University Place

cc >>

Tom Zinnen

Welcome, everyone, to Wednesday Nite at the Lab. I'm Tom Zinnen. I work here at the University of Wisconsin Madison Biotechnology Center and for UW Extension Cooperative Extension. On behalf of our partners, the Wisconsin Alumni Association and Wisconsin Public Television, thanks again for coming to Wednesday Nite at the Lab. We do this every Wednesday night, 50 times a year. Tonight we have an extraordinary presentation about an extraordinary invention of people here at the University of Wisconsin Madison. And it's not every day you get to introduce somebody who works on the single largest scientific instrument ever built. And when I was at the National Science Foundation, starting in November of 2008, IceCube came up on several occasions as a great example of some daring research. And I was very pleased to be from UW Madison when people were talking about it in that way. Tonight's speaker is Stefan Westerhoff. He joined UW Madison physics faculty in August of 2007. His research area is astroparticle physics. It's a new area of science at the intersection between high energy particle physics and astrophysics. After graduating from the University of Dortmund with a diploma in 1993 and the University of Wuppertal in 1996, Professor Westerhoff worked as a post-graduate researcher at the University of California in Santa Cruz from '96 to 2000. And he was on the faculty at Columbia University in New York City from 2000 to 2007. Professor Westerhoff is currently a member of the IceCube Collaboration. IceCube is a cubic kilometer sized detector at the South Pole consisting of several thousand small detectors frozen into the Antarctic ice at depths between 1500 meters and 2500 meters. IceCube is designed to study the origin of high energy neutrinos and, thus, open an entirely new window in astronomy. Please join me in welcoming Professor Westerhoff to Wednesday Nite at the Lab.

APPLAUSE

Tom Zinnen

>>

Stefan Westerhoff

So, is this working? >> Yes. >> Okay, thank you very much for coming, and thank you very much for inviting me. It's a pleasure to be here. And today we'll talk about the words neutrino astronomy, as we've already heard, the IceCube experiment at the South Pole. But you will get two experiments for the price of one. As you can see over here, I will not only talk about the IceCube experiment but somewhat related about the Pierre Auger Observatory which is also a very big experiment. And since the physics goals of these two experiments are somewhat related and I'm working on both of these, I'll introduce both of them in a more sort of comprehensive talk about our quest to find the sources of cosmic rays and therefore find the sources of the most violent processes that we know in the universe. So it's not easy to show you a picture of the IceCube detector. It is about a mile below this area over here. So as you can already see it's a rather unusual telescope. So let me give you sort of a data sheet of the IceCube experiment. IceCube is a cubic kilometer sized detector that's frozen into the ice near the geographic South Pole at a depth of about 1,500 to 2,500 meters. The detector volume itself has about a billion tons of ice, and in this billion tons of ice there are about 5,000 light detectors, which I will introduce in the course of this talk. The primary goal of the IceCube experiment is to detect neutrinos from extragalactic sources. And my talk will essentially be about these items, so introducing the IceCube detector and then the physics goals and at the end also some first results that we already have. By detecting these neutrinos we learn about the most energetic and most violent objects in the universe. And let me just add at this point here that this talk actually comes at a good time because about a week ago, 10 days ago, we finished a week-long celebration of the completion of this detector. The detector was completed last December. All these 5,000 light detectors were finally put into the ice which was a process that took several years. And we had about a week or so of collaborations and celebrations with invited guests here last week. So this is a very good time to talk about this experiment. Of course, now the work really starts when we have to look and analyze the data that we're taking with this experiment. So here's another picture of an airplane approaching the Amundsen-Scott South Pole Station which is where the detector is. This is the runway over here. And IceCube is buried roughly about here. This is the shape of Antarctica, the South Pole is about here, and IceCube is about a mile below this area here. So this is what it looks like. There's and instrument with 86 of these strings with light detectors on these strings at a depth of about, as I said, 1,500 to 2,500 meters. There are also detectors at the top of the ice that sort of complement the detector. You can see the size of the detector by, usually you put a person next to a detector to sort of get a scale of how big the detector is. We can't do that here so we put the Eiffel Tower here. So you get roughly a feeling of how big detector actually is. So the question you might ask at this point is, why would one do that? Why would you instrument a billion tons of ice in Antarctica? And I'll give a short outline of my talk here. I'm going to talk about astroparticle physics or particle astrophysics in general, that is the science of searching for the most energetic sources of the universe. I'll talk about cosmic rays in the Pierre Auger Observatory in Argentina. That's the other experiment that I want to introduce here. And then I will talk about IceCube toward astronomy with neutrinos. So here's a short summary slide of what astronomy is, sort of at all wavelengths. So you have classical astronomy. That basically deals with the electromagnetic spectrum from radio to x-ray. So optical wavelengths play a big part but also, in the last century, radio and x-rays became very important. Over the last sort of few decades we have learned that the light particles that are emitted from some of these sources have energies that are about 10 to the 10 times larger than the energy of photons from optical light. So the spectrum of some of these sources extends in energy to energies 10 to the 10 times larger of what you would call classical astronomy with light. And we have also learned that there's not just photons or light particles around but also other particles. For example, protons and heavier nuclei that constantly rain onto the Earth from outer space. And they have energies up to several joule. These are the highest energy particles we know in the universe that have already been observed in the universe. And tightly connected to these cosmic rays, as we will see, are neutrinos. These particles are neutral, therefore they are not deflected in magnetic fields like cosmic rays are. So they should, in principle, be easier for astronomy, but we will see they come with their own set of difficulties. And this here is going up in energy is not quite true because these particles could be a higher energy than those. But you get the general idea that astronomy is expanding to higher energies. We are not looking at just photons and light particles over here but also at real particles, therefore the name particle astrophysics. And the experiments that particle astrophysicists deal with look very strange. They don't look like your average telescope, maybe this one does. They can look at this. This is a detector of the Pierre Auger Observatory in Argentina, and I'll get back to that. This here is an experiment that I worked on as a postdoc. This is one of those experiments that looks for high energy gamma rays, and it's just a -- field water pool filled with water. Light detectors inside and there's a light-tight cover on top of it. So it's not your average telescope. This is a telescope that also looks for high energy gamma rays. That almost looks like a telescope I would say. But the other ones look rather strange. So this is somewhat a strange area of astronomy where you encounter experiments that don't look like you usually see in astronomy textbooks. So let's talk a few minutes about neutrinos, what they are and why they are so strange. They were, as you know, the fundamental building blocks of atoms, and therefore everything in this room and everything we know in the world, are essentially neutrons, protons, and electron. And neutrons and protons can sort of turn into one another in nuclear reactions. So neutrons can turn into protons and vice versa and the process is called beta decay. And here's a picture of what happens. You have the original neutron and that turns into a proton and an electron. And this was noticed early on in the last century that the electron would go this way, the proton would go this way, and if you believe in momentum conservation this can't be all because this thing is at rest and then something goes off in this direction and something goes off in this direction but nothing goes on in this direction. There's nothing here but there should be if momentum conservation is valid. So something is missing in this picture. Or total momentum is not conserved. That's not an option for a physicist. So Pauli, Wolfgang Pauli came up with an idea and said there must be something there. Something that we can't detect. And he called that a neutrino. So to fix this idea of beta decay, Wolfgang Pauli, who is shown over here, invented a new particle that is very light and has no charge. Neutrino means the small neutral one. And of course since we don't see it, the principal feature of neutrinos is that they usually escape unseen. They interact very little with anything so they just go out and you can't see them. They move almost at the speed of light, and they're very difficult to catch. And there's a famous quote by Pauli that says, "I've done a terrible thing. I've postulated a particle that cannot be detected." That actually would be a good thing for a theorist, really, because you're safe for a while until the particle...

LAUGHTER

Stefan Westerhoff

But this particle was actually detected. That's even better for a theorist, I guess. The neutrino was eventually detected by Reines and Cowan in 1956 in a nuclear reactor. That was three years before Pauli's death. So he lived to see this and his response was everything comes to him who knows how to wait. So neutrinos are everywhere. They are produced wherever there are nuclear reactions. For example, in the sun. There are more than 50 trillion, that's 50 times 10 to the 12th, solar neutrinos going through your body every second and you don't notice that. So they really do not interact very lightly. So here is a picture of where neutrinos come from. I just mentioned the sun. Nuclear reactors, this is how they were detected by Cowan and Reines, so they built detectors around a nuclear source. There are, of course, decay processes in the human body so we can do all sorts of neutrinos. The Earth is, actually. Of course, accelerators, like Fermilab's and KEK. In 1987 we observed neutrinos from supernova 1987A. These, apart from the sun, are the only extraterrestrial neutrinos that were ever detected. There's a relic neutrino bath all around us from the big bang. So lots of sources of neutrinos are around. And here's the thing about neutrinos. As I said before, they rarely interact. They just zip through almost everything. Therefore, very hard to detect. But you can see over here in a picture is one of these nuclear interactions. And you can see actually something is happening. So these are all sort of particle tracks over here that are made visible in this special detector. And you see that something must be coming in here because suddenly three particles are produced here. And, in fact, what is coming in here is a neutrino and what it produces is a pion, a muon, and a proton, three other sort of particles. So my point here is you can see the pion, you can see the muon, you can see the proton, you cannot see the neutrino. So neutrinos are very, very hard to catch. This will get us to the size of the IceCube detector. So, why do we actually expect to see neutrinos from astrophysical sources and what do they tell us? Well, that gets us into sort of another area of astroparticle physics that are already mentioned and that is cosmic rays. What are cosmic rays? Cosmic rays are particles that rain on to the Earth's atmosphere from outer space at a very, very large range of energies all the time. And the lower energy ones you can actually see they produce things like this aurora seen over the South Pole here. The more energetic ones you can't see. These particles are a hundred-year-old mystery. They were first detected or discovered about a hundred years ago, and it is fair to say that, apart from those that we know come from the sun, we have no idea where they are coming from. This man who discovered them in 1912, Victor Hess, and I will tell you a little bit about how these particles were actually discovered and a little bit of the history of cosmic rays because that leads us to experiments like the Pierre Auger Observatory and the IceCube detector. So, 1912, this is the heyday of radioactivity. People, physicist discover radioactive elements and they make experiments with radioactivity. And a detector to detect radio activity is this electroscope. Electroscopes are usually in the presence of radioactive material with discharge. What people like Victor Hess found out is that electroscopes discharge even if there is no radioactive material in their neighborhood. So you just have an electroscope, you charge it and you usually would put the radioactive material next to it and it would discharge, but even if you don't put that radioactive material next to it, it will discharge over time. A little bit slower but it will eventually discharge. So people thought, well, it must be the Earth. So there must be some radioactive radiation coming from the Earth. The Earth must be radioactive. And Victor Hess figured the easiest way to test this, well, easy, is to take one of these electroscopes and go up in a balloon and check whether at a certain altitude they don't discharge anymore. So if the radioactivity is coming from Earth, if you go up into a balloon, it should get weaker. That's what he did. Here's a picture of his measurements. He went all the way up to 17,500 feet without an air mask or anything. That's commitment. But what he found is something very interesting, especially if you looked at the previous plot where he had actually a tie when he went up there on the balloon. He found that the radiation level doesn't go down, actually it goes up. So that means this radiation cannot come from Earth, it must come from outer space. Thus, the name cosmic rays. So he had discovered that there is a constant rain of particles on to the Earth's atmosphere from outer space. So now comes a little bit of history. People at the beginning didn't know what are these particles. So are they nuclei, are they maybe electrons, maybe they're photons. Robert Millikan, very influential physicist, a Noble Laureate at the time, thought they were photons, and he coined the name cosmic rays and it stuck. So these particles are not rays, they are really particles. And the name cosmic rays just stuck. So after they discover the chemical nature of this cosmic radiation was unclear for some time. And this name, cosmic rays, reflects Millikan's belief that they were gamma rays from space. However, his student, Arthur Compton, also a Noble Laureate later, discovered that these particles are mainly energetic particles, so real particles. And we know now that most high energy cosmic rays are protons and heavier nuclei. So just to give you an idea that at the time, this was in the '30s, this type of science was still on the front page of the New York Times. So here's an article from a physics convention in Atlantic City, where it says "Millikan retorts hotly to Compton," his student, "in cosmic ray clash. Debate of rival theorists brings drama to session of nation's scientists. Their data at variance. New findings of his ex-pupil lead to thrust by Millikan at less cautious work." So a day later, Millikan denies a clash on theories, and it says, "Scientist protest that the word 'incautious' was not aimed at Compton." He disclaims any coolness. And this is his letter to the editor of the New York Times. Now, a couple of years later, the worst thing that can happen to somebody like Millikan happened to him and that is his own data proved Compton right. So Millikan's data confirmed Compton. There's other cosmic ray studies at Panama that tend to back rival's ideas. So he was a fair player so he admitted that cosmic rays are actually particles. If you go to the New York Times archive in the '30s and you look for Millikan, search for Millikan, you find tons of articles. He had a very good connection to the New York Times, apparently. And one of the more interesting articles that I found quite funny is the fact that the Hayden Observatory in the American Museum of Natural History was actually opened by a cosmic ray particle. So it says here, "Cosmic ray to open planetarium tonight." It says "a cosmic ray messenger from interstellar

space will switch on the stars tonight promptly at 9

00 o'clock in New York's first artificial heaven." And I like this paragraph over here. "So far as is known, this will be the first time that a cosmic ray will be made to perform a task at the bidding of man."

LAUGHTER

space will switch on the stars tonight promptly at 9

So, what do we know about cosmic rays now? Cosmic rays are charged particles, mainly protons, heavier nuclei, that continuously rain down on Earth from outer space. A small fraction of them have energies in excess of several joules. That makes them the highest energy particles in the known universe. And the question is, where do they come from? Do they point back to their sources? Remember they are charged so they should be deflected in magnetic fields on their way. We don't know how strong these magnetic fields are so we don't know how much they are deflected. But our hope is that we can, at the highest energies where the deflection is not that large, do astronomy with these particles. And the question, where do cosmic rays coming from, where are cosmic rays coming from, was one of the 11 greatest unanswered questions in physics in this issue of Discover magazine. Still unanswered so don't expect the answer in this talk. One of the things I should mention, because I'm going to show a couple of graphs now, is that we don't usually, for elementary particles, use the unit joule for their energy. We use a different unit because typically particles have an energy that is a tiny fraction of a joule, like 10 to the minus 18 or so. And we don't want to carry the 10 to the minus 18 around all the time. So what we do is we use a different unit for energy and that is called the electron volt. And that is essentially the energy that an electron would gain if it goes through a potential difference of one volt. So if you have a one-volt battery over here and you have an electron here and let it go, it will, of course, be attracted by the positive part here, and on this path it will gain an energy of one electron volt. One electron volt is therefore 1.6 times 10 to the minus 19 joule. So that's a very convenient unit to describe the energy of elementary particles. It turns out, for cosmic rays, we were will a little bit too smart there because this is the cosmic ray energy spectrum. So this the is flux of cosmic rays as a function of energy. So it goes from 10 to the 9 electron volts all the way to 10 to the 20 electron volts. So now we have, because cosmic rays have such a high energy, now we again have these high powers of 10 because they almost have an energy of one joule. A cosmic ray at these energies that comes in and, for example, hits you would have the same energy as a baseball. It's not the same effect. It will just zip through you so you don't have to worry. But it has the same energy has a baseball hitting you. So it's quite a lot of energy. So these particles hit Earth's atmosphere with this wide range of energies but the flux goes down very, very quickly with energy. So at the low energies you have a lot of flux and at the high energies the flux drops very, very quickly. As a matter of fact, the flux on this range over here where particles are essentially, in this range we know that these particles are coming from the sun. In this range over here we have no idea where they're coming from. Here we have about one particle per square meter per second. Here we have about one particle per square meter per year. And down here are the very highest energies. We have one particle per square kilometer per year. So you can imagine that you need very, very big detectors to detect these particles over here. So, as I said before, so far the origin of these particles is completely unknown. There's been know astrophysical object ever positively identified as an accelerator of high energy cosmic rays. So somewhere in the universe there are sources that can accelerate particles to energies that are equivalent to the energy of a baseball in full flight. So, this is actually the energy frontier in physics and that is also a good place to look for new physics at these very, very high energies. And that is the reason why we are so interested in these particles. So this brings us to Enrico Fermi, a very, very famous and bright physicist, who was the first to propose a mechanism how these particles can actually be accelerated to these energies. And that was as early as 1949. And his idea was, in this famous paper on the origin of cosmic radiation, his idea was that these particles are not accelerated in one go but in a very tedious process, that I will explain on the next slide, that accelerates them a little bit and again a little bit and again a little bit. In about 10 to the 7 or so times you do a little bit and you get a lot in the end. So it's a very incremental process. And the way it works is these particles are accelerated when they are reflected by moving shock waves. I will explain that on the next slide. That the process is called shock acceleration and this is how it goes. You have a cosmic ray particle and that hits some area where you have a strong magnetic field. And then this magnetic field gets all scrambled and then it leaves in this direction. And usually if this was at rest, the incoming cosmic ray and the outgoing cosmic ray would have the same velocity. So no gain there. But Fermi's idea was that if this magnetic field blob is actually moving and this particle is going in there and this magnetic field blob is moving against it, then it comes out with a little bit of a higher energy. That's the same idea, it's not quite right but I will say it anyway, if you throw a tennis ball against the wall it comes back with the same velocity. But if this wall was moving towards you, it would get a little bit of the kinetic energy of the wall and be a little bit faster when it get reflected. And that's the idea here. So the particle goes in, gets reflected by this sort of magnetic wall, and then comes out again with a little bit of higher energy. There needs to be magnetic fields here too so the particle bends around and goes against this wall again and again and again, 10 to the 7 times or so and gains a little bit of energy in all of these processes. So after 10 to the 7 of those crosses of this wall over here it has enough energy to explain the highest energy cosmic rays. If you're interested, you can probably Google it and find it very quickly, in the, I don't know, '50s, Frank Capra made a bunch of movies about signs, the sun and so on. He made one about cosmic rays and has a very funny cartoon about Fermi acceleration. So maybe you can find it. I have it on a different computer. I didn't think of copying it over, but it's a very nice 10-minute or so movie about how this goes. So the question is now, this is a nice idea but is there anything like this in the universe? Is there a moving shock front in the universe that could do it? The answer is yes, there are a lot of them. Here's one example. It's a supernova remnant, Cas A. This is a picture in x-rays. And of course, you know what happens in a supernova remnant is that you have shocks going out. This thing explodes into space, and here's where it explodes and plows through the material over here. Here is where you would have Fermi acceleration, when this front from the supernova sort of plows through the material that is around here. So supernovae, therefore, can account for cosmic rays all the way up to energies of about 10 to the 16 electron volts. But at that point, they run out of steam. So we've seen that it goes all the way to 10 to the 20 electron volts. So the last couple of orders of magnitude we can't do with supernova. So above this energy we think that cosmic rays are accelerated in more extreme sources, most likely outside of our own galaxy. And here's one example. This is what's called an active galactic nuclei. This is a super massive black hole that sits over here. It has 10 to the 8 solar masses. This is an artist's rendition. There's material accumulating onto that black hole, and by a mechanism that's not yet understood, two jets are coming out, one in this direction and one in this direction, and in these jets material is ejected. So you have material coming out over here, and when this material goes along the jet that's one of these shock fronts where we can have Fermi acceleration. So we think that active galactic nuclei are possible sources of the highest energy cosmic rays. I will not get more into this source business and this theoretical business because it is really not known whether these sources actually do that. This is just an idea that we know of sources that could potentially do that. By the way, these jets have been seen, of course, at other wave lengths. This is one of these AGN, and you can see one of its jets over here. And they extend quite a bit into space. You usually only see one because the other one goes in the other direction, and you only see the one that goes in your direction because it's Doppler boosted. So if you go back to that sketch of the cosmic ray flux as a function of energy, we think that in this range over here cosmic rays are of galactic origin, they originate from somewhere in our galaxy, for example in supernova. And here the highest energy end they come from sources like these active galactic nuclei, these supermassive black holes at this part over here. By the way, if you think about sources like the LHC and Tevatron, man-made accelerators, these people usually tell you they are the energy frontier of physics. As you can see over here, we can do quite a bit better. So the LHC is still affect of 16 or below the energy of what nature can do as far as particle acceleration goes. So, let's get to the experimental part then. How can we find the source of cosmic rays? And this is, of course, the most important question we have to ask is, do they point back to their sources? If they don't, there's no point in detecting them on Earth. If they get all scrambled in magnetic fields, then they don't point back to where they're coming from. So making a map of these particles by detecting them wouldn't be any use. So we have to answer the question, is astronomy with charged particles possible? And if it's true what we believe about magnetic fields, then only at the very highest energies, 10 to the 19 electron volts, 10 to the 20 electron volts, in this tail over here where we have very little flux over here, only in this range over here will these particles point back to their sources. So these are the cosmic rays that we need to detect, the ones with the very, very highest energies. So this is a worldwide effort. You can see a bunch of experiments that try this kind of physics or related physics over here. A couple of experiments that try to detect gamma rays sources like VERITAS in Arizona and HESS in Namibia, named after Victor Hess, the discoverer of cosmic rays. Some experiments try to detect these cosmic rays directly. They're cosmic ray detectors. That's the HiRes experiment in Utah and the Pierre Auger Observatory, the biggest one of them, which I will introduce next. And then the experiments that try to detect neutrinos. One of them is IceCube, of course, and the other one is currently built in the Mediterranean, that's the ANTARES Project. So I'll talk more closely about the Pierre Auger Observatory and the IceCube detector. So it's going to be a tale of two experiments. There's a picture of the Pierre Auger Observatory, and there's kind of a picture of IceCube. So let's look at that flux plot again before we start. Once again, it roughly looked like this. By the way people have come up with funny names for this little feature here. This is the knee and this is the ankle. Sure. Then the flux here, as I said, is one particle per square meter per second. Here, the knee, is one particle per square meter per year. And the ones we're interested in, the ankle particles over here, are one particle per square kilometer per year. So that's a very, very tiny flux. The only way you can beat that is to wait a hundred years, not a good idea, or to build an experiment that has many, many square kilometers. And that's the way we're going now. So, there is another problem with that and that is that this flux over here, one particle per square meter per second, you can set up a satellite above the Earth's atmosphere and detect these particles. They come at, a satellite is one square meter, so you get one particle per second so you get a lot of statistics quickly. Here, you not only get the same statistics, you would have to send up a satellite that is a square kilometer size. That's not possible, so we have to build these experiments on Earth. You would say, why not? Why is this a problem? Why would you send up a satellite if you don't have to? Well, you cannot detect the original cosmic ray particles on Earth. And the reason for that comes here in a computer simulation of what happens to a particle that comes in. So let's say this is a cosmic ray proton coming in, going into the Earth's atmosphere. And at some point it's going to hit and interact with an air molecule. What happens then is it splits this air molecule up, and there's a little bit of a mess created at that point, many particles. These particles have still very high energy. They can, again, interact with air molecules. And again and again and again. So you're going to create what's called an extensive air shower which is about 10 to the 10 particles made out of water. Of course they have all lower energy so they energy balance is correct. But you can't see this particle over here. You can't see this mess. So you have to reconstruct the properties of that particle up here from this mess. At sea level, the only particle that's leftover are the ones in green over here. These are muons. And you probably know that apart from being bombarded by neutrinos all the time you're also bombarded by muons. Again, you won't notice that to don't panic. The muon flux at sea level is about one per square centimeter per second, one through your hand per minute. One through your hand or so per second. These are the only ones that are survived to see the world. So you can detect these particles, you cannot detect that particle over here. So the question is, by detecting these particles, can we get any information about those? So what you have to do then is you have to build a detector that covers a very, very vast area. There's a picture, it's an artist's rendition again of the Pierre Auger Observatory in Argentina. Each of these dots is one particle detector. And you can see here this image of an air particle comes in here, creates this extensive air shower, and then the muons are left over and they hit the ground and they detect it by these detectors that sit on the ground over here. And this whole thing has an area of 3,000 square kilometers. And each one of these dots is one of these detectors over here. It's just a water tank, essentially. It's a tank with 11,000 or so liters of water. The other thing you can do, these particles don't only hit the ground but while this extensive air shower goes through the atmosphere, it also excites air molecules and they fluoresce back, so they produce light. And you can detect this light with light detectors. So the Auger Observatory also have, at four corners, sitting and having these kind of cameras that overlook that array for these fluoresce photons that come from this air shower. I'll show you pictures in a minute or so. It's an international effort. It's about 350 scientists at 72 institutions in 18 countries that are listed over here. You can go to this web page to learn more about this experiment. It's situated right here. Buenos Aires is over here. Santiago de Chile is over here. So you typically fly into Santiago, then to Mendoza, and then you have to drive about four hours to get to this experiment. Mendoza is the wine growing area of Argentina. It's a very nice place to be. Again, this is an experiment that is very difficult to photograph. Here someone tried. These tanks have a distance of about a mile. So you can see four of them here lined up, and I will explain a little bit what these detectors. So it's 3,000 square kilometers area. Each of these red dots is one of these detectors. It takes you quite a while to get to a detector that is broken or something. 3,000 square kilometers, that's the size of Delaware. Now, for a state that's not very big. For a physics experiment it is very, very big. This is one of those tanks. Each tank has a name. This one is "dad". This one also found a friend.

LAUGHTER

space will switch on the stars tonight promptly at 9

Here's a picture of what happens on one of these. So this front of air shower particles comes down and it's a very thin pancake of particles and these tanks over here get hit first and then these get hit later. And from the relative arrival times of these tanks, you can reconstruct the direction where that shower or that pancake came from. And that is also the direction of the original cosmic ray particle. This is essentially how this works. The fluorescence detector, that's a picture of one of the fluorescence detectors, they can only observe at night because there's too much light around during the day. You can't see the faint light that comes from this air shower. So you can only do that in dark moonless nights, actually. Even the moonlight is too abundant for this. So they are here in these six buildings. Each one of them has a telescope in it. And they are closed during the day, and they are open during the night. And this is one of these cameras. So the air shower comes down over here. The light gets reflected into this mirror on to this camera of light detectors which is shown over here. And you can see the air shower going down in this camera when it happens. I'll show you a event, a typical event. So this is a picture of one of these ground detectors and one of those fluorescence detectors over here. What the ground detector will typically see is something like this, the footprint of the shower on the ground. The size of the blob is how many particles hit that particular tank. So you can see that the shower hits somewhere around here. And from the relative timing you get the arrival direction. And what the fluorescence detector sees is this track. It sees the track of the air shower in the atmosphere. And that's enough information to reconstruct the direction of the original cosmic ray particles. So the goal of the experiment then is to make a map of these particles. At the very highest energies we get about one of these particles per month. At the lowest energies where the flux is higher, we get a couple of them per second. We're only interested in the highest energy ones, as I said before, because these are the least deflected so they might point back to their sources. This is a picture of what, sort of on a sky map, what these arrival directions look like. And the game now is to find the sources by looking if there are any clusters here in arrival directions. The first thing you see is this looks pretty isotropic. So it looks like particles come from everywhere. These particles come from everywhere. I mean, if there had been one particular spot only on this map with all these dots on that spot, you would know that is the source of cosmic rays. But this kind of map doesn't give you the answer. It tells you they are coming essentially from everywhere. So the question is, are there sources everywhere and that the reason they come from everywhere or do these particles just get deflected in magnetic fields to such an extent that they are scrambled when they arrive here? So we have just started this experiment so the idea is that over time maybe some sort of hot spots in this sky map start to show up. So regions in the sky where there are more particles coming from than from other regions over here. By the way, this is the northern sky, which you can't see from Argentina, so this is the region of the sky that is accessible for the experiment. You can see some of these hot spots but they will also, of course, just happen by chance. So you really have to do this experiment over a couple of years to see whether any of these spots turn out to the sources. There's an interesting sort of first result that came out a couple of years ago with the first couple of events we had. This is, again, the same sky map. The cosmic rays are the circles over here together with the error bars so they have about a couple of degree error bar. The red spots over here are these active collective nuclei, these supermassive black holes that I explained earlier. And you can see that they seem to accumulate around here, and the cosmic ray arrival directions also seem to accumulate in this region. By the way, this white dot is the closest of these supermassive black holes. So us that's Cen A. And you can see that there seems to be a clustering of events in the neighborhood of Cen A. This has to be watched in the next couple of years to see whether that is actually true. But you see this is a very tedious game. The answer is not obvious. You have to run this over and over years and try to find out if there will be any clusters of cosmic rays. And as I said before, it is not very clear that these particles do show, point back to their sources. As I said before, we don't know how strong magnetic fields are so we don't know how strong deflections in these magnetic fields are. It could be that even in 20 years of running this experiment, the sky map is still isotropic and we have not identified the sources of cosmic rays. That would also tell us something, for example, about the strength of magnetic fields because that means even from the nearest sources of cosmic rays there would be so much scrambling that we can't detect them on Earth. But it is a risk so the question is, can we do sort of better than using the protons itself. The protons get deflected so we're always at risk of not seeing the sources because scrambling in magnetic fields is too strong. So to summarize that experiment, the first indications that the cosmic ray flux might not be isotropic at the highest energies, this Cen A region, is an interesting region that we want to watch over the next couple of years. But, again, no source is obviously visible on the sky in this map. And the question is, maybe these particles do not point back to their sources. Galactic magnetic fields and intergalactic magnetic fields might scramble the arrival directions even at the highest energies. The good news now is that cosmic rays are not the only particles you can look at. They are not the only messenger particles from these high energy sources. Where there are cosmic rays, there are always also neutrinos, and that brings us to neutrino production. Neutrinos are the ideal messenger particles, obviously. They propagate in a straight line. They are not affected by magnetic fields. If there is something in the way, they will just zip through. They're not easily absorbed. They can get away from the source. They are produced wherever cosmic rays are produced. So in this sketch over here you can see that this is one of these particles that does Fermi acceleration, so it gives back and forth of this shock wave. So we have an accelerated proton over here. But typically you have a lot of light around these sources. So this proton will eventually hit a photon, a light particle, and produce all kinds of other particles. For example, pions, and whenever you have a pion they decay and produce neutrinos. So whenever you have a proton, you also have a bunch of neutrinos. So maybe it's a good idea not to look for these protons themselves but to look for the neutrinos that will go with them. As I said before, there are a couple of advantages to these neutrinos. They are not deflected. They are not easily absorbed. But the problem is they are also not easily absorbed in your detector. So they zip through any environment that they want to escape from. But they also zip through your detector. So you have to do something about that. You need big detectors. You need cubic kilometer-sized detectors. So let's list the requirements for a neutrino detector. If you want to detect these particles, you need a large detector volume. Cubic kilometer. If you want to detect a few neutrinos from astrophysical sources per year. You can't see the neutrino itself. So you need this neutrino to convert into something else before you detect it. You've seen this picture, and I will show it again, of this bubble chamber picture at Argonne lab. You don't see the neutrino coming in but you see the muon or pion or whatever it produces once it hits something in that detector, some atom or molecule. So you want a huge detector material where the neutrino that comes in actually interacts. And once it interacts and produces muon you can detect that muon. So you need a huge detector material so the neutrino can actually interact so you can detect it. So, you would typically have the neutrino interact and produce a muon but muons and neutrinos are not the only ones around. There's an incredible amount of muons, as you have seen, all around. So you need to shield your detector from that one muon going through your hand every second, large muon flux that hits the Earth. So you have to build this detector deep down in something, in a mine or wherever there's shielding material above, some absorbing material above. However, if you build your detector in an absorbing material, you want to detect that muon that is produced by neutrinos and you can only do that, for example, by the light photons that are producers. But these photons have to go through your detector to be detected. So you want an absorbing material that absorbs the neutrino, that shields it against the muons, but light has to travel through that material, nevertheless. And there's really only one material that does that and that's ice, water too, but let's start with ice. Blue light travels 200 more meters in ice without being absorbed. So here is a medium where a neutrino can interact and lots of it. If you built your detector deep in the ice and shield it against the muons in the atmosphere that rain down on the ice from above and you have this feature that light travels so you can still detect light that's produced by that muon. So that gets us to IceCube. Here's another picture of IceCube. This time all the 86 strings are shown over here. So IceCube has about 5,000 or so light detectors buried in the ice and 86 strings. So each of these lines over here is a string. There's nothing in the first mile but then between 1,500 and 2,500 meters, every 30 meters, sorry every 17 meters, you have one of these light detectors over here. And the distance between these strings is 125 meters. You can see that on the top over here. So that gives you about a one cubic kilometer or one billion tons of instrumented ice over here. So light that is produced in the ice you will detect with these 5,000 modules. So this is one of them. I'll get into the details what this is. They're all hanging on these strings so this is one of them, and then you go 17 meters down there will be the next one. And each one of these is one of these strings. So, again, when you hear a talk about particle physics, these days, LHC frontier physics, they show you, I think it was the ATLAS detector. Here you can see two people sitting next to it. But just for you so you get the idea, this is the IceCube. If you want to scale them, this is what it will look like. So this is definitely quite a bit bigger. So, IceCube was built during several austral summers, and then we took data during the rest of the year until the next austral summer production would continue. And that gives you a little bit of a time line of IceCube. This is the first string. It was built in the 2004-2005 season. Not much you can do with one string. It's just it test. It doesn't work at all. And the next year we added nine more. So we added eight more to get nine strings. That's called IC9, IceCube 9, and we run that in the 2005-2006, we built it in the 2005-2006 season and then ran it for a year. So there are already physics results out for that tiny little detector over here. Then in the next season, '06-'07, we added several of them, and so we ended up with IC22, 22 strings. And then in the next season came IC40. And then IC59. That was in the 2008-2009. We are currently analyzing data of this particular configuration. I'll show you a little bit of that towards the end. And then we had IC79 and there were only five or so left and they were finished last December in the '10-'11 season. And now we have IC86, and that is the completed detector. So let's talk about the IceCube concept one more time. Again, this is IceCube. So it's a lattice of photomultipliers. Photomultipliers are devices that can detect light. They can actually detected single photons. Your eye can, too, actually. But these are usually used in physics to detect light particles. So what happens? You have this shielded and transparent medium in the ice over here and a neutrino comes in. Here we go. A neutrino travels through the Earth. Most of them won't do anything. They will just travel through the IceCube and don't do anything and we won't see them. Some of them, a very, very tiny fraction but remember there are billions of those so we can wait, will actually interact somewhere near the detector in the bedrock or in the ice. And they produce, in their nuclear interaction, they will produce a muon. And that muon, muons are essentially electrons with a heavier mass, will just travel through the ice. Muons travel very large distances through material. So the muon will travel through the detector over here, and it has the same direction as the neutrino that came in. So from the direction of the muon, if we detect the muon, we can detect the direction of the neutrino. But we cannot detect the muon we have to find directly, we have to find some way to detect it. Luckily, what's happening is that a particle, like this one that is charged, goes through material at a speed faster than light will produce what's called Cherenkov radiation. Now, you will say that's stupid, nothing can be faster than the speed of light. That is true, nothing can be go faster than the speed of light in a vacuum. But the speed of light in water is less than the speed of light in vacuum. The speed of light in water is about three-quarters or so the speed of light in vacuum. So a particle can be faster than the speed of light in water or in ice. So a particle, like this muon, will typically have a speed that's pretty close to the speed of light in vacuum. So it will be faster than the speed of light in the ice. And what's happening then is you produce a Cherenkov light. This, by the way, this picture of this interaction over here is a little bit like what I showed you earlier. I can rotate this and put it over here. You can see the neutrino, you can see the neutrino coming in over here, but then it produces these three particles and one of them is the muon. And that's this particle over here. So it's exactly the same interaction, more or less, than the one I showed earlier in this bubble chamber picture produced at the lab. So the muon produces Cherenkov light and these photomultipliers are light detectors, as I said, so they will detect that light. So this is what Cherenkov light sort of looks like. It comes in a cone and these detectors will then detect the light. And again, from the relative timing, these tubes getting hit earlier than these ones over here, you can get the direction of this muon and, therefore, of this neutrino in the ice. So Cherenkov light is a little bit like you have an equivalent for Cherenkov light when you have a supersonic boost. So this is an airplane going faster than the speed of sound in air it will produce, essentially, the same thing in acoustics as this Cherenkov light is in optics. So Cherenkov light is essentially what supersonic boosts are in acoustics, transferred to optics. So you can actually see Cherenkov light. This is a nuclear reactor and there's water around it. So there are lots of neutrinos coming from this, lots of muons coming from this and other particles coming from this nuclear source over here. And you can see the Cherenkov light by this blue glow that you can see around nuclear reactors. So, here's an IceCube event. So this is how this goes. The particle comes in and the light hits these tubes and from the relative timing, this one gets hit early than these ones, you can reconstruct the direction of these particles. In this particular case, one out of 48 of these 5,000 optical detectors were hit, and you can reconstruct the arrival directions of the original particle to about a degree. That's not good for optical astronomy, but this is not optical astronomy. It's a new area of astronomy with particles, and for that it's actually pretty good. So this is the IceCube collaboration. It's nine countries, 36 institutions, about 260 collaborators from all over the world, the US, Brussels, Switzerland, UK, Germany, Japan, Sweden. And this is the collaboration. So, once again, in order to get these strings into the ice you need to drill holes. So this five megawatt hot water drilling system was actually developed here at UW Madison at the Physical Science Lab which is towards Stockton. So what you need is a hot water generator. Hot water is pumped into the ice. Then this string with all these tubes attached is put into the ice and then it refreezes. And if something broke, there's no chance that you will every get to that tube again. Luckily, less than a percent of these are actually broken at this point. So this is a very safe procedure, and this whole drilling procedure was a real success. So this what it looks like. So this is how this string with tubes is entered into the ice. So this photomultiplier tube starts its journey to a depth of 2,500 meters. So this is what these optical modules, these photomultiplier tubes look like. This is the actual photomultiplier. So this is the light sensitive part. This is the photomultiplier tube. But you need to get this signal that this detector sees all the way up to your computers on top of the ice. So the signal is actually digitized right in this glass sphere over here. So what goes up along the string is the already digitized signal. So there's no analog signal going up. It's all digitized right here and then sent up. So everything that has to happen to a signal happens already right here and then it goes up. And the event is sort of put together on computers at the surface. So, why is this photo tube pointing down? Why is it not also pointing up? Well, IceCube is actually, this will be surprising, it's at the South Pole but it actually observes the northern sky. It's a detector that makes a sky map of the northern sky. Why is that? The reason is because the particles that we actually do want to detect come on top of a huge background of other particles. So, for example, if a cosmic ray at any energy, at low energies for example, comes into the atmosphere it produces this huge air shower that I showed you on a computer simulation earlier and that produces lots of neutrinos. These are not neutrinos that you actually want to see because they come from cosmic ray air showers that come from low energy particles that got scrambled all the way. So you really want to see the neutrino coming from the source, not the cosmic ray. The idea was to look for the neutrino because that one does not get deflected. But these cosmic rays produce neutrinos in the atmosphere. So you gut swamped by these neutrinos in the atmosphere. So what do you do about that? Well, so here's IceCube in the South Pole. Everything is flipped around. So there are lots of neutrinos produced in the atmosphere and they rain down on to IceCube and they produce a lot of muons in the atmosphere, too. And all of these muons are background for IceCube. But those muons that are produced on the other side here, they get all filled up by the Earth. So you want to look down with your detector because that is the region where you have much less background from muons. So you want the Earth as a filter against atmospheric muons, muons that are produced in the atmosphere. And they are produced at a rate about 1,500 per second. So if the particle you're looking for sends you one or two neutrinos per year and you get 1,500 per second from this background, you want to make sure you get rid of this background. That's why we are looking down. So these muons that are produced on the other side of the Earth or anywhere around here, they get filtered by the Earth, and you really only see the ones that are produced by neutrinos. However, there is still another background so you can filter these out by looking for upward going events in your detector. But that's why the predecessor of IceCube, AMANDA, had this nice little sort of picture of a penguin looking down through the Earth. So it watches sources in the northern hemisphere over here. But there's one unavoidable background and that is if you have neutrinos produced in the other side, they will, of course, go through the Earth. So you can filter against those muons produced over here but not against the neutrinos. That's a background of about 10 per hour that you have to live with. So you have to find your astrophysical signal on top of that background. So in the last 10 minutes or so let me introduce a couple of results that we already have. Once again, this is a game of producing sky maps. So making maps of the arrival directions of neutrinos and then looking for hot spots. And this is one of these maps. This map actually runs counter to what I said earlier, that they only observe the northern sky, you can see the southern sky in it, too. We have actually also gone to making maps of the southern sky, but the sensitivity is much less than in the north. So this is all, essentially, background, and we apply really strong cuts in the data to reduce that background. We are not very sensitive but we nevertheless make a picture of the southern sky, too. But now you can look for, again, hot spots of arrival directions and you might say well there is something maybe here. But then again you have to ask how often does that happen by chance in a map that has so many events. And this particular one over here has a chance of happening about 18%. So this kind of thing happens all the time. We need something that is hotter than this particular spot. But this is just the beginning. This is one year of data when the detector was not complete. I think this is IC40. So, again, over the years, this will accumulate. And we will try to find hot spots of arrival directions in this kind of map. How do we know that this detector works if we don't have a standard source? Astronomers, when they build something they build a new telescope and they look at one of the billions of stars that they have. And they can detect it, is it at the right place? Great, the telescope works. We can't do that because there is no established source of neutrinos. So, what do we do? Well, there is one interesting thing and that is the fact that cosmic rays actually get blocked by the moon. So if you have a flux of cosmic rays here they can get blocked by the moon. So if you look at the cosmic ray flux, the down going muon flux in IceCube, you should see a shadow at the position of the moon. And we actually do see that. So this is the picture of down going, so this is not upward going neutrinos, the ones we are after, this is down going cosmic rays that we also see because of the muon content. You can see that there is a dip of a couple of thousand events in the direction of the moon. And that gives us confidence that we're actually looking in the right direction. And that our programs that calculate the arrival directions of these particles actually do work. So this is not a standard -- not a source for us but this is one of the things that this kind of experiment has to do at the beginning, and we have to watch and monitor this moon shadow over the years to know we are still pointing in the right direction. But, of course, we will soon find the source of cosmic rays, of neutrinos, and then we'll use that as the standard count. But before we have that we can use the moon shadow. Let me give you one other example of a result that was completely unexpected that we already have in the first couple of years of data. That has to do with the fact that I already mentioned a couple of times that IceCube is not only a neutrino detector, so we're not looking for upward going neutrinos in the detector from the northern sky, but, of course, we also see muons that rain down on to the South Pole from cosmic rays in the southern hemisphere. So, again, this is one of these huge air showers produced by a cosmic ray above the South Pole, and the muons that go down over here, they will go through our detector and usually we will toss these events out because these are background events or we're looking for something that's upward going. But we keep these particles and there's a good reason for doing that because these muons over here essentially point back into the direction of the cosmic rays that produce them. And they come from cosmic rays that have about 10 TeV or so in energy. So they're very low energy cosmic rays. So what we're doing is we're making sky maps of these cosmic ray arrival directions. So we're making a sky map of the southern hemisphere of what the sky looks like in cosmic rays in at TeV energies. And you would say, why is that? TeV energies, for sure these cosmic rays are completely scrambled in magnetic fields. I said before, TeV is 10 to the 12 electron volts and the ones we are looking for are 10 to the 20. So we are eight orders of magnitude below. So sure these particles will be scrambled. The interesting thing is what we wanted to sort of see in -- that there are hot spots on the sky, which we didn't, we see in these cosmic rays. And this is the arrival direction distribution of cosmic rays of TeV energies in IceCube. And you can see it is not isotropic. There are hot spots. There's this region over here, for example, where you get more events than from the other regions. And there are sinks. These blue regions are regions where there are not that many cosmic rays coming from. It's a tiny effect, this is the order of 10 to the minus four of total flux. But it's pretty significant if you have 32 billion downward going events per year at this detector. So we can see in the cosmic ray sky at these energies hot spots of cosmic ray arrival directions. That's not entirely unexpected, that's why we keep this data, because experiments in the northern hemisphere have already observed that in the northern hemisphere, too. This is a picture of the Milagro, that's the water tank that I mentioned earlier. They can also detect cosmic rays in the northern hemisphere. And they see these regions over here of sort of increased flux of cosmic rays, and they seem to continue into the southern hemisphere. So we now have sort of a, with IceCube entering the game here, we have a complete picture of TeV cosmic rays and they are not isotropic as they should be. And nobody has any idea why that is. So I cannot tell you why this happens. One of the most important supernova remnants sits right here. But it's 300 parsecs away and scrambling radius of particles of this energy magnetic field is 0.001 parsec. So you have many, many, many orders of magnitude smaller in your scrambling than the distance to this nearest possible source of cosmic ray. So that can't be. It's a coincidence. So we don't know where these particles come from. It might have to do with galactic magnetic fields that sort of funnel particles in certain directions, but it's a completely open field and it's a completely serendipitous discovery of this detector. It's built as a neutrino detector but it also sees cosmic rays so you can make a scan of cosmic rays and you see the surprise. So we are hoping for many more surprises like that. I'll give you a list of sort of IceCube physics topics. Of course, we're going to find neutrino point sources. We want to find the origin of galactic and extragalactic cosmic rays. We can also study the atmospheric neutrino flux that has implications for particle physics. Hopefully, every 30 years or so a supernova goes off. Maybe we're lucky and we'll see one of them. Gamma ray bursts are very high energetic explosions in the universe that have been seen at other wavelengths. It will be a very important discovery if we see neutrinos from gamma ray bursts. The more you go down this list, the more speculative it gets. There's dark matter particles over here, relative magnetic nanoparticles, other exotic phenomena. So don't put your money on these but put them on those particles over here. So it's for a nonstandard model, neutrino interactions and so on. But the most important thing, and I will end my talk with that, is that, of course with penguins, but also with a quote by George Hale who said that, about a different telescope, "If we knew what the discoveries were likely to be, it would make no sense to build such a telescope."

LAUGHTER

space will switch on the stars tonight promptly at 9

Thank you.

APPLAUSE