 And so if you don't know, Jim is a professor in the Astronomy Department and part of the Astrology Program. All right, take it away, sir. OK, thank you for letting me barge in and give a talk on this. I've been keen to talk about some of the SETI projects that we've been doing for the last few years since I sort of got involved with AB. If you've heard me talk in one of these settings before, I've probably talked about stars and the active stars and all the things that makes Miles job hard, because I like the solid activity, which makes the data noisy. And so I'm sorry that my stars are so noisy. But it makes them exciting, yes. But I'm going to take a different approach today. And I want to talk about the SETI ellipsoid as one of the techniques that I'll share, which is sort of part of a larger project to bring SETI into the sort of modern era of what we call survey or data-driven astronomy. And so this is work that's being done with colleagues at sort of all the big SETI institutions that have sort of cropped up in the last few years, the Penn State SETI department, the SETI Institute, of course. This yellow ball here is the Berkeley SETI research program, which is mostly funded by the Breakthrough Listen. And recently, we also have a good collaboration started with the LST Solar System Science Collaboration, which is a mouthful. And so let me talk a little bit about this. This is the big motivator for me, which is this incredible facility that is being built in Chile, which is the Vera Rube Observatory, pictured here in CG, which will go online in about two years. First light will be under two years, like a year and a half, something like this. This is an 8.4 meter diameter monster telescope. So it puts it in one of the top 10 biggest pieces of glass that we've ever built. And it is a very short telescope, meaning it has a very wide field of view, which means it is used to survey large patches of the sky repeatedly for 10 years. So you've got a big piece of glass whose only job is to take repeated time series measurements for a long time over a large area of sky. This takes a huge volume of data. And so it's going to be good for all of your favorite astronomy topics, supernova, variable stars, maybe even exoplanets, and then everything else as well. And one of the big things we do here at UW, especially in the Dirac Institute, is work very heavily on the software that will drive this facility, both that will actually take the exposures and process them in real time, and then that will be used to make lots of catalog level explorations. So I hope you go tell all your friends, LSST, that's what the name is, the survey on the very observatory is the greatest thing that astronomers are going to be doing in the next few years. That's my biased opinion. You've got a 10-year survey that takes 30-second exposures. And so if we're thinking about SETI, the search for extraterrestrial life, intelligent life in the universe, so-called intelligent life, or technosignatures, as Rory said, there is a volume of parameter space that LSST and the very observatory would be very good at probing. And that is everything on the right side of this diagram. Anything that's longer than 30 seconds in timescale and up to many years in timescale, so here is up to millions of seconds, this program, the Legacy Survey of Space and Time, that will be conducted on the observatory, should be a great platform to study. Now this is a figure from a white paper about why we should look at high time series, so like nanosecond timescale stuff. There's lots of great astrophysics reasons why nanoseconds are interesting, but I'm not going to talk about that today, but I stole this figure from Shelley. So the right side of this diagram is what we want. You've got this massive data set. You've got sparse time series, but you've got 20 billion stars in this data set. This is like a profoundly large data set to study. You've got many targets over a long period of time, 10 years of study, and we have the capacity to do real-time monitoring. Within 60 seconds of the shutter closing, this facility will broadcast changes in the sky to the world. So we'll be able to do real-time follow-up and coordinated monitoring. This means it's a great observatory to use to trigger your follow-up facilities, your radio telescopes, your infrared telescopes, even J2ST at some point will be triggered based on this facility. So we need SETI projects. We need technosignature projects or life in the universe projects that can use this data. This is terrible data if you're used to things that astronomers have been doing like J2ST or Kepler or TESS or Spitzer, any of the things that we've talked about in this room with regards to life in the universe. This is like the worst data. The data, like the sun comes up and the data is only 1% photometry and it's in these big crappy filters. Like this is not good data if you really want to study the atmospheres of planets, for example. It'll be useless for you, I apologize. But it will still be a mountain of data for us to chew through. And so this is the motivating factor for me. We have this data. What are the techniques that we can find that might make use of it so that we can leverage it for the search for life in the universe? And then there's one good visualization of this from Jason Wright's work. This is the nine dimensional search parameter Haystack. Radio telescopes are somewhere down in here with 10 to the minus 18. I don't like this visual. Whatever, LST's on the top, fine. This is a better way of looking at it. Jilt Harder famously said that if all the parameter space we need to search for detecting life in the universe is equivalent to the volume of the oceans and water, then what we have searched for, at least as of 2000, was equivalent to a pint glass of water. That's how much parameter space we have actually searched when trying to find life in the oceans. And that is a precious small amount of water to infer the existence of like whales. Probably we're doing a bit better than that now. That was 20 years ago. Probably we're doing, Jason Wright says we're up to maybe a hot tub worth of water. So that's promising. Like you could actually find animals in the hot tub worth of water, probably not a whale. And I would argue that with some qualifications, this new generation of computerized, automated, large, data-driven surveys might get us up to a swimming pool sized volume of water. This is still tiny compared to the volume of parameter space you could possibly search in time and space and frequency and duty cycle and recurrence times. But I would put a lot more faith in a swimming pool than a pint glass. So I think this is an important step forward. Now there are many challenges when you're dreaming up a SETI program, as Woody can tell you from years of experience here. You have many things to look for and a range of believability or plausibility which you might search for. Sophia Shake has this great nine axis of signature merit that she describes. Things like are the signals long lived or short lived? Are they repeating or single events? Are they ambiguous or are they unambiguous indicators of intelligence or life? The same kind of arguments we make about biosignatures. Are they unambiguous biosignatures or are there many pathways to form them? Part of this also at the top of this nine axis are feasibility. It's no good if the signal you look for requires all of the money on earth to search for. That would be a prohibitively expensive search and so it's probably not a good signal to look for. So feasibility, cost effectiveness, can we actually do it? Do we have the data? These are important things. So there's what kinds of signals to look for but also important is when and where to look because it is an awfully large sky and we don't have unlimited amounts of time on telescopes. And it's that first bullet point. When and where should we look that the SETI lipsoid gives us maybe some help? This is a matter of signal coordination. If you were an extraterrestrial agent who wanted to get someone's attention and you were trying to beam messages in the hopes of being noticed, how would you get someone's attention? If all you knew was that maybe there's life out there, humans out there looking up, how would you try to rally their attention and stand out from all the other things that are happening in the cosmos? This idea of coordinating signals with natural phenomenon as like a game of sort of cosmic Marco Polo, if you will, has been around for a long time and this is one of the first papers I could find on this specific method from 1975. But let me put a cute animation that we have made up. Let's see, does this play? Come on now, this animation's more effective than it plays. Okay, okay, so you have, let's say, a supernova. Some kind of conspicuous event goes off. That information reaches another civilization and they say, great, here's something that we can piggyback and message on. Then the supernova reaches us. We say, great, a supernova. And then at some well-known amount of time later, this conspicuous other signal from another star system arises. So let's walk through this geometry one more time. You've got a conspicuous astronomical event. Let's say a supernova. You've got an extraterrestrial agent or civilization on some other star at some measurable distance away. And here we are with our radio telescopes or observatories watching the whole sky all the time looking for conspicuous events. We know the distance to things in the cosmos now. So we have some estimate of the distance to the supernova. We have an estimate of the distance to many of the nearby stars. Now I'll talk about that some more in a second. And so there is this growing ellipsoid with time where conspicuous signals might arrive. We call this the steady ellipsoid. Let me see if I can, there. There's this moment in time where if a signal were to arrive from another star at this specific delay time, that would be a very conspicuous coordination in time and space. And then this ellipsoid grows at time, eventually approaching a sphere. Okay, so here is the more technical schematic that was in one of our papers. All right, so again, you've got some coordinating source event in green here, earth here on the right. And there is this ellipsoid, this three-dimensional surface that grows with time that when it intersects stars, nearby stars, if a signal were to show up from that star, whatever the signal is, again, this is a when and where problem, not a what problem. If a signal were to show up, then it would be very conspicuously timed with regards to the supernova. Now these kinds of ideas are called shelling points. These are ideas about how you get someone's attention. It comes from game theory. If you had to find two people in a world, a video game, and you didn't know where each other was, you'd try to find some point of common reference, or commonality, or common communication. You'd find something to reference your attention around. And that's called the shelling point. And so this is one example of a time, a temporal, spatial shelling point. The upside of this idea is that we have all the pieces already. We see plenty of conspicuous events. Now there's a bit of observer's bias here, or human anthropic bias here. Like what do we consider rare, or conspicuous, or noteworthy? Probably things that happen once a day are not that conspicuous usually. We like supernova in our work because they happen relatively rarely, and they're very bright. They're good signal lamps to rally your attention around. They happen over one per galaxy per hundred years or something. But there are many other possible coordinating events that you could rally around. And it gives you, whether you like this idea or not, it gives you a very clear when and where to look. You still have to argue about what to do and how to get telescope time to look at stars. But at any given moment, there is a specific when and where you should look to see if there are conspicuous signals. And this cuts down parameter space a lot. The biggest uncertainty in this whole scheme, historically, has been the distance to other stars because we are governed by light travel time. We depend on understanding that the speed of light is the same in every direction. And so we need to know the distance that the light needs to travel so that we can build this triangle, this ellipsoid with time. And until recently, this has been the biggest issues that we don't have the distance to stars measured precisely. But thankfully, the Europeans have solved all of our problems and launched this beautiful telescope about a decade ago now called Gaia. And Gaia uses this very cool set of orthogonal telescopes that are fixed at a 90-degree angle from each other. It does this repeated scanning of the sky over a long period of time. And over many years, well, it doesn't really unfold the sky, but over many years, it creates the most precise, amazing map of the nearby universe of the Milky Way. And importantly, its whole job is to measure distances to stars. And it does it really well. It does it so well that 54% of the stars nearby have distances that are less than one light year. And astronomers don't often talk about light years. We talk about parsecs or other units of measurements, but I like light years because light year translates to year of uncertainty. If something has one light year distance uncertainty, that means we have a one year uncertainty in signal timing. And so our distance uncertainties directly translate into timing uncertainties. And so for a small percentage of the most nearby stars, we have distance uncertainties that translate into timing uncertainties down to weeks to months, which is actually really good. So the preferred target for this scheme over the last 30 years has been Supernova 1987A. This is a really important event. It happened in the nearby Large Magellanic Cloud Galaxy. Here is the most recent beautiful photo of it from the JDBST facility. You can see that it's beautiful. There's all these cool echoes around it from the interstellar medium where the light is reflecting off of stuff. There's cool shock physics going on. I'm not a supernova astronomer. That's everything I know about it. But what's cool is the geometry that we use to measure the structure of these rings and echoes and structure around it is the same geometry. It's the same math that we use in calculating the steady ellipsoid. But instead of light echoing off of dust, we're asking civilizations to echo signals at us. I like this figure because it's rare we get to see things so distorted to scale in astronomy. So if Supernova 1987A is at the center of this circle and we're at the right side, the information front, of course, is circular. The Supernova information front is a sphere. And the steady ellipsoid is this teeny, narrow, skinny oval. Okay, so here is an example of the steady ellipsoid for Supernova 1987A near us. Let me walk you through this diagram. In red are the stars that don't know that near us. Within a couple hundred light years, the stars near us that don't even know Supernova 1987A has happened yet. Like they haven't even seen it yet. You can see the curvature of that sphere as it's radiating outwards. Roughly half the stars in our neighborhood don't even know the Supernova's happened yet. It was only, you know, not even 40 years ago. This is like recent news in the galaxy. The stars in blue have seen the Supernova. So those are the ones, right? Obviously closer to the Supernova. We're at the center of this diagram at zero, zero. And the stars in pink are the ones that could have sent a message to us, right? They have seen the Supernova and enough time has passed that a message then could have arrived. And that is the ellipsoid. And these black dots are the ones that are on the surface of that ellipsoid. And so at any given moment, there are something like 140 stars intersecting that three-dimensional space. And we know their distances very precisely, better than a few percent. And so we can target those stars, their telescopes. So if you have telescope time and you want to conduct SETI searches, this would be one way to go and know which hundred stars to look at. Don't look at the ones in red. They don't know anything about Supernova 1987A yet. At any given time, you can go look up, we can calculate very quickly which stars are intersecting this three-dimensional ellipsoid. Yeah, I like this point. 1987A, this Supernova is still super recent news. Most of the stars in our neighborhoods still don't know about it, which is amazing. Okay, this is just geometry, right? We're just drawing circles and ovals in space. And so what's great about this is we can calculate far into the future when this ellipsoid intersects more and more stars, right? And you can calculate this propagation to the shape in time. And so it won't be until hundreds of years later, 25, you're 2500, that most of the stars in our neighborhood have actually seen this. This means that any given time, even right now, even though 1987A was in the 80s, four decades ago, it's still a hot target to look for SETI signals coordinated around. It means that there's always gonna be lots and lots of stars that are just now seeing this signal and could be broadcasting signals in coordination with it. Ian, why the wiggles on the 2000? Oh, this, I just took a slice through the middle of it. So I'm just smoothing through the stars that are available in like a little narrow slice through the sphere. So it should be a perfectly smoothing, but I got lazy when I drew the figure. That's why I'm telling you. An honest answer. Yes, as an astronomer, I am always lazy in my figures, but colorful but creative. Now again, we haven't made any prescription about what the signal should be or what kind of stars should be sending these signals. And so if you take the astronomer's classic reference diagram, the so-called Hertzsprung-Russel diagram, what we call a magta diagram, you see the main sequence here is the big line in the middle. The sun is somewhere in the center of that. We've got main sequence stars. We've got massive stars, m dwarfs. There's even a few white dwarfs that are intersecting this ellipsoid right now. Again, we're not making any statement about what kind of star or what kind of signal you should be studying. There are arguments to be made about what the signal should be. We're just saying which stars to study. Okay, so that's the idea. This is the idea of the ellipsoid. It's a when and where kind of situation. Good for planning observations. So what do we do? Let's go make some observations. And so that's what we've been doing for the last few years. We've been going through databases that are active right now, such as the test mission we've just submitted this paper, going through tens of thousands of light curves that intersect the ellipsoid from the test mission, 20,000 alerts from the Gaia mission, 10 million light curves from the Gaia mission. So far we have found nothing interesting, spoiler, but we haven't found anything interesting. But what we have done is develop good code that can run quickly through lots of light curves to figure out when and where you should look through both patchy sparse light curves from Gaia and also exoplanet light curves from the test mission. So we are actively trying to deploy this code through other surveys that are running right now and are excited about the Rubin Observatory in a couple of years, where we're gonna have to be able to do this on billions of stars every night. We're also not limited to Supernova 1987a. Now I've just talked about this because this is the sort of hallmark event that has sort of captured our attention, but there's many other events that might be conspicuous and interesting. And in fact, there is another coordinating event just this last year that we've started to get excited about. And that is Supernova-223-IXF, the memorably named Type II Supernova in the M101 Galaxy. Now this was an exciting event if you care about Supernova, which I don't know anything about Supernova, so it was just another event for me. But it was very exciting for the Supernova astronomers because it was the nearest Type II Supernova in a decade. And so it's been very well studied. People pointed all the big telescopes at it to get benchmark data on lots of Supernova physics stuff. And so we said, this is great. There are people who are looking at this Supernova already with all of the big expensive telescopes that we can't get time on. And they're looking in the right patch of the sky. Let's figure out which stars within a few days of the Supernova occurring, which stars are going to intersect that ellipsoid. And so because we had the math already done, we calculated which stars from the nearby Gaia catalog, and that's these black dots. They fall in this little patch of sky, as you would expect, because the ellipsoid is very narrow because the event has just reached us. But there are already hundreds of stars. At the time of publication, there was 108 stars and a little ball right around the Supernova. And that ball will continue to grow for the next few years as people hopefully continue to study it. And so what's great, I think, is this is a model of how we piggyback, right? We're doing SETI work essentially for free with big error quotes on the word free because people still need to get paid to do this work for all kinds of other science. But we are doing SETI, or we are getting observations for free from the community who are looking for other reasons. So lots of other telescopes went and imaged M-101. It's a beautiful galaxy. Lots of other telescopes at different wavelengths imaged this and got lots of nearby stars. Now, we are also observing it with radio telescopes. And so we started with the SETI Institute and the Alan Telescope Array. We started a campaign to monitor all these stars for a few months. And so that work is ongoing. And again, I'm here to sort of advocate for SETI as a discipline with lots of surveys, with lots of studies, with lots of facilities. We should be running this on every other facility that you can think of. There is code. This is only one idea of many. This is part of a larger project to do what I would call Optical Survey SETI, or Survey SETI. Again, SETI, using these big data-driven time domain surveys. And so this is a new slide that I just made, which I'm really proud of, which is that we've got lots of papers now from many students, from many different projects, not just on the SETI ellipsoid, but also for other projects to see Kepler data, looking for laser lines and apogee. We're already starting projects with Annie Zanadakis, looking for rare stars in ZTF. We have lots of projects. This is just a slice of what's coming to publication now. And there are many future exciting projects with these surveys. One that I think will be a really big hit here at UW in the next couple of years is looking for unusual moving objects. We have a huge solar system group within the Dirac Institute that studies lots of asteroids and comets. And we'll be on the forefront of looking for interstellar objects, comets from other star systems. We know of two of these already. There is debate about whether or not one of them was artificial. I don't want to get into that debate right now. We can have drinks and hash it out later, if you'd like. But the point is, if a comet or a interstellar object makes a right turn, we should be there ready to watch for it. We should be able to determine that that thing made an unphysical change in velocity or in its trajectory. We should be able to detect that with our observatories and we should be able to track it. Similarly, if stars start exhibiting truly strange variability or patterns of variability or started blinking morris code to us, we should be ready with our algorithms to troll through our databases to find these kinds of signals. And again, I'm very excited about this project, which is harnessing huge spectroscopic databases that we already have. This is an example of high resolution spectra from a sunlight star in the infrared, from the ground. And looking for, I have colored on, an anomalous emission line. We have, from the Apogee survey, which UW was a founding member of at the Sloan Digital Sky Survey, we have something like 700,000 stars in our galaxy with high resolution spectra that we could use to look for anomalous lines. Now, of course, there's lots of caveats because the atmosphere is in the way. You're looking for weird things while you're looking through an atmosphere. So we just heard this is a problem, but we have a huge data set that we can draw from. And again, lots of good code that we can write to do this quickly. And so that's my conclusion. Is that the SETI ellipsoid is perhaps an efficient way to select which stars to monitor. I'm not arguing what you should monitor for, but if you want to do SETI monitoring, this is a good way to pick some stars to go after. And it keeps providing stars over time. Gaia is the transformative technology that makes this feasible for nearby stars because we can finally, with precision, measure the distance to nearby stars. Supernova 1987a, while it's old news to us, is still relatively hot, exciting new developments within the galaxy. And we haven't ruled out SETI signals from even stars than 100 parsecs within a few hundred light years of ourselves from coordinating signals, right? Only a small percentage of stars could we say haven't sent a signal. And I think this is just one of many, many projects in this new era of survey SETI. And I'll take any questions.