a planetary transit. So in principle, if you look at a load of stars and you look for where these dips occur corresponding to when a transit has gone across a star, you are well on the way to discovering an exoplanet. It turns out that you can get a lot of information from these dips, from these transits. And if you start analyzing that very, very carefully, even though you can't even see the planet in any direct way without a lot of drama and a lot of sophisticated equipment, it is none the less possible to infer the mass of the planet, the size of the planet, the density of the planet, the atmospheric composition of the planet, as well as its orbital period around its central star. The duration of a transit shown at the bottom in this solid black curve here combined with the periodicity, in other words, the loop repeat of these dips as the planet's orbit around these stars, gives you the orbit of the planet around the star. And hence you can get the mass and the radius of the planet. Once you've got the mass and the radius, you've got the volume, you can get the density. And the densities of the exoplanets that have been discovered vary from having the density of rocky planets like Earth, like Mercury, all the way up to planets that have the density of polystyrene. Things that are even less dense than the gaseous giants in our solar systems, such as Jupiter and Saturn. It's worth pointing out that you can get even more information if you've got an extremely well calibrated light curve. So a light curve is the term that we use for, how does the brightness of a body change with time if you take lots and lots of successive images? And it turns out that if you pay really close attention and you've got very well calibrated photometry, in other words, you know exactly how much light you're getting at a given moment in time, you don't just benefit from the transit when the planet is transiting in front of the stellar disk, but also you see a slight boost in brightness as well. That's because you see the planet, you see the day side of the planet reflecting its star's light towards Earth. Whereas when the planet is going across the disk of its star, we see its night side. So we just a little bit less. So as you can imagine, there's quite a lot that can be inferred about the nature of exoplanets just from studying these light curves. So how do you go about getting these light curves? Answer, lots of telescopes that are dedicated to the purpose. One of the amazing projects that was responsible for discovering a lot of the early exoplanets around normal stars, rather than pulsars, was a project called WASP. WASP was an acronym standing for the Wide Angle Search for Planets. SuperWASP is kind of the successor of WASP, only better, shinier, more lenses, better lenses, better cameras. This is the location of one of the SuperWASP observatories. I visited this one in South Africa, and there's another one in the Canary Islands that's able to look at slightly more Northern skies. This is a photograph of the apparatus itself, and it comprises a bunch of Canon lenses. These, I think, are 200 millimeters in focal length, F1.8 for the photography aficionados. And these give quite a wide angle view on the night sky by the standards of astronomy, quite a few degrees in extent across an image. So as you take successive images, you get light curves from thousands of stars. Then you calibrate the data really, really carefully, and you plot out those light curves. And you ask the question, which of the stars exhibit dips? Which ones should be followed up to confirm or refute the presence of an exoplanet in orbit around them? WASP was responsible, this image, this graph is from a few years ago, but WASP is responsible for the discoveries of all the exoplanets indicated with royal blue symbols here. The ones indicated by the pink symbols correspond to some of the early exoplanet discoveries from NASA's Kepler Satellite. This satellite was launched in 2009, and sadly it ran out of fuel a few years after it was predicted to in 2018. So that's when this satellite was retired after nine years of amazing service. It discovered many exoplanets, and the most common size of exoplanet that it found was, guess what, different from the planets in our solar system. Kepler found a lot of planets whose masses are in between the size of Earth and Neptune, yet still rocky and not gaseous as the more massive planets in our solar system, Jupiter and Saturn are. Kepler has been succeeded by TESS. TESS stands for the Transiting Exoplanet Survey Satellite. You can see people love acronyms in astronomy. This is run by the Massachusetts Institute of Technology, by MIT. And during its lifetime, it is scheduled to cover 85% of the sky, an area of sky 400 times larger than the area of sky studied by the Kepler Satellite that I just mentioned. It's designed to search for as many exoplanets as possible within a radius centered on Earth of about 200 light years. So SuperWASP, Kepler, and TESS all use the transit method of finding planets. But there are some other methods as well, even though the transit method is winning the numbers game. More of that in a little while. There's another way in which you might hope to discover planets, and that's using the technique, not of photometry and light curves, but the technique of spectroscopy. This technique makes use of an important phenomenon in physics, which is that if a moving body is emitting in a wavelength-specific way, then that movement, the speed with which its moving, is entirely measurable on the basis of the Doppler effect. If that wavelength's specific way in which it's emitting is in the audio, such as the siren on an ambulance, then we change in pitch or frequency of the audio signal of the siren. Shorter wavelengths if it's coming towards us, longer wavelengths, a lower frequency, a lower pitch if it's racing away from us. The same is true of light with light, of course. Different elements across the periodic table emit different wavelength-specific signatures, recognizable spectral fingerprints that enable us to identify which element is present and how fast it's moving. So if we attach a sodium streetlight to the back of a spacecraft and launch it, skipping over the technical details, we can totally measure the speed at which the spacecraft and the sodium streetlight are moving away from us. I explored the usefulness of this technique a bit more in one of my lectures last year in the series on Cosmic Vision in the lecture called Unraveling Rainbows. If we use the technique spectroscopy, then we start to study the motions of stars, which are caused by the orbits of planets on around them. Let me show you how this works. From the perspective of a star with a planet in orbit around it, let's imagine that the star is colored red and the planet is colored green, in the star's frame of reference, the planet is just orbiting around it, and the star is stationary. Now, in a different frame of reference, let's say the frame of reference of Earth, then the reality is that there's no sense in which the planet is orbiting around the center of mass of the star. On the contrary, the planet and the star are both in orbit around the center of mass of their system taken as a whole. They are both in orbit around their common barycenter. So if you start doing spectroscopy on the star and you look for any particular spectral signatures in it, and there is a planet in orbit around it, you will see subtle solitary motion arising because the star is in orbit around the common barycenter. It's worth pointing out the perhaps obvious point that what makes planet hunting so hard is that the light from the stars around which the planet's orbit are so bright and so overwhelming, it is not trivial to see planets in any direct way. And so that's why we either use the method of transits or the method of spectroscopy on the central star, which is perfectly bright enough to take a spectrum of, we take it at successive epochs and we see these specific wavelength filters oscillate with time, we are well on the way to saying that may well be oscillating because it's got a planetary system orbiting around it. At the very least, you've got yourself a candidate exoplanet, which you can then do follow up observations of and either confirm or refute the reality of there being a planetary system around your star. One of the follow-up methods that's rather fun is the method of direct imaging. Now this isn't trivial because you've got this incredibly bright star and you've got these, oops, sorry. You've got the planets in orbit around the star that by and large don't emit any light for themselves, or if they do, it's at an extremely low level relative to the stars. So what you really want to do is to be able to eclipse the bright light from that central star, and then go really, really deep in your observation so that you can see whether there are indeed any orbiting planets present. Well, the only way that you can arrange eclipses in this way is manufactured eclipses with something called a coronagraph. A coronagraph is a bit of metal that doesn't let light through, and you line that up with a star and then you can see what's around it. So let me show you the case of a star whose name is HR8799. These are a set of successive observations taken a few months and years apart by Jason Wang and Christian Marois of this particular system. So the black circle in the middle has blanked out the light from the Sun. And the kind of flashing light in the middle is just stray light from the central star. But I hope you can get a sense that there are four stars in this stellar system that are rotating anti-clockwise. There's one really close in at about three o'clock. And then there's the two brighter ones on the right-hand side. And then over at about 10 o'clock, there's the faintest one of all. So direct imaging, although it's quite hard work and you need a coronagraph, can be marvelously reassuring about validating or refuting the suggestion that your candidate exoplanetary system does indeed have orbiting planets. This particular example has the orbital plane of those planets parallel to the plane of the sky. But if you were to rotate this system by 90 degrees, you still stand a chance of seeing the planet if you take multi-epoch successive images with a coronagraph. So this is now similar kind of imaging sequence of the star known as Beta Pic and its planet known as Beta Pic b. Its orbital plane is perpendicular to the plane of the sky. So all you really see is the line of that star going behind the star now and reappearing. It's only for a fraction of the slightly over a quarter of an orbital period of Beta Pic b in orbit around the star Beta Pic. So direct imaging is marvelously informative, but quite hard to do. So what types of exoplanet do we find via these various means? Answer, quite a lot of different sorts. Are the planets the same as the ones that are in our solar system? Well, no, but the ones that you find tell you to some extent about the technique that found them. So one type of exoplanet that is found a lot is the type of exoplanet known as a Hot Jupiter. When someone says you've got a Jupiter-like planet, what that is shorthand for is you've got a gas giant, quite massive but gaseous, not such high density as the rocky planets like Earth that are full of iron in the middle and things like that. If you're talking about a Hot Jupiter, what that means is you've got a gaseous giant'cause you called it Jupiter, and it's hot because it's orbiting really close in to its central mothership star. And indeed, that was one of the very first exoplanets to be discovered around the very first one. In fact, around a normal main sequence star 51 Pegasis. The Hot Jupiter, the first planet 51 Peg B, in orbit around the star 51 Peg had a really short orbital period. And by short, I mean slightly over four days. That means if you were living on the planet 51 Pegasus b, you would have a birthday every four days because that's how fast you would be orbiting around 51 Peg. If you've got a star that's got a really short orbital period, then for a given mass, it's orbiting very, very close in to the star that's at the center of its orbit. If you are really close to a star, think sunburn, you're going to be very hot. So that's what a Hot Jupiter is. So like Jupiter in our solar system except really, really close in. Another type of exoplanet that's been discovered is the so-called Brown dwarf. Now a Brown dwarf is, in some sense, midway between a planet and a star. It's more massive than Jupiter, considerably more massive than Jupiter, but not massive enough to be a proper star. Not massive enough for it to collapse efficiently so that the temperatures are high enough so that we can get fusion of hydrogen and everything else that follows from that in the normal sequence of stellar evolution. But in a Brown dwarf, the fusion, the burning of deuterium, also known as heavy hydrogen is possible. Deuterium burning means that it's responsible for giving off some light. Often this can be at infrared wavelengths more predominantly than in the optical wavelengths. So this diagram from NASA illustrates that planets and exoplanets can in many characteristics resemble those in our solar system. But to resemble those in the solar system, they're not going to have much more than about a dozen times the mass of Jupiter. Brown dwarfs will have a few tens times the mass of Jupiter if they're going to be massive enough to be hot enough to give you deuterium burning. More massive than about 80 times Jupiter's mass and you will have a normal star. So that's the sequence of these orbiting bodies that have been discovered in the course of searches for planets. I'm now going to show you a really messy graph, but it's an interesting messy graph. So give me a moment to just explain what it is we are looking at. This plot shows the detections of all exoplanets, all planets, actually, that had been discovered up to a couple of years ago. So the planets that are in green circles with a little letter, those are in our solar system. So the way that they get found is completely different from the exoplanet searching expeditions that I've described in the first half of my lecture. The bright magenta crosses on the top right hand quadrant of the graph, those are ones that have been found by direct imaging. Those are all the ones. If you look now at the X-axis, the X-axis is telling us about the orbital distance. The distance a planet is orbiting around its central star. So right in the center there, where it says 10 to the power of zero, 10 to the power of zero is a grandiose way of writing the number one, that's one astronomical unit. And so you'll see that the green circle with a letter E in it symbolizing Earth is at one astronomical unit. But all those magenta crosses that are on the right-hand side are at distances of 10, 100, 1000 astronomical units away from their central star. Now it would be a mistake to draw any profound conclusions from that other than the fact, it in direct imaging, you're not going to get the very closely orbiting ones, the ones orbiting at a very tight orbital radius because they would be submerged underneath the coronagraph with which you're trying to block out the central star. So that's why you will only tend to see exoplanets orbiting at the largest orbital distances using the technique of direct imaging with a coronagraph. In contrast, the red points are the ones that correspond to transits. And these are all at really quite small orbital distances, very, very close in, but a range of masses. And that's because the ones that are closer in, the ones that are most massive are the easiest to detect via that method. It's again, part of a selection effect. If they're close in, they're orbiting very rapidly, it's much easier to stand a chance of picking up that dip in the light curve corresponding to the transit. And then the Doppler effect ones, the spectroscopy ones corresponding to the Doppler wobble. They're a bit more in between, but again, you get a more dramatic signal for the oscillation of a star orbiting around the barycenter of its system if you're talking about massive stars really close in. So that's why there is a truckload of blue points at really quite high masses at a thousand and getting up to 10,000 times the mass of Earth. So this is quite an interesting plot, but it tells you as much about the selection effects of the techniques used for planet-hunting as it does about the properties of exoplanets that are out there in space. Can we see planets forming? We absolutely can. It turns out that almost all stars that we've looked at that are younger than a million years have a dusty proto planetary disk in orbit around them. And I'm going to show you an image now from the ALMA radio telescope, which is in the north of Chile of a star system known HL Tau. The central star is unresolved at optical wavelengths. But if you look at radio wavelengths, sub millimeter wavelengths, then what you see is a concentric set of seemingly dusty rings, absolutely concentric with the central star, lots of gaps in between those rings. How do we think those gaps are formed? Rocky planets that have snow plowed out the paths that they follow somewhat analogously to the way that moons have carved out gaps in the rings around planet Saturn in our solar system. So that's HL Tau. And that was particularly exciting because it was the first resolved image of a protoplanetary disk giving strong indication that there were planets information orbiting around, tracing out, clearing out those gaps in a way a bit like the movie that I showed you very early on at the start of this talk. This is another such system. This is TW Hydrae. It's much closer to Earth. Hence we see it in much finer detail, but exactly the same phenomenon. And through now think over a couple of dozen such examples of this. I want to show you now a rather more complicated example. This is a star known as GW Orionis, and Stephan Kraus and team, based largely at the University of Exeter, used an instrument called GRAVITY on one of the very large telescopes in Chile. What they were imaging, so if you look at this system on the right, you can see that there's kind of a sense of dusty rings, but it's a little bit unclear on the basis of preexisting images. But that image with the GRAVITY instrument brings into sharp focus the idea that there are noncoplanar dusty rings. How do you get non-coplanarity in forming planetary systems? Answer, a multiplicity of stars within. And in the paper in science, which Stefan Kraus et al published nearly a couple of years ago, they present very strong evidence that this is a triple star system with protoplanets forming these noncoplanar rings around them. It's an amazing system and definitely one to watch. For anyone who's feeling a bit unnerved by the thought of a triple star system, let's take things down a notch and just think about a planet orbiting just two stars. So a circumbinary planet. I want to illustrate to you what I mean by a circumbinary orbiting body. So I'm showing you the identical system on the left and on the right, but from two different perspectives. The one on the left is face on with respect to the orbital plane, whereas the system on the right is edge on with respect to the orbital plane. And what this illustrates is that if you got a circumbinary orbiting test particle, by test particle, I mean it's mass is very, very light in weight. It will process in and out of the plane of the circumbinary, the binary star system within. So that's what a circumbinary planet is. A former graduate student of mine, Sam Doolin, and I explored the dynamics and the stability of circumbinary orbits of test particles ran different synthetic binary star systems, finding that you can get stable orbits of particles around an inner binary star system even if that inner orbiting binary has a very eccentric orbit, even if the masses are very different, you can get stable orbits if you are sufficiently liberated about your definition of stability to mean that the orbital plane of the circumbinary body, the circumbinary planet will actually process in what's called the longitude of the ascending node of the orbit. It was great fun. And that behavior really jumped out of our calculations because we did it fully in three dimensions. Previous calculations exploring this had largely been in two dimensions. And, of course, you don't get such clear behavior of things moving in the third dimension if you don't do the calculation that way. So this is a slightly busy plot. I won't go into it in too greater detail other than to say each sub-square in this overall plot corresponds to a particular hypothetical binary star system of a given eccentricity and of a given ratio of its two masses. And within that, the different colors correspond to prograde orbits, where the orbiting material is orbiting the same way as the inner binary, retrograde orbits when it's going in the reverse direction, and polar orbits when it's going in a completely different plane altogether. Sam and I were delighted when three weeks after we published our publication, the first discovery, the discovery of the first-ever circumbinary planet known as Kepler 16b, was reported by NASA using the Kepler satellite, and it was exactly in that green location down there, an area that we had posited, simulated would give a stable circumbinary orbit. We were terribly pleased. This is a slightly busy plot for which I apologize, but let me take you swiftly onto NASA's PR about the discovery of Kepler 16b."Relax on Kepler 16b," the poster says,"where your shadow always has company." Two shadows, of course, coming from the fact you've got two stars, and the planet is orbiting around them. Well, the final wild system that I want to tell you about now consists of a dwarf star, an ultra-cool dwarf star and seven planets. So this particular system is located in the constellation of Aquarius. And if the light curve of this particular system is plotted, it looks, well, busy and a bit confusing, and certainly quite hard to interpret. The green points that you can maybe see in the top half of the plot correspond to individual measurements that make up the overall light curve as measured from the ground. Whereas the black points correspond to individual measurements in this overall light curve as measured from space. Now this may look to the untrained eye, extremely noisy, But careful calibration and insightful analysis reveals a remarkable system. If you zoom into individual ones of these dips where the, you've got a sort of very sparse lower level of the black points here, if you zoom into one of those, then they don't look quite so noisy after all. They do look a little bit like dip, superimposed on dip corresponding to transit of one planet superimposed on transit of another planet, a transit happening at the same time of the same star. If you fold the data very carefully on all the different orbital periods, corresponding to all the different planets that are orbiting around this ultra-cool dwarf star, things start to make a bit more sense still. Now, I indicated earlier that if you think about the periodicities of the dips in light curves and their durations, you can get to the radius, the orbital radius of a given planet around the central star. And, of course, it's orbital period coming directly from the periodicity with which you see these particular transits. The radii of all seven planets in orbit around this particular star, this particular star's name is TRAPPIST-1, I'll explain where the name comes from in a second, but all seven planets in orbit around TRAPPIST-1 have a radius, an orbital radius that, I'm sorry, planetary radius that varies between about 0.7 and just over one of Earth's radius. But they're much more packed in. They have much smaller orbital radii than the planets in our solar system. So these planets are numbered 1b all the way down to 1h, the lowercase b is always the first planet discovered in a planetary system. And then you just successively take all the other letters going through the alphabet. Goodness only knows what we'll do if we discover a system with more than 26 planets in it, but that's a problem for another day. If I now just plot the orbits, the comparisons of the orbits of the planets around TRAPPIST-1 and the Sun. So TRAPPIST-1 and the Sun is superimposed on the left, and then we've got planet 1b going all the way to 1h their appropriate orbital radii, they are well within 1/6 of the radius at which Mercury orbits around our Sun. They're a little bit more comparable to, or rather, they're really in between the orbital radii of the so called Galilean moons that are in orbit around Jupiter. You can see those plotted on the left there, then in the center, those are the orbits of the planets around this star, TRAPPIST-1, and Mercury's orbit is way above. This is another visualization of that particular orbital system. And this perhaps shows you a little more clearly what those orbits are. Now, in fact, the time taken for TRAPPIST-1b, the innermost planet, to orbit the star 24 times is about equal to the time taken for the next planet eight, TRAPPIST-1c, to orbit it 15 times. And the that's the same time as the time taken for TRAPPIST-1d to orbit the central star nine times, and so on. It turns out that the ratios of the numbers of orbits you have to do from ever so many close in, to much fewer, much further out are in the following ratios, 24, 15, 9, 6, 4, 3, 2. These are orbital resonances. And this is a sure sign that there's been a bit of shimmying around as those planets have settled collectively into orbit around the Sun, sometimes slowing one another down, sometimes speeding them up. And as those planets have migrated together into the orbits that there on now, they have attained these orbital resonances. Well, the final exoplanet system I want to describe to you this evening happens to be something really rather remarkable. Everything that I've talked about up to now in this lecture is an exoplanet or an exoplanetary system within our own galaxy, the Milky Way. But last year, a particularly successful lockdown research project identified what is most probably the first extragalactic exoplanet. So just to be clear about words, we said exoplanet is a planet outside our solar system. Well, we are no longer talking about a planetary system that is outside of our solar system. We're talking about an extragalactic example, an example that is outside of our own galaxy, outside of the Milky Way. This particular exoplanet is in the galaxy of M51, the very popular Whirlpool Galaxy. It's a very popular galaxy system for amateur astronomers to observe. And indeed this particular image came from our back garden in Oxfordshire. It's an image, the exposure time is 1200 seconds, 20 minutes, and you can see very clearly the features that are familiar to many amateur astronomers. If you look at this exact same galaxy system, not with a telescope in someone's back garden, but with an X-ray satellite, you do see something of a similar picture. The budget for this is rather higher, I can assure you, this is the CHANDRA satellite launched by NASA and ISSA. So you can just about make out the spiral lamps of M51 there, but at the center of that box, there is a prominent X-ray source. And that prominent X-ray source had been studied multi-epochs successively with CHANDRA, measuring the X-ray luminosity coming from a particular X-ray source believed to be known as what's called an X-ray binary star system. And in such X-ray binaries, there's usually a compact object like a neutron star or a black hole, and then a normal star. So that's the kind of binary star system that this has been identified as. But the thing is something very strange was seen in its X-ray light curve, specifically a possible exoplanet candidate. This is the paper published in"Nature Astronomy" last year by Rosanne Di Stefano and coworkers. And what they found was that when they looked at the X-ray light curve from the object that was in the box on my previous slide, its characteristics fully resembled the known and fairly well understood X-ray characteristics of an X-ray binary star system, except it exhibited dips. And this dip has been... So the signal to noise is not as superior and convincing as some of the WASP-like curves, or some of the transits with Kepler or the transits we with with TESS, but nonetheless, it's a very convincing dip. And in the paper, they take the reader through different alternative scenarios that could be responsible for the dip. And they present a reasonably compelling case that the most likely explanation is that a planet in orbit around that X-ray binary star system, so circumbinary planet around an X-ray binary pair is responsible for this dip in the light curve. It's a compelling paper, and it's a very impressive analysis. And naturally, ongoing X-ray observations are being pursued in order to identify the periodicity with which these dips occur so that analysis and deductions about the characteristics of that exoplanet can indeed be inferred. Well, what of the exoplanets that have now been discovered? Until the last decade of the last century, we knew of none. We only knew about the planets in our solar system, but a very special milestone was reached in Wednesday of last week, which was at the milestone of having found 5,000 exoplanets had been attained. About 30% of these are gas giants like Jupiter or Saturn. About just over 30% of those are known as super-Earths. So planets a bit like Earth in the sense that they're rocky, not gaseous, but the super bit means more massive than Earth. There are about 35% that are deemed to be a bit more like Neptune or Uranus. So meaning fairly massive, fairly hefty, but often quite icy, although not always. More rarely, they can be really quite warm examples. But about 4% of the exoplanets that have been discovered thus far are what are referred to as terrestrial. Small, rocky planets around the size of Earth, or perhaps a little bit smaller. At the risk of stating the obvious, a gaseous planet or an icy planet is not likely to win many points on the habitability front. But the habitability of exoplanets and the search for life throughout the universe is the subject of my next Gresham lecture on Life In The Universe. But today's lecture, The Planetary Universe, I really hope has been of interest to you. Thank you.(audience applauds)- Is there a correlation between the stability of a solar system and the number of stars that solar system has?- That's a very interesting question. Again, I think I want to come back to the remark I made about stability as being something you may have to be a bit more open-minded about than when we thought that planetary systems could only exist in a two dimensional plane. Actually, even the orbits of our own solar system aren't in a perfect plane. Neptune is doing a few peculiar things, for example. But if you've got a number of stars inside a planetary system, then it's a really good question to ask, are those planets stable? Is the system as a whole stable? And the answer is it isn't necessarily stable for billions of years, but it might be if you allow for the fact that some of the planet's orbital planes might themselves be processing, the configuration of stars within might be processing. This precise question about stability is something that is amenable to making modern numerical calculations and simulations of the behavior for millions of orbits because simulations are a wonderful way of being able to run experiments, admittedly, subject to the assumptions that we build those calculations on, and then sort of roll the clock forward in time in a way that we can't really get a window on with stellar systems. We've seen planets orbit around, but we don't see the evolution of planetary systems on many, many dynamical time scales because certainly, thus far, human existence is a bit too fleeting.- [Audience Member] Thank you for an excellent lecture. My question really is on solar system formation. And you mentioned that the rocky planets tend to form closer to the star and the gaseous ones outside. Is that what you'd normally expect to see with exoplanet systems? And if that's the case, are there any examples where, for example, the reverse is true?- I don't know of any examples offhand where the reverse is true, but that's not to say they don't exist. It might be to say it would be really quite hard to discern their existence. I mean, there are certainly systems where you have gas giants really, really close in to the central star, 51 Peg b being an example of that. But I think it would be really quite difficult to discern the presence of rocky planets on much larger orbits. That's not to say they're not there. They could be there. There are reasons to think they might be there. For example, the asteroid belt, which is sort of in the middle of the rocky planets and the gaseous planets in our solar system. And then the Kuiper belt objects way further out the Neptune. There's a lot of rocky stuff around, Predominantly, as far as our solar system is concerned, we think those are dwarf planets, but the very fact that you can get rocks far out of the solar system does suggest that it might be possible. Has a lot to do with the formation mechanisms and the evolutionary scenarios that happened. I imagine it's possible, but I doubt it's within easy reach of current technology.- [Audience Member] Oh, thank you. I was just wondering, this is all quite recent, a lot of these discoveries you've mentioned, what was the specific kind of technological breakthrough that allowed this to happen? And further to that, what is the barrier that is stopping more? Is it computational? Is it telescopes? Does that makes sense?- That is a most interesting question. I think what made the difference wasn't so much technology as a combination of human imagination and commitment. So the very first exoplanet that was discovered, PSR 1257+12, that was actually discovered serendipitously. So the people who did the analysis on that couldn't make sense of it in the context of pulsar timing behavior that had been observed in the past. And it was only when they did, as I say, the very careful analysis, the follow up analysis with the VLA radio telescope that they were compelled to conclude on the basis of what you saw was very, very high signal to noise timing data, the existence of at least two planets subsequently confirmed to be three. That was published in January of 1992. And I think just the conviction that exoplanets were out there, not that people necessarily by that time were thinking,"Oh, the solar system is so special. There just won't be any planets elsewhere." I don't think that was at all the reigning paradigm within the science community, but somehow the conviction that if they're there, we need to search for them in such a way that we'll find them if they're there. I think that was really the change. I mean, WASP, the forerunner of super-Wasp was a bunch of camera lenses on a mount that could track really well and really stably in the sky. That's been around for a while. So I wouldn't say it was particular technology, except maybe I should qualify that by saying that perhaps detector technology made a bit of difference. The fact that you could get relatively high signal to noise light curves with relatively short exposure times with very high fidelity, so low signal to noise, because if you've got a dip that's fairly subtle, but the arrows on all your measurements are this big, you wouldn't discern it. So if there was a technological breakthrough that made it all possible, I would say it was the CCD detectors, which, of course, now have been themselves superseded, but that was probably, that and human imagination and human commitment.- I'm afraid that's all we have time for tonight. Professor Blundell, thank you very much for a wonderful lecture and thank you all for coming this evening and your attention, both people here in the audience and those joining us online. Please join me in thanking the professor for her lecture.(audience applauds)