 talking to us about his current future research and any astrophotography at any time that he can change. Yeah, there's a camera. For his astrophotography, we'll be able to see this later as well. Excellent. All right. It is a really big honor to actually stand up here in front of the people who taught me everything I know about astronomy. That's probably not an unfair statement. So thank you for inviting me. And thank you for coming in the post-AS haze of 1,000 emails that we all have to answer. My talk today will feature lots of familiar themes to many people here. Lots of familiar words like Kepler and Galaxy and stars in my conference. But really, this is a thinly veiled excuse to try to get people to come have coffee with me, which is really my passion. And I really want to highlight the work of the students that I have the joy of working with and working for. Both here and at my previous institution at Western Washington University. So this whole slide deck is basically just me stealing cool stuff that I'm happy to be associated with. And we'll talk about a couple of recent papers that have come out. But I will return to this point. And so there is a coffee machine in my office. And please do come and talk and share in this talk. Because that's the reason to keep coming back to this university like I have so many times. And I think that's ultimately the reason to have this job. OK. Broadly, my science is interested in things that come from objects like this. This is a sun, an ultraviolet. Thankfully, this is not what your eyeballs see. This is probably closer, not exactly what your eyeballs see here on the right. And on the very interesting things that we can see on the sun, things like magnetic activity, flares over here on the left, the science explosive event, star spots wandering, evolving, moving, spinning, as we see on the right. And broadly, I'm interested in how do we take the amazing things that we know about our sun, known for a long time, 400 years in some cases. And how do we apply that? Or how do we measure those things around other stars, using missions like Kepler, missions like TESS? So a primer, just to sort of set the stage a little bit for the data that I'll be talking about. And the data that's sort of entranced to me for the last almost decade. The Kepler mission, study one field in the sky here in the little statue numbers. Right here, above the plane of the galaxy, 10 to 40 degrees out of the plane of the galaxy. And it just stared there for 40 years with a nearly unblinking eye, taking exposures of about 200,000 stars, imaging. What are you? Can you turn the lights off? Yeah, sure. So it stared at basically the same patch of sky for four years. Oh, that's not, that's wrong. Ooh, that's too extreme. Try to undo the point. There it is, the happy middle. Perfect, thank you. Thank you. Okay, so four years of data on one field of view. And then tragically Kepler died, and like all good things, all good villains never died once. So it was reborn into what's known as the K2 mission for Kepler 2, where they could keep this spacecraft quasi-stable for roughly 90 days at a time, and thus you could stare at fields for about 90 days, four fields a year, as it processed around, and this is an older graph, they're observing almost 20 fields of view over another four years. So nearly 10 years of operation, and in total about a half million stars, with a revolutionary data set. Now Kepler was designed from the ground up to do this. And I remember being at the WAS in DC in 2010 when this figure came out, and it was like that packed first day everybody actually got up at 8 AM to go see this talk. Which I apologize, I didn't make a bunch of 8 AM talks, Julie, I'm sorry, I didn't do that. But last week. But this was really exciting. This is the first exoplanet discovery announcement from Kepler, where they showed five transit planets in the actual data here, and we all marveled, this was from 30 days of observation, we all marveled at like, how clean these light curves looked, how a little scatter there was. What is this interesting variability at the bottom of transit? There's lots of excitement here about this. But this wasn't exactly my jam. This came out a little while later from Gabor Bosry at Berkeley. This is cool, this was comparing that same 30 day window to the sun, or sunlight stars. So here we have solar light stars, these are so-called quiet stars compared to the sun on top. And then these are the spotted stars again compared to the sun on top. You can see there are stars that are very, very quiet. Quiet is in very little modulation in their brightness over time. And stars that are very, very loud, or very full. This was cool, but what really changed the course of my career was this paper by Lucianne Wachowicz, who was a former grad here as well. And she put this early release paper about just the magnetic activity as viewed by Kepler. And here we have, as an example, four K-stars with 30 days of data from Kepler where you see they're all wildly very big. There's quasi periodic sinusoidal modulations. There's little outbursts. This is in fact exactly what I showed in that first video of the sun. This is flares erupting, socastically all over the star. These slow sinusoidal modulations are star spots rolling in and out of view of the star. Here we are observing four random K-dwarfs and they're all going just totally wild in this data. And this, this was like a light bulb that went off in my head that said, this is what I want to work on for the next, well, as long as Kepler lives. The takeaway point that I've come to in the last few years is that Kepler data, while measuring all this amazing variability, is really more akin to taking a high resolution spectrum of a star. So here's what we traditionally think about a spectrum of a star, obviously a very detailed spectrum indeed, where we have the brightness of the star as a function of wavelength or the frequency of the photons that reach our instrument. Here we have some absorption line and if we think back to our astronomy, 300 series classes we can think about the width of that line might tell us something about the surface gravity or the center of that line might tell us about the radial velocity, right? This tells us about the physical, fundamental properties of the star, the shape of this continuum, tells us something about the temperature of the star, things like this. And the Kepler data really provides, I think, an analogy to this, and then provides what I call, or what is called the power spectrum, right? So instead of photon frequency, we have time frequency, how often things happen. And here we have power, which is kind of like flux, it's like how much light comes as a function of this frequency. And so when we look at this, this power spectrum or periodogram is sometimes called if you show wavelength or period instead of frequency. This peak would be like the rotation period, like there'd be a lot of power, a lot of modulation that correlates with that frequency or that period. That would be like the rotation period, for example. And in fact, this allows us to do really interesting stellar astrophysics using just the brightness of the star measured over time. So we can measure fundamental stellar properties just like in a high resolution spectrum. And then then we can extend this analogy or abuse this analogy a little bit. And we can measure, we can determine things like the spectral resolution based on the cadence and what wavelength or frequency domain you're sensitive to based on the cadence and the baseline. I'm sort of waving my hands at this point because this is not strictly true. But the similarities are actually many. And what I really like about this analogy is that Kepler with its 500,000 stars, half million stars observations is actually really competitive with some of the biggest spectroscopic surveys done today. When we think about things like Apogee, which UW has had a big part of building, and other medium to high resolution surveys, the biggest spectral surveys of stars have half million, maybe a million stars with good fundamental parameters measured. And so Kepler and K2 together have moderate to high resolution spectra, if you will, for a half million stars. And with tests, we expect lowish resolution spectra, again, Eric votes, for something like 20 million stars. This becomes a really competitive data set if you can figure out how to pull out the properties of the stars from these light groups. I think that's really cool. Okay, yeah, tests is like low to medium resolution. We'll talk about tests more. As an example of some of these fundamental properties, this is not work that I've done, but I think this is really cool highlights of things that have really changed the field in the last few years. On the left, we have astrophysic, astro-sizemology or astero-sizemology, seismology being just like vibrations, just like in geology, seismology of the earth, and astro being stars. Hopefully everybody here knows that. And from the oscillation patterns that we see in the stars, especially for giant stars here, you can back out the stellar mass and radius directly. There's been similar work by Fabian Bostan and collaborators about how the higher time resolution or the sum of flicker, she here measures the eight hour flicker or the sort of unclear variability on eight hour time scales, how that correlates to log G or surface gravity of the star. And so we've been able to measure surface gravities just from light curves for 100,000 stars or so. And there's even indications here. This is a new paper, which suggests that we can actually constrain stellar inclination based on the very specific beat frequencies and things that are measured in the power spectrum. Very fascinating work that I think is really cool. Closer to my wheelhouse would be things like stellar magnetic activity. So here I think, I'll generalize and say our three seminal works on stellar magnetic activity. I think if you haven't read this paper by Brett Morris, a current grad student here on hat P11, this is a spectacularly cool paper, which I'm certain most people have here talked about, where we actually have a planet crossing its host star and we can use that to constrain the latitudes and longitudes of the star spots. And from that Brett has been able to extrapolate what the global spot pattern looks like on the star and shown that the dynamo must look something like the suns, that there are preferred latitudes that star spots show up at, things like that. Really cool, really fundamental astrophysics, stellar astrophysics coming from this system. Koski Anakata, who visited actually, you know, earlier this year, or I guess in 2018, has been doing some really cool work in measuring star spot lifetimes. And then I will talk more here about stellar flares and while Kepler has been an absolute revolution for measuring stellar flares. Something that I think will become seminal in the coming years is what Kepler and Tess will have to say about, say, for example, massive stars. Ends of the HR diagram we don't usually associate with these same kinds of variability. This is sort of the wild west right now. There's not a lot known about these stars at these time resolutions. And so Trevor, also the current battle right now, has been doing really cool stuff with the very earliest data out of Tess. I'll let him talk about that more, but I'm really, really excited about that. So with a half million stars along various lines of sight here in the artist's rendition of the K2 and the Kepler campaigns pointing to different pencil beams through our galaxy, we can start to do what is called galactic archeology or understand what the structure of our galaxy looks like as understood through the constituent stars that make it up. Now usually when we talk about galactic archeology, again, we talk about spectral catalogs, things like apogee where you're measuring the chemical compositions of stars or the alpha abundances of stars, things like that. Here we have enough stars with ages, with radii, with fundamental properties, we can start to do galactic archeology. And the figure that I want you to keep in your head for the rest of this talk is this well-known relationship from many years ago and even creating this, I just like this particular version which is stellar age as it correlates to magnetic activity or the surface magnetic field strength, something like that. Where we have very young stars. So in this case, we have five sort of benchmark stellar clusters. We know the age through other mechanisms and we have measurement here, in this case, it's spectroscopic measurements of the calcium flux due to the chromospheric emission. And this correlates with magnetic field strength and so we see here young stars on the top left, my glacial point is dead, young stars are very active, have a strong magnetic field and old stars or older stars like the sun are considered inactive. So despite seeing flares and spots, the sun is considered a quiet and pleasant place to live. And along with this evolution, we see and was broadly theorized that we should see more flares when they're young, that they should rotate faster and they should have bigger star spots. Stronger magnetic field should drive all these things and should be related to all these things. And similarly, at old ages like the sun, we should see slow rotation, the sun rotates in something like 25 days. Fewer flares, smaller star spots. Again, a much more pleasant place to live, all around. And understanding this relationship, how narrow is this line, is a lot of what I think we can do with Kepler. You want me to fix that? Sure. Okay, so here is an example of a flare star that shows both flares, the explosive big things on top, and the slow sinusoidal modulation, which is the rotation. This is an active, fast rotating M dwarf called GJ 1243, which I spent a long time working on when I was here as a graduate student. This is a rapidly rotating M dwarf, 14 hour rotation period. And from Kepler, we had 11 months, 300 days, of one minute cadence data, where we were able to measure light curves that look like this for 11 months straight. Now this is just two days, two random days, truly just random days on the star. It's incredibly active. The biggest flares here are as large as the biggest flares ever seen on the sun. So this star, off seasons off once or twice a day. It's really, really active. And from this, we've generated the largest catalog of stellar flares for any single star besides the sun. And this was fun. We knew this was an active star going into Kepler. So for example, this two day window has like 48 flares, 50 flares, something like that. So it's flaring roughly once an hour in Kepler. But this is like the boutique analysis. This is what you do when you have brand new data. You comb through it and look for every little weird thing and count all the butterflies. What we'd like to do is find all the flares in all the stars. And this is what I spent my postdoc at Western working on. And so without getting into any of the details which are interesting and tedious, we searched through all of the light curves from Kepler. So that's something like 200,000 stars, both the one minute and the 30 minute data for those playing along the column. And the biggest effort, of course, is removing all of the garbage, be it astrophysical garbage like the star spots that are interesting or just systematic garbage, like problems with the spacecraft. So a lot of effort was spent doing that. And the goal was to find every single flare. Here I just highlighted two examples, the rapid rotator and a moderate rotator. You can see the red stuff is the flare. So good, it seems to be working. Now, this is time consuming but not difficult. The worst part about this is you have to determine your incompleteness. And so what we spent most of the time doing was injecting fake flares back into the data and trying to figure out at what point did we no longer reliably recover them, which is exactly analogous to what's done in the exoplanet community. Okay, so again, going back to the view of one star, here is the metric of interest, the diagram of interest for flare studies. This is the flare frequency distribution. So on the X axis, we have the event energy. So the biggest flare ever observed in the sun is something like 10 to the 32 works. So we see this is the same star. We see that it has events that reach out two orders of magnitude, at least, greater than the biggest event seen on the sun. And the Y axis is the cumulative number per day. Yeah, that's right, cumulative number per day. So at 10 to the 32 works, again, about what we see for the sun, we see a few of these per week. They're at least that big. And then like many things in nature, you see this wonderful parallel relation, a sort of straight line in log log space, which tells you that there's lots of little things, lots of little flares, and very few big ones, which is good. So this is what we'd have for one star. And again, the goal was to do sort of the industrial scale analysis of all the stars. And so this was the result from the paper that just hit the archive last week, which is when you look at many stars, here in this case, these are a bunch of K and early M stars. We can make that same diagram, flare energy versus occurrence frequency, but we can sort them by the rotation period. And I told you in that one diagram keeping your mind that as the star ages, it slows down, like a top, losing angle momentum over time. It gets slower and slower and slower. And so the older or blue lines here, the slow lines, should have lower magnetic fields and thus lower specific rates of flare. And what's so lovely about this is that it's very easy to see. There's clearly a rainbow, I guess unless you're colorblind, it's not as easy to see. But in general, it's fairly easy to see that there's red stuff at the top, fast, rapid rotator stars that have high flare rates and at the bottom, slow, inactive, less active stars. Somebody called this the unicorn plot because they looked like the main of a unicorn they said, so that was cool. Okay, and we can write down equations. They're not fun to look at, but you can write a power law equation that includes all of these different lines and tries to fit the evolution of these things, blah, blah, blah. And the point being that we have a tool now that populates this space of cellar mass versus cellar age, where we're making some assumptions about age. And the point is we have a new metric that shows that the M-warves here, the low mass stuff here around mass 0.5 and lower, stay active for a very long time. Their activity rates here, that's the color map, decline very slowly. Whereas the solar type stars here are a mass of one, but in solar units, decay more rapidly. And this suggests that we could actually use cellar flare rates as an age indicator, which is, I think, really cool. So here we've made a very specific measurement using the time series where you count up the number of flares as a function of energy. And what you get out now is an estimate. It's not a great estimate. I'm not showing you the error bars here, which are very big. But you get an estimate of the age, just from counting up the flares. That's quite cool. Okay, but that was just for the Kepler field. And as I said, K2, the rebirth of Kepler, doubled the size. And so, not only did it double the size, it goes through different parts of our galaxy. Here's the glass plane in this gray line. We have fields up here, like field 17, which is right towards the North Galactic cap where the old stars or the halo stars live. The flare rates should be very low up there. And then down here, field nine, smack in the middle of the Milky Way plane, field 11, field zero over there. The flare rate should be very high. So this is gonna provide an independent test of the flare rate as a function of age when we look along different lines of sight of the galaxy. And of course, the future is bright because tests exist now. Tests launched mid last year. This is the sort of successor to Kepler. It's a smaller telescope, so it's not getting as faint. But as you can see through this sweet 3D animation or our graphic here, eventually it will survey more than 80% of the night sky. And so that, as I said earlier, will amass 20 million stars with time series. And this is, I think, important. It's simultaneous with the ground. So in fact, starting later this year, it will survey the Northern Hemisphere. Starting later this year, in cycle two, so it did the bottom hemisphere first, this year it will do the top. And so we'll be able to observe these stars at the same time, say, from a patch of point, which is really exciting. And indeed, the very first data is coming out of test now. Spencer, I put your name on here because I think later this year, Spencer and I are gonna look at how we mined these light curves, like we did with the Kepler light curves, to try to find the flares. This is my two-day hack of trying to put this together. And this is an obvious, easy case. But you can see the data looks qualitatively similar. So a bright future of tests, 20 million stars we can hunt through, which is all over the sky, which is exciting. And then I'm really, really excited about this. This is a tweet that Tom Barkley, at the test, guest observer office, put out last month. He said, look where NASA tests is gonna be pointing. And for those who don't know, this is the Kepler field. And so he tells us here, July 2019, so market calendars in July of this year, tests will revisit the Kepler field that had four years of observation. And importantly, 2019 is about 10 years after Kepler first observed the Kepler field. And so this lets us think of all kinds of cool, and going back to the sun, stellar astrophysics that we might investigate on, say, 10-year time scales. For example, something I'm very excited about, Brett Morris just published his new paper just this week, where he created a synthetic liqueur for the sun. This is the famous butterfly diagram that discovered the solar activity cycle. We see the latitude of sunspots migrates with time and seems to have a repeating sort of butterfly shape every roughly 11 years. And he creates this synthetic liqueur for the sun. There's no least re-liquid when he can recover as expected, that nice 11-year peak. Good thumbs up. We haven't been wrong for 400 years. What's really interesting is this 10-year liqueur span that we'll have available from the very first pointing of Kepler and the new pointing of tests allows us to poke at these things. So for example, we know from looking at the sun. So here is another way of looking at the butterfly diagram. This is just the integrated sunspot area. And again, it looks like an 11-year modulation. Boom, boom, boom. Here's four cycles to look at. Here is, at the same time, the parallel index or the intercept of that parallel law of the flare rate. And you can see it goes up and down by something like a half or an order of magnitude over the solar cycle. That means, here's another way of looking at it, that means this flare frequency distribution should move up and down with time over a decade, which is really exciting. And so if you can measure the specific flare rate within a year, say, this is just one example of a Kepler where we have lots of quarters of data to look at, we should be able to actually constrain activity cycles for stars just by counting the number of flares as a function of time, which is really exciting. And so I have a student who's been starting with this in Kepler, looking at the four-year data from Kepler, looking for stars that might have evolution, if not a full cycle. And so here's just two candidates that he's pulled out. These are Matt Sagan's and Kevin Covey, a former, he's a professor now, former student here at UW. These are two candidates who have colored the flare frequency distribution by time. And again, you can see the nice sort of uniform plot that I'm always looking for, nice rainbow where we see blue over here, red down here, so I think you ordered it blue-red at time. So this is one where it's decreasing sort of monotonically at the time. This is going to be the first candidate we go back to in Kepler and say, what is the new flare rate now 10 years later of the star? Has it gone down even further? Has it gone up? I think this will be really exciting. Okay, so if this kind of thing interests you, again, put a pin in your calendar for July 2019, later this year, and I think we'll be able to very quickly follow up some of these things. This I think is a very exciting project we can work on. So if this interests you, come talk to me. There's a lot more to be said about flares that I'm like sweeping under the proverbial rug here. One example, we've just been counting flares, like just butterfly collecting, counting them, but they're actually interesting events in their own right, with lots of cool high-energy astrophysics going on. One cool observation that Kepler makes very routinely, and Tess does as well now, is that many flares don't look like this normal simple explosive rise and sort of thermal cooling exponential decay, but instead they have this complex multi-peak structure. There's maybe an indication of like oscillations going on in the tails. There's a lot of like detail in the so-called morphology of these flares or the shape of these flares. And something that has been on my back burner for a long time now is going through and looking at details of this morphology, deconstructing them as I happen to sort of cartoon here. And this is something that I would love to talk to anybody who wants to think about fun statistical methods for decomposing these events, using interesting samplers to decompose these events. I think there's some fun work to be done here, and the sample isn't massive. We could do some fun stuff. Okay, that's all I'm going to say about flares. And let's spend the rest of the time talking about star spots. Now, again, this is that same animation of the sun over a couple days as the sun rotates. This is an example of a large star spot, though not unusually large. It produces spots about this large, maybe once per decade. But I think George told me once that he saw this spot naked eye. And like during a forest fire or something. And so I've always thought that's really cool to be able to see a star spot on an R-star without a telescope. I don't like to do that. But don't look at the sun. That would be bad. I have to say that. Also, it was in 1947. So if you Google like sunspot 1947, you'll end up with lots of funny stuff about aliens. So it's kind of a hard thing to Google. That's the same year that Roswell happened. Okay, star spots, again, are very interesting astrophysical laboratories. They're regions of strong magnetic fields poking through the surface. They inhibit convection and make a little, little cool spot. Hence the name spots. Very easily, monitoring the brightness of a star in my little animation here, gets you something related to the spot size, or filling factor if you will. So a bigger, darker spot will make a bigger modulation in time. And a faster stellar rotation will give you a faster sine wave. Now the picture becomes very complicated. You have many spots constructively and destructively interfering with each other. But to first order, this is a very good measurement of stellar rotation. Much more difficult, but is being done now, to some degree, with Kepler, is moderating a different rotation, as we see on the sun. And the spot lifelines, as I said, with Koske Namakata's quick new paper, measuring the evolution of these things in time. And then here I highlight two things that I think are sort of open questions, which are RV work done, but are very difficult. The first is stellar activity cycles, as I previously mentioned. The second is the change of the physical spot properties with stellar age. So the spots are the spots darker or bigger as a function of stellar age, over a giga-year time scale. These are very hard things to measure. Sort of unanswered questions, but active work. Lots to be done with star spots. And what makes me so excited about Kepler was the thesis work from Amy McQuillan out of Oxford, this is her 2014 paper, where she measured, I think they call this the shrimp diagram, because I think it looks like a little cocktail shrimp. This is the largest catalog rotation period ever amassed, homogeneous rotation periods ever amassed. Before Kepler, it was hard work measuring stellar rotation. There was something like order a thousand stars, maybe, with different methods, different sources of data, different surveys. And now just in Kepler, we have over 30,000 stars, where she's colored by different samples all over the block. But the point is, there's this really fascinating structure. Well, there's lots of structures here, in fact. Let's talk about them. The first is the overall distribution. So here we have stellar mass in the normal astronomy backwards units. And then here we have rotation. So fast rotators down here, again, young stars. And old rotators up here, the slow stars. And there is this theoretical three-dimensional manifold-shaped surface, whatever you want to call it, which we call gyrochronology or spin clock. Which says that a star will slow down, and the rate at which it slows down is predictable. That it loses angular momentum at some predictable rate, and thus slows down over time. And the point here is that you can draw these lines of constant age, also known as an isochrome. They should be older stars, and this is a cartoon, so the shape is not quite right. But this is the general picture. And if this picture is to be believed, and there's lots of reasons to believe it, then you should be able to just look at this diagram for a star. But given mass, let's say a solar-type star, you measure its rotation period and you have some estimate of the age. What a useful and promising tool we thought. Can the mass come from astro-seismology? In this case, this is color because these are dwarfs. And astro-seismology from dwarfs, while not impossible, is still very, very difficult and limited. You need only the brightest stars to do it. It's still very hard. But they're pretty accurate. That's right. Well, yeah, sure. They're as accurate as we can be, I think, in this case. Yeah, probably. The masses are good, but probably 10%, something like that. The rotation periods are good to a fraction of a percent, because we have in this case, years of data. So the measurement error, very small, usually here. And this is a great illustration. I love this idea because this promises that if you just stare at this guy long enough, you'll be able to measure the ages of all the dwarf stars. And that seems great and straightforward and doable. But of course, nothing is ever that easy. And that's why we're employed. And this is great work with Rory and David Fleming. We just wrote an NSF about this. I'm really excited about this, where we take into account this other well-known property of stars that a lot of them are not alone. It's like that lots of stars are actually in binaries. And in fact, we're very familiar with tides. Here in Seattle where the tide comes in, the tide goes out, thanks to the moon. Tidal forces between binary stars can really wreak havoc on the rotation rates. Perhaps not surprisingly, but up until recently, up until Rory and David Fleming started looking at this, really, I think, underappreciated. And what this results in, okay, now these people are theorists, and I don't understand this. They put massive increasing order left to right. They're foreign to me. But actually, I couldn't find the figure he sent me the other way around. I'm sorry, David. But the result is this is like a scam. You have rotation, and you have the orbit of the two stars. And this exerts tidal forces on each other, and this totally messes up the rotation from the prescribed spin down that you would expect. And instead, wherever the orbital period is, you end up getting tugging and often tidal locking or resonance locking, things like that, interesting things that can cause scattering in this space, which means you don't actually predict the age correctly. That's a problem. We think there's hope. But the NSA should give us money if they ever go back into business. One of the other really cool features of the shrimp, of this age rotation mass diagram that got me really excited was an observation by, again, annual film from Oxford, that there is a bi-modal distribution here. So this is just the end dwarfs here, the tail of the end dwarfs here. You see almost like a forked tongue of a snake. You see these two little populations here. And when we do a histogram, you can see there's two bumps. Okay. A period bi-modality. Interesting. That could be interesting. They wrote a subsequent paper where they were able to show that this wasn't just in the K, but the end dwarfs, but also the K dwarfs, the bi-modality. And it's still kind of in there here at the K dwarfs. And then I wrote a couple of papers where we actually found, with Gaia, thanks to Gaia, we realized that a lot of the G stars were actually subgiants. And when you throw out all these subgiants, here's our color magnitude diagram we were all indoctrinated with at birth as astronomers. The subgiants here, they have their own problems when it comes to rotation. Funny things are happening inside the star. And so the spin-down clock gets broken. And when you remove the subgiants, what you see is the shrimp actually has the bifurcation or the bi-modality extend all the way down to the G dwarfs. So the mystery has deepened. Now we know that there is a break here. And if we think this is an age feature, this correlates to something like 600 million years. This means that there are not none, but very few stars in the solar neighborhood with a specific age of 600 million years if this clock is to be believed. Interesting. You know, don't believe me, but interesting. There's two possible explanations that have been proposed. The first is that this is actually just an age feature. We're just tracing the stellar star formation history of our nearby Milky Way. So this dip, this bi-modality or this dearth of stars at this age represents a sharp decline in star formation at this one very specific time. That's interesting, not impossible, but unlikely. And I say unlikely because we have work from nearby galaxies, from M31 and others, which have star formation histories mapped over a few hundred parsec scales, just like Kepler looks at, and they don't find modulations that are this sharp over these hundred nanoscales. So this would be a little surprising and a little hard to explain in a galactic formation standpoint. But it is an explanation. The other is that something happens, and that's kind of as far as we've gotten, that something weird has happened in stars, they spin down, the clock is working, and then they just like jump forward in this time space. There is some preferred rotation rate that stars don't like to have, and they jump past it. And the mechanism for which this occurs with this is totally unknown, which is awesome, but troubling. Like what is it about this rotation period that stars don't like? I have no idea, but there's lots of reasons to believe and instead it is totally a viable explanation. We just don't know what is happening. So this is really exciting. Now there's various tests, we can look in different lines of sight, and of course we're doing that. One interesting hint of this, again going back to mapping stars with guide, as in this bio-modality here, if I bend it up by height above the galaxy along the Kepler outside, I take little slices in the galaxy. Now this has been known to correlate with age, that older stars live higher up in the galaxy, in the mid-planations in the mid-plane. But the bio-modality is more prevalent in the mid-plane. Okay, so there's more young stars down here. But what's interesting is that this sort of seems like a pivot point in all of these, for all of these heightings. So this does seem to be a spectral rotation period. Stars exist there, but they go through there very quickly. So something very strange is happening here. This is something that I want to keep working on. So we have work, we have a lot of work to keep going on this. To extend both the sample rotation periods with K2. So I have a mass proposal with Ruth Angus that was funded last year. Which I think we're going to manage a rope in Tywin Gordon to work on a little bit, which I'm really excited about. We're going to measure rotation periods for all the stars along all the different lines of sight in K2. And see if we actually see this shrink or this bi-modal feature show up. And then as ever, we can actually measure stellar rotation for 20 million stars perhaps. Well, it's probably not that many, but something like several million stars with tests. And then Gaia, of course, if you have played along at all with Gaia, this is a groundbreaking mission. One of the cool possibilities here is that with all of these different missions measuring rotation in different directions at different distances, we might start measuring the age distribution of our galaxy just from rotation alone. To the point where we could actually measure age differences in leading and trailing ends of the spiral arms, for example. Or we could measure clumps of ages, perhaps from waves of star formation. So there's lots of cool things that we can do when you start surveying and do galactic archaeology. Okay, one of the things I'm excited about as an example is that we've started this as a dedicated K2. I'm a first student, Zoe Bell, and I've got an undergrad here at UW, Sam Quist, who have been working on measuring rotation in the Everest data. So this is the K2 reduction that Rodrigo Luger at all produced. And I can highlight it again because it's a little hard to see, but this bio-minality does seem to exist in a couple of the other lines of sight. So this is really exciting. This means that this feature is generic. It's not just something weird about the Kepler field. We're seeing it along different lines of sight. It's something weird might be going on with these stars. It might not just be ages. One of the offshoots that I'm excited about that I'd like to work on in the future, and so I have a student that's going to be starting this this quarter, but it's an invitation for people who want to think about this, is the rotation of binary stars. So Gaia provides a CMD, a color-match diagram that's so precise that we can pick out obvious binary stars. So here is a line. This black line is the zero-age made sequence. And we can pick out stars that are a factor of two brighter and show that they are very likely to be equal mass binary stars. And one of the interesting observations is that the rotation period distribution of these stars doesn't seem to be exactly the same as down here. Now this is further evidence that maybe tides are important, for example. So this is something that we're going to be working on for the rest of this year, 2019. And another observation that came out of this most recent paper is the possibility of measuring the ages of stars just using the main sequence location alone. So here I've blown up this one little patch of the main sequence here. So I've chopped out the obvious binary, so it's actually just this little piece right here. And here I've changed the color scheme to hopefully dry your eye a little bit easier. The blue stars are the rapid rotators, and the red stars are the slow rotators. And what we can see is there is a clear gradient here from rapid to old. So this model that stars are changing brightness suggests that these must be the young stars and these must be the old stars. Sorry, they change the rotation rate, excuse me. And indeed when we look at a theoretical model of how stars change, this is from the Mesa isochrome hybrid. This is tracking the evolution of a cave war over 10 dating years. There is a slight change in the brightness and the color. The isochrones actually move diagonally to the main sequence over time. And so we think this is showing that from Kepler plus Gaia alone we can actually measure very precisely the ages just using the Gaia isochrones. Now there's a big caveat here that metalicity is a huge problem. Metalicity actually will draw stars the opposite direction. The old metal core star should be over here. And so in my mind one of the outstanding questions is why is this signature actually as strong as it is? Why are the old stars as high up as they are? Why aren't the old stars over here where they should be metal core? So an ongoing project is to try to figure out the metalicity of all the stars in the Kepler field and see if we can explain why maybe one example is that metalicity is going over there and that this is underestimating how much the isochrones should move with time, for example. So if anybody knows how to use the Maze isochrome grid, I think there's a wonderful project to explore different evolutionary tracks or do any stellar population synthesis models of the Kepler field that include rotation and trying to model this interesting period gradient. Okay. This is a 10,000 foot conclusion and that is that Kepler has been the tide that has lifted all stellar astronomy boats. Kepler was launched as you may have heard in fact I mentioned earlier Kepler was launched to find exoplanets find a planet like Earth and so it had to stare at stars for a very long time and in doing so it learned more about stellar astrophysics than almost any space mission with very few exceptions like anything else in the last decade. Kepler is actually the best stellar astronomy mission in the past couple decades and we basically got it for free because we convinced the community to pay for exoplanets which is great, right? Because exoplanets go around stars and galaxies are made of stars so everybody wins by us measuring exoplanets. This is good. And then in more detail Kepler is finding that stellar flares are perhaps a viable agent here. Because we're not going to get a Kepler light curve every star but we are going to get an LST light curve of approximately a bajillion stars and the stellar flares show up while these small modulations from the star spots are very hard to measure. These flares especially in the blue bands the LSTU and G bands can be very big, sometimes magnitudes large and so this flare rate is perhaps a viable age metric for the younger stars right out of the box from LSTU we'll be able to measure ages from stars out to several kiloparsecs of LSTU just by measuring ensemble flare rates. Star spots themselves very interesting and very useful for constraining the physical properties of stars and maybe the ages we imply from these rotations can help us map the age distribution in high fidelity of our nearby Milky Way possibly making analogous star formation history maps what we see with the fat survey and the ox survey, things like M31 so I think that's really special and then my final point which I've been trying to make is that time domain astronomy particularly fueled by Kepler and by TESS especially when supported by things like Gaia and soon LSTU offer a new opportunity to study black archaeology from time series data alone we find interesting trends that might reveal subtle details about the lives and evolution of stars and we might find maps of ages out to several kiloparsecs which would help us map the entire age distribution of our galaxy and then my last advertisement, shameless advertisement to come have coffee or come talk with me is that we've been holding a group a pseudo group meeting at Mondays at 10am so if you're interested in these topics statistics there's several ideas for SETI which I think would be really interesting with the data we have in hand if any of these things interest you please come chat with us Mondays bright and early at 10am and with that, I'll take any questions