Alan Weinstein - “Recent results on Gravitational Waves from LIGO and Virgo”

thanks Carter let's see if this is on can you hear me yep great it is a pleasure to be here I have to say I was at slack in the 80s I see some of the great gray beards in the audience that I remember as as good good old friends from those days and it's just always a pleasure to be back here at Stanford okay so I'm going to I've got a lot of things to talk about about some of the recent observations of gravitational waves made by LIGO so let's get started with gravitational waves 101 because I assume that not everybody here is an expert on the subject so let me I just need to remind you of some of the properties well what gravitational waves are and the properties that we expect them to have so uh matter according to Einstein matter and energy cause space and time to curve that's manifest at manifesto Avot II and if matter and energy are accelerating then they will change the gravitational field dynamically and the news of the rapidly changing gravitational field will propagate outward at the speed of light as gravitational waves gravitational waves have nothing whatsoever to do with light so if they travel at the speed of light that tells us more about space-time than it does about matter energy and the form of the energy in other words it travels at the fastest speed sustainable by space-time and light light shares the same property the the equations of general relativity let's see I think I have a decent the equations of general relativity these are Einsteins field equations are pretty darn complicated to solve in the even in the absence of matter on the right hand side in general relativity matter and energy tells space and time how to curve and space-time curvature tells matter and energy how to move even in the absence of matter in of matter so this right hand side is zero you still have dynamical space-time curvature and gravitational waves from black holes are vacuum solutions to Nine's field equations some of the basic properties as predicted by general relativity is that gravitational waves will propagate at the speed of light they stretch and squeeze the space between objects so by measuring the distance between objects we can see the strain which is a unitless quantity that they experience due to gravitational waves due to the curvature of space and time between those objects and the the disturbance propagates at the speed of light in the language of quantum mechanics the graviton the carrier of the gravitation of gravitational waves in the same way that the photon carries electromagnetic radiation the graviton would be massless the space-time distortions are transverse to the direction of propagation so if the waves are propagating along the z axis they stretch space in the x axis squeezed space in the y axis it's a it's a wave so it alternates by stretching and squeezing and because gravitational waves are described in general relativity as tensor fields as opposed to vector fields the pattern of their polarizations there are two of them plus + cross and that are related to one another by a 45 degree rotation and that's in contrast to dipole radiation from electromagnetic sources in which the two polarizations are 90 degrees apart because electromagnetic fields are vector fields and so this is a quantum language saying that the graviton is a spin to field a massless spin-2 field would have to illicit YZ plus or minus 2 and linear combinations of them are these + + cross polarizations I'm telling you all this this is all theory I'm an experimentalist I trust I'm Stein but I will verify okay so the field equations admit two wave-like propagating solutions the tensor the space-time tensor here's the time axis and here's the spatial part of this space-time tensor has these two polarizations plus + cross and what they do is stretch and this is a strain and what they do is stretch and squeeze space in a 45 to the two polarizations in this pattern and the strength of the gravity amplitude of the gravitational wave is of course proportional to G and inversely proportional to the distance to the source and it's also it's proportional to the quadrupole moment the acceleration of the quadrupole moment its quadrupole radiation not dipole radiation and and that's due to conservation laws actually and this so this strength can be calculated so you know just how strong a wave would be from a distance a source that some distance R with acceleration quadrupole moment and one of the best sources of accelerating quadrupole moments are dumbbells or two stars just orbiting each other and when those two stars orbit each other they obey Kepler's laws to lowest order which for a given mass relates the the frequency the orbital frequency to the distance between the two stars capital R so when the two stars are close together very close together their orbital frequencies are very high and the gravitational wave frequencies it's quadrupole radiation and the gravitational wave frequencies is twice the orbital frequency and so in order to get gravitational waves with frequencies in the the band accessible to the detectors are ground-based LIGO detectors that I'll talk more about in a minute we we need to look for gravitational waves with frequencies above say 20 Hertz which corresponds to orbital orbital frequencies of 10 Hertz so imagine stars orbiting each other 10 times a second by Kepler's laws they have to be just tens of kilometers apart so these are not ordinary stars these are compact stars neutron stars or black holes that are formed when massive stars burn all their nuclear fuel and collapse into a compact object blow out the outer layers in supernova and leave behind this compact object and if there are two of them in a binary then you will end up naturally forming these kinds of systems and they're very efficient gravitational wave emitters but they are pretty rare imagine that they're probably therefore very far away so if you put them at us distance on the order of two stellar mass objects separated by tens of kilometers orbiting by Kepler's law tens of times or hundreds of times or a thousand times a second and you put it far away like in the Andromeda galaxy maybe a mega parsec away you're talking about strains of ten to the minus twenty-one it's a very very small number gravitational waves do come in all kinds of frequencies not just the ones accessible to ground-based detectors lower frequency detector so the Miller Hertz band lower frequency gravitational waves can be detected by detectors in space this is a consent artist conception of the laser interferometer space antenna Lisa which we hope will fly in the next um in the next ten years fifteen years or so the LIGO detectors have arms they're Michelson interferometer with arms that are four kilometers long Lisa has arms of four million kilometers long on the other hand you can make even longer arms sensitive to lower wavelength gravitational waves if your arms are as long as many many light-years from for example distant pulsars and use them as timers and look for deviations in the timing associated with passing gravitational waves so this is Arecibo I think it's still there although the hurricane damaged it a bit but pulsar timing is sensitive to gravitational waves with nano hertz type frequencies corresponding maybe to black holes supermassive black holes in the core of orbiting and merging galaxies and at the very longest wavelengths lowest frequencies and longest wavelengths associated with the whole size of the universe if these kinds of waves were around during the early moments of the universe then they would have frozen in to the the Cosmic Microwave Background and be detectable today now there was some hint perhaps a few years ago of of a detection but it turned out to just be magnetic dust in our galaxy so the hunt is still on in all of these cases we think that we are on the threshold of discovery except for one in which we are past the threshold discovery that's the one I'm gonna concentrate on in this talk okay so the two LIGO detectors LIGO is the laser interferometer gravitational-wave Observatory we're built by the by Caltech and MIT for the National Science Foundation they are three thousand kilometers apart one is in southern Louisiana and the swamps of southern Louisiana the other is in the high dry desert of Hanford eastern Washington State it's where a lot of the work associated with nuclear power and nuclear weaponry was developed so it's one of the most polluted places on earth but it's still a fun place to visit I encourage you to do it anyway and the LIGO Laboratory is part of the LIGO scientific collaboration which has grown to be over a thousand scientists from 80 institutions 50 countries 15 this is big science now when you get to big science you no longer show a picture of a thousand smiling faces you just sew the logos but the logos are here actually and so here's Caltech and here Stanford Stanford's been involved since the very very beginning and has played major roles in many many aspects of developing the LIGO detectors which I don't have time to talk about and so I'm not going to tell you much about how the LIGO detectors work in fact there are people in the audience who can't talk about it with in greater detail than me I'll just remind you that they're Michelson interferometer with the laser and a gravitational wave goes by and it stretches one arm and it squeezes the other arm and the arms are nice and long so that the strain produces a measurable displacement of the masses on the order of 10 to the minus 19 meters okay and then we detect it and we are limited only by the noise in the detectors and here's a the frequency spectrum of the strain equivalent noise it's a you know a spectral density so it's in units of strain per root Hertz and it ranges from 20 Hertz to to a few kilohertz and it goes real really lousy at low frequencies due to all the motion of everything around in the material and the environment and it gets worse at high frequencies because the laser starts getting resolved into individual photons and the quantum shot noise of the individual photons shows up as noise at higher frequencies you can see lots of spectral lines the ones you can see are all well understood there things like the wire the the fibers that were used to hold up the mirrors have make these lines over here these violin votes there's 60 Hertz and multiples that somehow get in and there are calibration lines so all the lines are understood and can be modeled and don't really disturb our ability to detect gravitational waves very much at all we've been operating the the LIGO detectors now for quite some time on the almost twenty years for the first ten years or than 2000s we operated the initial LIGO detectors and for ten years we observed and observed nothing detected nothing other than noise that didn't dissuade us we knew how to build a better detector we spent five years building and installing the advanced LIGO detectors we turned on in September of 2015 detected our first event occur a week or two after that and we have detected a bunch of fence since then this is our first observing around our second observing runs ended in August and I'll be telling you about some of the results from these runs gravitational waves are emitted by anything including my my fist right now but I'm pretty weak gravitational wave detector but any kind of extremely energetic strong Quadra accelerating quadrupole moment will produce gravitational waves I will be focusing on the ones that we've detected so far which come from binary black holes and binary neutron stars we are also looking for one of each neutron star orbiting a black hole or vice versa other sources would be spinning neutron stars like pulsars we're only sensitive to them in our galaxy whereas these objects are usually very far away and the galaxies very very far away spinning neutron stars will produce continuous gravitational waves core collapse supernovae in our galaxy would produce beautiful very short duration gravitational wave bursts and ultimately we hope to find gravitational waves from the early universe as well and that will be quite a challenge I'm going to focus on the compact binaries so compact binaries binary neutron stars and binary black holes are of course really interesting in the case of black holes it's telling telling us about general relativity all these come from very massive stars that have collapsed we think so I'm the binary black holes will tell us quite a lot about general relativity and about the properties of gravitational waves in general and binary neutron stars that's much more complicated and I'll have more to say about that shortly in all cases what you have and maybe I'll just illustrate it here's the wave form which is a starts out being a sinusoid sweeping up in frequency as this saw has the sources orbit each other lose energy by gravitational radiation fall into smaller orbits move faster so the frequency some picks up and the amplitude picks up as well until finally there's a merger and maybe the best way to see that is with one of these little videos let's see if I can get one of these little videos to work so this is actually not an artist's conception there's a supercomputer simulation it's a solution to Einstein's field equations out of numerically it's a very non-trivial thing to be able to do here are the two black holes that's those are a cartoon and you can see them in spiraling here they're coming very close to that you can see the curvature of space you can see the flow of space and the flow of time right about here they merge together from two black holes into a single one if you've been following on the bottom you see the in spiral wave part of the wave form of the merger part of the wave form and then the ring down as this newly formed black hole emits sort of a ringing signal as it becomes quiescent so we observed for six months in 2015 and 2016 we observed a whole pile of events this is the loudness of the signal a whole pile of events that were consistent with our estimated background from instrumental noise you're gonna have instrumental noise but then we found three events that were all consistent with being high mass binary black hole systems and they were standing out above the background at least two of them this one is way above the background is what our first five signal event please don't ask me to explain what this is I don't have the time it's not interesting anyway this guy also turned out to be a 5 Sigma event and the number the nomenclature is gravitational wave discovered in 2015 September 4 14th a day that I will always remember of course this was Boxing Day the day after Christmas and this event because it was not sufficiently significant to call it a 5 Sigma or even a 3 sig move it happens to be a 2 Sigma event in statistical significance so we called it like over go trigger terrible name sorry about that ok three events from initial ly goes from the advanced Lagos first observing run the first one was the loudest it was spectacular got us one of these little things and you've probably seen this plot lots of time so I won't belabor it here's a time frequency spectrogram and here's two tenths of a second ok and here the signal is seen to sweep from around 30 Hertz to 500 or 400 Hertz and then died away and that's exactly what we were looking for and again it corresponds to the in spiral and then the merger and then the ring down of the final black hole and what you can see over here is that the velocity of these stars went from a third of the speed of light to 2/3 of a speed of light before you stop calling it a velocity anymore because they merged together so that's pretty good stellar-mass objects the size of Palo Alto orbiting at this at 2/3 of the speed of light before they smash together that's pretty good and here's the distance in separation in Schwarzschild radius down to one short-tailed radius these three events are kind of interesting in their own way are they're all black holes as far as we can tell they're all high mass systems as that if the black hole masses are get larger that means that their intrinsic amplitude is higher because mass is the source of gravity so the higher masses produce higher amplitude gravitational waves and these are these are standard sirens there they're not standard candles their standard sirens they not only have well-defined luminosity or energy release they have well-defined amplitude like sound if you like so we call them standard sirens so we know from how loud they are how far away they they must be so the more mass of a system is the more the louder it is also the more gravitational radiation therefore it emits and therefore the quicker it loses or energy and orbital angular momentum and the faster it in spirals and sweeps from low frequencies to high frequencies and the shorter it lasts in the LIGO band okay furthermore the larger the mass the bigger these Schwarzschild radii are and that means that they will they're further away when they start to touch than lower mass systems and by Kepler's laws that means that when they touch there they're orbiting at lower frequencies okay so a higher mass system is higher in amplitude sweeping faster in the band and then it merges this happens to be a lawrencium which I know all of you know and love this Lorentzian ring down happens at lower frequencies lower mass systems have lower amplitude this one happens to be closer sweeps over a longer period lasts longer in band over a second in our band and merges at higher frequencies okay so that's how we can by observing these waveforms and observing the frequency evolution of them we can measure the masses or some combination of the masses and this illustrates that but I think I already said it so we measure the masses here's m1 and m2 or m1 is greater than m2 so that's why this is grayed out and you can see that our highest mass system we mainly see the merger in the sweet spot of our sensitivity and the merger tells us that what the total mass is whereas for the other systems we observe the in spiral more of the in spiral and the in spiral is governed by this weird combination of the masses the so called chirp mass it's something like the product of the masses rather than the sum of the masses and so what we measure well in these contours is the product of the mass is something like that so that's why you see this sort of hyperbolic kind of locust and we measure the mass ratio poorly okay we can also measure the spins they have a subtle effect on the phase evolution and the spins are interesting if they're aligned with the orbital angular momentum which they might be if both stars formed in from the same swirling gas cloud that formed the massive stars that supernova and form the massive black holes then they're there their spins would be roughly aligned with the orbital angular momentum on the other hand if their spins are not aligned with the angular momentum then there's a spin orbit coupling similar to what you see in the hydrogen atom and that spin orbit coupling will torque the orbital plane and as a torps torques the orbital plane will see something face on edge on face on it'll be modulated and we can sense that modulation and measure the the this torquing and measure the spin and the precession of the spin very poorly very poorly unless we have very high signal-to-noise ratio so for our this is basically telling you that the two objects before they merged have spins that are maybe mostly pointing opposite to the orbital angular momentum may be perpendicular to the angular momentum but it's pretty poorly measured okay from these standard sirens we can measure the distances to these objects in there about the these two these left to loud ones were about five hundred megaparsecs away or about 1.3 billion light-years away and the more the less some the less loud one more like a Giga parsec away of course if they merged 1.3 billion light-years away it means they merged 1.3 billion years ago the universe was a little bit younger and maybe things have evolved a bit since then and that's a good Ashtyn physical question to ask another interesting thing that happened is that when these two black holes merge together is that the final black hole mass is less than the sum of the two original black hole masses in the case of our loudest event 1509 fourteen by three solar masses where do those three solar masses go they were radiated away in gravitational waves that's an awful lot of energy to be radiated by anything okay three solar masses is about 10 to the 54 herbs or a tenth of 55 Birds and that's a lot in gravitas it's 10 to the 80 gravitons and it corresponds to converting on the order of four or five percent of the total mass of the black holes into radiated energy that's an extraordinarily high radiation efficiency for about two-tenths of a second the luminosity at peak three point six times ten to the 56 urge per second of this event briefly for two-tenths of a second outshined all the electromagnetic radiation of all the stars and all the galaxies of the universe by a factor of 50 or so it was the most energetic event ever observed by humankind since the Big Bang and we haven't noticed these things before we're only just now starting to notice these things so here's we discovered two more in the second observing run and there's more to come by the way there's more to come so now we have started to build up if you like a mass spectrum okay one event is a you know is a discovery second event is a confirmation a third event is a distribution a fourth event is background and fifth event is calibration so we're moving along here are all the binary are the black holes that have observed from x-rays up until the era of gravitational wave astronomy somewhere between five and 20 solar masses our block the black holes that we observe are maybe because of Mom Alquist bias they're more massive because the more massive ones are louder easier to detect but it is still not surprising to see such high mass black holes we're not quite sure where they come from and we'd like to know and the way to find out these things is to collect more events and start to piece together a puzzle the way astronomers and botanists do by filling in something like a hertzsprung-russell diagram of by of black holes and that's the kind of thing that we're working on with more and more data but I'd like to focus on this guy over here this was the last one he just observed it a couple of months ago it was special for a couple of reasons one reason why it was special was that the our our sister detector in Italy Virgo had turned on just a week before they were not quite at the sensitivity of the LIGO detectors but they were on and they sought with a stat with a signal-to-noise ratio on the order floor nowhere near as loud as as the the more sensitive LIGO detectors but they saw it and it was in fact a good fit so this was a great triumph for the the Virgo detector and it started the beginning of a of a network of detectors of gravitational wave detectors and one of the reasons why it's important to have this network is because different detectors will have different orientations with respect to the polarization of the gravitational waves and by measuring the pattern of the amplitude of the signal in different detectors correcting for the sensitivity of the detectors you can infer the polarization of the gravitational waves remember I showed earlier the gravitational are predicted by Einstein to be tensorial that is to say plus and cross polarization transverse to the direction of motion well let's see if that's the case we now have three detectors and can start to disentangle this so there are other possible polarizations I pointed out these two these the ones that are only the only ones allowed in general relativity and they correspond here in this space-time tensor to these two the one one and one two and one two one and two two components of the spatial 3x3 metric tensor but there are other possible polarizations here's just the spatial part of the tensor you could imagine not just + + cross but you could also imagine x + y vector polarizations and those vector polarizations which stretch and squeeze space longitudinally as well as as well as transversely and then also breathing modes in which you're stretching space like in both x and y direction the same amount as well as longitudinal pure logic dodol modes in which you just stretch space along the z direction of motion so there's interesting patterns here and actually these two are degenerate in a quadrupole antenna so there's really five independent polarizations possible this would correspond maybe to a massive graviton spin to graviton that would have 5 velocities not just do and they have different and the are a quadrupole antenna gravitational wave detectors have different antenna patterns sensitivity here are the two arms of maybe a LIGO detector the LIGO detectors most sensitive to plus polarization at its zenith it has no sensitivity to vector polarizations at its zenith our order scalar instead it has more sensitivity either at 45 degrees with respect to the arms or at along the arms in the case of scalar type polarizations so by by we assume that these are the correct antenna patterns for gravitational waves is described by Jen Relativity but what about these maybe they're possible and maybe these patterns are possible you can do this now with three detectors with two detectors you cannot but from the amplitudes the pattern of amplitudes the ratios of amplitudes in the Hanford detector here I'm putting little little antenna patterns of the Hanford these are quadruple and tenant patterns these one these four plus polarization for Hanford Livingston and Virgo gravitational wave hits all of them and produces a different response and all of them according to their antenna pattern from this information first available only on October 14 we can now actually disentangle some of the ambiguity and actually make the first experimental measurement of the polarization of gravitational waves and what we find is that tenser pure tenser gravitational wave polarizations is favored over a pure vector by a Bayes factor of 200 that means it's really strongly favored pure tensor is strongly favored over pure vector or pure scalar these are extreme there's no good there's no theories that predict pure vector or pure scalar polarizations so these are pretty extreme cases we will need five detectors in order to disentangle fully of the polarization but this is the first time we've been able to make some statement about the polarization of gravitational waves from observation not just theory okay that's important to me we can do something else with three detectors we can locate the source of the sky very accurately so for example here Hanford and Livingston and vertigo detectors and the line between them is bisected by a circle on the sky and the timing between Hanford and Livingston tells us that the source must lie somewhere on this circle okay so you know these are antennas they're not telescopes we're not pointing in a place in the sky we can't point but we can use three of these antennas in order to do timing localization and we can use amplitude information to improve and find out where on this circle in the sky it is and if we have three detectors that gives three circles on the sky they'll all the linter sect in two places but from amplitude information we can disentangle it and find the location of the source now we're never gonna find the location of the source with great accuracy because these waves have have wavelengths of you know tens of thousands of kilometers or hundreds of thousands of kilometers and just from you know basic optics we know that the resin that the look the accuracy with which we can locate sources in the sky will be limited by the wavelength of the other of the of the waves that we're detecting and that corresponds to in our case on the order of a degree or so we'll never do better than that even with many sensitive detectors on and loud signals in the case of GW 1708 14 we made a ring in the sky between how we do something more sophisticated than that it's all Bayesian but you can think of it as there's a ring in the sky between the Hanford and living climbing have it conferred and Livingston timing the amplitude information locate localizes it to this this banana shape in blue and then we also have rings in the sky from Hanford in Virgo and Livingston in Virgo and that gives us the sky localization that's way way way better than we've been able to do with any of our black holes why is Sky localization important we want to tell telescopes to point in that direction of the sky this isn't a great localization this is 30 square degrees okay your typical powerful telescope you know has a field of view of one square degree or less okay so take a while for one telescope to go through that whole area and try to find something but they did we pointed dozens of telescopes at that direction in the sky and they look for a signal and you know they saw nothing well these are black holes you know black holes don't make light your telescopes are only sensitive to light sorry so that was very disappointing well I should say it was disappointing to me but because wouldn't it be great if black holes when they merged together did make light but there are things that do make like when they merge together and it happened three days later so here Wow okay well it really started I'm doing 30 million years ago and I'm ten thirty million years ago this happened music can't even turn down the music two neutron stars already each other they're about the smashed together don't worry about it and when they do they make a crapload of light and a huge amount of material that's all right shut this thing up okay and then comes the logo okay let's go back [Music] Thank You NASA let's go back I'm not gonna I'm going to turn this off and I'm gonna see if I can show you that when they merge together they make a lot of light they make relativistic Jets which start shooting out they eject an enormous amount of material dynamical ejecta and this is Neutron rich material okay flying out and the gamma rays smash through this dynamical ejecta and make and then it actually prevents it from smashing through and produces what's called what people call a cocoon producing a weak gamma ray jet at least that's the idea okay so let's see if that's what we found well in on August 17th I woke up to this this is another time frequency spectrogram now spanning 30 seconds and from about 30 Hertz on up to about 500 Hertz and maybe you see this thing coming all the way up like this and that is low-frequency sweeping to high frequencies it's a chirp and it happens for a long time which means this is a low mass system we see the same thing in LIGO Livingston and nothing in Virgo okay nothing in fir go still when I saw this my heart pounded so historically that I couldn't see I couldn't I couldn't think straight yeah I got on the telecon everybody else was like I could feel they were so excited on the telecon they're shaking so much that I swear I detected the gravitational waves that they were made or maybe it was just me I don't know everything was vibrating we were just we couldn't believe what we saw actually I'll be honest here this is not Louisa this is what we saw on Livingstone and Livingstone over a second or two here's that trap and here's this thing what the heck is that actually we know exactly what this is this happens all the time in our detectors it's a blip glitch our detectors have this is it in time series okay in the time filter time series a blip glitch happens probably about five or ten times a day in the LIGO detectors there are nasty things we don't know how to get rid of them they're almost I think almost certainly associated with scattered light you know we've got kilowatts of light circulating in our detectors 100 kilowatts and some of it may be a part in 10 to the 15 you know like Pico watts of laser light get out bounces off the shaking beam tubes because it's all in a vacuum system comes back manages to find its way back in to our photo detectors and makes this we see these all the time we got rid of it in real time took about 10 or 20 minutes we just simply snipped it out you can see the track underneath it we knew something was real there anyway another problem the Virgo data wasn't there it was delayed there was some network blip glitch or something and the and the Virgo data wasn't there so we had to scramble to get that it took us a good 15 20 minutes and we were just in a huge rush because we wanted a point the telescopes but there was something else there was also a gamma-ray burst 1.7 seconds later okay our heads were exploding I really have to tell you I was on shift that day was my job to tell the astronomers where the point but by now I just I couldn't even hold it in so here is our signal once again combined after in Livingston and here 1.7 seconds later from Fermi and also from integral SPI ACS a little excess and that it's a it's a um so you didn't have more stuff here it's um it's a clearly a gamma-ray burst but it is kind of a a wimpy one it's not it's not like what we expected because a gamma-ray burst that we normally detect there's okay once you've seen one gamma ray burst you've seen one gamma ray burst they're all different but for the most part when we detect gravitational when Fermi detects gravitational wave our bursts they come from may be binary neutron star mergers producing a relativistic jet and we're looking down that jet and it's far away like Giga parsecs away and it's we can detect it from across the universe redshifts of one two and and beyond okay if this gravitational wave signal is as loud as we think it is this is a much much closer thing so much closer that this gamma-ray burst should have been three or four orders of magnitude bigger than this and it wasn't okay we'll talk more about that later though the gravitational wave signal we know that binary black holes they're well modeled by these templates which come from you know supercomputer simulations solutions to Einstein's field equations but these ones happen to be but from the simulating extreme spacetimes sxs numerical relativity group and there are you know typically short for binary black holes for binary neutron stars the wave form as predicted is a lot longer it starts out at low frequencies but it sweeps to higher frequencies much much slower because they're lower mass and take a longer time so it takes a while for this signal to UM actually sweep from lower frequencies you were actually playing it in the audio but only the whatever the dogs in the audience whatever I don't know what you know the people who are sensitive to really low frequencies can detect it when it's approaching merger even maybe it'll be right in the sweet spot of normal people's ears and you might be able to actually hear the signal directly and actually you know this is a simulation of the signal not the actual data which would be noisy but when it actually does finally merge it tells us where the mass of the system the total mass of the system it's what it is extremely well you know I've spent 15 years waiting to hear that from nature and I did on August 17th so it was a real thrill to me this is the longest loudest and closest signal that we've ever reserved in Lika does the signal and the data match this prediction this template well once we adjusted for what the masses must be yeah here's the signal here's the data and with the glitch removed actually and here it is with the signal removed and it's nothing but noise and so that tells us that at the lowest order we can trust our template which is based on general relativity and it tells us therefore that general relativity is a good description of what's happening so this is actually the strongest test of general relativity that has ever been done because we're studying general relativity in the very very strong field very highly dynamical these things are moving at half the speed of light regime where it's never been tested before whereas you know generality is very nonlinear you can't even solve the equations they're so nonlinear and yet things work pretty well so that's remarkable but more on that later we can measure the masses the masses of the stars are measured mainly in this combination this chirp mass combination so the masses M 1 and M 2 are constrained to be in this in this banana shape sort of that's the the 90% confidence level band something like that there's also the spins they're hard to measure and they're roughly degenerate with the masses so we can't measure spins very well in this system but we can ask ourselves what if the spins in these black in these neutron stars are typical neutron star spins I mean we detect neutron star spins from pulsars if they're typical then they're they're dimensionless spin amplitude and general relativity is small and if it's small then we constrain the masses to be really in this region somewhere between so here's the two masses somewhere in between the 1.1 to 1.6 solar masses which are typical of pulsars of neutron stars in binary systems so this does look like a relatively typical system there's about a half a dozen of these systems known in the Milky Way galaxy this is not in the Milky Way galaxy it's in the galaxy rather far away so now that we have the masses we can put them on this diagram and we see that we're finding a whole different population the binary black holes are very massive five solar masses on up this is kind of a logarithmic net solar mass scale the lower mass systems are neutron stars and only one neutron star has been well measured to be around two solar masses these guys that are more than two solar masses they actually have big error bars so we really don't know what their masses are but more on this in a bit but what we do see in this pattern emerging here again these are the neutrons these the black holes observed by x-ray these are the black holes that are observed by light from lago in gravitational waves these are neutron stars usually observe as pulsars what we see is that these stellar-mass black holes are typically above five solar masses the neutron stars are typically below - salar masses this thing over here some kind of a mass cap maybe there are theorists who believe that there's good reasons for there to be a mass cap we'd love to know whether we can fill in this region for observations in nature or not and in fact what it did happens to these two neutron stars when they merge together here's the sum of the two masses what is this object at the end well turns out that's kind of hard to tell with what we have now with what the information we have now and I'll try to explain why in a minute okay I mentioned that when these two neutron stars as opposed to black holes merge they don't just merge together into a single black hole they release an enormous amount of dynamical a Jacke ejected even before then they so this is totally ripping apart the neutron stars does leave behind stuck together some remnants which might be hyper massive neutron star and then all this stuff might fall back in or some fraction of it fall back in and then that final object might collapse into a black hole along as it goes it releases a great deal of light and emits gamma rays gamma ray bursts that's an artist conception by the way this is not an artist conception it's a visualization of another supercomputer simulation of what happens when neutron stars merge together trying to put it in as much physics as possible so you can see they're starting to distort each other this is called tidal Distortion so the gravity of this star is pulling on the near edge of the gravity of That star more than the far edge so it's tightly distorting it inducing a mass quadrupole moment that enhances the gravitational wave radiation and enhances the in spiral so it happened it hit dumb it inspires even faster they merged together they spew out all this material a lot of the material just just flies out into the interstellar medium that's um you know just pollutes the interstellar medium with very heavy you know Neutron rich material a lot of it so that becomes unbound but most of it remains bound in in some kind of an accretion disk which eventually falls in and turns whatever's in the center here this hyper massive neutron star props perhaps into a black hole and it powers a a relativistic jet these neutron stars are incredibly complicated enriched objects I don't know if you appreciate just how amazing neutron stars are just as objects by themselves okay they're formed they're the dead remnants of massive stars that core collapse their outer layers blow away in supernovae leaving behind a dead neutron star they're just at the Tomic nucleus except instead of having 57 protons and neutrons they have end to the 57 protons and neutrons okay in in such a system all four forces of nature the strong and weak nuclear forces and electromagnetism and gravity are all at their most extreme beyond anything that we can do in the laboratory so um so there's an enormous amount of nuclear physics that one can do by studying the material inside of neutron stars how do you study the material inside of neutron stars the way particle physicists do it smash them together okay you don't need an accelerator nature does it for us just very infrequently and what happens well we don't know because the because the inner core of neutron stars are super nuclear densities we really don't know they how they behave so one way of asking the question is to say that neutron stars well maybe they have you know if quark gluon plasma inside maybe they're hyper ons inside strange quarks inside all kinds of things we're not quite sure because we've never no one's ever studied a nuclear matter at these kinds of densities however these can all be the matter can be characterized by an equation of state which relates the pressure to the to the density and these neutron stars are held together by gravity pushing it in trying to make it collapse and the pressure is holding it up okay and so depending upon the nuclear equation of state which is a property of nuclear matter that we dearly nuclear physicists would dearly love to understand we know that equation of state we can tell it tells us something about the neutron stars and vice versa so for example if neutron stars have strange matter inside that softens the equation of state allowing it to compress more okay it will hold up less matter before it collapses into a black hole and so the equation of state produces a mass radius relationship relation for a neutron star that looks sort of like these these green curves for strange quark matter they never get up be of beyond about one and a half or point one point eight or so solar masses we happen to know that there's one very well measured neutron star mass at two solar masses so this rules out strange quark matter at least this equation of state as describing neutron stars it's a pretty strange equation of state because of course it's got strange quark matter but it actually is kind of normal because when you go to higher masses you go to larger radii like ordinary matter does more mass bigger right and by contrast in these more intermediate stiffness equations of state when you have more mass you should make a bigger star but you also have more mass to compress the star gravitationally and the balance between those two things produces in this case for the case of these equations this range of equations of state produces neutron stars that have a radius of about twelve kilometers whether the mass is a half a solar mass up to two and a half solar masses still twelve kilometers okay no matter what and the very stiff equations of state can support even higher masses and bigger stars as big as fourteen or fifteen kilometers that's a big neutron star the question is what our new trip what is the mass there's only one mass radius relation there's only one nuclear equation of state which one is it okay with the way to tell is to see whether you know the more massive the stiffer equations of state will have more will produce larger stars that means that they will hit each other when they're further apart or at lower frequencies so to see where where the merger happens you want to see the frequency at which the merger happens and that tells us about the radius of the of these neutron stars but here's the problem in this simulation this is just the simulation here is the Freak the the the amplitude versus frequency or the spectral amplitude density versus frequency for the gravitational wave signal it sweeps from low free can see the high-frequency like the ones I showed you earlier it actually falls even though the amplitude increases because the time spent in a given band Falls because it spends less you know it's moose weeping from lower to higher frequency faster and about a second before you might start seeing this deformation this tidal deformation and that will distort this signal and then when they finally smash together that's tidal disruption and this might be the signal of a hyper massive neutron star that's resonating with body modes at 2 3 kilohertz something like that and to see this would really tell us about the nuclear matter out of which this stuff is made but here's the problem here's our noise curve it rises and all this is happening below our noise that is we're really not resolving it we're barely resolving any titled information at all in fact our title until we have better detectors our time we have a hard time measuring the tidal deformation or tidal disruption what we have for the signal is from 4gw 1708 17 is just barely a hint that there's some tidal deformation between a hundred Hertz and about seven or 800 Hertz just barely some information and that allows us to limit the tidal deformability of these stars here are some equations of state these are the stiff ones corresponding to bigger stars less compact stars here are the soft ones corresponding to very compact stars and the color is the consistency with the LIGO data and what we see is that were consistent with soft or intermediate black holes of neutron nuclear equations of state corresponding to neutron stars with radii less than 13 or 14 kilometers so we're starting to contribute to learn about the the nuclear equation of state okay I'm running out of time I wanted to talk oh let me just say a few more words okay this is my favorite slide actually because I'm a fan as I say okay first thing we've learned is that the gravitational wave signal is really fully consistent with general relativity over thousands of cycles is it consistent with your favorite beyond gr theory I don't know right now we've only compared it to general relativity okay the gravitational wave polarization has been measured for the first time weekly but measured for the first time and it is consistent with being tensorial plus and Krauss not at least pure vector or scalar these extreme alternatives okay but measured for the first time the tidal disruption if it's there at all is weak that tells us that the nuclear equation of state is not very stiff and it tells us that neutron star radii are less than 14 kilometers or so here's another thing the gravitational waves and the gamma rays arrived within two seconds of each other now the gamma rays traveled at the speed of light maybe a little bit less because there's an interstellar median essentially the speed of light for our Ford from well 40 mega parsecs away that's some 1.3 that's 130 million light-years so the gamma the gamma rays travel up for 130 million years before they got to us the gravitational waves got to us 1.7 seconds before that that tells us that gravitational waves travel at the same speed is like to one partner 10 to the 15th now Einstein would say yeah I told you that a hundred years ago to which I would say yeah it's what you told me but I found that it it's true and I didn't know that and you didn't know that until August 17th of this year also by looking at the phase evolution the frequency evolution we see no evidence of dispersion dispersion might exist if these waves were carried by massive gravitons which have a de Broglie relativistic debroglie dispersion relation no evidence of that that tells us the graviton if it if it has any mass at all it's very very small consistent with zero okay we were also able to test Lorentz invariants violation we see no evidence of that it's constrained to one part in ten to the thirteen there's another thing look for the last I don't know a thousand or so years of their travel there 130 million years of travel to get here the gravitational waves and the light fell into the gravitational potential well of the Milky Way galaxy okay that's sort of like you know Galileo dropping a stone and a wooden ball from the leaning tower of pizza and finding that they fell at the same time except nature did it with gravitons and photons and we got to observe it we tested the equivalence principle in a totally new way it's merely the Shapiro delay for the experts in aliens we can also test up that the gravitational lensing of the photons and the gravitons are also the same to a very good level of accuracy but for the astronomers we need to see whether any light other than gamma rays were released so we had to find the source in the sky and if we could find the source in gravitational waves in x-rays and gamma rays and in optical ultraviolet and infrared and in radio and maybe even in neutrinos that's what I call multi messenger astronomy so let's look let's see if I can get this to work did we see let's turn this off so let's see here is this localization confirming the localization from integral the localization from the LIGO detectors from the Virgo detector which is interesting thing more about that let's zoom in here oh that's a big patch of sky look there's a bunch of very massive galaxies in that patch of sky let's see if we can find anything in that little patch of galaxies how about this one this one is called NGC 4990 3 anything new there let's see we located the source in the sky about 28 square degrees and 3 dimensions so we also know was 40 mega parsecs away Virgo didn't detect anything that told us that here's the LIGO ellipse that told us that we must have been in a null in the Virgo antenna pattern the all of these quadruple antennas have these antenna patterns with nulls in them LIGO does just like Virgo does but fortunately it wasn't it in all of the LIGO detector but it was in a null of the Virgo detector which allowed us to locate the source in the sky really well from the lagoon from the Virgo non detection okay that told us that let's see what do I want to say oh the three-dimensional localization told us that it's about 40 mega parsecs away and 40 mega parsecs is essentially a redshift of zero so on this redshift versus energy release gamma ray energy release plot and here's all the short gamma rays and long gamma rays many of them that have been observed and whose distance has been measured or redshift has been measured here's our guy it is the closest and dimmest gamma ray burst ever observed well dimmest with a redshift ever observed we'd like to understand why I'm running out of time here's another really cool thing we hope we can detect we could hope we could find all this dynamical ejecta is Neutron rich it probably produces very high mass atoms like gold and platinum and uranium and plutonium and those things in fact here's the periodic table and there's sort of a concordant picture of where all these atoms come from so the light is hydrogen and helium were produced in the Big Bang that may be a little bit of lithium and the intermediate ones here come from exploding massive stars that's what we're made of we're Stardust everybody knows that but all these heavier elements we're not quite sure where they came from but maybe a very good possibility is that they come from merging neutron stars like the one we just saw here are the cosmic abundance of the heavy elements in terms of mass number and really you can't explain these very high mass things from super nodes from exploding massive stars it's really fall short by a factor of two or three whereas here and here it is a close-up of this region over here and these guys modeling the process by which this dynamical ejecta produces heavy elements through rapid process nucleosynthesis our process nuclear synthesis these these dashed and dotted lines here are different models that semi quantitatively describe the spectrum of the solar abundance of these heavy elements in the solar system the abundance of these heavy elements in the solar system so maybe we're onto something so we really gotta find this thing I already ran through this we looked at that I ran it all the way to the to that NGC 49 93 that galaxy right over here is there anything there Oh got it okay the next morning waking up I woke up to this picture this is a new star in a galaxy far far away 40 mega parsecs away that wasn't there before and furthermore a lot of other people saw too I think Swope a DLT saw it first and now you know just after a few hours and this was like eight hours or so after the signal and after the gamma-ray bursts so here is a gravitational wave signal and the gamma-ray burst and some number of hours we all this optical and infrared and ultraviolet poured in and even eight days later the Chandra x-ray Space Telescope's saw it but it took eight days to see the x-rays okay and the light curves are shown there we have seen light at every wavelength ultraviolet infrared and radio even in the radio it just started to rise about ten days after the merger that allowed us to identify the host galaxy from its distant from that host galaxy and from the fact that we measured the grep from the gravitational waves for it for it to be 40 mega parsecs away from the host galaxy we know the redshift and now we can put it on a Hubble diagram so here is the distance 40 mega parsecs here's the redshift or the recessional velocity over here and look it lies right on the line that we expect for a constant nearby Hubble parameter here is the posterior probability distribution for the Hubble parameter 60 70 80 and 90 this band over here is the Hubble parameter as measured by the Planck satellite that is a cop that is from Cosmic Microwave Background this band is from nearby supernova nearby being with a non-jew couple hundred megaparsecs okay they're both pretty soft precise much more precise than us and they disagree with each other oh we would dearly love to settle that disagreement right now we can't with one event we can't I am at the end almost er least I should be the suit the star went from not there to blueish to reddish just as you might expect from models called Killa Nova models that I don't have time to talk about it but you can ask about it later is there any evidence of heavy element production this is kind of hard to see but here is a spectrum at infrared wavelength this is 10 microns up to 24 microns all right no sorry this is in angstrom sorry ten thousand angstroms up to 24,000 angstroms the these are the data in these these this gray and then there's a smooth fit to it in this black these are regions in which the atmosphere makes it very difficult to measure the spectrum and the red is a model of a binary neutron star merger and the production of heavy elements you know heavy r-process elements in that regime the agreement is not great but it's not terrible either it's semi qualitatively there these two peaks over here are associated with lanthanides neodymium and the like even you know gold and platinum so it looks like this might be the site or a site for the production of heavy elements is it the site is it where most of the heavy elements come from well we can look at the quantity of ejecta and then we can look at the rate of mergers we measured the rate of mergers based on one event okay you take what you get okay based on one event and it's in the right ballpark it's really in the right ballpark this is consistent with explaining all of my wedding ring I mean I'm Stardust but this this is neutron stars Wow okay I am so out of time all I want to do is say by skipping a whole pile of slides that the future of gravitational wave astrophysics is bright thank you very much gratulations the discoverers congratulations for a wonderful talk it conveyed not just the physics but also especially the excitement of this business I'm still trying to come down still calm down yeah I have a high I'm a high Q us live yes only a few hours like that yeah the first question think five years down the road what is the most exciting line you're looking forward to no I didn't have time to tell you about okay five years down the road it's almost too much to ask to detect the gravitational waves from the from the from the inflationary period of the Big Bang that's a long shot it would be really great I hope to see a lot more of course black holes and neutron stars maybe that's not so exciting maybe it is depends on what we learn from it but what I'm really hoping for is something that we've of course never even thought of because we're opening up the sky to a totally different and new form of radiation and the unexpected will be the most exciting thing for sure other things okay you guys ever seen that movie beetlejuice that's a great movie you can see it okay it's about a ghost and you the ghost um to some of the ghosts you say his name three times okay now there is a star out there that's gonna go supernova in our galaxy any minute now of the same name but please we're not up right now we're down right now we're gonna we're upgrading our detectors when we come back up in the fall of 2018 you can say that name three times okay don'ts okay I'm a little bit superstitious about this and nothing else and the point is is that a core collapse supernovae in our galaxy when we detect it is going to give us of science you know we do use our our solution Stein Stein's equation as templates to pull weak signals out if the signals were completely not describable by these templates well the template based search would miss him but we have non template based searches which are less sensitive templates really help which would find these things our detections stand independent of the theory they are real measurements of strain and they are from extraterrestrial sources okay extraterrestrial strength we call that gravitational waves we think about it in the context of general relativity you don't have to this is extraterrestrial string lots of xenon lots of xenon oh yes yes you're said giorgio you said you're so see this is great yeah actually people back to your slide was it 40 or something no but so your equation of state it seems like you would like to be able to go higher frequency in some sense does this mean that these are you window opportunity for the old bar detectors so there do exist gravitational wave detectors which are bars and which are most sensitive up at the thousand Hertz or even you can you know the smaller they are the higher frequency they respond to yes their sensitivity of the best cryogenic bars in the world today is tortas magnitude heartless that's not doesn't mean you can't figure out how to do it yard you'll please I don't know how to do it Peter you need to improve the coatings on your mirrors in the laser power he's taking a different approach thank you very much we will turn up the laser power which is gonna break on bears but we not we're gonna we're working on it we do believe and look this left level here is 10 to the minus 23 this is you know here's where we're really shooting I got a go to slide 902 this is 10 to the minus 24 and 10 to the minus 25 this is the best we think we can do with a four kilometer long detector like the ones and the vacuum tubes that we've built and we want to use we could start skimming 10 to the minus 24 and even better but with longer and bigger detectors we could go to a couple of a couple times 10 to the minus 25 that's what we're trying to do if you have the money for us we will take it okay this is not going to be cheap great wife said that he's worried that the charge of the engineers of might be responsible for your excel or what Gary knows more about this than I do yes so one thing we do do is discharge our mirrors remember this is kind of a random noise it's not gonna unless you know some very large charge object whizzes by this is not gonna produce a false event like this but the charge can be the way we've been reducing the charge besides for you know what didn't work was flashing UV on it that didn't work what does work is simply reversing the the the the the sign of the electric field and then it builds up again and then reverse it reverse it again and what we found when we do that when we when we reverse it so we're sort of kicking all the charges out it doesn't change the noise at all that's what we'd like to do but if you do coat our mirrors it ruins their cue and we don't want to ruin the cue so we have to figure out a better way now in the future we'll probably go to cryogenic detectors made out of pure crystalline silicon and they'll have a different charging problem but maybe we'll be able to solve yeah they're such conservative bozos well sorry we're such conservative people that for our first two observing runs we were hesitant to send out low-latency alerts point here to just any old you know person instead we set up memoranda of understanding with several dozen of our close astronomer friends and we promised that after observing four events we would stop being so closed about it and send those alerts out to everybody guess what we've observed six events and there's more coming by the way in the start of our third observing run in fall of 2018 our low latency alerts are going out to the public promised by the way all of our data associated with these events are available LIGO Open Science Center lost LIGO org analyzed away find echoes [Applause] thank you very much sir thanks very much

Channel: Stanford Physics

Views: 6,702

Rating: 4.8571429 out of 5

Keywords:

Id: -i_ARhHfbpg

Length: 72min 56sec (4376 seconds)

Published: Wed Nov 15 2017