 Yes, I am. Okay, okay, let's slide in. Okay, so we might be live. I think we have this 17 seconds change. Okay, welcome everyone. Thank you for joining us for today's webinar. My name is Alejandra and I'm going to be today your host. Today we're presenting investigating the variability of a Christian binary black holes by Scott novel. Scott is an assistant professor at the University of Tulsa, and a senior NASA postdoctoral fellow since 2017. He got his bachelor in physics from Caltech, and then he did his PhD at the University of Texas in Austin. Then he moved to the University of Illinois at Urbana Champaign, where he did his first postdoc. And then he moved to John Hopkins and RIT Rochester Institute of Technology as a research scientist. It is very nice to have Scott, he is an expert in computational methods of general relativistic magneto hydrodynamics that's very complicated work, also known as GRM HD, which he uses to simulate accretion systems about single binary black holes. He's also very interested in exploding modern high performance computing facilities and techniques to solve astrophysics problems in the strong field limit of gravity. So remember that you can ask questions over email through our YouTube channel or Twitter, and then we will forward those questions to Scott at the end of the office webinar. So without further ado, I'll turn it to Scott. Thank you. Thank you very much, Alejandro. And thanks for the opportunity to speak today. I'll be talking about creating supermassive binary black holes and in particular looking at the variability of the emission from these objects. So these are black holes that formed and developed in the centers of galaxies and the galaxies merged and all the gas around them accretes onto the black holes and that accretion process then lights them up. And here what you're seeing here is the X-ray emission from such a process where the black holes are pretty close to each other. They're emitting gravitational waves and they're emitting high energy luminous X-rays. So we're hoping to be able to see both gravitational waves and X-rays vary as they in spiral and merge as black holes in spiral and merge. So this is, as Alejandro said, it is a rather complicated process. You have to do these supercomputer simulations and you're using pretty novel techniques to solve these new problems and so there's a large team of people that is assisting me and also innovating new techniques for the team and I list them here. And mainly we're based in Johns Hopkins RIT and NASA and we're collaborating as part of this network. And if you want to see more papers or such for me, you can follow this link, this web link here and follow me on Twitter. Though I have to say I'm not too active on Twitter but I hope to change that in the future. So it, coincidentally enough, it's Black Hole Week here at NASA. It's a social media campaign to just communicate our excitement for black holes and NASA has created this really adorable cartoon shown in the right-hand corner here about an extraterrestrial tourist that wants to go visit a black hole. And so the narrator kind of teaches the alien how to, you know, be careful around a black hole and maybe advises against it. But anyway, so it's kind of a fun little set of media elements that kind of communicate some of the aspects of black holes. If you teach younger students or even a fun undergraduate course in astrophysics, it could be kind of an entertaining way of introducing black holes. And here I've listed some links and such to them, to the materials. But and also if you have a story about black holes that you want to share, please tweet it and hashtag Black Hole Week and that way we can see what you're doing with respect to black holes. So, yeah, I'm really interested in the strong field regime of general relativity. So general relativity is Einstein's theory of general relativity and it tells you how space-time curves, mass curves space-time and space-time controls how matter moves, gravitationally speaking. And it doesn't, it's not a force, it's really a curvature of space-time. You know, mass flows down or is, you know, is affected by this curvature and that's, that is the manifest that's, that's what we see as, as things being affected by gravity. And so this, this curvature can be actually time dependent. Right, so you, so bodies can move. So on the lower left hand corner is a simulation of two black holes. It's an early simulation of two black holes in spiraling, spiraling closer, spiraling and getting closer and closer together. And they make gravitational waves and these gravitational waves are ripples of space-time. So these are curvature waves in the space-time system and they are, they can emit energy and angular momentum away and allow the black holes to finally merge. And the merger process results in a burst of radiation that is very luminous and we can now see, we've seen this happen thanks to LIGO. And so gravitational wave is a transverse wave and it has, it's a little different than an electromagnetic wave because it's a, it's a tensor wave. And so if the wave is going into the, into the screen, those dots in the upper hand corner show you how those dots would, would deform as that wave passes by them. So there, there, there can be two polarizations. The left and the right polarization here. And that, that's how gravitation waves affect things. They elongate in one dimension as they squish in the other and then vice versa during the next phase of the wave. And so we can use this to explore systems that are otherwise hard to see with light. They're often dark or small and not very hot. Except for maybe the supermassive binary, binary black hole case. In any case, it allows us to test general relativity in the strong field but regime allows us to explore our, the theory of gravity in a way that no other means can. Right, so this is the most extreme limit of GR that we have available to us in nature, and we can make use of it by measuring these gravitational waves. And we've, we've, we've done that. So, I think this is coming through my headphones, I don't know if it's being transmitted but this is the sound of gravitational waves and it was first heard by Lego, or by anyone in September 14, 2015. And in the upper left hand corner here you see the actual gravitational wave signal detected by Lego Hanford and Livingston these are two gravitational wave observatories interferometer laser interferometers. That's what Lego stands stands for, and they, they, they detected the signal in the lower right hand corner is is just a simulation of what it would look like to see a field of stars behind the black holes merging so the black holes are so compact. They feel their gravity so strong that they curve space time, and they distort the lens they gravitational lens in a strong way, the light that that is emitted behind them. And so, the field of stars is distorted in a really weird way. And you can see the two black holes in spiral and then eventually merge, you can see neat things like eyebrows so like secondary echoes essentially of the lens system of light behind it. Note that this is a simulation. But it does highlight the fact that there's probably not a lot of gas around these black holes. Okay, so I'll come back to that but of course this this was a. Incredible discovery. People had foreshadow that there would be gravitational waves but it was really engineering and scientific feet to to design and create an instrument sensitive enough to detect gravitational waves and for that. These three individuals who helped lead the mission. They received the Nobel Prize in physics and yeah so it's incredible incredible discovery and it's turned into actually a wealth of astrophysics. Since then so this these are the mergers that LIGO has LIGO and Virgo Virgo is a observatory in Italy. And together so those three observatories can can now search for gravitational waves together in an even better way than just like alone good. So they've seen now. More than here but this is just a sample of the first 11 events that LIGO Virgo saw. So, so there are black holes in the upper loss my cursor, but there are in the upper region. These are the black holes. So these are all the black holes we know about. So there are EM black holes. These are black holes identified with electromagnetic radiation, and then there are black holes identified with gravitational wave radiation. And so we've seen most of these LIGO gravitational wave events have involved slightly heavier on average masses than EM black holes. And then there's there was one neutron star merger event in the actually it's called O2 that second atmosphere patient campaign of LIGO. And so anyway so I just want to highlight that these are these are black holes that involve five to I guess 80 solar masses. Okay, so these are considered stellar origin black holes. Black holes that came from massive stars that evolved and went supernova and then formed either neutron stars that eventually formed black holes or actually just right out of the bat just formed black holes. So, however, we'll be talking about supermassive ones. So they're kind of two kinds of groups of supermassive black holes that we've seen. So these small black holes are sometimes called stellar mass black holes or stellar origin black holes. They exist and they can exist in X-ray binaries, which are which is an animation as shown in the lower left hand corner of a star orbiting a black hole. And those are tend to range around three to 100 solar masses. And the ones that we've seen in light are usually galactic. So they exist in our galaxy. Those that we see with LIGO or extra graph galactic and they can actually form these long these powerful jets. So an example of the jet is in the lower right hand corner but this is for a supermassive one. However, jets are seen in stellar mass black holes too. So anyway, black holes are great for understanding how stars evolve the stellar mass distribution of systems. They also tell us how matter behaves in the strong field regime of gravity. Supermassive black holes. These are black holes that have maybe formed from super from really massive stars and then have accreted a whole lot of gas over cosmic time to get to the size they are. We don't quite know how they form and evolve. So that's why we want to use gravitational waves on EM to find how they evolve and we're still doing that. And but these things are 10 to the 5 to 10 to this 10 solar masses. And as I said that there's at the center of galaxies they form jets. They can tell us how galaxies evolve. There's feedback. So black holes jets and outflows from black holes feedback into the galaxy alter star formation and do all sorts of crazy stuff. So, but really I'm interested in understanding plasma physics in the strong field regime of gravity because there's nowhere else in the universe that you can do this. And so more about supermassive black holes well so yeah so they're they're the only system about black holes that you you can you expect to have lots of gas around. They're they're going to be emitting a lot of light. So there's an active galactic nuclei they're active because they're very bright and luminous and they're shooting around lots of gas. So we want to look for those systems because we want to explore. We want to find them in the electromagnetic regime and sector and also the gravitational wave sector. And primarily I'm focused on Lisa sources so these are. So there are a range of gravitational wave signal waves that you can look for those that vary at, you know, that have frequencies of Hertz to those are Lego sources to Miller Miller Hertz systems which are Lisa sources, and then there are these things called this this type of detection called pulsar timing arrays. And I'll go into that next but anyway, so we're going to be looking for these things in the with the with light, and we are going to take advantage of these new generation of telescopes that are robotic. They can currently scan the sky, and they scan it every few days so we have like a movie of the whole sky and we can see how things are varying and we can search for these new types of systems. So we haven't discovered a supermassive binary black hole in the inspire regime yet. So that's that would be a new discovery, and we're looking for them because we're interested in these systems and also future Lisa sources. So, which I'll talk about soon. So LST here I mentioned is the large synoptic survey telescope, and it's a big robotic telescope that's going to detect hundreds of thousands of agent. And many of these, we hope are going to be binary agent. So we, we hope we have enough statistics to find these binary agent. And however, we don't know what to expect. So we need, we need to do accurate simulations to really know what to look for so that we can pick these these needles out of the haystack. So as I mentioned, there, there's one way of hearing gravitational waves called pulsar timing array. So there's this movie on the bottom that shows a neutron star that is emitting this, this, this light, it's called a pulsar. And this, this light, it's so neutron stars spin at a very even rate. Okay. And this lighthouse effect of the pulsar, then sweeps this light at us at a very constant rate at a very predictable click. And these clicks can be recorded, we can observe a pulsar, we can measure the space, the time between these clicks, these pulses, and we can see if they vary over time. And this variation could be due to a gravitational wave transiting the space in between the pulsar and Earth. And if we have a network or an array of pulsars, then we can use this, we can use this timing offset to, to hear to notice the gravitational wave passing one pulsar and then the next and then we can use that to triangulate maybe the source of the gravitational waves. However, we need a lot of pulsars to do this. And so right now, there are these radio telescopes that are searching for more and more pulsars so that we can keep on here, listening to them and see if we measure any of any time offsets. So pulsar timing arrays happen at the nanohertz regime. So this is a noise, this is a characteristic strain so this is these black curves are the sensitivity curves of these instruments, and the color color shapes are signals from different sources. So pulsar timing arrays are sensitive to 10 to the nine to 10 to the 10 solar mass black holes, binary black holes, whereas Lisa it's sensitive to around 10 to the five 10 to the six solar mass black holes and Lego is around, you know, 10 to 10 to 100 solar mass black holes. So they, they explore different regimes of black holes and different, you know, with their different regimes of gravitational waves. And how, however, we're likely to see counterparts to pulsar timing arrays sources but we're not a lot. We're not really expecting to see them merge because they evolve too slowly. You know they're too massive. We hear them likely it's going to be too far away for it to inspire and merge in our lifetime. However, Lisa sources, if we hear them, they're likely to merge in our lifetime and in the lifetime of the experiment. And so we'd be able to see the whole range of evolution. So Lisa so Lisa is a constellation is kind of like Lego in space essentially but instead of just two arms it's three arms forms a equilateral triangle, and it's going to be an orbit behind the orbit of earth. So it's not going to be an orbit around earth is not like your, your regular telescope but it's going to be really big it's going to be, it's going to span millions of kilometers to detect these really long wavelength gravitational waves. And in the lower left hand corner here we've got a strain plot that shows tracks of in spirals so this is the signal gravitational wave signal of a 10 to the seven solar mass binary black hole, and it in spirals merges chirps and rings down. And so you can see these at different masses. And also they're going to be sensitive to white dwarf binaries in our galaxy and that's going to be in a background source that is going to confuse us a little bit, but, but we're going to, since Lisa so sensitive, we're going to see them so we're for sure going to see gravitational waves. And there's some verification binaries that we will, we know these binaries exist and so we can look for them. We can pick out their signal and test the antenna with them. But we're expecting to see lots of order 100 of these sources well above the background. We can even see Lego sources that that that start in the Lisa band and then go and then merge in the Lego band. That may be harder to find though. So, so one, one thing that Lisa is great at is it can explore black holes evolve and binaries form and merge throughout cosmic time. So this is a pot of the signal to noise of black holes from binaries as a function of redshift and mass. So if you're at around 10 to the 510 to the six solar masses, you, we may be able to see them with so the color contours are different signal to noise ratios. So 100 is the cyan color. So a signal to noise ratio of 100 is pretty good. You can get a lot of information out of that. And so you can see that out to redshift 20. That's crazy. Now if we want to see like light from it to we want to actually see it nearby. So we're hoping that some of these mergers happen nearby. But you can use the ensemble of binaries that you see to differentiate between models of formation and evolution between them. So these different tracks show you how black holes form differently and evolve differently. And if your population doesn't match with one evolutionary track, then it'll support maybe another model. So that's, this is going to be a really exciting time for astrophysics because there's really it's hard to do this any other way. So what's the electromagnetic evidence for binary black holes? Well, we see them sometimes have these. So the nuclear regions of galaxies can sometimes have these double knotted features in x-rays. And so this is thought to be these are two black holes accreting. And black holes often are, if they're accreting, they're often bright in x-rays. So this is a way, and those x-rays, even though there's lots of gas in the nucleus, those x-rays can penetrate all that dense gas. Just like x-rays can penetrate your skin in the doctor's office. So you can also use radio waves. So in the upper end corner and actually all the plots on the right use radio waves. And they can show radio wave antennae are really big and you can make rays that are really large, just like we're using the event horizon telescope. And you can resolve really small angular features in the system. So these are actually radio is the best way to really see the smallest angular features of systems out there. However, sometimes radio is obscured by nuclear gas. So anyway, there's evidence in the radio for double nuclei. However, all of these systems that we've seen so far are don't get any smaller than around maybe 0.1 parsec. So that means that it's not really close enough to be really bright in gravitational waves yet. So we want to actually see things that are closer together. And that probably means we won't be able to spatially resolve them. So we have to look at their modulation, how their light changes over time. So people are doing making catalogs of these binary candidates for dual agent. We're actually interested in binary agent, which are even closer than duals. In any case, there's this whole campaign going on. There are people looking at the light curve how the light varies. And these they have identified modulated signals and they with this modulated model and assuming it has to do with the orbit of the binary then you can say well it's this particular system as a separation of 140 to 1400 M. So we use units of M things scale with the mass M is the mass of the system. So the that's a natural scale of this of the determines the dynamics of the system. So in any sense, in any case it. So this system would inspire in 10, you know, in a million years, not really what we're looking for but still this this could be an interesting system to follow up on. However, oh and another group found one that I was actually going to spiral in seven years and it's restaurant. And that's really exciting as a period of happy year. We were really, really excited about this over then this group followed up with with more data and found that the longer baseline data did not match their fit so that so this is highlighting the problem that you actually need to to to see more than five cycles of an oscillation in order to really convince yourself that it's not due to just statistical red noise fluctuations. Okay, so. A GN actually very in this red noise sort of way, and that red noise can look like a periodic signal at long wavelengths, or low frequencies. And so you have to make sure you actually cover many cycles in order to to really convince yourself so there's no really convincing binary object yet. Duals, they're all out there, and we're pretty convincing that exists but not something that's going to merge in our lifetime yet so we're, we're inspiring a field that is going to search for these things and in particular. So the European Space Agency is thinking of launching Athena and Lisa together. Athena isn't the next generation x-ray telescope. They're thinking of launching them together so that they can use the x-ray telescope to look at these counterparts. And I just want to put this out here. The Astro 2020 Decatur reviews underway to and this is a big theme of the Decatur review, please my point of view. Obviously that we're using to simulate these things well we want to take, we want to start from the black holes in the inspire regime and have them go really close to merger we're using post Newtonian techniques. So we don't really solve Einstein's equation the full numerical relativity, but we use post Newtonian techniques so that we can really just focus on the gas dynamics. So, and then we know from numerical relativity simulations what the waveform looks like. So that's been solved for a long time and well for at least about a decade and but the matter dynamics is kind of the frontier so we start with something in the, we assume that there's been this evolution the black holes have come from those black holes settled into a core, and they then eventually enter the inspire regime. How they got there is still some of a mystery but we think that it's been solved by non non spherical symmetric distributions of stars. We then will copy or transfer our data to a numeric relativity calculation week so that we can do the merger and and see what the merger signal is. So the key challenge is really to resolve the dynamics from the black holes scale to the scale of the circum binary desk so here you're looking at top down views of the gas. The density of gas, moving around the black holes and these are the white dots around the black holes. Here we've, we've kind of ignored the gas near the black holes and here we're covering the gas near the black holes, but not the gas in the center into that. So black hole accretion disk theory so got a jet corona that emits x rays, and you have the bulk of the disk that emits x rays and UV light. You need mhd ideal magneto hydrodynamics gr radio transfer to carry the photons to a simulated camera and in the, in the, in the distance. And you also need, you know, to understand how the temperature of electrons and protons vary. However, we're just like focusing on the mhd gr and, and the ray tracing. So the, so the magneto rotational stability allows gas to lose angular momentum and to transport it to a larger radii. And it can, it's the mechanism that explains how systems accrete efficiently at the rates that we observe them. There's no other mechanism that we know that is so efficient at accreting at getting gas to accrete. So we really think that in order to accurately model these systems we need to evolve magneto hydrodynamics so that we can study the instability. Here you just see field lines. There's an annulus of gas that is orbiting into the board on the left hand side and out of the board on the right hand side. So we're moving along this, this disk of material, and disks, if they follow a Kepler and rotational law, they, their orbital velocity increases as you go to smaller radius, the specific angular momentum decreases as you go to smaller radius. And so two fluid elements connected by a coil, which is really a magnetic field line. It wants, it wants to move faster. So it stretches that field line, that tension in that field line, then pulls on this one. It's getting more angular momentum to this one so it moves outward, and this one is pulled backward, and it's given less, it's, it's taken away angular momentum so it wants to move inward. So then the tension in the field line increases only exacerbating the problem. So this is a nonlinear, eventually reaches a nonlinear phase and then inspires turbulence, which you can see in this figure. So this is a single black hole. This is an edge on view of gas, of log of density of gas, creating onto a black hole. So this is a really thin disk, high resolution, thin disk simulation, one of the first that we did. And it shows you just how turbulent and chaotic these disks are. But you have to do all this crazy dynamics to accurately model how accretion happens. So the post Newtonian scheme that we use is developed by Nico Eunice and collaborators, and it was really designed for numeric relativity simulations. And these numeric relativity simulations would start with this data as initial data, and then they would evolve Einstein's equations. However, we want to actually use it to evolve the space time forever. And we can do this because the black holes that we're starting from are not really close together. We're in the post Newtonian regime. So post Newtonian just means we're we're perturbing, we're finding solutions to Einstein's equations of GR to some perturbative order, 2.5 post Newtonian order. And what we do is we perturb in the quantity m, m one over our m two over our so m is the mass of one of black holes and ours the distance to that black hole, or, and we also expand in V over C. So those are the expansion parameters, and we expand to those orders and then truncate it and that's our, our, our approximation to the space time. This allows us to give get like a closed form expression. It's just a function call in our program that says, give me the space time and it returns the space time and then we use it however we want. So it kind of eliminates all the, all the difficulties of evolving Einstein's equations for us, and we can focus on the matter. And but it also includes all the rich dynamics that includes radiation reaction terms so the black holes do still in spiral and everything is handled self consistently. There are different approximations in different regimes so near the black holes, their post cur or post short shield. So it's a black hole with perturbations, the perturbations are from the other black hole, and that is the inner zone here in the shown in the lower left hand corner, represented by I z. There are two inner zones for the two black holes and there's a buffer region that kind of interpolates between the inner zone and the near zone and near zone is the region that surrounds the black holes immediately. And then the near zone has a buffer region buffer zone that connects it to the far zone and the far zone is the wave zone, which is it, which is where it which is the approximation is post Minkowski is just flat space with with waves on top of it. So we use this to for a space time and you can see, we've seen, we've measured how well it is that preserving expectations of what a black hole space line should look like so the Richie scalar, which is shown in these colored contour plots on the right. So the Richie scalar distributed over space at two different orders first order on the top and second order on the bottom. And the, the left and right columns just showed different like focus on different regions of the space time. And you can see that the Richie scalar so in gr the Richie scalar should be zero everywhere. So it's, it's, it's, it's small, it's smaller for the second order more accurate case which is what we expected so that's good. And it's only large in, as you get really in the buffer zone as you get to the black holes. So this is, I mean at the second order scheme only gets to be about 10 to the minus three. That means that there's some fake mass in the system, but it's distributed smoothly around. So it probably doesn't have much of a dynamical effect. And we've actually measured this in different ways. And we're pretty convinced that this is working pretty well. So what this allows us to do is simulate hundreds of orbits of black holes so the numerical relativity calculations can do that efficiently. And so, especially with matter. And so this allows us to let the disk settle into a steady state. So we were pretty confident that this is a natural or realistic looking accretion system. And these costs usually like a million CPU hours on computers. So they're pretty expensive, but we can do them with the supercomputers we have today. This one simulation that I showed you here. Actually, we discovered a novel periodic signal in the light. So what we did was we cooled the gas that gas was cooled by by radiation, we measure we recorded the radiation that the gas emitted. And we integrated it over the volume to get the light curve that light curve has a signal as a had a had a periodicity in it. And we so we took the Fourier transform of it. And we see here, it happens at around one and a half binary frequencies binary orbital frequencies. And we think, well, hey, you know, the period periodicity of the light signal should be close to the period of the binary, right, because it's that's what's mixing things up. But we found that actually it's one and a half. And it's because most of the light signal comes from gas that is tidally torqued. The gas is wants to fall into the binary, but the binary has this this quadruple moment that's time varying and this turns into a title torque, so it torques the gas. Some of it is expelled out into the circumbinary doubt disk. And the rest is is accreted to the black holes. So that that stuff that's flung back out that's torqued out is crashes into the disk. And that's that's triggering the creation. So here, on the right hand side is a movie in rotating with the lump. So the lump is an over density region so this is density of the gas. So this lump develops it's the right, it's the red, red to white feature, and it orbits with the with orbits around the black holes at a rate a little bit less than the black holes. And so we're in the rotational frame of that lump, more or less. And so we see that the streams keep on shocking against that lump, and that shock heats a gas, and then that hot gas then radiates the light. So this is the brightest feature of this of the disk, and that that that. So what happens it, what, why it's one and a half times the frequency of the binary orbit is because it's a beat frequency between one of the black holes, passing by the lump. And it's two times because this happens twice because this is an equal mass binary black hole. So this is a non trivial manifestation of gas dynamics around a binary black hole that we wouldn't have discovered if we didn't do the simulation. So in order to actually carry the gas to the black holes. So that there we, we, we just ignored the, the, the gas that fell into a spherical region around the black holes, but in order for computational efficiency. So we want to see how the gas actually falls onto the black holes themselves so we need to create a grid that resolves the gas dynamics near the black holes and we do this using this, this work or work to grid scheme that so we warp or focus the cells. So we want to see which the equations are solved. And we focus the cells around the black holes so that we can really pick out the, the, the small scale features that happen near the black holes. And we've done this. The paper was three orbits that we could, we could do and we could afford because it is an expensive calculation. And we were at the top movie here is a 3d rendering of the density and the bottom one is, is the same rendering but further out, and the white lines are magnetic and this was the first realization of mini discs forming around black holes fed by and, you know, with mhd with ngr. And we found that the mini discs, even though we started with them if they quickly depleted, but then they were replenished by the circumvinary gas, and this replenishment happened periodically as the black hole passed by the lump. So, as it left near the lump it, it depleted, right, the black holes, you know, the gas accreted onto the black hole. And then, and then as after it was depleted it passed by the lump again and then fed from the lump and was repleted. And so this created this, this is seen here where we plot the mass fraction in each mini disc, we see that as one is increasing in mass, the other is decreasing and vice versa because they're out of phase with the lump. And so this, this modulation is also wasn't expected. And so is a manifest, manifest from the fact that there's this lump of material that orbits the black holes, and that lump of material modulates the mass accretion rate and likely the light curve. So in order to calculate the light curve we do these this general relativistic ray tracing, where we transport light, a full spectrum of light through the disc from the photosphere through the corona. We have a corona model, which is a non thermal model where we assume that the chrono is at a fixed temperature. All the details are given in this desk ollie paper. I won't have time to go through all this but anyway we look at we want to calculate the whole spectrum of light coming through this and because we have a simulation we can look at how it varies in time. And so here this is just a demonstration that we can show you we can calculate the ultraviolet light and the x ray light and we see here the x rays come predominantly from the mini discs, that's where the hottest temperatures are. Notice also there's this what what may look like a third black hole but this is not a black hole this is just a cutout. So because of how we've designed the coordinates we can't actually cover this region. Okay, but there's very little gas rate there. However, we want to change that in the future and we and we have, but for this calculation which was a first of its kind. We were still limited by this this this fact. So don't get confused by this black spot. These are the black holes over here and over here. The ones that move. So this is the spectrum. So this is the first spectrum calculated from that ties to creation on to binary black holes. And we broke it up into different components and circum binary disk component, the mini discs that are orbiting the black holes individually and the creation streams that connect the circum binary disk to the mini discs. And then there is just the total. We see that the mini discs spectrum actually varies the most right because it's the most dynamic one. And we're, we're hoping that this, this, this, this modulation even though we didn't have much time because we only have three orbits of evolution that this, this modulation will give us a light curve that we can use to identify these systems. So we can measure the spectrum of AGN and see if it matches this spectrum. So there, there are key aspects that we can look for are, while binaries are expected to be brighter in x rays than a typical AGN, and it's also supposed to have a slightly broader thermal peak. So a binary single AGN would would drop like so maybe maybe peek at a higher temperature, but it won't be as broad as this because of the mini discs. So the presence of the mini discs and they enhance the shock that happens between mini discs, the creation streams falling onto the mini discs that will broaden this thermal peak and may actually make this little notch here. Apparent. So that would be also a signature to look out for. So here you can see a time average of that. And the width of the curve is proportional to the standard deviation of the variability of that curve. So you can see the mini disc accretion is really the most variable. And so we hope to look for that variability. Also angle dependence. So, since we didn't have a whole lot of time to integrate over. We were using how, you know, if we saw this at at a high inclination if it was edge on how much it would vary. We see that there is an angle dependence. So those that are more edge on we see greater variability and that's represented in the right hand plot. So theta cam is the angle from the z axis that the camera is at. The angle is when you see the binaries in edge on view and they theta zero is face on. And so, and then fi cam is the, the angle around the orbit of that that the cameras at. So we're like moving around in the same way as the orbit, and you can see if the black holes are lined up, then one is lensing the other. You can see the signal from the background black hole and then increases light there. So that orbital modulation will be present but probably not as strong as the accretion rate modulation. I'm used to it's NASA. A movie of our simulation work and you can see the whole movie here at this link below I, I don't have time to show you the whole thing but I encourage you to pause the movie, copy this link into your browser and look at the whole thing, because we're pretty proud of it. And, but we now have longer simulations so we did trouble orbits using the same grid we just ran the simulation longer, and we saw really strong evidence of this depletion repletion, process of the mini discs. We found that it kept on going. Just to remind you the grid looks like this. And we now have more more enough time enough data to get a really good light curve out of it and that's being processed right now. And we see this the plot of the mini disc mass for each mini disc as a function of the orbits. And the right is a Fourier transform of it. And you can see how different components modulate at different frequencies, and how it relates to the phase between the phase difference delta five between the binary orbit and the lump orbit. Each component has has that component with the phase of the orbit but not all the components and there's some that are have that double beat frequency power, but not, not everything so so we're still. So in the next spot we see this is plotting the mini disc mass of one to the total and you can see how it really, it really increases to large fraction of the total mini disk mass. And then it quickly creates and then it increases again so this is a pretty strong periodic signal. We expect that to be really dictate the light that we see from them. Now, we've eliminated the the weird circle at the center of our grid, and what we did was we covered it, we have two patches of two patches of coordinates. So we have that a spherical like grid around the circum binary disk and then we have a Cartesian grid around the black holes. So that allows us to get rid of that that circle and cover the whole domain so. And it's in it's more efficient so we've been able to do 30 orbits. So we even have more data and we can see how, because there's no gap in the middle there. Decretion streams can flow right through from one mini disk to the other and we're thinking that the this mini disk mini disk interaction will then also create an interesting light signal. And unfortunately I don't have those results ready yet we're still, these are relatively new simulations and we're still processing them with our ray tracing code. This is the first of a kind where there's no gaps in our volume, we're covering the whole thing and we're doing it completely well. And also I want to point out, these aren't just thin disks, they have thickness and this thickness varies. And so there can be a creation coming kind of above the black holes onto the black holes, because the streams are kind of relatively thick compared to the black holes. And we're now working, this work is in progress, working on analyzing that. However, it compares well to our previous work with the warped coordinates. And however, we didn't start with mini disks in this, in this patchwork run, in this latest run, we just let the circum binary disk fill in the mini disks, and we still see similar periodic structure. But slightly, we have, so this is the old power spectrum. And you can see that the new power spectrum, there are new features at the low frequency. And the frequency at one and a half omega bin seems to be double peaked now. And so we're trying to figure out where these new features are coming from. The more detailed power spectrum is due to the fact that we have more orbits of data, more time of data, our data set is longer in time. So we're still trying to figure out if the sloshing between the mini disks, sloshing of gas between the mini disks is the cause of this or some other feature. So I just want to quickly think I've exhausted more than enough time. Supermassive binary black holes are really exciting. They're important gravitational wave sources for Lisa, and we're hoping to find them with light very soon, and so that we can study them both in both, both sectors, light and gravitational waves. And welcome any questions you have. Thank you very much. Hello, Scott. Thank you very much for this nice webinar. It's impressive the amount of work and the simulations and the pictures and the videos are amazing. We have been doing a very great job. We have a question from, well, a couple of questions wanted from our YouTube channel. I'm not here on the chat, but I would read it. And it's related to the introduction of your talk. So they say if we detect an offset in pulsar timing, how would we know if it was caused by gravity wave versus pulsar physics. That's a very good question. And fortunately, I'm trying not the best person to ask, but I know that we know a great deal about pulsars. We know more about them and we can study their pulse history. I mean, they pulse very rapidly, I mean some pulse with periods of milliseconds and most longer, but so we can, when we observe them, we can get a bank of, you know, 1000 pulses. We can study their history and see if these, these changes in the timing are due to glitches in the neutron star and that's been observed before. So, but those glitches are also associated with changes in the pulse structure too. So I think with enough statistics and understanding of the pulse shape of each pulsar, we can then differentiate between a glitch and a transiting gravitational wave. That's my relatively novice answer to that question. But I, if you're really interested, I can put you in touch with people that are experts in pulsars. And, but yeah, that's a really good question. Okay, thank you Scott. And also for our viewers, we have, we know that there's a 20 seconds delay. So if somebody wants to make a follow up questions or something. We have to wait a little bit and then it will be read to Scott. And I also received another question. Well, I'm going to try to do my best to read the question. But in the sense that I could say, what would you say to young scientists or like researchers, particularly in South America, about your field in the sense that they, in order to do those simulations you need not only researchers and high-performance computing facilities, but also the codes, okay, the codes are like very, like complicated nowadays and advanced. So how would you say like for young scientists, how can they, like, start in that field? Can they move to that field or what would be your advice? Yeah, no, I do acknowledge that it's challenging. You have to learn general relativity, fluid dynamics, and then you need to know how to program, learn how to use supercomputers. I don't get overwhelmed, do one thing at a time. Start using computers as early as you can. That expertise builds upon itself over time and you just become more and more confident and you can, it is something that I found using computers is that time and experience does help. I mean, you get better and better and that experience feeds off itself. Don't be afraid to tinker with programs and play around with programs right your own, learn about operating systems, you know, knowing how a computer works. I think probably that right now I see as a big challenge in the younger students that our computers are pretty easy to use. So I grew up in an era where computers were hard to use. So you had to, you had to often get in there and figure out how to fix it, you know, re-program it yourself. And so that experience then was like, okay, when I had to then go write my own program, I kind of knew how to do it all. Anyway, so for right now, you know, you click on an app, it works, you know, you can send a picture to someone on your phone really easily. But all that before we had to like upload, you know, pictures to an FTP server, we had to figure out what the command line of that was. My sensor is turned off. Hope you can see me. And so, but yeah, so I like, if you want to get into scientific computing, I would encourage you to get a Linux distribution. And that'll allow you to more easily program, in my opinion. You can use Python. Python is good entry code, entry language to start programming in. But really, I think in order to do scientific computing, you need, you know, either C or C++. I would not encourage you to learn Fortran, but some people use Fortran. And I use Fortran in grad school. So there's nothing wrong with it. It's just that I think more people speak the language of C and C++ these days. But yeah, just start doing it. Get in contact with people that, you know, whose research you find interesting and get their advice and maybe work with them. You know, you can, these allocations, people can use, we have teams, we have students, grad students in Argentina that are working with us and are using our allocations on these big supercomputers. So we, we, it's possible. So you don't have to be in the States to use them, you, and you can team up with people that have these resources and use them. And there, and in some codes, there are communities that you can use to learn from. So the Einstein toolkit is a numerical relativity code. And it has a pretty extensive and welcoming and nurturing community that can help you with problems, as lots of examples and things. So I would encourage you if you're interested in that kind of feel to go that route. Okay, thank you Scott. So, again, if, if there's a follow up question from the person you can, you can do it. I'll tell him to Scott. So do we have questions from our physics coordinators. Yeah, I have a small question. I'm very impressed with all the work that you have done. I mean, with all the collaborate, of course, the one question, because they were like two or three years ago when when they started all the signature from gravitational waves, in which there were some coincidence signals that Fermi ice cube, so on. Because you talk about mostly relative process transfer, but could you also extract signal from the neutrino neutrino flaxes from the, from the disk decoration these can so on principle I mean like to add an extra observable to the signal. Um, neutrinos. I mean just to finish the, yeah, and is it shouldn't be difficult to implement in your code or is mostly straightforward. Yeah, our code doesn't necessarily like our code doesn't isn't a function of the mass of the black holes, per se. So we can use this our simulations to model stellar mass black holes or like a black holes. I mean, the radiation does kind of it is parameterized by the mass but we could we could readjust the parameters and redo the ray tracing to to get a new light curve. So that's with the light. So, in order to look for that so for me saw what a possible coincidence for me is a gamma ray telescope NASA NASA gamma ray telescope, and it saw a blip right coincident with the first LIGO event. However, it was the line between the LIGO event and for me went, it just skinned the atmosphere of the earth. Okay, and so that they think that maybe the reason why they think it may not be real is because integral which is another gamma ray telescope and orbit would have seen it and did not. And so they think that that for me just picked up some gamma rays coming from the atmosphere. But lots of their there, but from those for me results, people came up with lots of different theories that could predict it. Like for instance, actually kind of my favorite one was one where there are these dead disks around the black holes. So the gas is just orbiting each black hole but not accreting onto it so they, they're, they're, they're relics from a prior time. And then as they get closer, the gas kind of gets heated up more and becomes more magnetized, and then allows it to accrete rapidly as the black holes merge. I'm not really sure if the matter dynamics is really consistent with, you know, we know about matter dynamics and about that's dead disks and why they would be there and why it would be stable for so long. I'm not sure anyone has done a really consistent calculation of that but I thought that was neat. It's maybe theoretically possible, but it would have to be these objects are also really pretty far away. So the luminosity also scales with the amount of mass you have there right mass of the gas, the mass of the gas around these systems is going to be a fraction of the mass of the black holes. Right. And so, these things are so far away, and they're not very massive they're not going to be very bright signals, even if there was, you know, a bit of gas there. Neutron stars are, you know, we, I go saw the GW 17 0817 event, it was pretty close. We see them gravitationally, and also electromagnetically, because it formed a kilonova signal and a gamma ray burst. It wasn't even a pretty powerful gamma ray burst, you know, because it wasn't really pointed at us. But it's proximity helped. And, and I think the neutrinos, we did not see neutrinos from that. And though it was, you know, there was a lot of nuclear material there was a lot of nuclear decay. And I think, I'm not really sure how, how we can improve our chances for seeing the neutrinos from these events. Now for binary AGM, so they're just past year or so they saw a blazar. So blazar is essentially a an active galactic nucleus black or supermassive black hole that's a creating and has a jet that's pointed right at us. And they saw a, they saw a burst in blazar activity, coincident with a neutrino burst. So ice cubes saw energetic neutrinos come from the same region as the blazar at the same time as the blazar ramp up. And they think that that was a multi messenger observation of a gene activity. So, one could imagine if there are two blazers, inspiring and merging you can see the gravitational waves, the neutrinos and the electromagnetic waves, all at once. So it would be a three messenger event. So that I mean that's, I think that's why supermassive binary black holes is so exciting is because, you know, there's going to be gas there's going to be gravitational waves there and there can be even neutrinos from the blazar jet processing so you have to be lucky right. But yeah, I don't know I the neutron star mergers and neutrinos I think maybe there's a chance there by I need to look into that. Okay, and do we have any other question from the coordinators. Yes, one from one from me. Okay. So, so, so I do particle physics right so I'm going to ask some very naive questions. So, so following the robertus robertus question. I was wondering if a you could get a signal for instance dark matter on the creation. Because in principle you should have it right there. Right so if we could annihilate in some very specific channels. So maybe wondering wondering if you if you could actually say oh yeah we're having some extra contribution to this x-rays or whatever we're getting so can we be able to tell that. Yeah. Yeah people are looking into that the theoretical predictions of dark matter. the notion of annihilation because black holes are, you know, they're, they're good at attractors of particles so they, they can actually increase the efficiency of collisions of dark matter. And so, because the, if you have a cloud of particles around a black hole the density increases as you get closer to black hole. And the scattering cross section of dark matter interactions scale that, you know as And so that then naturally means that these collisions are going to happen more and more closer to the black hole. So, yeah, actually my colleague across the hall from me, Jeremy Schmidtman, has calculated this and there have been estimates about the flux levels of Sagittarius A star at the center of our galaxy, whether or not we can see dark matter signals from there. I think there was actually a paper that supposed that what they saw was actually from Sagittarius A star. I think all these signals are observed in the gamma rays because of the rest frame, the mass of the supposed mass of the black of the dark matter candidates. I think it's in the GEV scale. If I'm not right, if I'm not wrong, MEV or GEV, no, it's in the GEV scale. So that would make it a gamma ray. So that's why I think people are focusing on the gamma ray emission. And there, in order to really be convincing and that's not some sort of like jet consequence, you would probably need like really high spatial resolution. But and gamma ray telescopes don't usually give you that. In terms of binaries, yes, certainly if that would that that could even accelerate further the interaction rates of of the dark matter can particles. I don't know if anyone's looking at the dark matter project in the regime of a binary. That could be interesting. It'd be harder because you have a dynamic space time. But we do have codes to do that sort of thing. So to follow GD6 in a binary space time. So yeah, that would be really kind of interesting. But yeah, I think there are theorists doing this. There are people looking at data too. So this is certainly on people's radar. Super. And they have another naive question. This is even even more naive. So you're going towards the merging of two black holes, right? And so in principle, nothing can cross an event horizon, right? So so what is going on with the event horizon when two black holes merge? I mean, we we we in principle see it, right? So so wondering what's going on with the event horizon that allows the two black holes to actually merge. So yeah, so we see everything outside the two event horizons. But as two black holes get closer, there's a actually the way this happens is really interesting. You can get these caustics. So the event horizons is a three surface, right? It's like a topologically a sphere. But it can as they get closer, they can touch at a point. In order to look at the event horizon, though, you need the full 40 space time of data. Because the event horizon is defined as the last surface of scattering of photons that reach infinity, right? Future infinity. That means you need all those photons at future infinity to integrate backward in time to see where that last scattering surface is, right? So so it's in practice, that's not what's done. What's done is you you calculate something. You have a spatial slice of your space time and you look for parent horizons. And those apparent horizons are guaranteed to lie within the event horizon of the black hole. And so you look at how these parent horizons form and merge. There's a parent horizon. Yeah, parent horizon is where the light rays kind of bifurcate, right? It's a surface at which the light rays bifurcate. So you can do this in a way that is is only determined by conditions on a spatial slice of the space time. So you can you can phrase this as like a Cauchy problem as a initial value problem. And so you can look at these, it's easier to look at the parent horizons forming and merge. Usually what happens is the parent horizons get close and then suddenly the outermost apparent horizon actually encompasses both of them. So the universe is always kind of being shielded by all the weird stuff inside. So there's usually this discontinuous jump because all these things are done with computers, everything's discrete too, right? So there's always that discontinuous jump between the two parent horizons and one apparent horizon. If you want to look at the continuousness of it, you have to look at the event horizons. And then you have to do the ray tracing from distant future observer backwards and track all those, how those light rays travel and escape and arise from the black walls. And that determines the event horizons. And there you can have them form these really cuspy shapes that merge and create this long filament structure. And then and then that that that touching point grows. And then they create like this dumbbell shape. And then they form more or less a one black hole. So there's nothing, I guess, nothing that breaks that condition that nothing can escape an event horizon. One could say, well, where does the gravitational waves come from, right? Nothing can escape an event horizon. How does that, how do gravitational waves arise? Well, that's because there are these two point masses that are churning up space time. It has nothing to do with event horizons, really. It's just any sort of two point masses on an orbit forms this quadrupole potential and this time dependent quadrupole potential then emits gravitational waves. Anything makes gravitational waves that moves us certainly. Okay, I don't know if I answered your question. Yeah, I understood it was even more much more complicated. It was not naive whatsoever. Yeah. Thank you very much. Welcome. Thanks. Okay, thank you Scott very much. And if people want to ask him questions, they can email him or follow him on Twitter. And let us see us in two weeks with another webinar. And let's thank again Scott. Great. Well, thank you everyone. And yeah, this was great. Thank you very much. Okay, bye-bye.