The Dream of Fields: Magnetism in Cool Stars, Brown Dwarfs, and (Eventually) Exoplanets

so welcome to today's CFA colloquium which is going to be given by Peter Williams from the CFA as some of you may know Peter was an undergraduate student here at Harvard graduated in 2006 and when he was here he did junior thesis with Dave Charbonneau and a senior thesis with Matt Holman so he had an interest in exoplanets and planetary bodies and things like that for a while but when he went to graduate school at UC Berkeley he worked on exploring radio transients with the Allen telescope array so it kind of took a very different direction and then in 2012 he took the courageous step of coming back here to Harvard to work with me um and that wasn't a courageous point working with me the courageous point was that he decided to shift direction quite seriously from what he was doing for his thesis to come here and work on what he will talk about today which is the study of magnetic fields and magnetic activity in brown dwarfs and in starting to explore the idea of doing this for exoplanets as well and Peter used a wide range of techniques and observational facilities to do this work so he'll talk today about radio observations that were carried out with the VLA but he's also been using Chandra and xmm-newton Scott in telescopes in the optical and recently all mine it'll tell you about some of the results in his spare time Peter is still interested in radio transients and one I think very notable result that came up from that work fairly recently was his ability to debunk the claim the extra galactic origin of fast radio bursts within 48 hours of one paper appeared in nature so he's he's fast and he's nimble and today he's gonna tell us what he dreams about yeah thanks you know I'm glad to be here talking to everyone about what I've been doing and this is actually my fifth year as a postdoc at the CFA supported by IDO and I just want to say as a scientist and also as a human being it's been quite nice to have that kind of stability the postdoctoral phase so if you do know me you know that I tend to not always maintain consciousness during talks even ones that I'm quite a big fan of so I just wanted to try and get some of the takeaways out front before if you are like me you know occasionally other things intervene so all the points really want to make is that to understand stars and what they do to their environments magnetism is an essential component of understanding what happens there and people have been studying star magnetism since we know basically since the foundation of modern astrophysics and lately though we've been extending from stars like the Sun to things that are smaller the very smallest main sequence stars down into the brown dwarf regime and well things that we've always seen in this field is that observations lead Theory you know the generation magnetic fields is fundamentally due to convective dynamos turbulence is just really hard to deal with theoretically but with modern observations and of course our theories gotten better as well we're seeing that at this transition point at the bottom of the main sequence you know we see a lot of interesting new things that you don't see in some like stars and my claim is that in the broad brush they are essentially planet-like rather than star like phenomena which yeah you know if you if you study things smaller than galaxies the F dimension habitability for why you do things and you know this is directly applicable to both habitability and I'd like to try and turn persuade you that just it's intrinsically interesting as much as you can call anything that also even in the next few years like not not in the next decade but literally the next two or three years a bunch of new things are coming along that will really transform our ability to to understand how man eight dynamos work at the bottom the main sequence so I'd like to start with the most massive stellar things or stars that I'll talk about cool stars of course this is just going to be a very small piece of this field there are people in this room have been literally studying it so it's longer than I've been I'm not gonna try and claim that recovering everything but I want to start out with stellar magnetism in cool stars to try and motivate things a little bit and so okay this is the Sun it's largely pretty boring you know yeah the only interesting things going on are these dots and I'd like to pay attention to this corner here this is just the Sun milling along my its own business with this nice set of sunspots forming in this particular video from Florida napkins in purgatory and you know this is pretty much invisible light this is what's going on and you know why are they this big why are they forming in pairs why are they have this structure these are basic things where the Sun is more photons you know enough photons to power our entire civilization we still don't really feel like we understand those basic questions of course being astronomers we know that visible light is not all that there is if you look at the Sun and something like x-rays the importance of magnetism is even more clear so we've got these amazing loops and prominences and all sorts of activity and there's a really great one that turns or comes into view right around check out that you can see at the edge right there no this is this is sort of the dominant you know what's going on the surface of the Sun is dominated by the role of its magnetic field and of course finally track the lower left corner here yeah if we can get to play I mean this is the best one all right well there was gonna be a large flare you can see that's the loop that's about to explode and it zooms out and you can see that it blasts out you know this coronal mass ejection which propagates through the slower system and does all sorts of nasty things do anything that happens to be orbiting that star so magnetism is responsible for that all this we can generally say that drives activity you know it's non thermal heating of the air layers of the star and fortunately for us magnetic activity is associated with a mission across the electromagnetic spectrum so of course at the photosphere you've got you've got broadband variability do those spots that we saw on the Sun karma spheric lines ultraviolet emission from transition region and radio and x-ray emission from the corona and we've got dense hot gas and then energetic you know non thermally energized particles that spiral ound the magnetic field lines they're creating these beautiful arcs so magnetism is important for stars as we have known for a long time and we've heard even and I think multiple talks given this week at this one institution main activity is coupled to both rotation and age in stars this is everyone points this Gamache pod at a great you know three data points more or less we see that as stars age so their lithium abundance goes down especially there's an exponential drop-off there the rotation slows down and their calcium plus chromosphere commission goes down and for something like 40 years we've had this cartoon picture that stars are magnetic their rotation drives the creation at mimic field as they rotate though that magnetism leads to mass loss they lose angular momentum they slow down and so you get a feedback process that caused them to gradually die down in their activity you know we we know more than we did 40 years ago but in a way a lot of things are still you know still comparable to this level of sophistication so for instance one thing that we've established now is that this relationship between rotation activity saturates so here we're plotting x-ray emission over a metric luminosity analog scale versus what we call the Rossby number which is the rotation period over the convective turn over the time scale for convection in the star so this is basically telling you when you have a complex convective upwelling how much time the Coriolis effect has to induce rotation into it which is going to tell you we believe that's relevant for a dynamo action you see as you rotate faster which is going up this line x-ray emission and we believe my negativity increases and that saturates of this canonical 10 of the minus 3 level you know there's still argument about what exactly causes this saturation effect and then you know so this convective turn over time it's important to emphasize you know this has motivated an idea of what's going on the fluid dynamics but the way that this number is determined in practice is basically plotting up all these measurements and then finding a finding a functional form that makes us have low scatter if you just plot it in terms of rotation period there's a lot more scatter here's the Sun by the way and then these are our stars of different masses and people basically make a mass dependent magic you know to make make a mass dependent function for this that gives you something that looks nice you know the numbers that you get we believe are reasonable but essentially you know it's not like this number is something they can just pull out of theory directly in fact the lowest mass stars it's really not a well-defined quantity furthermore we don't really understand this rotation ActiveE relationship in detail in other ways so a lot of people it's gotten a lot of attention recently there's a big interest in this idea of gyro chronology if you maybe we'd like to know stellar age as well and if stars spin down as a age if you measure the rotation rate then maybe you can learn how old they are that would be great there's a lot of you know it turns out to be a hairy subject one recent result I want to talk about Jennifer Vaughn stars have paper in nature looking at the relationship between rotation and age in stars of different temperatures so this plot here so we have stars that are we know their ages either they're in clusters or through asteroseismology and so this y-axis is the predicted rotation period from models of stellar spin down over the observed rotation period so the one-to-one line is the models are correct and the data seem to be showing things diverging for older stars so this means that these stars are rotating more rapidly than expected the the grayish area is a model that says basically at a certain age they stop spinning down as efficiently so this is stars are see around 6000 Kelvin this is a set of stars that are more some like temperatures and these are low mass stars and so there seems to be some the it's not visible from these data but it's consistent where if you said something happens at Rossby number equals 2 so the the Rossby you hit that Rossby number at different ages for different stellar masses or temperatures meanwhile some of the work coming out of here suggests is that when stars do slow down rotationally as they age they do so very suddenly which is not necessarily inconsistent with this but you know you'd have to have stars that are breaking and then they break less efficiently and that's something they break very efficiently and people have hand-wavy explanations for that why that happens but it's hard to be more than hand-wavy about it now one thing you'll notice here is that we have a very few number of data points this is probably statistically indistinguishable from a constant this particular data set and a lot of the work in this field has been hampered because we just need a lot of data and fortunately one of the things that's happened over the past decade plus is we're getting a lot more data so large format CCD cameras especially are changing how things work here and especially for measuring star rotation periods from their photometric modulations is something that we can really start doing in kind of an industrial scale that was not possible before so this is a little bit of a cheat because I'm not showing this full dataset but here's so stellar mass versus observed rotation period for a sample in a 2011 paper and here is what we get from analysis of Kepler data I've done my a meme quote and you know the key is you go from a bunch of stuff to really being able for instance this gap very much appears to be a real thing this vertical one is a function of the assignment the stellar masses and is not Astrophysical so we're renting you know we're entering a new regime in terms of being able to do bulk analyses of these kinds of you have of stars so in particular here at Harvard I've used the time domain groups copy of the pan-starrs medium/deep survey data set to do our own analysis so this is a project done with Aaron cada Phong is a Tufts undergrad who is applying to grad schools this fall and so the pan-starrs me and deep survey for those of you who aren't familiar so it's got ten seven to seven square degree fields spread out and are a they're observed essentially every night rotating through gri zy filters so it's deep it's red sensitive and the survey lasted five years so on average we have something like eighty epochs per filter across that five year time span so a lot of data points and people usually show the pan-starrs telescope but I always love this picture of John Tom are you just like staring at the F to see CDs menacingly so we use the imaging that we have a copy of here at Harvard extracted I did some stuff with big data so in the end in those seventy square degrees you pulled out something like four million things that look like stars about half a billion photo metric measurements and then after you do some analysis probably something like 180 thousand ones that are actual cool Dwarfs in these seventy square degrees so because we have five filters and we have these deep stack images where anything that you can study and in an individual epoch is you know in the deep stacks it's amazingly well constrained so we can measure the colors of all the objects we're interested in and we can make color color diagrams and do things like get approximate stellar parameters so this left-hand panel so we've gotten our iiz color color diagram in that domain stars are a nice straight line so what we've bought here is the grayscale is the full sample of four million star-like objects and then the ones we basically just you know fit a line to that spine with a certain air bar and select those ones as in this color color plane they might be stars and we just reap lot them in a color you know I think it's probably that much hated jet car scale just emphasize them so because we have five filters we have four colors and three color color planes if you take the intersection of all those this is plotting the same data but now we perform the same kind of cut in the other two planes and take the intersection and you can see this nice spine of stars and it's a little bit hard to see here but there's a nice little red giant branch that pops out so we have that we can identify things are probably stars we also collaborated with Andrew man recently had a very nice analysis of measuring the fundamental parameters of M dwarfs and then building up synthetic spectra and basically giving polynomials that relate their colors in various filter sets to their fundamental parameters and so here ear and as analysis for the pan-starrs filters so we have some nice polynomials in the paper that uh that will let you map from pan-starrs colors to things like effective temperature approximate mass radius etc and you know pan-starrs it's a giant data set and it will be around for a while so so these could end up being very useful I should mention this paper was accepted just a couple of days ago so from all those stars I'm going to skip over the details we do a sort of periodic analysis try and find things have periodic modulations so you know this is kind of a first cut project so in the end what we did is we had some very generous you know these are our you know this is a buy I fit we did some very generous cuts it's narrow it down to things that we definitely feel good about definitely have significant you know clear rotational modulation that ends up being 270 objects now I have to say one of the hopes in this project was that we'd be able to go to things that are cooler lower mass then done in previous surveys like the mirth survey those ended up I mean even seventy square degrees ends up being close to a pencil beam I mean to the for you know to they get good precise photometry on an elder Wharf you need to be very close and there's only so many of those on the sky so this mess of blue X's is the Kepler data the pinkish diamonds our Elizabeth Newton's work from the mirth survey and the green square the green circles are the stars that we measured so mass decreasing on this axis and log of rotation period increasing here so you'll see we largely overlap the mirth survey we go to things that are a little bit more massive these bars here are error bars on the mass measurements you know if you don't know the distance you don't know the k-band absolute magnitude mass from color is tough and so you know keep in mind that all these points have a lot of wiggle one of them but we do seem to show that the Kepler data had this they don't show any stars of rotation periods larger than about 70 days and we support the finding from Elizabeth mirth work that it does seem like there's some systematic in the Kepler data that's preventing people from finding those and there does seem to be this population of extremely slowly rotating low mass stars so here instead of that fairly uncertain mass parameter we plot the pseudo Murph minus K short magnitude so a color that is a you know somewhat better constrained and is decent proxy for the mass so we also cross match these objects to x-ray catalogs so you add some more items to the rotation x-ray diagram things like that one thing that's interesting is so this plot is the amplitude of the variability in the Z band which is where we have the best sensitivity versus rotation period and so one thing that Elizabeth found in the mirth work was a plot essentially very similar to this one where you've got low amplitudes at the very short rotation periods the rapid rotators at the long rotation periods but you know we would be sensitive to stars that were low amplitudes with moderate rotation periods we only the only ones that we detect there have fairly high amplitudes so to get a little hand wavy if as some of the previous studies show stars do spin down very abruptly then these intermediate rotation period one are the ones that are in the course of doing just that and so maybe the fact that we see high amplitudes here shows that you've got a lot of spots you've got a lot of contrast and so maybe that's indicating that this that you've got sort of a high-level amount of magnetic activity that is driving this rapid spin down of course the thing that's always you always have to keep in mind here is that you know the variability amplitude is a function of the spot you know asymmetry you know if the star was all one big spot you wouldn't see any variability either so you know a small amplitude might correspond to a lot of spots they might just be very axisymmetric ly distributed so this is the first thing that we've done with this large data set of half a billion data points and if anyone is interested in doing more certainly happy to share although so I am told that the official pan-starrs public data release will actually really definitely for real totally be coming out in may ish so hopefully hopefully in a certain sense our data set will be being superseded by the you know people have you know people complaining because it's taking a long time for this public data release to work out but the people who are doing it have been pouring their blood sweat and tears into it it's been a matter of lack of resources not lack of dedication so be great to see that actually coming out sorry this is the first thing to down this data set you can definitely do more just as a quick example something yes we found a few eclipsing binaries in that data set the different colors here are from the different filters and you sort of normalize them all and obviously if we're thinking about stellar activity flare so the other thing to think about the tricky thing here is you know because it's a nightly cadence most stellar flares are gonna be a single data point and trying to identify the what are the real outliers nude flares are things that are due to random data errors that's that's always tricky but I do want to emphasize so people think that you know Ellis is TV it's coming along right around the corner but really you know LSD first light is 20 22 and so if you want to build up a five-year data set for instance like the pan-starrs survey that's twenty twenty seven that's a deck from now at that point like j2 is t is probably going to be on this way out so this Panthers dataset is really not going to be surpassed for a decade you know we should all be getting ready for Alice's tea but this is we have the the nightly monitoring the five filters it's it's not Elsa tea scale but if you want to get ready for this is the day set to use and I'm not aware of anything else that's going to come close until until that firehose turns on in 2022 so that was all I wanted to say about four cool stars obviously as I said before there's many many more things that could be said but now I'll turn to the very coolest stars and down into the brown dwarf regime so to give you some context here it wasn't expected that things like ma dwarfs brown dwarfs would have magnetic activity so the term of art when you is here are the ultra cool dwarfs so anything with spectral type greater than m7 or so you start seeing molecular lines in the atmosphere their laser atmosphere starts you mean neutral so and of course we should keep in mind that say an m8 dwarf could be either a star or a brown dwarf through even the very lowest mass stars would have spectral types of elf or so so in this you know there's a regime where they overlap in spectral type and then once you're into the mid ELLs those are all definitely brown dwarfs anyway the key difference is once you get to past spectral type 73.5 or so these things become fully convective they don't have a radiative core like our Sun and as we've believed for a long time that the interface between the radiative core and the convective zone of the Sun is important for driving its magnetic activity so this is kind of a profile of the sun's rotation and you can basically see these contours are showing you there's strong shear here and there's been believed for a while the shear is important although there's more and more more and more people are starting to question this assumption there's a recent work for me from folks here looking at activity and things that are fully convective and arguing that because their rotation activities can are like those of higher math stars maybe it's the same kind of dynamo and so this is this was not even settled again even the Sun we we don't feel that we fully understand this so besides the lack of tackle Kline it's also true that ultra-cool dwarf do have these neutral atmospheres and when you talk about driving things like the Sun's corona we believe that the convective motion the outer layers they drag around the foot points of magnetic loops they twist them up we believe that those are important for driving the energetic coronal phenomena that we see ultra-cool dwarf because those outer layers the atmosphere are neutral the convective motions are not going to drag around MINIX in the same way and that would also be expected to you know the magnetic field might be there but you might not see the kind of energetic activity that you see in a s'more sun-like star well obviously this turns out not to be true true so the first detections indicating activity and ultra-cool dwarfs come from things like x-ray and UV observations so this is just um observations that Geoff glinsky did of VB 10 where it was observed in the UV and 10 you know 10 observations basically no signal at all and then one observation shows spectra that you would associate with say the solar activity the offset is just for visibility so this was the first tenth and indeed we've borne out that these objects do seem to store and release magnetic activity in the way that the Sun does sometimes so it is true though that when you go to these very low-mass objects the way that you probe activity in a sun-like star many of those mechanisms become less useful so for instance x-rays are you know quite justifiably one of the standard ways of understanding manic magnetism in say and earlier made m-dwarf but if you look at x-ray emission as a function of spectral type it really drops like a rock so these are you know so m5 mid m dwarf L dwarfs so they means their log scales and you know note the size of the gaps here so once you get to spectral type m4 so every subtype that you go down your typical x-ray luminosities creasing by an order of magnitude so you know you've barely any detection is here so this is a this one detection in the elder or fushime is the key Lu one binary which i think is for plus or minus two photons detected I've heard tell that we might have four more photons from a different L Dorf but this is x-rays they're just not going to do you any good in this regime the same can be said the same girl trend happens for H alpha emission and then spectroscopic methods often end up having issues because these objects tend to be very rapid rotators so lines get smeared out and then of course in general these things are getting very very faint polar metrically and you know that's never a good thing if you wanna detect something the exception to all this is radio emission fracture them about ten percent have persistent radio emission and from that right you know right there we know that they have magnetic fields and they also have processes are accelerating particles to non thermal energies so this plot shows fairly well-known in stellar circles little bends relations so the x-axis is logarithm of x-ray luminosity and this is a spectral radio luminosity so herbs per second per Hertz and it's been long established that uh so these are things like active binaries these are things like M dwarf stars these are things like individual solar flares where you see this pretty close to linear scaling between x-ray and radio luminosity and your typical pictures of what's driving activity where you have some kind of reconnection event that unleashes a burst of energetic particles that make sense and of course you can you know log logs AXYZ obscure many sins but uh it's a pretty good scaling so these colored things here are what we see in m6 and later type dwarfs where I've emphasized there's a lot of upper limits I've emphasized the ones that are Texans and so we have things that are relying off the relation you know consistent with a scatter and then you know these things are x-ray do detection and x-ray upper limits so they're somewhere off here just well well above this relation and what this means in practice would be another way to think about this is the radio luminosity of these objects see to kind of bottom out at this 10 to the 13 herbs per second per Hertz number so as you go cooler and cooler x-ray is falling off a cliff but radio dust is persistent and if you went to my ITC lunch talk we can detect radio emission from t dwarfs and things like that so this is you know this was the only tracer that will work in something like a t dwarf so that's kind of one thing that starts getting different once we get to this ultra cool regime if we look at the rotation activity relation it starts getting a little funky so here I've game pod this Rossby number rotation period normalized by the convective number and then normalized x-ray luminosity so again these colored points are the m6 and layer dwarfs and we have some upper limits so there's a sample here that you know or seemed pretty consistent with the standard relation which is the black points and then there's things which are definitely well below and a good number of upper limits that are making it so I wouldn't you know you wouldn't be able to say that there's definitely a gap there or not but regardless of that we definitely see much more scatter than you do in this otherwise saturated regime so that's one thing that's different and this is going to get a little bit speculative but there's been this idea if you look at the edge of the black points here there might be a little bit of a turnover there's been this idea of what we call super saturation which is maybe the very fastest rotators of the x-ray emission starts decreasing again and these are fast rotators they do have lower x-ray emission you could say maybe this is more evidence for super saturation of course so a lot of the mechanisms that people propose will drop x-ray emission by a factor of a few but not by more than order or magnitude I'm sorry I forgot to say that this is work so work that I did with Ben Cooke Whitney's an ru student here and of course now is one of our many talented graduate students so if you uh if you look at this rotation x-ray emission so what we needed here was reported the x-ray against V Sinai because that's the measurement that we usually have so now faster rotation is on the right instead of the left anti-correlation so the particular thing we've done here the problem is is that temperature and rotation rate are correlated so we've we've cut down to a very narrow temperature range where there doesn't there's a bunch of correlation there and you know so what do you want to say is if you look over here there's upper limits that are well below this 10 to minus 5 level you don't see that I mean we have less constraining upper limits but you don't see that here likewise you see objects that are up here in this half of the diagram but you don't really see them up here this is clearly not conclusive but we might be learning something that this dynamo you know we the standard hydra-matic demo we do expect that greater rotation leads to greater dynamo activity we might be seeing evidence that that dynamo behavior is shifted in this regime but again this was clearly we need some more data to really figure out what's going on in terms of what we see in the radio emission I haven't really talked about that in a detail yet we do see several new things so in the non-flowering mission so we see both bursts and fairly quiet and emission and the non flaring mission you see very flat spectra so um we expect we believe this emission is due to synchrotron where you have a peak in the spectrum I'll show that in a minute and then we have objects like this one is one of our poster children where the spectrum is essentially flat from you know 1 to 25 gigahertz UV set and m6 you actually see a rising spectrum up there and you know it's very hard to explain staring theories we also often see periodic modulation so just to show the spectrum so what I've got here this is these are two spectra of somebody's ultra-cool dwarf we've got a two mass 13 14 and LP 495 this is kind of a standard synchrotron spectrum and the main thing is so there's a few parameters that set the shape of the spectrum so you know if you set the magnetic field that essentially sets a spectral peak and you moves up and down like that so the key thing is so then the sloop here is set by the kind of power index of the electrons and so you can make it pretty steep you can make it flat you know comparably flat to what you get up here but this is the optically sick part it's a new the 2.5 there's nothing that changes that you know if you set the density optically thick you can't see anything more so that just moves things up and down like this and then this is sort of the characteristic length scale of what you're seeing and that kind of sets the overall amplitude but these are the shallow declining parts of the spectra you know it's just you can't get that with synchrotron so that's one puzzling thing we also see the emergence of periodic strongly polarized bursts and these are that's a typo these are so intense that they have to be due to some kind of coherent emission process where you've got electrons moving in bunches so this is a light curve of an l-3 dwarf to meso seven four six plus 20 so this is over about 10 hours you can see this is Stokes total intensity and you see these big spikes and this is Stoke circular polarization and the fact that the amplitudes are about the same going both up and down tells you this is essentially one her percent left circularly polarized emission this is kinda the nicest light curves of that but we see is phenomenology in a few objects so the interpretation what's going on here is that the emission is coming from this thing called the electron cyclotron maser and stability so it is one of these coherent processes it's basically due to a resonance of the cyclotron motion of particles around a magnetic field line with their with their actual with their emission frequency so it gives you this strong circular polarization a very important thing is that the emission occurs essentially at the cyclotron frequency so if we see a mission at say five gigahertz you can immediately know that the the region that gave rise to that mission had an ambient magnetic field strength of you know a couple kilogauss so this is great so the radio mission immediately diagnosis the ambient magnetic field strength it's extremely bright so we're characteristically see bursts lasting serve the 10 100 time scale because the emission goes above the cyclotron frequency cuts off very abruptly spectrally so usually you sort of span about a decade of bandwidth with a fairly flat spectrum and this is kind of a cartoon model here where you know if you have a dipolar field with field lines converging the pole and in the earth and the planets that's where the emission comes from and it usually ends up propagating out somewhat perpendicular to the field in a very broad cone and if the magnetic pole and the rotational axis are misaligned you get the lighthouse effect like a pulsar and you see these periodic bursts yeah and usually it originates fairly close to the surface so if you have if you're assuming a purely dipolar object you you see the cutoff of the mission you immediately know the dipole field strength so as I alluded to before we actually see these bursts an extremely cool object so the record holder right now as an object killed to mass 10:47 plus 21 it's a t 6.5 dwarf so effective temperature of 900 Kelvin we don't know the mass cuz it's a brown dwarf and there's a jersey between the mass and the age so it's first detected with Arecibo we a little while ago did a follow-up study la so it's another 10-hour light curve where the top curves are showing the periodic purse and this is a reference object showing that we're okay when it comes to cymatics so we're splitting between a couple frequencies and I do want to emphasize that things like this might not look so so thrilling but those are significant I think this is a three and a half Sigma or so we furthermore detected burst so this is this is that five and seven gigahertz we also detected bursts at frequencies up to ten gigahertz which applies so this this object is you know a cool small dwarf eight times the temperature of Jupiter that's got a three-and-a-half kill Goss field compared to Jupiter's field of around 1214 gas so you know this this object is somehow generating it's got a very efficient dynamo and furthermore it's doing something to accelerate electrons to non 30 you know to near relativistic velocities so pursuing that so we actually got all my observations of a few of these objects and indeed you can even see signatures of their synchrotron emission at Alma at all my frequencies which implies that yeah you're talking MeV electrons you know there are some how be accelerated by an object that's much cooler than the Sun so here is this the discovery image this is the first ultra-cool dwarf to be just to be detected by Alma except for disks which many people here do a lot of work with and so here's a spectrum where this is kind of existing data and here are the Ouma data points where you can draw one of these classic synchrotron spectra through this so this is not one of these objects of one of these weird weirdly flat spectra but it is true that if you just apply a standard synchrotron model you have to have in a very efficient acceleration process and what's promising is that these objects you know at these energies the electrons cool extremely quickly the pooling timescales something like hours so by making monitoring observations at these high frequencies you can hope to probe the mechanism that's accelerating the electrons in these magnetic fields this green line is showing the sort of spectrum that you get from the maser mission so you know there's a lot of variability but so if this were a maser process they need be assuming an insanely strong magnetic field 30 kilowatts but there we don't believe that's at work here so this is something we don't Alma as I mentioned monitoring this would be nice it turns out that Alma's support for time domain type observations is is not great so I actually probably don't need to belabor this too much because I sent a book an email to everybody last week or two weeks ago there's a lot of things you know you know the turnaround time for targets of targets of opportunity is 15 days they won't let you try and do a simultaneous observation with another telescope at least that's not a guaranteed capability just I once had a proposal rejected because I wanted to look at something for four hours straight and that was deemed technically impossible and it would be nice for science cases like this and that turns out solar system planetary scientists also are very interested in these kinds of capabilities it'd be nice from almost better about this the impression that I've gotten is that you know Alma does have development budget it's very very unclear how the use of that is prioritized and so community pressure to emphasize that there are a lot of people who are intersted in these capabilities I believe is very important yeah so if you're all interested I hope you all sign up for this exceed all my time domain special interest group email mailing list and hopefully it will require very little effort on your part you just have to be willing to sign your name to say oh ma please do this of course if you want to uh you know do things will have to write white papers and things like that someone will have to do that but not everyone but I think you know 15 days is kind of silly we can do much we can do better and we can get great science out of it so switching gears a little bit one of the ultra-cool dwarf that's really the the poster child for these kinds of studies is a binary called NL tt-33 370 so this is a short period visual binary of two objects that turn out to be nearly identical so we can detect one of them in radio emission with VLBI which means you have insanely precise astrometry of the binary which means you can measure the masses to exquisite precision so here you know ninety two point eight plus or minus point six solar masses for one ninety one point seven for the other we got the distance to fraction of a parsec and then you know if you assume your favorite model I don't think anyone is gonna claim that we actually know the effect of temperatures of these things two three Kelvin but you know these objects are extremely low characterized and you'll note that they're nearly identical I mean there are binary so we have to assume they're coeval same composition masses you know yes that is correct yes that would be very different yeah so I mean their masses are identical to to the air bars and so these are essentially the same object but kind of crazily only one of them seems to be active one of them is very bright and radio emission the other one we detect nothing so with this with the ephemeris for this binary it's been visited with the VL ba in many epics and so this is um taking taking all the VLBI data and stacking them on the predicted position of the a component and so over the different epochs the B component is moving its orbit and you just don't see anything at the a component furthermore you know so they've got the same mass same age same composition and we believe that they've the same rotation rate so in a comprehensive study we did I used data provided by the mirth survey and I think that in in the data we can't resolve the two objects but we do see strong evidence for periodicity as of almost identical values of just more than three hours furthermore we recently got resolved Keck near spout data and there v-sign eyes are almost identical at 39 36 kilometers a second so same mass same rotation rate we believe same compositions same age and yet one is a one of the brightest ultra-cool radio emitters we have and one yeah so we're limiting it right now to thirty times fainter than the other that is really weird and I mentioned before that we only detect ten percent of these objects in the radio and you know you can argue maybe it's an age thing but really you know whatever is going on you know it seems to it seems to be something that's not related to the fun mental prowess of property maybe this one has a little like IO moon that's energizing the system and the other one doesn't so secondary is extremely active again it's the most radio bright so this is kind of a schematic summarizing uh the intensive observing campaign we did so this is a time in days and different sorts of facilities so in 2013 we've got two consecutive nights of intensive VLA monitoring with a lot of contextual data from Earth and also MMT and then a year later we got an intensive night of ela plus mirth plus swift plus Chandra so that gives you you know very messy data sets like this one we're up here we've got a time series of the radio mission so we've got Stokes total intensity Stokes circular polarization we've got the mirth broadband photometry sampling of ultraviolet from Swift and ultraviolet and x-rays from Swift and Chandra and this is so this radio spans about 10 hours and so you can see that there's slight phase shift between the maxima and the x-ray in the optical unfortunately it looks like we missed some kind of big x-ray flare but you'll see you lower-level things and then you see that you know you have this variability which is not reflected in the circular polarization at all where something like this big flare that does seem to be intensely circularly polarized like those major bursts so this is just one of the three nights of observing if you put the three nights all together so this is one day this is the day later and this is a year later so we've got three 10-hour light curves so this is the unpolarized component I've subtracted out the polarized flares this is a polarized a circularly polarized residual and so you'll see that you have kind of consistent pattern and then finally the the great the black vertical lines indicate the optical Maxima so unsurprisingly the things got rotation period of four hours after 24 hours it hasn't evolved too significantly a year later so the you know the the characteristic behavior of the circular polarized thing seems pretty similar and a phasing of that relative to the optical emission seems pretty similar but then the unpolarized radio emission is just completely different and so this blue line is basically showing the average of what we had before now of course it's possible that the phasing you know you can reface a little bit and it will look slightly less different but but I think I think this is very interesting that the polarized and optical modulations do seem consistent so in the paper we've got sort of spin out a theory of maybe what you've got here is sort of dense plasma torus and bright plasma trapped at the poles that are giving you out of phase you know absorption in phase and out of phase emission so there's some kind of stability but when you get to the radio the employers radio machine there's also even evolution we have recently acquired Alma data of this object as well so again this is it's very flat spectrum extending to about hundred gigahertz this was a nominal light curve over I think it's it's like more than rotation some like five hours so it's very bright even on the wavelengths this is Miller Jansky which is oma for bright also it seems to be about 30 percent circularly polarized which so circular polarization analysis is not officially supported by oma because they expect most things that have circular polarization of about you know half a percent 0.1 of a percent and you know it's hard to characterize your instruction know what well enough to know that you're seeing anything real but I've you know we've looked at this carefully and it really does seem to be intense circular polarized Mission and again it's hard to imagine this being this major process because you would need may not feel much stronger than anything that we've seen any other kind of star and in principle synchrotron emission can give you this intense circular polarization if you're looking like right down the barrel the gun but but that's where the emissivity is lowest and I don't know how to explain this you know there are people have fancy secret ROM codes where you can evaluate things without the kind of classic approximation formula maybe but just for them fundamental physics I don't know how you get this intense submission at this wavelength if anyone has any ideas I would love to hear them so summarize some of the things that we see that are interesting are these very flat spectra these periodic modulations and these maser bursts so investigating these kinds of objects or is what I've been up to for a little while looking forward so I think what these data are telling us is that the magnetism of these objects is like planetary magnetism for instance in that non bursting emission spectrum and the variability very much resemble what you see from Jupiter so this is a movie of imaging of Jupiter apparently that's weird it wobbles around periodically this is really neat this is um if you observe an object that's close enough with the radio interferometer you can actually reconstruct the depths axis in the server the same way that you can reconstruct the image so Bob salt is the compactor age reconstruct this is the radio emission from Jupiter's Van Allen belts in a 3d reconstruction so basically this key projected and the kind of double doughnut shape that you see here is due to Jupiter's rings where energetic particles diffuse inwards and the radiation belts and then when they hit the Rings they get a significant amount pitch angle scattering which kicks them up and gives you this characteristic two-component structure and the physics of the diffusion of the particles in these rings I think is really interesting now I mentioned I mentioned these auroral bursts and I didn't really go into the fact that you know roll makes you think of planets and indeed in the planetary case we think of you know we think of irori as being something that you kind need to be driven by a star so for instance you've got the Sun well it's turning my neck feels there's the earth is my net field if it gets hit by coronal mass ejection you know there's this impact on the front you get you know magnetic field lines you get flow around to the back in the Earth's magnetotail where you see things the field lines are pinched down these are facing different directions and so you can get reconnection or energized particles they tunnel into the poles and here we're seeing the visible Rory because people are wavelength biased but of course there's copious radio emission as well and the phenomenology of that overall mission is exactly like what we see from the ultra-cool dwarf but of course ultra-cool dwarfs are not being driven by a big solar wind it's far from clear what's driving that but the fundamental process of accelerating some electrons to flow down that dipolar field line seems like it must be operating but what does this mean is that by studying these objects we can really study the phenomena of extrasolar planetary magnetism today we can detect these objects and these are Rory are part of these general rich complex magneto ionosphere current systems so here is a diagram of Jupiter's current system where the planet self is there and you may have heard that Jupiter's magnetosphere is largest coherent structure in the solar system it's like the galaxies clusters of the source them and yeah this is the fact that you're impeding the solar wind on it drives all sorts of really neat stuff where you get you know basically you get kind of a someone should make an awesome movie of this because it's really hard to explain but you get these by conical tails and all sorts of currents flowing around and of course IO is driving things - people spend their lives studying this and Jupiter is particularly complicated because things like IO are driving all the activity but there's law of energy which is eventually getting funneled onto the atmosphere and also the the dynamics of these systems are really interesting and of course one thing that is interesting about this so if we can measure the radio emission of a planet you can measure its magnetic field if it's that maser emission you can measure its rotation period and of course the habitability is believed to be important magnetism is believed to be important in controlling capability this is kind of a overwrought public relations image from NASA but so this is Mars being bombarded by the solar wind without a magnetic field goodbye atmosphere and this is Earth where this magnetic field being kept safe and then you know this is this I mean again this is a PR image but the fundamental while the key science results for maven is that they believe that Mars lack of magnetic field led it to lose its atmosphere I would say you know people write a lot of papers about this I would say this is topics that were just beginning to understand there's gonna be a lot more to learn and you know but I it is we can be confident that's important we don't we don't know how it works so that's what science is for then one side note that I really like is the key difference between why we have a man at field and Mars dozen might be plate tectonics so to create a manic field and something like the Earth's dynamo you need convective motions so basically you need to be heat differential on the core so it needs to be warm on the bottom and cool the top plate techtonics you have mental convection you're taking away heat from the top and so you can drive that convection in the core that creates a magnetic field things like Mars are in the stagnant Lin regime where you've basically got a bike blanket on top of the core you don't get the heat differential you don't get the convection you don't get the magnetic field and so you know if you detect the magnetic field of a planet that might be telling you that it has active geology or about its composition so in the next few years we're gonna learn so much more about it'll be great it'll be the greatest so the upgrade VLA is a hundred times more sensitive already you know we're doing as much as we can with that right now the guy I data released came out right now this will give us you know essentially complete characterization of you know we're looking at things within 50 parsecs gaya will take care of them all a little bit underappreciated the meerkat telescope is a radio telescope being built in South Africa it's an ska Pathfinder by this time next year we expect a 64 element version to come online it's got a 15 Kelvin system temperature it's basically a 50 times sensitivity improvement in the southern hemisphere and in this regime where we're looking at individual objects spread all over the sky that doubles our capability to study nearby interesting objects so I think meerkats gonna be great of course things like tests and Rosina will give us all sky monitoring of activity signatures for wide variety of stars so for sort of like you know the things that will be possible in next few years you know you can just do volume and survey if something like you know something like 800 systems with fully convective objects in a 15 parsec ball you could just observe all of them I mean you just do it and then of course age is clearly an important axis so things like target surveys of young clusters you know this is this is especially meerkat I'm excited about will really open up our abilities and then in longer-term things like Asti and ska will you know increase things even more yeah I do want to emphasize though that you know we think of brown dwarfs as stars with sunlight magnetic fields I think the way to really think about them is actually with magnetic fields like this I believe this is based on actual data it's a little bit hard to tell from their descriptions but you know this is Earth's magnetic field it's being impinged on by the solar wind so again that's not gonna be relevant for for brown dwarfs in the same way but you know this is very different than those churning coronal loops that I showed you at the beginning it is true actual exoplanets based on what we think we know right now are not yet within reach so here these are upper limits from various studies as function of frequency this is a luminosity space so the limit depends on how close the thing is you're looking at this is where Jupiter is these are where we are right now and so the slope of death is because as you go to lower frequencies the sky brightness gets higher and higher and so at fixed collecting area your ability to detect faint signals gets worse and worse and so like you know fairly dramatically so you basically need to row really against that curve you need a lot of collecting area to have the sensitivity to check something like Jupiter that being said you know Jupiter is our backyard example but uh we know that T dwarfs have kilogauss magnetic fields maybe there are planets that have magnetic fields that are much stronger than we expect my personal hope or belief is that the Hera telescope will be great for this so the key is you need to go to low frequencies we that's probably true you know may be able detect plants out here but we probably need low frequency sensitivity so Hera is actually a cosmology instrument the hydrogen epic of reorganization array so it's basically Arecibo chopped into little bits it's about three meters across it's got a bunch of Z is pointing individual dishes it was recently funded to ten million dollars from the NSF to build out to most of this 270 dishes so it's a collaboration led out of Berkeley with lots of American universities involved and you know we're under construction now I'm I've written the software that has to deal with a 1.4 terabytes of data Knight is there the paper dishes in the background sorry the Hara dishes in the background some preliminary dipoles appear and Aaron you will WAIS making things happen so even though this is cosmology experiment I think will be great for time domain astronomy the key is the fact that you've got these boring ZF pointing dishes you have a lot of collecting area at a small amount of money and you know we're always constrained by the amount of money so if you want to get more area you have to build it cheaper the same things that make you one for the epoch of realization science you want a very stable instrument you won't understand it extremely well for detecting rare time domain signals that's what we want to and finally you know Hera is it's it's not a general-purpose Observatory its experiment it's got a few jobs which I've I've experienced the other end of that spectrum I think having that focus is really important so because we're playing at the zenith we can only see a certain declination strip but there's something you know within within 25 parsecs something like 600 systems things including formal hop and it's not unreasonable to think that you could detect something like a 10 times Jupiter 10 times the field 10 times the luminosity at 10 parsecs very long term so I showed you Jupiter's emission I didn't tell you about the atmospheric cutoff so we need to build a low long wave leak telescope on the moon because they asked for your block signals below 10 megahertz so if there's rocky planets will probably have weaker fields and so we need to go to space all right that's it some specific things that I think are neat so even very very cool T dwarfs can Jerry very strong magnetic fields this ml tt-33 370 binary I think doesn't get enough love it like it's amazingly well understood and it's just it's a facet like this this this issue of why one has strong magnetic activity and one doesn't and it's one of the fundamental things were grappling with and it's like the perfect case study for understanding that I think the way to think about I mean besides magnetically these things are fully convective they're gas balls they've got clouds they're basically you know brown dwarf is basically a really big planet and finally I like this idea that if we detect my neck feels in rocky exoplanets that may be telling you something about it having active geology I'll put up a broader summary and take any questions thanks big the point that in the in the binary that you detected the three three thirty seventy that you two stars behaving like differently and I'm wondering in the early part of your talk where you showed the correlations amongst you know thousands of objects how do you know that some of those aren't unresolved binaries or how do you deal with that where do you ascribe the properties for a single object when in fact it's two unresolved objects I mean I would say there is any generic answer there you know it's something you have to be concerned about and it's something you know depending on what regime the problem space we're looking in me you've got different ways to mitigate that well it's certainly true that uh and we know that and we know that binaries do have enhanced activity so yeah I mean you do whatever you can to understand that but I don't think I would say there's one thing that you do yeah I mean you know obviously you can for instance kind of by analogy with Kepler people have spent much much effort you know studying the things that look like exciting candidates to really reject companions to very - very good constraints using other techniques and you know if you have one object turns out to be a bit of a linchpin to your science case then obviously you have to you have to devote investigating that I was also curious you mentioned the base or a mission and you observed that with Alma so presumably you have velocity information where where the the emission you ascribe to masers also evident in the very narrow line widths that you see from these objects so we don't I don't believe that the homo mission this amazing mission but anything yeah I think that's that's it from but so what you do see is if you look at time frequency diagram subbies maser burst you see frequency drifts and we believe that the correct way to interpret those is the major emitting region you see they're propagating up or down magnetic field line and so the ambient density is changing I feel circle is changing but you can infer the rate of propagation the rate of physical movement from that frequency drifting this pulses so you can try and understand I mean the the the emission regions of these Mason's are believed to be you know we're scales they're they're very manageable so they would be very narrow I mean one way to nail it would be to actually resolve the longings yeah so that's I mean so others there's a very nice Arecibo studies where you know I have these these bursts where they're like two points and light curve together and then if you look at him with Arecibo you see this forest of striations and and what we detect with the VLA is as all as there's a whole complex of major sites that we're all blending together so yeah people do do that with Arecibo it's challenging you need you need a lot of sensitivity to really pull out the individual aspirations so well classic way of getting the magnitude of magnetic fields a month's money it wasn't in any of these objects so in the higher s ones the that believed that that technique currently is breaks down around spectral type m7 or so this for the same on Doppler imaging where people do time-resolved Azeem on observations you can use that to reconstruct the field structure that that they haven't gotten to to work below and six or so and so things are going to think they're gonna be rapid rotators which blends out the same oscillating so it's it's challenging there is an instrument coming along called spare room which is a spectral parameter in the infrared which should be able to extend the same on studies to lower mass objects to some extent but yes I don't they believe that they can do things with rotation periods longer than eight hours something like that but what we see typically it's one two three four hours so yeah this month this is one of the reasons it's one of the ways in which these things are hurt Aprilia by just about any other means besides radio so one thing that nicked I found in his paper was there's an interesting trend with the x-ray luminosity and the co rotation radius for the star the temple area no rotation reason story and he looked at it suggested that the rusty month number might not be the full story and have you looked at other properties of these stars and see if they had trends with the experiment honesty as well yeah so I was actually an immense paper we investigated a they offered a couple explanations for for maybe that super saturation effect and if I recall correctly I think the core edition raises well on ones that definitely did not explain what we see in the ultra-cool dwarf convention there should be really constrained which will you know even though you don't have a type of line you're gonna have structure in the angular momentum distributions up in the interior of the star so it may be that it's not really as alpha square type of Dynamo is as we'd like we initially think yes wondering whether you know if you have those structures then maybe there are certain parts of the star in which you have the equivalents of portal holes things are able to get up but other parts all the loops are kind of close and what might be a coma sphere and whatever activity is going on there is never going to be generate a lot of energy is no chemistry you ever see a Chandra so it may be that these aspects are yeah it's definitely true that the convective in these things you get you get these columns I don't know if we're Kesha's here but and we've got people people do a similar demos in this regime and and you know the convection it's not radially symmetric boiling so you do I mean whenever I read the papers about these things they don't talk about the shearing and the internal structure there is as as being important to the Dynamo that I I think I could say that but I said for instance you know while all these things have very planetary like mind is in some ways for instance you do see large x-ray flares from nine quarks occasionally as far as I know that's nothing that Jupiter does and so you know they're not you know they do seem to be intermediate in all senses of the word Alma has very similar bureaucracy Eisel but it's complicated by the international partnership the only the only way to push something an official is to act by a user committee to try to organize pressure from the European side and you'll be successful I mean we have we have members of the Omaha Science Advisory Committee science advisory committee doesn't have the same power it's a user comedian that has to push things on yeah yeah I mean I do believe you know poor on white paper we'll see if that suffices if there's really lots of dragging then yeah you know it's time to bring out your bureaucrat ninjas and extortion finally we'll find a way to make it happen but it is I have heard for instance that the Science Advisory Committee makes recommendations and it's very unclear how those record you know what the board actually causes to happen the relationship between that the recommendations is very ambiguous so you know it's a large international billion dollar project its workings are going to be mysterious cater your ear identical twins object what constraints can you place on some some variety of leaving defects in that object it's very bright it's this ridiculous back thirty difference to essentially identical objects which a word meaning could it be the magnetic axis as somehow removed that's something we thought about so I think there's two lines of argument where I don't think that's relevant first of all we have the rotation periods and we head out or I believe nothing doesn't rotation period to get right over to solve them so we're not sure we have a V sine hi measurements and those are about the same in both cases so sine I is about the same so if one of them had a low sign I then then maybe you've got something that's nearly Pole but they both have relatively high V sign eyes and so the clinicians can't be that different and also so if you do one thing I didn't mention is we have is ten percent success rate and detecting things in the radio but there was a nice study last year looking at objects that were pre-selected either from a chav emission or optical very optical IR variability and in that pre-selected sample there's a five out of six various fortifies for success rate so that could also fold in the inclination factor I suppose I think that's some evidence that you know if if the electro dynamic engine is running if you've got the manic field and and the currents that seems that it's it seems there could be detected fairly independently of inclination oh good oh I think we say yes yes no I'm very spinner to be some yeah yeah all the time I had that I think the fact that we can detect the things of the radio reliably given other indicators suggest that equation lots of more effective again this is not what we are detecting is dominated by the emission that expect he wants to travel if they ask good point

Channel: CfA Colloquium

Views: 9,826

Rating: 4.5 out of 5

Keywords:

Id: SsYy78m9-yM

Length: 72min 19sec (4339 seconds)

Published: Thu Oct 20 2016