NASA's Heliophysics Summer School - August 17, 2024 - Ben Lynch

Okay, um, good morning everyone. Uh, my name is Benjamin Lynch. Um, I am currently a researcher at the EPSS department at UCLA. Um, before that, I was at UC Berkeley, the Space Sciences Lab for many, many, many years. Um, I did my undergrad at UC San Diego. Uh, I got a degree in electrical engineering. Um, but I didn't want to go build cell phone antennas, so I added physics as a second major to try to bump up my GPA to get into grad school. And that tells you a little bit about who, you know, kind of what I do, so physics was easy and engineering was hard. Um, okay, so I went to graduate school at the University of Michigan. Um, and that was exciting, and I did my thesis research with a group down at DC. Um, and then I went to Berkeley for my postdoc in 2006, so I was there basically for 15 years, more or less. Um, okay, so part one and part two of the lectures today are going to be kind of—anyway, this is this is this is going to be an experience. Um, I definitely want questions and interruptions. I have no idea what any of these slides are going to be because I uploaded them, you know, 15 minutes before walking over here. So this part one is a collection on solar wind type slides, and part two is going to be a collection on CME type slides, and they're going to be a mix of kind of super basic stuff that you're like, "Oh my God, everybody knows this already," and cutting-edge simulation research results where I'll be like, "You know, I don't actually know what's going on in this simulation," but um, so we will watch the movies and try to figure it out together. Okay, that's got nothing to do with what I'm going to talk about, so um, everyone's familiar with this. I I've seen it drawn on at least one of these papers.

Okay, so the large-scale magnetic field structures the solar wind. So the large-scale magnetic field of the sun structures the solar wind, and that large-scale field evolves in time. We have the solar cycle. During, during solar minimum, they have a very dipole-like configuration of the global coronal field. So there's closed flux here at the equatorial plane, more or less. There's lots of open flux here. You have tons of fast solar wind coming out of this open field region, and then lots of kind of junky, dense slow wind coming out of this low-latitude region. So this is very structured at solar minimum. This is, is about as simple a magnetic configuration as the sun can get. Now, as the activity ramps up and you get a whole bunch of active regions emerging and things, um, the magnetic configuration of the Sun goes really crazy. Then you have, which is essentially a complete mix of open and closed field sort of all over the place. And so this structuring—this is solar wind speed um, in, in height or radius on these plots—um, and red and blue are polarity. So you have an open magnetic field of positive polarity, like the North Pole, and an open magnetic field of negative polarity, um, like the South Pole. And so these are very well separated. And so, you know, if uh, we remember from, you know, first-semester E&M, if you've got oppositely directed fields out here and there's got to be a current sheet in between them, and that's the heliospheric current sheet. And so during solar minimum, the heliospheric current sheet is just flat, kind of extending out of these streamer, streamer regions. Once you've mixed up all that topology, you've got all of this open and closed solar wind intermixed, and so you've got bits of fast and slow, and you've got mixed polarity kind of all over the place. So that's interesting. We're going to zoom in and look at some of that later. Um, and then later on, on as the solar activity calms back down, you return to that more relaxed state. So this is the basic picture of the solar wind in the, in the minimum configuration.

So if there was no solar wind, you just had the magnetic field of the sun, it would look like this; it would just be a dipole. And that's our kind of fundamental unit of magnetic field, fundamental structure of magnetic field. So here's a dipole, but the reason why, but the sun doesn't look like this. The sun doesn't look like giant closed loops forever. The sun looks like open magnetic field, right? This—so the way you go from a dipole—so this kind of configuration where a bunch of this overlying field has opened up—is you put gas in this configuration and you heat it up and you make sure there's a vacuum over here, and you've got a big ball of matter here, and then you've got the wind outflow. And that solar wind carries all of these high, high-lying kind of weak field loops that wanted to be closed as part of this dipole gets dragged out by the momentum of the plasma and creates the solar wind. And so here you've got that current sheet in between positive and negative polarity. Now, Parker worked all this out, and in fact, here's a set of, you know, hand-drawn curves from 1950 something, um, where in fact all, all you need to do is you can assume the plasma as isothermal, and as long as you've got basically no density here or no, no pressure at infinity over here and you've got a big, you know, solar mass here, you heat, you heat the corona to a million Kelvin, and you get the supersonic outflow. And so depending on the temperature in this very simple isothermal model, you can dictate the solar wind speed, and this works surprisingly well. It's not the whole story, obviously, but even since the 60s, we've kind of understood what's going on on the, on the basic, the basic large-scale picture. So the sun is more complicated than just this.

So what's the second most complicated closed field structure? So this is what a, this is what a helmet streamer looks like with a real magnetic configuration. This is a PFSS, so that stands for potential field source surface reconstruction of the magnetic field, which is derived from the photospheric magnetic field distribution. So it's, it's a kind of low-order spherical harmonic—that's another one of these things that we did in E&M that turns out to be super useful in real life. I was very surprised about that, but you solve those charge state distributions on a, on a sphere, you have a boundary; it's a, you know, you derive your, you, you expand your potential as a sum of spherical harmonics, and then you solve for all the coefficients. And that was a Jackson problem that everyone loves, and now it is also a solar physics problem that we've been doing for, you know, since, since the late 60s. And this also works surprisingly well. Um, so this is, this is the helmet streamer with this open field being dragged out, um, and if you, and the defining characteristic here is that the open fields on either side of this closed field system are of opposite polarity, but if they're of the same polarity, you have a different structure in the closed field system. So if over here you can see the blue field lines, let's call them positive here and positive here, so you've got this funny sort of flux distribution where this is positive and this is positive, and so there's no current sheet here; there's just some field lines, but there's this weird interaction boundary between the open field and the closed field. This is called the separatrix, and these are properties of magnetic fields; they're basically where um, the boundary of flux is contained. And so what do I mean by that? I mean there's a neutral line down here; there's field lines going from positive to negative over here, and if you integrate from the neutral line over this way and just count up the flux, you're going to get some amount of positive flux up until the edge of this thing, and if you integrate this way, you'll get negative flux up until the middle of this thing. So all of the flux above this polarity inversion line right here is contained within this separatrix volume. And the reason why that's interesting is because if you've got a field here and a field here and they're connected in different places, is you can, you can build up different conditions in the plasma on field that's adjacent, and so that means on the open, say on the open field here, you can have lots of fast outflow, and on the closed field you can have a lot of material accumulated. And so now there's a big density jump in addition to the magnetic field topology jump, and that's interesting because the minute you do that, all of the things like wave dispersion and um, you know, mode conversion and like density gradients make are good things to scatter waves off of, and when you apply some energy here or even here, um, you can create some shear between these flux systems.

So what, what, what do you think would happen if I basically increase the field strength of this closed field system? Like what would happen to these field lines and how would they interact with these field lines? Anybody have a guess? Yep. Yep. So, so definitely, definitely these will inflate. And so what will, what will happen to this boundary? Yep. Yep. So, but it'll be a different shape, right? Because now I'm expanding the field outwards, and, and say I energized this by spinning it. Now I've added a component of the field coming out of the plane, right? And you could imagine just this thing sort of coming out of the plane and going back into here, so it's like a dome, and I spin it like this, and I concentrate that spinning around the polarity inversion line so that I'm slowly adding basically whatever the component is out of the board. So that means after some amount of time, I'm going to have a field with, say, okay, let's say out of the board is a y component, so positive y component here, negative y component here, but my outside field is still potential; it didn't, it doesn't have any of those components. So now I've got a discontinuity across the separatrix boundary, and every time I have field that's not lined up, I've got some current. So the neat thing about these separatrix boundaries are they're where electric current accumulates, and so there's a way in which you could say all of the energy that I added into this system by doing something to it—either having strong fields bubble up, having some sunspot twist around, all of any kind of anything I do to this—will create current along the boundary of this flux system and whatever this flux system is up against. And so you could say, well, my energy before I did anything is like E naught, and then I energize it; that energy, that energy difference is contained in J, and J lives on these separatrix boundaries. And I think if you, I think you do the integration, you can write it either, either way, um, as like J squared or J squared. Basically, you've got, you've got this stress, and you've got this current sitting here on this separatrix. And so if you continue to stress it and you continue to drive that current intensity up, now you're, now you're getting into the regime of being able to kind of push these fields up against each other and initiate reconnection. And so this pseudo-streamer topology is very favorable to reconnection, um, and that, that turns out is going to play a, a reasonably large role in generating kind of the junky solar wind from bright streamers, kind of dense and junky. Yeah, it does, but the reconnection is a little bit different because the topology is a little bit different. So we will come back to reconnection at the helmet streamer, but yeah, so there's a, there's similar dynamics, but it's the topology is a little bit different. So instead of having an X point, you have a Y point, and so you have that connection to the heliospheric current sheet already, and so now you're stressing a system that already has kind of a permanent current sheet attached to it, and that means the dynamics are going to be a little bit different. But that is an interesting question because you are like, "Well, why would reconnection create one kind of outflow in one situation and a different kind of outflow in another?" And lucky for us, the solar wind is so complicated that we all get to spend our lives looking at it.

Okay, so this is back to—seen all this—this is the cartoon version of the helmet streamer and cartoon version of pseudo streamers, also sort of solar minimum and solar maximum. Um, if you plot those magnetic field solutions for the corona um, in a, in a magnetic field configuration that's like a dipole, you're going to get something like this. And so this is a, this is what they call a synoptic map. Um, beneath all of these field lines is the flux distribution; can't really see it because I drew too many field lines on here, but over here you can kind of see it, um, but this, all these green field lines are closed, closed field. And so in fact, the ones that I've drawn—these particular field lines that I've drawn—are the last closed field line before the, you know, the source surface. So this is all, it's all PFSS fields. Um, so what do you notice here? You notice that the sort of the North Pole is red polarity, red; it's either plus or minus. Um, South Pole is blue, same thing, and basically the entire equatorial region is closed field like this. This is exactly, you know, the thing we looked at before, and this is an example of what that field looks like when it's really crazy and kind of all over the place. And you can see that it's not that different, but it's getting more and more crazy. Um, so the big, one of the big things you notice is now there's this whopping great bend in the current sheet. So instead of just being flat along the Equator, you know, it's up at kind of plus 45 degrees, comes down, lives at, you know, minus 50 or 60 or something, and comes back up plus 45. So it meanders sort of all over the place, and you notice that the structure, the level of complexity in these, you know, red and blue regions, which are the open field, open field lines that connect down to the, the surface, um, you've built all of this weird structure over here, like this is all open field, but it's open field in these very weird patterns. And that's, that's because in a map like this, if you don't have a color, that color means this, this part of the field is open to the heliosphere. So this gray means it's not open; that field is not open, but it doesn't reach the source surface because it doesn't have any green field lines drawn over it. So it's closed flux that doesn't reach the source surface, which means all of these gray regions are pseudo streamers because if it was open field, it would have been red or blue, and it's not, but its closed flux is not underneath the helmet streamer. So you can see the difference between basically solar minimum and solar maximum is that there's a ton of pseudo streamers up, a lot of, a lot of the surface of the, of the Sun. And so that's partly why during solar maximum it kind of looks like there's stuff, radial striations everywhere, and there's sort of slow, junky wind everywhere.

Okay, this is, um, this is going to confuse everybody, but at Michigan, one of the things Michigan is famous for is looking at um, solar wind composition. And what solar wind composition is is basically the minor ions, the kind of rare elements in the solar wind—the heavy elements of carbon, oxygen, silicon, iron, neon, argon—um, and those, the, the charge states of those elements tell you something about the conditions of the corona where that solar wind was formed. And why they do that is because the density drops off so fast that for ions like carbon and oxygen, um, you know, charge states or like C6 have a formation temperature like around a million. And so any carbon that's sitting around in the solar atmosphere that's above a million Kelvin wants to be sitting at C6+, okay, or C5 or C, depending on where you are on this temperature gauge, but they, they change, they change ionic states by collisions and running into um, you know, each other and electrons and radiation. And so there's a whole set of continuity equations you can write for kind of every one of these minor ions to track how much of them are in the 8+ or the 7+ or the 6+ or the 5+ kind of bin for everything. Um, and if you plot those things in time, you can watch basically the amount of, the proportion, you know, how much, how much of this carbon is in C6+, how much of it is in C5+, how much of it is in C4+, you can watch those things change with distance because these things that determine how much there are—the ionization and recombination rates, photo, photo, photoionization rates—all of that depends on density and temperature. And the density and temperature when you're very near the sun, it's large, and so those rates are large. And then as you accelerate with the solar wind and get out into space where the density is basically zero, you no longer have collisions, and, and whatever, whatever the last kind of state was in the solar atmosphere before you um, basically shut off the ability of carbon or oxygen or iron to change the, to change their electronic configuration, um, that's, that gets frozen in. So what you can see down here is that basically all of these elements are more or less—these curves more or less straighten out after, you know, four solar radii, three solar radii, maybe six, seven, eight solar radii, but definitely by about 10, kind of every heavy ion charge state is set more or less. And so that's an interesting way to kind of back out some piece of solar wind's temperature history because if I've got a piece of solar wind that has iron 16 in it, it must have been very, very hot and must have come from some very different conditions than a piece of solar wind right next to it that had iron 9+, like massive temperature difference. And so one of the things that people want to do is they want to take um, remote sensing observations of these charge states, which you can get with spectroscopy, and try to match them with the solar wind that you measure in situ with a spacecraft um, when you've got instruments that can, that can resolve these, these charge states. So composition basically means you can tell if there has been heating in the corona, and so I'm going—my view is that that kind of shows reconnection or can often be related to reconnection—um, and it tells you how long the material's been sitting in a closed flux region um, through, through elemental composition, but may or may not talk about that.

Okay, Parker spiral—that's that. So we know it has the structure because the Sun rotates, and all those radial field lines that open up into the heliosphere get this shape. And so we can model that because all you, all you have to do is you have to have a magnetic field model, and then you spin it like the sun spins, and you can watch all of these field lines create this um, spiral structure. And somebody tell me in like two sentences what your guys's project was on the solar wind. Did you do that yet? Yes. Yeah. Okay, that was a lot. So what did you plot? You plot density and temperature and polarity. Mhm. Temperature is a tough one because, anyway, none of the models get it right yet. I mean, they don't actually, the models don't get anything else right either, but um, we do, we do at least know what's going on. Okay, so yes, we get the steady-state solar wind picture, and here is what I was talking about in terms of um, the temperature creating distributions of charge states. Um, and so here at like Michigan, for example, they've built in all of these charge state rate equations into their MHD code, and they can run it in a mode that calculates this kind of at every time state, and that's pretty impressive, and that's, you know, I think worth doing. And so if you look at this thing like 07 to 06, um, this is a ratio of hot oxygen ions, um, and you can see it's, they're basically red in the streamer, so that's the hottest material, um, blue in these open field, kind of the deep open field coronal holes, sort of yellow at the boundaries of the coronal holes, and then kind of uh, you know, green, sorry, green transition into yellow, but notice over here, here, this is kind of a pseudo-streamer structure, um, and it's also yellow, so that's kind of interesting. And then similar sorts of things for C5 and C6 and for the average charge state of iron. Um, so you can really tell that if I've got some solar wind that has a 07 to 06 of, you know, one—

Had to have come from from these these Clos field regions. Now, if I'm measuring it with a spacecraft at all, that plasma had to have somehow figured out how to get from this closed field region onto some open field Lins, and that's where reconnection comes in. And so that's why that question about what kind of reconnection happens at these sorts of interfaces is also uh very important.

Yeah, yes, some of them are in later slides. Um, and there's also there's also I mean, okay, the problem with real data is you can find every example of every everything. So you can find some pseudo streamer solar wind that looks exactly like helmet streamer solar wind, and then I think you can find some pseudo streamer solar wind that looks nothing like helmet streamer solar wind. Um, and why those two things happen would be an interesting thing to figure out.

Okay, so if we if we know if we know this kind of structure and we have our oneu spacecraft measurements, we can kind of use um what we've known to figure out where all the solar one came from. Um, so what I've got here is I've got Parker streamlines basically mapping one month, one caring in rotation worth of Sol wind um from 1 Au back to some position at the sun. Um, here's solar wind speed measured by the a spacecraft at 1 Au, so that's out here at 215 solar radi. Um, I've colored it by speed, and if you take basically This Time series and run it backwards, you're plotting in Carrington herenton longitude coordinates. So so if I'm sitting here measuring some bit of 500 kilometer uh per second solar wind speed, I know that the Parker spir Ral at 500 kilometers a second has a certain shape, and I can just draw that shape all the way back to the Sun. And when I do that, I end up at a different carington carington longitude which corresponds to a different date. And so in fact, the sort of the two or three or four days it takes solar wind to get from the Sun to one auu is the two or three or four days you go backwards in time by doing this tracing. And so that's your guess at basically where the solar wind comes from in terms of position in caring and longitude, and you have to you have to pick a distance to stop tracing this spiral back at, and you might as well pick the potential field Source surface of your magnetic field model at the Sun.

So now we're going to talk about analyzing that structure of the magnetic field back at the Sun using um very unhelpful MTH. Okay, here is we've seen these two before. So this is the solar magnetic field, the coronal magnetic field, turn solar minimum, more or less flat. Here it is at a more active period, kind of a big fat warp in it and a bunch of open field structure um including all of this pseudo streamer stuff. Um, if I wanted to describe how complex all of those magnetic flux systems were that were interacting and causing the solar wind to come out, um one way to do that was invented by this guy um off um and he said, okay, I'm going to take a magnetic field line and trace it, but I'm not going to do just that. I'm going to take a whole Locust of points around the beginning of that magnetic field line, and I'm going to track the shape of that locus of points as the field line as my model field line goes through the atmosphere. And then at some outer surface when I stop tracing my field field line, I'm going to look at what shape those points are. And so on this cartoon you can see if you start with a circle in this particular example, the shape that I've got out here is it's kind of massive ellipse. And so there's a couple things that are interesting about that: one is the that describes that the basically the difference in area between this circle and this circle kind of at the beginning of your flux tube and at the end of your flux tube, those differences in area tell you how the magnetic field changes because you've got conservation of magnetic flux. Um, so the same amount of flux that goes through this circle is going to end up going through this giant oval over here, so that tells you about the shape of the field distribution. Um, it also tells you how distorted and how smooth basically the magnetic field uh solution is because if I move just a little bit off One Direction of this field line and I come up here, okay, maybe I didn't maybe I've only only moved a little bit dire of off the output, so it's sometimes I can have a kind of Delta thing over here just ends up with Delta over here, but sometimes Delta over here ends up in a hugely different place. Um, and so you can see when that happens in this map of this quantity, um these derivatives are like position like field line position, so you basically take the ending end point and change the beginning point a little bit, make the end point change the beginning point a little bit, you do it in all the in all the coordinates.

So what is this? So this is this is what they call a qmap, which is basically just the logarithm of this squashing factor, and there's two signs, so there's two colors. Um, you've colored it basically on the polarity of v, and where it changes color you have the most uh Divergent kind of situation with your magnetic field. And so the most Divergent situation with your magnetic field you can have is basically the helmet stream because on one side my field line goes up to one pole, if I move a little bit of Delta now on the other side of the current sheet F line goes down to the southern pole, so even though out here one a u they were very close together or three solar radi they ended up very far apart. So these derivatives on position are huge because now you're like on other on other side of the Sun. So that's why this line between the two colors is basically black. Um, log of Q is like infinite I think actually at at singularities. Um, but there's this other set of features here which are kind of like these dark arcs that come out of the heliospheric current sheet which is this sign change and then kind of connect back to them um somewhere else, and it turns out these dark arcs are exactly the separat structure you'd expect from pseudo streamers because there's no sign change, but it's the same kind of thing where if you move a little bit your field line comes down, you end up on one side of your closed flux region, and even though we're now all the same polarity, you move a little bit kind of end up on your other side. So all of these arcs and you can see okay here um okay for this top example this blue arc has got something to do with this patch of gray right here, and so you can see uh you look really carefully there's a couple of blue dots here. So in fact this coronal hole is not uh is not just in the southern the South Pole, it's got a little tiny extension up here, that little tiny extension maps to this Arc, and in fact same thing happening over here, this extension of the North the red Polar Polar parental hole coming down creates this red Arc here, and down here in the more complicated case basically every time there is a unshaded region you've got an arc above it. So there's a whole mess of like these kind of narrow open field corridors down here, and that's generating all of these ARS. So there's kind of big changes in connectivity uh sort of associated with being mapped back to to these pseudo streamer regions.

And you might say, okay, well that's that's a cute mathematical exercise, um but what does that mean for the solar wind? Um, and you can if anyone wants to download these things they're available here. Um, okay, so this is what this is what this stuff looks like in terms of coronograph measurements. So here's the sign change, and you get exactly this helmet streamer structure. Here's one of these big arcs, and you have these kind of thinner pseudo streamer rays in the crog graph images. And if you look sideways at what this stuff kind of looks like, um so for example this is a the they didn't color it with red and blue to show polarity, but it's the same it's the same kind of thing. Here's the heliospheric current sheet, here are all these arcs, and if I take this surface where I've defined my upper boundary in my coronal magnetic field bottle and I move it down, you can see the tip of the the helmet streamer which is this singular line all the way around here, I'm now intersecting a bunch of this closed flux. So I took my my Surface which was at 2.5 up here with my sort of One Singular yellow line all the way around here, and if I move that surface further down to here, now all of this bright stuff is closed flux. So this is the top half of the helmet streamer that follows this pattern, and now we're starting to get into the close field regions of the pseudo streamers that live beneath these arcs. So remember from the beginning when we were talking all about the edges of these flux systems, these boundaries are the edges of the plux systems, these are all separator surfaces here, and so we'd expect a bunch of current to be on basically all of these surfaces, the boundaries of all of these surfaces, um and that and that will and that will cause reconnection. And so that's this cartoon is basically if you have an exchange of topology, you have a bunch of current built up on these boundaries, and you get reconnection, your magnetic field lines change their connectivity, and you can jump open field around.

And so what does that look like um in real life? Well, you stress one of these pseudo streamer configurations and you build up current and you allow basically blue and red flux to come together and change their connectivity and become green and yellow, which is this is called interchange reconnection, and it's basically moving blue flux from one side to the other, and you get all sorts of outflow into the green and into the yellow region. And in fact those boundaries um it's not just the pseudo streamer boundaries where that happens, it's it's also the helmet boundaries. So there's a couple different ways to perturb the helmet streamer boundaries as well, and so here's a couple of simulation results that show the structure that appears um when you do things like spin up the open closed field boundary. Um, and so an example of that is kind of here if you had some shuffling of the boundary around here, you would get interchange or connection and some material that was previously closed to then escape on the open field. And so this is the kind of uh okay the color here is basically where these where these field lines map to, um so this is a kind of a geometric um mapping another kind of geometric mapping visualization. And so this is a narrow coronal hole, and if you spin up the whole thing, you put all of this structure into the open closed boundary, and so all of the stuff that's purple is um open field, and all the stuff that is gray is closed, and now I've taken what was a a very smooth separatrix B and I put a ton of structure into it, I've created this whole like Spiral of intermixed open enclosed field, and this is what that looks like if you simulate the entire sun and keep track of all of the places that um reconnection happens in your simulation of the solar wind. So you're not doing anything except letting the solar wind run, and there's still reconnection along all of these separator boundary layers. And so here is I don't know one of these things, the heliospheric Curren sheet is somewhere in here, but all of these other things are the separatrix boundaries of pseudo streamers kind of being jostled around. And so if I said okay everywhere that there there's a big difference in this of these okay the difference between the eved flux and the actual flux for my field line tracing in this movie, um you'd say I I'd expect Slow Wind everywhere that I'm seeing big changes in reconnection kind of all along these boundaries.

Yeah, 30 minutes, great. I have no idea if I'm close to being done or not. Um, so this is this is what some of that stuff looks like in observations. So here is um you can trace basically a bunch of this outflow stuff from helmet streamers and pseudo streamers in density in coronograph images and even in heliospheric images. Um, and so a bunch of this stuff makes these streaks, and when you come up here and you look at the actual like density and velocity of things that you visualized up until the spacecraft interception, um it turns out a bunch of the structure is reflected in the insitute density quantities, the velocity quantities, and even the magnetic field quantities. Um, I think this is PSP um looks like okay I'll find the reference update the slides, it's I remember the paper that this is from, I think it's rard 2020. And this is another way of doing that, you can look at density blobs coming off of chonograph streamer Rays. Um, and if you track these disturbances, this is just stream after stream stream of nid stuff just coming out of this slow solar wind all the time, coming out of these helmet streamers all the time, the St you can see that's Blobby and got a kind of signature in white light is pretty much always slow um because the fast wind is just that much less dense. I mean the fast wind is not there's a ton of structure in the Fast Wind as well, but it's a different kind of structure, it's more sort of waves, it's more turbulence, it's more um pristine kind of um that was just accelerated and had a smooth ride out from the open field regions, whereas this stuff that comes from the boundaries is all sort of gunky, it's slow, it's messy, you're you know it's got It's got all the same waves or at least a lot of the same waves that the fast wind has and then it's got all of this other stuff too, so it's just kind of messy, but it's what we can see which is interesting.

Okay, and this is more examples of this structure stuff of the Slow Wind in data. So um down at the bottom here is like a histogram of small flux ropes, what a small flux rope is is a kind of a coherent magnetic structure. Um, uh here's an example of basically periodic density structures that you can get some some characteristic size scales and characteristic sort of uh periods out of. Um, and you can see these basically uh what looked like sort of sequences of Peaks. Um, and it turns out some of these statistics seem to match up between the insitu stuff and what they see in in the remote sensing um over here, here of paper where they also see this kind of structure in those same composition quantities. So you know carbon oxygen iron um and they have sort of similar types of widths to the density structures and similar types of widths to the magnetic field features. Um, and up here at the top is super thermal electrons which give you a sense of field line conne activity. And so whenever you have a a just a basically uniform strw either sunward or anti-s sunward that means you're on comfortably open field, um the electrons are just going One Direction, everything's fine. Uh, when you get these patterns that have basically electrons going in both directions, that's a good indication that probably your magnetic field line is connected at both ends, you got a little bit of closed flux in the middle of your open flux solar wind. Um, and that happens a lot in slow solar wind, but you can see it's not uniform and not continuous because then all of a sudden everything drops out. And so that's an indication that your field Line is now disconnected from the Sun, both end points touch the end of the heliosphere. Um, so you basically lost your source of energetic electrons, your connection to that source, and then they're back again, and then you lose them again, and then they're back again. And so what that the impression that gives is chunky solar wind, I'm connected, I'm not connected, I'm connected, I'm not connected, and it's just a total mess, but it's a total mess with some structure because you can measure these characteristic widths and you can measure you know um well it's it's it's nice that they all seem to line up, not all the time, but um if if you were to really take this to heart, you would say there's some correspondence between density and field structure and velocity and composition, and all of that is related to the origin and release and acceleration that that solar wind has experienced. And so that's why I'm so interested in all of this reconnection stuff because that's how I think you can get some magnetic field signature and some density signature and presumably some set of um composition signatures kind of in the same process. Um, it's not definitively worked out yet, and so it's a it's a it's an area of research that lots of people are still um working on which is makes it interesting which why I'm talking about it.

Yeah, yeah, not just in CMEs. Um, Okay, so we tried to simulate some of these processes like reconnection at the helmet streamer tip, um and see that's going to go okay, and this is kind of this is one example of uh what this Dynamic solar wind release looks like in simulations. Can you can you just replay that movie a couple times back? Okay, so this is two viewpoints of um basically a set of reconnection events which take the last little bit of dense okay, so this is velocity to my left and density to the right, um and you can see the the density has accumulated in the Clos field region. You just play the movie like on repeat somehow, um and you can see stuff falling in and flowing out and pinching off, and if you play it one more time, sorry, um you're going to see these kind of spiral structures that get built up here, built up here and ejected. And so that's the that's the density component and the magnetic field component of what is basically a streamer blob, these helmet streamer blogs. Um, okay, sometimes so the the the thing is okay if we play this one more time, um if you okay yeah you're not um there's a way in which okay can you pause this movie? Should I run back there and pause it? Um, how about right here? Okay, yep, perfect. Okay, so this thing you see where the field lines are all kind of spiraling around, so so the view point here is a little bit Edge on, the current sheet is kind of in this plane, and here you're viewing it U sorry this is face on, this is Edge on. So here here are the sheets basic right here. So this little Decon this this disconnected thing which looks like a tiny spiral in in field lines when they're plotted like this is actually if you look at these same field lines, the green is connected down here, comes up it's spiral Spiral spiral Spiral and then okay it could either connect back down here somewhere in which case this blob be closed or this one of these spinning field lines could just act up over here in which case it would be open. And the problem with the helos current sheet is all of that changes all the time. So the the minute this guy runs into something else, all of these field lines have the opportunity to sort of rearrange themselves. And so yeah yeah it this is the helmet streamer version of that, but yeah it's the it's the I mean okay so eing Wang and Neil Shey and the chronograph guys would call these the sort of down down flows streamer down flows and they see them in white light and and that is just a kind of larger less energetic version of yeah what they what they call basically super arcade down flows which is the same exact reconnection process just in a CME you know. And so um there is in fact observations of a bunch of stuff kind of coming back down on the periphery of these helmet streamers and stuff. Okay, but that was a good question, and I think there's more I think there will be some more simulations. So okay so this is this is a different simulation, but it's kind of showing the same thing. Um on the left here I'm now looking at basically 30 solar radii worth of this is the equatorial cut in a in a simulation where basically the sun is just one of these solar minimum dipole configurations. So this whole half plane here is the current sheet at the equator, and what I'm plotting is basically the sign of B Theta which is kind of an annoying thing to plot, but the reason why I'm ing that is

Because if you have a flux rope, and the component of the twist is in spherical coordinates Theta, and it's down on one side and up on the other side, I now have a bipolar signal in Theta. Where I've got whatever negative positive or positive negative, so I've got a kind of white blue or white black signature in Theta. If I had a bunch of spiral and field lines, so if I look at this, I say I've actually got that white black signature all throughout the entire plane. So my entire heliospheric current sheet is nothing but a whole bunch of flux ropes. That doesn't quite seem right, but in fact, it is.

So here's what happens if I just play that all of this stuff is streaming out; all of this junky stuff is constant outflow from the helin streamer. So you can see that there's a distance that it sort of starts, and that distance is uh one two three sort of four solar radii or whatever, but it's about the distance of the helmet streamer. So if you put some is on the chronograph images, yeah, sort of two three four five solar radii, um and all of the stuff is just kind of streaming up here. Now what does this look like? Okay, over here this little E1 dot is this guy here. And so if you watch just this passing by, and if you can play this this movie a couple times over again, um this is a stationary Observer just right here, and I'm all I'm doing is sort of plotting a bunch of field lines in and around and near this guy. And so as you can see this sort of grayscale stuff moving towards the edge is ex is exactly this stuff. Um it is in fact every one of these bipolar bipolar regions, or in fact these little these little blobs, these little island blobs in the Slow Wind.

So that's at least one guess about where a bunch of this structure and the solar wind comes from. It's kind of fine scale magnetic structure; it's no it's this weird it's this weird B Theta; it's basically sign of B Theta, sign the asmuth field. So it so it's you could Track by I sort of how many of these flux ropes. Yeah, yeah, you have spinning here. Yeah, yeah. So that that's an excellent point, um because this is in the middle of the current sheet. So the current sheet is basically where the field switches signed, so there's almost no magnetic field strength, so beta is super high; there's a ton of density kind of sitting there because it's in pressure balance with a stronger field of the you know. So it's it's fluctuations in this kind of relatively High beta region associated with the current sheet in the plasma sheet. CMEs just massively magnetic field dominated. Okay, boom. It's low density; field strength is super high. And so even though they have like almost identical magnetic field structure, the plasma is so different that yeah, that's the CME comes out and it's like you know blows out a huge part of this solar atmosphere; this stuff is tiny tiny little flickering. And so there's not like there actually not any energy being lost here; it's not a lot of field destroy. So 10 minutes good.

Okay, so here are simulations of Pudo streamer reconnection, um with a bunch of GD points, but it's a similar it's a similar type of thing. So so here's the here's the separator surface, and here's some observations, and you can see that they're you know identical because this magnetic field stuff works pretty well for explaining the sun. Um here we've increased the energy in this closed flux system; we've now got currents that have built up on the separatrix layer and we're reconnecting, but now we've got so many grid points that we're getting some structure in the reconnecting layers. And what I've got arrows pointing to here are all little pseudo streamer versions of the flux ropes being made at the helmet streamer. Now this is a much smaller spatial scale because this current sheet is is just basically right above the pseudo streamer, so these things are way smaller than the than the helmet streamer blobs, um and that has consequences like this; that means that when they open when they reconnect and open up there's no opportunity for that's or sorry that plasmoid structure to maintain its flux; it reconnects in the pseudo streamer and opens up. And that makes sense because there's no unlike unlike a helmet streamer in which you could kind of live in the current sheet and maintain your plasmoid structure, um there's no the current sheet only lives here, so the plasmoid structure is maintained in the current sheet, but by the time you're out into this open open field region you've basically got to got to decide, um you know your flux rope is either going to reconnect this way or reconnect that way; it's going to figure out how to open up at one end so that it's aligned the same way that all of the other open field is aligned. And that means that the twist part of it of that plasmoid has the opportunity to propagate along the open field. And so pseudo streamer reconnection ought to have something like this pattern of twist in the magnetic field if there's plasmoids being generated on the interaction layer between the open and close flux. And so if you took a cut through this and plotted the components of velocity and magnetic field, you know sure enough there's a whole bunch of fluctuations in B and V. So this is an alanic structure; this is basically an alane wave, an alane twist wave being released into the wind.

So from these simulations, one of the things I would say to the earlier question is the difference between helmet streamer solar wind and pseudo streamer solar wind may be related to this kind of structure. So I would say I'd expect pseudo streamer solar wind to have more kind of twist waves than helmet streamer Sol uh yeah yeah yeah yeah uh that's a good question, um yeah I I think so I think statistically you'd expect say Northern Hemisphere pseudo streamers to be left-handed and Southern you know southern hemisphere pseudo streamers to be right-handed. Um that's a good question, um I think that's totally open and I I think it would be interesting and possibly I mean it might be that we that we could still see it at Parker solar probe distances; maybe I would be interested in seeing it at Parker solar probe distances. Right yes yes okay. Um I I think it's going to reflect the driveing, so I think if it's a kind of continuous lowlevel sort of gradual driving then that's going to sort of um that's going to basically be modulated by [Music] some basically the structure of the pseudo stream. I mean there's some inherent kind of uh energy storage sort of time scale and then some energy release sort of time scale, and I think once you've sort of saturated that you'd kind of get continual. Okay y let's yeah yeah that's a good question; there's no shortage of waves at one auu whether we are whether we're able to sort of say these are exactly the kind of waves that come from this exact kind of Recon connection we're not there yet, but I think we are moving towards there, um but but yeah Marco's right it's not just the magnetic field structure here but there is also material. I mean that's that's this is this is density mass density over here, so every one of these twist waves has at least as it was as it was a plasmoid in the reconnection region had some Associated excess density with it, um and that is that is dumped into this outflow as well. And in fact you can see it here; it's a little bit offset; there's a big density Spike kind of over here, and as the reconnection proceeds the magnetic stuff sort of separates a little bit, um and that's that's kind of expected uh whenever you have a sort of disturbance that's got all different things you know it's going to spread out because different parts of that disturbance have different propagation properties.

Um okay, and so here is just another sort of a more um more or less the same thing, but instead of being an isolated suo streamer you can stick it into a more realistic field configuration, um you can do the same thing where you kind of stress the boundary and you you can look at that wave sort of propagate up through and live on these sweb arcs, um and you may expect there to be some signature in the electrons if this is if this is a kind of a big enough wave. And um okay so there's a whole inue example here that we don't have time for, but basically if you do the exercise of of the kind of heliospheric back mapping and you look at what the corono magnetic field source region for say this interval of solar wind this carent in rotation worth of solar wind, um you find that basically every period of Slow Wind and I've highlighted them in yellow and uh green and blue here, every period of slow wind is either a associated with the current sheet Crossing at the helmet at the helmet streamer or um are basically these pseudo streamer arcs in the in the Q map. So here you've got a couple of lowly coronal holes; all of this gray stuff is various pseudo streamer flux systems, and sure enough for this particular solar wind interval it turns out I'm I'm sampling in the you know at 1 Au sort of a ton of these arcs. So interval one is an arc; okay, interval three and interval eight are both of those helmet streamer periods, um but over here interval four five and four five six and seven are sort of all sweb Arc related. And if I go over here and look at Inu data, this is sort of a mega complicated plot; basically at the top is velocity; here we've got the superthermal electrons; notice they do in fact get kind of broader in each one of these pseudo streamer layers, um and then down here at the end we have charge States, and you'll see that every one of these selected intervals has kind of weird charge State signatures or elevated charge State signatures so in oxygen and carbon and iron they're sort of hotter. Right right. So that means that's at least indirect um evidence supporting the idea that there was some process happening back in the corona that was energizing some of this plasma, and reconnection is one of those processes.

Um you can do some other stuff like some funny spectral analysis on the magnetic field fluctuations and you find out that for the most part statistically slow solar wind from helmet streamers and the slow solar wind from pseudo streamers uh looks kind of the same with possibly one kind of bump in this hel streamer wind, and this bump is interesting because this bump is at about 90 minutes, which is the exact same Cadence that folks that were looking at streamer blobs coming off the helmet streamer belt, um that's kind of their characteristic um periodicity, although it's hugely variable; it was interesting that there's perhaps some indication that that periodicity shows up in the magnetic field structures as well as the composition and density structures as well. Um so there's some other stuff on back mapping more of this sweb, um there's an example of this exercise uh sort of about Parker solar probe data, so this is another thing where you're like okay here is here is the connectivity of PSP, and you can see that at some part of this orbit you jump from one kernel hole to the next kernel hole to the next kernel hole, and these jumps are over sweb arcs, and these jumps are between isol gral Hol. So this is gray, so this is a pseudo streamer flux system; this is a pseudo streamer flux system; there's kind of closed flux down over here too. And so whatever this period of solar wind that Parker is measuring in counter 10 has got some mix of very interesting coronal hole stuff and potentially a whole bunch of this kind of separatrix interaction stuff between between the two sources, um and if you look at the data okay you can kind of see a a ton of structure; here's like a whole set of Switchback patches, um there's interesting density structures here like an enhancement at one side and then sort of lower density, and then over here we have an enhancement at one side and then sort of lower density, um and then these Switchback things have a ton velocity structure Mass flux structure, um but these two these two arrows are basically the you know are basically these two s arcs kind of here and here. And so if you look at kind of the generic simulation results from earlier you had this kind of density Spike offset by a bunch of kind of magnetic fluctuations, and while we have you know it's not a one toone correspondence because this is a very idealized simulation and this is real life and messy solar wind, the fact that there's sort of a density Spike and then a trough and then a density Spike and a trough and the big sets of magnetic fluctuations come after the density Spike kind of in both directions, um at Le at least to me says we're sort of on the right track with this picture; we're not there yet, but I think we're sort of on the right track. And that's it for part one, so that was the solar wind; let's do a amazing.

So we went from talking about the separatrix which you defined very well, um then you started calling them sweb art. Ah yeah yeah yeah yeah could you clarify sweb separatrix web? How's that different from a separatrix? It's it's not different; it is just the map of all of those separatrices were given uh basically Carrington rotation worth of magnetic field in your extrapolation. So if I if I say okay I'm going to create a pfss field extrapolation; takes me 27 days worth of data to accumulate the magnetic field observations, um I then generate my sort of 3D Field Box in a box, um every separatrix quantity in that 3D data Cube gets plotted in those Q Maps. So I basically integrate a bazillion field lines through my mag my pfss data Cube; take all those derivatives of moving the things around and then plot that quantity, and that's that's that cute the cute thing that I was showing that's got a big spike at the at the where the polarity changes VR and all of those little arcs connecting to pseudo streamers and or neuroc coronal homes because those are kind of two parts of the same sort of topology; either have two narrow coronal holes for a pseudo streamer, but in reality I just have both of those things all the time. So any of those connecting arcs sweb yeah thanks and M web is kind of the whole thing, but yes any one of those is is an s with Arc or it's separatrix or it's a I mean there's a bunch of kind of jargon on the topology terms, but they all sort of mean that I mean if you can draw the draw the system you know what's going on even if a million other people call it different things. Thanks.

Okay, so the the first half was large scale coronal structure and small scale coronal transients; now we're going to do small scale coron structure and large scale coronal transients. Um okay, what is the CME? Why does it erupt? Um illustrative examples of solar flares and CME observations uh yeah maybe we'll do some of this um yep reconnection mhd modeling we good. Okay, so coronal mass ejections; large sudden eruptions of the solar atmosphere into in planetary space; a billion tons of matter at a million miles an hour. And I think here we go good I hit this again; I think you have to play it from the back or replay it from the back. Okay, so what are these movies of? Um EU emission, so this is Imaging the solar atmosphere at a certain temperature and density, um probably like some iron iron line, um what we're looking at is over here, and so this is a uh this is an example of a relatively energetic CME, um this was this famous September 2017 eruption, um there are a couple different things here, so on the left we've got just the straight sdo AIA movie, and over here on the right we've got a what they call a base difference version, which is basically if you take um one of these images at time whatever T and you subtract whatever the image was several hours before, um you've got basically a a a measure of how much the emission has changed between those two figures. And in cases for solar flares you know you'll see things like huge amounts of dimming uh which represent material having been expelled into space. Um so if we play this movie sorry one more time, um flares are brightening; there sort of electromag Rapid electromagnetic radiation over a whole bunch of wavelengths, um representative of massive energy release, and that massive energy release heats the plasma to create this bright emission, um and then the the eruptive part of that there's kind of two parts from the CME; the part that sort of stays in the corona which is the flare and creates this bright admission and the part that erupts which is kind of all of the magnetic field and plasma sort of up above this structure, um that really wants to be in space and so it does; figures out how to get there, um and it blows out a whole portion of the of the solar hemisphere. So this is a really weird thing that the sun does because how I mean the real question about CMEs is like how how and why is it energetically favorable to explode a huge portion of the solar Atmos like why would it why would the sun be like all of a sudden let me just explode and now I'm more relaxed afterwards like like how is that a decrease in energy when I've just expelled 10 to the 32 ergs worth of stuff, um and that's what we're going to try to get to in this talk.

This is the basic structure of what a CME is, so so over here, so the first the first movie was AIA, um so that's kind of this this inner part here, and then what I'm showing on the outer side here is the chonograph measurements in white light, so you can see even when you had this huge eruption low down you had this whole big exploding thing that very quickly became much larger than the Sun, and this is expelled into space at 2,000 kmers a second or something. Um over here is a cartoon, um by Terry Forbes, and this is a great paper; we're going to use a bunch of stuff from this paper, so I definitely recommend it if you're interested in seeing these, um this is the kind of structure that describes this observation, so there's a big what they call plasma pileup bubble like shape on the outside, then there's a darker cavity, and there's sort of bright inner there's a bright inner part that's dense as well, and this is this is called the core, um and it's typically thought that the that any kind of erupting prominence or filament material, um that you can in many cases often observe erupting into white light and becoming that cord. So out here you know this is the bright Leading Edge is the pile up; you can see there's sort of a dark cavity in the middle, and then there's sort of a bright Central core, um and so this is what they call the three-part structure of CMEs, um and this is you know pretty simplistic, but it works pretty well, and it does kind of explain most of the large scale features of the observations. Um so this is what they look like, and there's a relationship between basically the plasma emission and the the kinematics of the eruption, so um this is kind of adapted from Jang at all1, um but what I've got plotted uh both sort of here and here you'll notice this curve is exactly the same; this is the X-ray this is the soft x-ray emission, and so x-rays uh x-rays correspond to coronal temperatures of you know like 10 million or five million; basically this is uh if you're if you're keeping track of this is soft x-rays; if you're keeping track of soft x-rays, um there's basically a little bit of brightening and then sort of a rapid increase in brightness, and you have these bright x-ray arcades associated with the eruption, um so these are on the left I've got these colorcoded regions for basically the pre flare the impulsive phase of the flare and the Decay phase of the flare based on the soft x-ray profile. Now what you'll notice is I've also plotted on the same thing here and actually I didn't plot it that she saying plotted it um is the height of the CME that's going off at the same time as this soft x-ray flare, and so this height is of the Leading Edge of one of these euv and then white light bubbles erupting, and if you keep track of that Leading Edge you get this kind of height profile, um you can take a numerical derivative of these height height time plots to get basically velocity time, and then you've got these three sort of velocity versus time um intervals that line up surprisingly well with the soft x-ray

Emission. So if you look at the velocity, it's kind of low in this, in this what you'd call pre-flare phase in the soft x-ray emission. Right. So, so if you're describing this in velocity terms, you'd say this is the slow-rise phase. So whatever this energized structure that erupts is, first it rises very slowly, and that very slow rise is associated with a little, little bit of X-ray, enhanced x-ray emission. Then there's what they call the impulsive acceleration phase where the velocity goes up really quickly, and you can see this is where the the height turns the corner. So you go from slow rise to rapid acceleration. Um, and our velocity here goes from 1050, 100 km/s up to like a th000 km/s, and then it more or less trails off; it either stays the same or trails off. So you'd say this is the impulsive acceleration phase, and then there's a kind of propagation phase, a sort of constant velocity propagation phase. And the fact that these two sets of intervals are so closely related means whatever the mechanism of energy release associated with the flare curve has got something to do with the mechanism of energy release associated with accelerating and erupting this whole structure, blowing out the whole part of the solar atmosphere up above it. Ymhm. It is these, these points right here, this dotted, these, these points and these error bars and these dash lines here are the D derivative of these points. This, yeah, down here, up here, down here, up here. Yeah, so one, two, three. So it's, it's including some of these points too, but yeah, it's basically the the transition from sort of before the eruption to the eruption. Yeah, not really. I looks kind of the same.

So, so for the CMEs that are not associated with flares, they tend to be the slower ones. Um, and so it would look basically the same, scaled down by a factor two or three or something. So it would kind of, there would be a slow-rise phase, and then there would be a sort of acceleration phase, but it would be, you know, 100 to 400, and it would be kind of over the same or even longer period. And, and so it's a, it's a much more gradual kind of evolution, but it, but it's essentially the same picture. Yes. Do you, do you expect the same energy release mechanism? Um, the answer is yes, but it's going to have different visual manifestations related to how much energy it is. So if I have a ton of energy that I'm dissipating, a ton of stored magnetic energy, I'm going to get a lot of radiation, a lot of energetic particles, a big ass shock, you know, a lot of acceleration. The magnetic field is going to do exactly the same thing topologically during the eruption of a slow CME, but I didn't have those strong fields, and so I don't get the same magnitude of energy release. So I can erupt like a very slow CME at a very slow speed and have almost no energy release if I'm just kind of going along with the solar wind sort of speed. I mean, so I can inflate a streamer very slowly, and it will eventually pinch off like a streamer blob, but it will be very large, be a very, very large streamer blow up, and that's kind of what those slow coronal mass injections are. They're sort of in between the kind of regular small-scale solar wind junk outflow and these big active regions that are like huge flares below the whole streamer open. There's kind of a set of intermediate eruptions that are also large-scale that do not seem to have the flare energies, and those are the kind of CMEs that you typically see during solar minimum because there's not a lot of active regions; the sun's basically quiet dipole. There are still CMEs during solar minimum, but they're larger scale; they're kind of, they probably arise from longer-term evolutionary thriving. So the energy accumulation is slower in general, and the amount of volume you're storing that energy in is larger in general because you're basically only shearing the whole streamer belt. And so you have a kind of gradual, basically a very slow version of this, and this is just the very fast version of how you erupt the same thing from much lower down. Right, right, right. And in fact, anyway, there may be some slides on here where we go through some of that.

Okay. So if you're looking for what can cause that much energy release in the solar atmosphere, um, this was a very useful table, um, from this Forbes paper. So basically, you know, from the observations that you need something like 10 to the 32 ergs of kinetic energy, so that's the acceleration, the CME. You've got something like 10 to the 32 ergs in radiation and plasma heating; that's the flare emission. Um, you've got about 10 to the 31 ergs in terms of work against gravity, how much material you've erupted, and the volume of that material is basically like a, you know, solar radii cubed or something, um, is basically 10 to the 30. So if you take, take 10 to the 32 ergs and a volume of 10 to the 30 cm cubed, you get an energy density requirement for one of these big CMEs of something like 100 ergs per cubic centimeter. And if you look at various, um, energy sources or energy terms for, for kind of characteristic coronal values, um, you look at, you look at motions in the solar atmosphere. Okay, you've got a density of something like 10 to the N. Now, the photospheric velocities where you, the velocities where you see all those magnetic fields moving around at those speeds at the, at the photosphere are slow, so it's like a kilometer a second. This magnetic field evolution on the surface of the Sun is relatively slow compared to speeds associated with the eruption. So this, this kind of energy density, sort of this, this mass moving at this velocity is 10 to the 5. All right, you're not, you're not getting the energy from that.

Um, in terms of temperature, okay, we know the corona is a million degrees. Well, that's, that's a fair amount of energy, but the corona is large. So in terms of energy density, we're looking at 0.1. Uh, in terms of gravitational energy, okay, big coronal loops are kind of 10 to the 5 km. Okay, you solar gravity, you know, that works out to sort of 0.5 ergs per cubic centimeter. And the only thing that's got anywhere close to the kind of energy density you'd need to blow out a whole portion of the solar atmosphere is the magnetic field. So a kind of characteristic field strength of 100 gauss, which is actually pretty low; the sunspots are, you know, often kilogauss. Um, but even just taking 100 gauss means, you know, at the, the same kinds of coronal volumes, that's the thing that's got, that's got the energy. So the whole question about CMEs is it's a problem of magnetic energy storage and magnetic energy release, and what, why CMEs are impulsive is because the storage is slow; the time scale is a long accumulation of gradual energy, sort of, um, buildup and then a rapid, fast release. And so the difference in those time scales means the phenomena is as impulsive as it is. Um, and what magnetic fields look like on the sun. Well, here they, here's a, here's a magnetogram, um, for this same September 2017 source region, and this is in fact, uh, the active region 12673 that rotates to the limb and is then the source of that, uh, movie at the beginning. Um, and so what do we see here? We see, uh, very concentrated strong fields, sort of right up, right up against each other, and there's also some indication that there is intermixing of positive and negative polarity in various places. So that's going to create some structure in the overlying field. Um, but in general, this is the starting point; you need, you need strong concentrated field, and you need to do something to that field to energize it because I could take this exact magnetic configuration and say, um, let me derive, okay, this is one thing from last time that I should have explained more, but let me derive a potential field source surface. So what is that? That's a, that's a mathematical reconstruction of, um, the magnetic field in the corona based on these observations. So these are boundary conditions for that calculation, and so I could take this set of boundary conditions and create a magnetic field that is in fact unique to this set of boundary conditions, and that would be a field that has no excess energy in it because it would be a potential field. So there's no currents, there's no stress, there's no shear, there's no forces; it's just a potential field. Okay. So the what erupts is not a potential field. So if I start with that and I'm trying to create an eruption from that situation, I've got to then do something to those fields. Now, the, the sun does stuff to those fields all the time; these things emerge, they run, run into each other, um, you know, they get sheared, they get stressed, they move around, they diffuse. Um, so the sun has got no problem doing a bunch of stuff to these fields. Um, when we're trying to model these kind of things, we've got to approximate the, the motions that the sun is doing, uh, acting on these fields. So we, we either introduce our own kind of idealized shear to represent like differential rotation, um, or, you know, you can spin up one of these spots to represent the accumulation of twisted field. Um, often in, in flux emergence simulations where you see these, these sunspots come up, um, you'll get regions of very twisted field in between the two strong polarities. And so, um, anytime you basically take a field and put some energy into it, um, that's, that's what we do; you're going to get some kind of dynamic response of that field. Yeah. Okay. That's, that's a good question.

Um, so the distribution of CME energies is pretty similar to the distribution of flare energies. So the, um, the fact that flares, okay, and flare energies are, are typically done by looking at the emission of x-ray, um, and so you'll assign a numerical value like watts per square centimeters or whatever, um, and so X flares tend to be 10 to the 32 ergs, sort of M flares can be, you know, high 10 to the 31, C flares you're kind of 10 to the 31, 10 to the 30, and, and you move all the way down. Um, in terms of having, um, like a real CME eruption, if you make the flare too small, the, the amount of field that you're erupting isn't large enough or isn't strong enough or isn't energetic enough to make it through all of the overlying closed field up above it. So there's kind of a minimum threshold energy from like a detection standpoint in terms of, you know, creating big, big visible, you know, structures, plasma structures that you can image that you can see. Um, so there's sort of that, I mean, I, I think you could, you could if you were to believe that it's all kind of part of the same continuum, you could track CMEs from eruptive to non-eruptive to like jets, and you're just kind of changing the scale over the storage and release, um, kind of paradigm basically. So like tiny little jets are more or less the same picture, but the source region is tiny, the field strengths are tiny, they don't have huge CMEs that come out, but they have these sort of, it's more like the re-, the interchange reconnection and the solar wind type of, type of release. So you get this twisted, interacting field. What I want to do is spread that stress over as much volume as possible, and if you have one of these big active regions, the best way to do that is erupt the whole thing, but if you've got something that's like sufficiently asymmetric or kind of not enough flux, you can get away with just sort of opening up and releasing that stress in a, in a less kind of catastrophic way. Yep. Yeah. And so like, you know, uh, whatever an X flare is 10 times rarer than an M flare, and you know, then you can kind of just extrapolate it out, and so you get an X10 flare, an X100 flare, an X a thousand flare, X 10,000 flare, and you get some of these, um, very active cool stars like end dwarfs and stuff just flaring all the time with like 10 to the 34 and 10 to the 35 kind of eruptions, and you know, that's again, that's basically the same physics as solar CMEs, just scaled up.

Okay. So if we wanted to look at what the field structure, what a stressed field structure for a kind of active region like this or CME source region like this is, um, there's a couple different ways to do it. Um, observationally, you often see along these long polarity inversion lines, basically solar prominence or solar filament material. And so this is cool plasma, but why it's interesting and why these are related to CMEs is not because the plasma is cool, um, up above these polarity inversion lines; it's because the magnetic field is so stressed that these field lines are parallel to the magnetic polarity inversion line. So if we go back to our kind of cartoon from before, we just have a potential field, an arcade, and in order to energize it, one way of energizing it is to basically grab the foot points of that arcade on either side of the polarity inversion line and just stretch it, just put some shear into it. So I've added that much energy to my field, and now I've created a field line that comes in and out of the page. Well, those field lines look like this in 3D, so they're basically up above and stretched all along the polarity inversion line. And so these red and green dots here are actually magnetogram data, and so the red polarity is one sign of the radial field, and the green polarity is the other, and you can see that this stressed field lies right between positive and negative polarities. And so these, these polarity inversion lines in the photospheric magnetic field are, are sites of energy accumulation in the sun, um, for various reasons, but that's that basically always ends up being true. Um, and so it's this kind of stressed field structure that has the energy to power one of these eruptions. Um, and you can, I think in this particular paper, um, they tried to match some loop shapes with, uh, these field lines were from like a model, um, model flux rope that they tried to fit to, uh, this field distribution. So this kind of structure has the energy we need, um, and in fact, it's kind of a generic structure. Um, you get, you get a similar sort of configuration.

Okay. So this is the example of basically taking, an arcade and just stretching the field lines. So now you've got this kind of region of very stretched field lines along the polarity inversion line. Um, but if you look at active region fields, and this is sort of one of the first, uh, nonlinear force-free or linear force-free reconstructions of an active region field, um, you could say the same distribution of mag, you know, plus and minus magnetic flux here give me these potential loops, which would look like this, or if I created some current and did a different kind of magnetic field model, um, I could match the same dis-, vertical distribution with a horizontal distribution of field that looks like this. And so there's going to be a distribution of currents in this field configuration that changes the structure of the loops from kind of overlying potential field things to much more stressed, bent, sheared sigmoid-looking structures. Um, and so you'll see that this, you can imagine the polarity inversion line going basically somewhere in the middle here, and there's almost no component parallel to it, whereas over here, polarity inversion line, the field, the field up above it has a lot of component along it. So this is a stressed field, and this, this is the same kind of thing you get when you do these flux emergence simulations too. So this is like a very simplistic model of like an active region emerging. And so, yeah, if you have one of these kind of very twisted flux tubes beneath the photosphere, and part of it bubbles up, then you create these two sort of strong polarity regions, positive and negative, and part of this thing that kind of herniates up into the corona maintains this sheared structure along the polarity inversion line. So all of that shear and twist is what we're trying to get rid of in one of these CMEs. Um, and if you do kind of global relaxation kind of things, you, you get these structures basically on every polarity inversion line on your magnetic field model. So, um, so what does that mean in terms of, okay. So, so there is a, there's also, well, there's kind of a kinetic energy floor about how much energy you need to kind of get through the, get through the overlying field. There's also a magnetic energy ceiling associated with the global topology of your configuration. So, um, on the left here we have basically this simplest and the second simplest kind of, uh, magnetic complexity in terms of a closed field configuration that I'm going to put some energy into. So in a bipolar configuration like this, this is the standard helmet streamer. So here red and green show the polarities of B, so you can see it's, and I have open field of one polarity on one side and open field of another polarity on the other side. So this is essentially the, the same kind of configuration as basically a dipole. And so we're going to approximate the sort of realistic source region with an unrealistic source region, um, but at least we'll understand the physics that's going on here. So if I take this configuration of just a pure dipole and it's axisymmetric, it goes all the way around; there's only one open state possibility. And so what do I mean by open state? I'm going to say the as much, okay, this is a potential field. If you want to answer the question, how much energy could I possibly put into this magnetic field, the answer is I open every field line to infinity; I just open all of the closed field lines. And so that's this configuration here; it's all open at the top, it's all open at the bottom. And so here, because you've got field going the opposite way, you've now got a heliospheric current sheet stretched all the way out to infinity. So this configuration is the most magnetic energy I could put into this system. So that's a convenient thing if you're worried about how big is your CME going to be. Okay, I can do that calculation for simple configurations; real 3D configurations, uh, that's a little bit more complicated; there's a roughly analogous version, but it's tricky. So that's why it's useful to kind of go through the simple symmetric cases. Now, if I make it, if my make my source region a little bit more complicated, I've now got a bunch of these separator lines, and I've got instead of one polarity inversion line smack in the middle of my arcade, I've actually got three. So here's a multipolar flux system, um, and the thing that I get out of making my source region configuration just a little bit more complicated is now I have two options for my maximally open state or like how much energy can I put into this system. So my two different options are there's one option that looks a lot like this where I try to drag as much of the field out to infinity as possible. And so if I start with this, I'm going to say I'm going to grab all of the red flux out here and all of this blue flux and stretch it to infinity. So that's basically if I erupted this central neutral line and opened all the way down to it, I would have to open all of this blue flux and all of this red flux. Um, and so these quarter panels are mirrored; the southern hemisphere looks exactly the same as the northern hemisphere. Um, but this upper panel here, what I'm calling maximally open is the multipolar version of this. So now I've got oppositely directed fields coming out of this side arcade up here. So this was a closed flux system here, and now I've opened the overlying and inner arcade. And so all of this field that was here is still here; it's closed. This separatrix here has become deformed, but it still exists. Now that separatrix is here, and so, so this is dividing closed field from the open field. I've now got a current sheet up here, and I've got another one down here, sort of exactly equivalent. Um, so this is, this is a higher amount of energy than the other version. So the other version, the other option that I have here is instead of grabbing every one of these red field lines and every one of these blue field lines and dragging them all out to, what if I allowed some reconnection to happen? What if I allowed this red and blue field to reconnect and become field that lives in this arcade or field that lives in this arcade? If I do that, I don't have to open this flux; I can just move it out of the way. And if I can move it out of the way, that took a

Lot less energy than dragging those field out to Infinity. So I've got another open state, which I'm calling maximally closed. Why? Because I've moved almost all of the blue and all of the red flux into this green system. So now I've got basically a dipole on this upper half with like one single open field moment at the equator. And then, of course, there's going to be another closed closed field dle kind of down at the bottom. So I've got a single field line that I've managed to drag out to Infinity here, but all the rest of the field is just hanging around. So this is also a higher energy State than before I put any stress into the system, but it's much less energy than this state because there's fewer open field lines and there's less current in the whole system. There's kind of one sh instead of basically a sheet here, a sheet there, and a sheet down there. So that gives me the opportunity to build energy up towards basically store very slowly energy up towards this open State. And as long as I um am careful about how I release that energy, I've given my system the opportunity to go from this very unfavorable more energetic state to this much more favorable less energetic state by reconnecting and closing back down some of these open field LS.

In the bipolar case—now this is a sheared arcade simulation from 25 years ago—but in the bipolar case, you can build bu up this energy and you get to something that's basically very close to the open field energy. It's not quite there, but it's pretty close, and you've built up this huge current sheet as you've expanded all of that that closed field with dis energization. So I basically sheared the hell out of my field, of my polarity inversion line here, inflated all of this field, created a current sheet, and once that current sheet reconnects, I can then liberate all of that sheared field and drag it out to infinity. And so then my stored magnetic energy drops really rapidly, my kinetic energy Rises really rapidly, and I turn all of this stored magnetic energy into kinetic energy of the eruption. And if I had enough physics in my model, it would be radiation and particle acceleration in kind of all of that. So it's it's a turned stored magnetic energy into CME kinetic energy and flare emission. So this is the bipolar case in the in the multipolar case, it's exactly the same thing, but now we've got these two options. So I build up energy close to this state, but my system allows reconnection, and so I'm going to I'm going to go from up here to down over here somewhere. And so there's a kind of Maximum, you know, this is a sort of uh this is the minimum energy that I would have needed to get a CME out of this, and I can go up above it no problem, and reconnection will then let me go from a higher energy state to a lower energy State.

And so here's the simulation of this where I've got basically two current sheets in this particular configuration, one to help move the flux out of the way and then the second current sheet to basically help get rid of all of that Shear that I had put into the central arcade. And so again, you see the energy magnetic energy drops very rapidly and the kinetic energy Rises very rapidly. And so this is I would argue basically the story of every CME known to mankind. Yeah, okay. Um, ideal mhd versus resistive mhd. And so resistive mhd will have terms that allow magnetic reconnection and dissipation. Ideal mhd does not have those terms. And so in theory, you should never have reconnection if you're solving all of those ideal mhd equations exactly. The problem is—and it's not a problem, it's also an advantage—when you do it numerically, even if you're solving the equations of Ideal mhd, there's some numerics that is just sitting at the end of your equation causing basically the same effect as resistive terms. And so you can okay, that's called numerical resistivity. And if you were a you know, analytic reconnection person, you'd always be trying to get rid of it, but if you're a simulation person, you like it because it gives you an eruption and it creates the topology change you need. Um, and it means that for most of your simulation anytime you're not near a bunch of strong current your system is evolving ideally. So it's the ideal mhd Evolution and then only when you have a bunch of really strong current and the system is like force influx systems against each other and you've got those kind of separatrices being battered back and forth or kind of evolving in time, um, then all of a sudden you get the sort of nonideal uh component to the evolution sort of right around those current sheets. And that means that there's you know, that can create some um tremendously impressive plasma Dynamics because you get these big bends in the field lines and you have strong magnetic fields down here and large alphane speeds. And so when you've got reconnection Jets and sort of dynamic ganic response to the the reconfiguration of the magnetic field, I mean you get kind of massive plasma acceleration and big bursty jets and everything. Um, so yep, any other questions on this? Okay.

If I'm looking if I'm going to zoom in to basically what's going on down here, this is the solar flare part. Up above is the CME part that erupts into space. Down here is the solar flare part. And so what does the solar flare look like? Well, okay, here is an example, and here's the kind of 30y old cartoon or 50y old cartoon by now, um, but say I'm on I am an observer here, and I'm looking this Direction. All of these bright Loops are flaring arcade Loops. The Contours here are again positive and negative magnetic polarity. You'll notice that kind of right in between the positive and negative polarity is basically the polarity inversion line that's here. All these all these Loops are going over from one side to the other. So basically this down here is your flare arcade. There are ribbons which are sort of the bright edges of the flare arcade, and then up above that somewhere is your eruptive flare current sheet. So it's kind of up above this arcade. All of these Loops here are basically Loops that reconnected and got drawn down and energized and basically a bunch of um particles and flow characteristics kind of come down here get dumped into the denser of the layers have like particle beams coming out of the reconnection region. They hit the chromospheric plasma, they evaporate the chromospheric plasma and basically heat the whole set of arcade loops. And so you get this bright emission here, and of course, you can sit down and figure out kind of exactly what emission happens with um, and so this is the kind of um, say very fancy version of uh, the traditional 2 and a 12d kind of CSH KP sort of eruptive flare picture. So you have this big current sheet, you have the flare region down below, the CME region up above, and down below there's a ton of structure in this reconnected downflow. And so this was like the the comment about sads earlier. Um, so this reconnection that's happening in the flare current sheet has a ton of structure on it. You can see these field lines coming up coming down. This is density in the color in the sort of yellow to purple color scale here, and so you can see there's like these voids and these flux tubes being reconnected down um creating a ton of structure basically right in this cusp region above flare loops. And so all the interaction between the plasma and the field and the particles sort of up around here corresponds to a bunch of interesting plasma stuff that ends up in emission of various types. And so you can one of the ways they try to keep track of energy is by basically adding up kind of all the energy associated with x-rays with the flow speeds kind of the euv radiation the H Alpha radiation um you know, you you name it um you try to add all that up and estimate the total energy that you must have released over this eruption.

So here's an example of you know, how well this 2D cartoon explains observations. So this is that September 1777 flayer in a bunch of euv wavelengths, and this is a great example because it basically looks exactly like the cartoon, which means we kind of have an idea of what's going on. Um, these are all at different temperatures, and so you can see things like this sheer core of the erupting CME is both hot and it's kind of cold, and there's sort of Loops that are going up around it, and you can see a little bit of the flare starting here, and this is this is high temperature emission here. So this is a couple million Kelvin. Um, so this reconnection here is going to turn this bubble into an erupting flux roow. And in fact, you can sit here and you know, do some fancy Imaging with radio and back out some plasma properties and you see that like as this thing evolves you basically have the whole eruption up here, you have the vertical current sheet sitting here, and you have the flare arcade being built up underneath it, and you can see that these Loops Rising. So the reconnection moves up as the eruption moves up. And if you play this movie co and yeah, if we can play this movie, this is what a this is what a low energy version of that same eruption would look like. So again, it's magnetically exactly the same. So this is the slow rise phase as we're kind of energizing this arcade, this helmet streamer slowly. We've stretched it out enough, and this continued expansion drives the vertical current sheet formation, and reconnection allows all of that stress field to escape. So can we play that again? And so the colors on the field line are basically red is sort of the overlying unstressed helmet helmet streamer field, yellow is kind of the outer layers of the closed flux, and white is basically the filament or the pro you know, the large scale magnetic prominent. So this has got the The Twist on it, and there's even more twist once reconnection starts. So there there it is, and now I've ejected most of my stressed field, and the remaining field closes back down. And you'll notice that the size of this streamer afterwards is sort of comparable to the size of the streamer before we dumped a bunch of energy into it. So if you can play this just one more time, we'll be so now I'm inflating it, and I'm over the for this particular simulation this was a energization of like roughly equivalent to like two weeks of solar minimum evolution. And so after a while, all of that stress field is decided it's now more energetically favorable for me to to leave this thing blows out the material and closes back down.

Okay, so at solar maximum, it can be really quick because you can be looking kind of at this nice quiet patch of quiet sun. There's kind of no strong fields, and all of a sudden boom massive active region shows up. You know, you go from a situation where it's kind of salt and pepper plus or minus five gals to plus or minus 2,000, and all of that structure that you kind of brought up from the bodus spere has got a is introduced into the corona with a bunch of that twist and energy and currents kind of on it already. And so active region can can go from like not existing to existing and blowing off X x-class flares in like two days. I mean, you just sort of yeah yeah, that is a good question. I have some thoughts on it, and there are some slides on it too a little bit later on, um, but yes, in we've been talking about CMEs and kind of this 2D picture, but you're right, it's really 3D. And in 3D these field lines are connected at the sun on both ends. And so even an eruption like this, um, the field that erupted is basically you can you can kind of see it here. This this sort of j-shaped hook here is one foot point of the erupting flux rope structure, and this kind of J shaped hook over here is the other foot point, and you can see the flare is happening in between and beneath the two foot points is go. So so in this picture, you'd say this is kind of one leg and this is the other leg, and you can just see from the flare Evolution okay over here just a couple hours later than here. I don't you know, my these dim regions aren't in the same place, so those foot points have moved around a bit. Um, and that's also its own kind of interesting set of things, but but you'd sort of say okay well here it's still dark here and still dark here, so maybe the you know maybe the foot points are kind of shifted a little bit to to being kind of here and here, um, but you're right, even many days after that they keep moving around, and and so sometimes they can um, depending on how you rebuild the closed flux underneath the eruption, you can eventually end up being connected pretty far away from where the source region was kind of at the nearest coronal holes or whatever. So you can both maintain a connection at both foot points, and also it is very common that one of those legs is reconnected up, and so that then you have what looks like a whole big CME, but you no longer have that connectivity. Um, and that happens basically all the time too. Um, so the thing that I think I wanted to say here was um, so reconnection is important. Reconnection in CMEs is important. It would be nice to know how much magnetic field was coming at you when the CM erupted. Um, now of the things that you can do to to calculate that okay you don't have you don't have measurements of the corono magnetic field, which is unfortunate, but you do have measurements of the photospheric magnetic field, and you know from this kind of picture that all of the reconnection that happens at this vertical current sheet which you can't see because it's up above these images somewhere, um all of this reconnection topologically Works to move these ribbons around. So the evolution from this image to this image is create more and more of these flare Loops under the reconnection region and just kind of build up the posteruption arcade or say post post flare arcade. Um, the outer boundaries of that arcade tell you the size of this flux system. And so if you watch the arcade grow in time, you can just basically measure how much flux the ribbon has swept over. And in ideal mhd and especially in this 2D sort of picture, um however much flux this separatrix in cases must be the flux you started with plus all of the flux you processed for reconnection. So by measuring the growth of this flare arcade, you've got some estimate of how much reconnection happened up in the corona during this flare, even though I can't measure the coronal fields. And the reason why you care about that is because how strong the fields are is going to be directly related to the flux content of the of the eruption. So how strong the fields are that you measure at spacecraft or how strong the fields are in the CME that runs into your planet or kind of whatever is is going to be related to to these field strengths. Um, and so what what what a real flares look like uh when you do this exercise. So here's a giant active region 11429. Um, so here's a massive amount of strong field. Um, there's a ton of structure along this neutral line in between the positive and negative polarities. So all of the energy is kind of concentrated here before the eruption. Um, this kind of map the colors here is basically a distance map of the flare basically the area swept by the flare ribbons. So the green the green Contours are like over the whole flare just integrate the whole flare and basically every pixel that gets lit up during the flare observation gets colored at the time that it became bright. And so what what this kind of image of a flare means is the blue pixels were the first reconnected flux areas, and then as you go in Rainbow that that's time Evolution. So the flare ribbons start very close the poity inversion line, they kind of rapidly form all the way along the poity immersion line, and then they slowly sweep away. So there's kind of two stages of reconnection. There's a very very rapid like I've picked the entire length of the filament that I've decided to erupt or the my entire flux rope or my entire stressed field region like this. So I rapidly extend parallel to the polarity inversion line, and then I less rapidly broaden like this. And so that's kind of like the initial impulsive phase. And as this whole thing erupts, I'm accelerating the CME the entire time that I'm doing this reconnection. And so you can see the kind of the structure here. And if I count the flux units of the magnetic field and area, I can get a number at the end of the day for basically how much flux um happened while this flare was going off. And okay, it's uncertain, and there's some you know, depending on where you put your thresholds you can get some error bars, but it's more or less a reasonable number um a lot of the time. And so okay if we can play this movie first and then that movie second, this is the image D version of that same Evolution. You guys want to okay. So that's the eruption up here, and this is the flare down here. If you can just play this one a couple of times. Um, these are the flare Loop, and you can see the little yellow dots right here are the instantaneous flare ribbons. Um, and okay if you have one of these simulations, you can create some synthetic observables, and you can kind of see the flare Loops in emission um following the the kind of the overall shape of the arcade development here. You can see that kind of in the middle you've got this sheared field, you've got these flare arcade Loops coming down here, and if you keep track of that flux that's what's in blue is that total reconnected flux. You take a derivative of that and get a reconnection rate, and that's in Gray. Um, and then the black and the red lines are the magnetic and kinetic energy of the whole system as well. And so you can see the the arcade shape of the flare arcade down here. The connection is happening up above. Here are the foot points of the CME on either side, and this reconnection is allowing all of this stress field to just erupt. You can see there's the drop in magnetic energy along with the increase in kinetic energy at about the same time as the reconnection uh total total reconnected flux comes up like this, and the derivative of that is basically the reconnection rate. So this this reconnection rate peques at about the same time as the kind of main acceleration part. And why I say that is because you could say here the kinetic energy of the whole system is going from you know, a slope like this to a slope like this. So you're really erupted right here. Okay. So so if we make the if we make the source region just a little bit more complicated, um here is here is basically our pseudo streamer kind of configuration, and here we're just adding a ton of energy to that same to that same situation. So we're putting all of the close in this close flux region. I'm just dumping a bunch of energy and watching on the right here I'm pling current density. Um, okay, so this is going to be velocity. So once I once I expand these side loes, I create a situation that is sufficiently stressed. I this myself I get a situation that's sufficiently stressed that it wants to just get rid of all of the shear I've put in one half that pseudo streamer and then the other half. And so depending on where I put my energy in, I can have a whole complex set of eruptions that are coupled to each other and are related to what else went on. Um, and so over here is velocity. You can see there's kind of a gradual uh um slowrise phasee, and now I've got reconnection moving flux around, and now I've got these big reconnection jits um during the eruption. And so just in case those movies were too fast, which they certainly were, um here's it here's a broken down into into bits and pieces. So there's kind of four things happening here um in terms of reconnection. So I've got first this is the same kind of Pudo streamer reconnection we talked about in the slow solar wind case, but now it's being driven by the fact that you've got something like a CME trying to go off under your under your closed flux system. So I've created a current sheet. I've got a lot of little plasmoids in that reconnection region, and eventually this expanding field builds up a vertical current

Sheet here, and you get your typical flare reconnection. You create a flux R up here; here you create a flare arcade down here. And this allows, basically, all of this twist, all of this stressed energy in the right arcade to erupt during the flare reconnection.

What's happening here is the flux on either side of that current sheet is being swept into the current sheet. And so when I'm doing that, I'm taking overline flux in this adjacent arcade—basically the other half of the pseudo streamer—and I'm eroding it. So I'm erupting this side, and all of this field that was just sitting here keeping this stressed field down now has an opportunity to move out of the way and become part of this eruption, and so it does.

After this half of the pseudo streamer erupts, the continued flare reconnection that's rebuilding what's left of the right-hand side lobe of the pseudo streamer arcade, um, basically acts as the facilitating agent for moving the restraining flux out of the way of the left-hand side. And so you get a second eruption in response to the first eruption.

Now, this only happens because, okay, I was in charge of the simulation, so I put energy into both halves. Had I not put an energy into this half, I would have just had the first eruption, and it would have been exactly like we looked at in the other case. But if I've got, if I've got a complicated region, I can then start stacking these things kind of closer and closer together. And so the, the same overall picture is true in that I build, I sort of slowly build up my magnetic energy. Um, this yellow part is that's kind of slow R phase, and then these blue parts are the rapid acceleration phases. So there's kind of this slow-rise phase and then the first massive eruption here. And so that means my separatrix is moving around first slowly and then rapidly. I'm rebuilding the, the flux system under, under the, under the first eruption here. And then when the second eruption goes, I'm rebuilding that flux system as well. And so if I keep track of the magnetic flux, I can watch it transfer out of one system into the other and then rapidly transfer back. And then these are the fluxes associated with the flare arcades, this one coming up for the first eruption and this one coming up from the second eruption. I, you could notice that at the end the flux is not that different than what I started with, but the fact that I had to move it all around meant I could liberate a whole bunch of that extra energy that I put in.

And so here's the reconnection rate; it's kind of low at the slow-rise phase. So that's reconnection happening in the weak field up above this, and then the two flare current sheets going off one after another have a ton of reconnection because they're trying to, they're trying to get rid of all the stress field. So I can take the same sort of cartoon that I had before and get something pretty complicated out of it, but I've really just taken the same thing that I knew what it was and done it, you know, put one right next to each other.

Now the Sun does this at all scales all the time. So during solar maximum, there's stuff like this happening all over the place. Um, this is another, this another version of the same movie, um, this one with a different color scale so that you can see more of these blobs coming out. And in fact, if you're really interested about reconnection, you can zoom in on these, on these current sheets, and you see them getting stretched out, and then they start to break and form all of the substructure in the reconnection region. And so this is the first overlying current sheet that's going on; you can see a ton of these island form; this is all a bunch of neat structure. Um, here the inflow basically sort of down and up into the current layer. And so this is like Y velocity, and this is X velocity—basically the reconnection jets going out one side and the other side. And okay, so that was the breakout sheet; this was the first flare current sheet like that, and this is the second flare current sheet over here. And so in every one of these, you're, you're creating a ton of structure, but all of this works to take the stressed energy from down here and erupted out.

And there's some interesting implications and consequences for things like how much structure you generate in your current sheet and when and why do you get this sort of substructured. Um, but it turns out, at least in the framework of MHD, um, there's a, there's a kind of over, uh, over-specified system in that all of these current sheets are responding to the global evolution of the atmosphere. So basically, how long the sheet gets and how much flux the sheet is trying to process through reconnection is determined by the whole kind of topology and the configuration and how much overall energy I put into the system before. But there's some geometry constraints here in that this current layer is very long and so has a bunch of inflow area; all the inflow is sort of up here and down here, and the outflow area is like tiny, tiny on either side. So you've got your system that's trying to process a certain amount of flux; there's, there's a lot of room to push it into the current sheet region, but there's only a tiny amount of room to get it out. And so the system has to figure out how to take all of this flux and all of this mass and get it through those tiny areas. And the only way an ideal MHD you can figure out how to do that is by breaking up the current sheet into lots of tiny little parts because now between every one of these blobs I've got another X line. So I can, in fact, turn sort of one, sort of one reconnection jet direction one way and one reconnection jet direction the other way into like 17 X points in a row. So I can actually reconnect a whole bunch of this flux, and I spit it out in these plasmoid structures. And so that's the way this system can figure out how to kind of resolve the contradiction of I have a global set of constraints over how much flux I've got to, got to, I've got to push through this current sheet, but I don't have enough physics in my model to kind of energize the particles in the way the Sun has figured out how to do it. So I've got to break my large structure into a whole bunch of substructure in order to get the mass flux and get the energy flux to get the momentum flux through this current sheet. Um, and that I think is a pretty universal property of reconnection. Five minutes, perfect. Okay.

And so in fact, we did this; you can, you can count and you can measure all the flux and measure the mass and get a kind of a reconnection rate sort of of stuff spitting out of these ends of the current sheets, um, and it all hangs together pretty well; seems to pretty good. Okay, so that was 2D; in 3D stuff gets a little bit more complicated. So here's some what I want to do with this; this is, this is a simulation of an eruption, basically, of the bipolar type where we kind of just, so it was this, it's, it's sort of similar to that earlier movie, um, but here we've, we've got an observed or sort of a more complicated magnetic flux distribution. So here I am sort of stressing some field reconnecting and erupting. And so this is my kind of prominence field here; I get rid of most of that stress, and there's some leftover kind of low-line stuff in the wake of the eruption. Um, and this is what it looks like in, okay, so I've colored, this is the flare reconnection pattern that's happening here on, on the surface of the simulation, and this is synthetic EUV emission, which is basically, in my case, I've just squared the density and done line-of-sight integrals. So, so this contour here is the extended polarity inversion line, and you can see when the flare happens, I'm now filling the arcade sort of starting from the left and progressing to the right. And so the, if, if you make one of these filament channels long enough, it doesn't just erupt all at once; it erupts on one end and sort of zips along. Um, so that, that looks something, and you can do this; you can take these flare calculations and basically calculate the flux and you, and you come up with some number, and you're like, all right, well, that's the magnetic flux that I had reconnect in my simulation. And if you're doing one of these real events, you can kind of do the same thing where you take a contour of the flare ribbon down here, and that's in red, and then I can plot it with these colors down here, and I can say, well, how close to the real reconnection estimate with my simulation? And okay, if I take a little box around my simulation like this, this orange, this orange rectangle, and I say I'm going to count all the reconnection flux in this orange rectangle because that orange rectangle is about the size of my observed flare, um, I actually have embarrassingly good agreement here, um, which was also kind of surprising, but I think it's sort of because this is generally a weak field region, so I didn't have to get any sunspots correct; I could just take a kind of a generic large source region here. But if I do that and I count up the flux, my modeling and the observations are, are pretty reasonable, um, in terms of observed flux in the flare and flux that ended up in the CME. Okay.

And I can do things like, okay, I guess we could play this movie. Um, so this is fake coronograph images of that same eruption that I showed earlier, which played a couple times, um, and you can see it erupts from the left first and then progresses over and erupts from the right later. Um, and this was an interesting high-latitude filament case because it kind of looked like there were two separate eruptions, but we saw from the field lines that it was really a single thing, kind of the size of the Sun erupting from the left and progressing all the way to the right. Um, so you can do things like compare the velocities, um, the acceleration, and you get sort of reasonable things, uh, this, but if you measure, if you, you know, if you, if you then take a look, so that particular CME did in fact hit spacecraft at L1, and I'm not sure what happened in the resolution here, but this, this region right here in the middle is the CME observed by, um, I think Wind at L1, um, and in fact, it was also seen by ACE, and, and this interval here is where the field is strong is where there's kind of a coherent, uh, rotation here are the magnetic field angles. This pink magenta thing is basically a flux rope fit to the ejecta. Um, this is velocity, density, temperature, and so my flux rope fit looks something like this. And so my spacecraft passing through this CME kind of looks like the very first fields were pointed up, and then they evolved the pointing down, right, or so, okay, in time they started down, and they evolved pointing up. So they started pointing down and evolved, and if you look at composition during the same period, um, just like the way we used composition to think about the solar wind charge states, it turns out that in these CMEs, often the whole interval that's associated with the magnetic ejecta where the fields are strong and you have these associated rotations, you both have kind of strange electron signatures—basically bidirectional electrons—saying probably your CME field lines are connected at both ends back at the Sun, but you've got all these composition enhancements as well: hot oxygen, moderate iron. Um, and then in this particular, uh, CME, because there was this period of southward field at the beginning, you even in fact had a reasonably, um, moderate storm for this, for this particular event. And so that's kind of a weird, that's kind of a weird thing for a, for an eruption that happened so far south. Okay.

So here's, here's an example of, of a coronograph observations and sort of some flux rope fits to that same eruption, and you can see the flux ropes are mostly below the equator, but we, we had spacecraft in the equatorial plane measure this CME, so something happened between here and when we measured to basically bring the CME up above the equator. And so what happened, of course, was it expanded, and, you know, colleague Jean ran a whole bunch of simulations where we tried to, uh, basically make this eruption look like the observations by taking it all the way out to 1 AU. And depending on the different CME models, um, we use to mimic the magnetic field rotation, you know, we got better or worse agreement. So again, to apologize for the resolution here, but this was a reasonable agreement, believe it or not. Okay.

So you could do the whole thing again with, um, you know, with basically the entire streamer belt, um, which we did this a couple years ago, and we said, okay, if, if you took all of the closed flux on the Sun and energized it all and then erupted it all, how big of a CME is that? And okay, you might say, well, this was a, you know, kind of fun exercise to do, and, and, and I can, I can calculate my, you know, my reconnection flux, basically, here's the entire streamer belt all the way around the Sun, here's my, um, arcade growth, let, basically recreate the entire streamer belt. Um, I can look at, you know, plot them, plot my numbers up here; my total reconnection flux is something like 2 times 10 to the 23 Maxwells. You say, okay, well, that's fine; that's a reasonable number, maybe. Um, how does that compare to other observed solar flares, right? So here's all the observed solar flares down here in gray, and then here's my awesome exploded every closed flux surface in the simulation up over here. So this is like an X60 flare, and you might say that's completely ridiculous; our Sun doesn't do that; we've never seen an X60 flare, which is true, but this is the exact kind of thing that we see in stellar flares all the time because the stellar flares have to be so big that we measure just the one, two, three, half a dozen photons from these things on other stars. Well, okay, once it's, you know, once we're sort of Harrington-class eruption, then yeah, this is, this is kind of what we're seeing, you know, in some of this exoplanetary habitability kind of questions is, is, is, you know, we know how we do with space weather, but if you, if you had a star that was spitting off X60 and X100 CMEs all, all the damn time, um, that's quite a different magnetic environment. Um, and so again, this is, okay, I don't, I don't exactly have time to go through all this, but you know, you can watch one of these massive eruptions like this and say, okay, what's the magnetic field structure of something like that, that, and I can fly synthetic spacecraft through all different places of my simulation for, for the fun of it. Um, and okay, what's kind of cute about this is these distances at sort of 10, 20, 30 solar radii are kind of what we're actually measuring now with Parker Solar Probe. So even though I've got this kind of bizarre simulation, I've got kind of realistic trajectories through those bizarre magnetic fields, and it would be interesting to see if the structure observed by spacecraft look anything like real magnetic, real magnetic clouds or real CMEs. And so you can do this exercise, and you can fit all of your favorite minimum energy flux rope models, the kind of twisted flux tube structures of these things, and at the end of the day, you know, you calculate twists and you calculate fluxes and you calculate kind of what you expect the reconnection, um, to be from, from the flare and what you measure the reconnection to be from the CME, and they're sort of related, not totally, but, you know, they're not too bad. Um, and so in fact, you do actually have some of this, some of this structure in space, in the, in the spacecraft observations. And so I, I guess for the last like two seconds of time here, we're going to go through one more example of a pseudo streamer eruption where this time we've got one of the legs disconnected. So if you look at these field lines here, I'm erupting here; got my slow rise, my rapid flare reconnection, and now, I mean, if you, if you're paying attention, where, where do these yellow field lines go? So they erupt once you got here, and they become this twisted thing, and now where, where are they? All right, so what's, what's going on there? Well, this thing over here is going on there. So if you've got this thing erupting in a pseudo streamer, um, okay, so I've got my erupting Slinky, the field of that erupting structure; there's an axial component and a twist component. The twist component is what you got from that reconnection; the axial component was like the spine of that sheared field structure that erupted. Um, and now I've got this twisted loop, and if I imagine this twisted loop rising in a unipolar magnetic field, field, so in open fields on either side that's all pointing the same direction, that's going to line up with the direction of one side of my CME but be the opposite on the other side. So if I've all got my open field pointing out and my erupting CME, I've got one leg of it pointing out in the same direction as the open field, but the other leg points inward. And anytime you have opposite field, there's a current layer in between. And so that's what this kind of blue isosurface is here is a whole bunch of current in between one leg of the CME erupting, and these blue field lines are like the open, adjacent open field. And so as this thing expands, it bubbles out, and it runs, you know, it's reverse; it's one leg interacts the opposite direction at open field here, and you get some reconnection that is different from the flare reconnection is happening here. This reconnection is just disconnecting one leg of the CME. And so then you've got this like super bizarre, like what is I, this is a question mark or something—the twisted fluxer question mark with a bunch of like big giant kinks in here. So if I flew through this thing like this, I would still see a flux rope, but my electrons would say I'm on open field. And if I flip through something like this, I would see a big change in BR. Um, so this is like, in some ways, the CME version of a solar wind switchback, but that's details on that later. So that's what this simulation looks like in EUV over here, here. Um, there's a ton of structure, and there's a ton of interaction between this erupting field that is in white, so this is the closed flux here, and the open field in yellow, and you see all of this complex structure here on the left-hand side is that leg disconnected. And if you look at, okay, so this is, this is information you got from the magnetic field; these are the same separatrix surfaces, um, that we talked about in terms of mapping the boundaries. And so you can, you can topologically do a whole bunch of very complicated stuff with looking at how the flux moves around between different systems, um, how that's related to the energy release. But okay, so over here, I'm looking at this whole eruption from the pole, so my pseudo streamer is like this, then over here I'm looking at it from the side, so it looks a little bit narrower this way and, and sort of broader, or there's a whole bunch of this kind of outflow at the end here, and that's, that's related to, you know, A, the system rebuilding the closed flux in wake of the eruption, and B, it's also related to continued evolution of the open field. Um, so this is what the, this is what that same eruption looks like in, in sort of fake coronograph observations—again, a ton of structure, even though this is a, this is an isolated idealized pseudo streamer, and all I've done was just put some energy into that closed flux system, and I've generated a ton of structure because that's, you know, what space plasmas do evidently. Um, and in fact, you can even kind of fly through these simulations and make fake PSP, Whisper kind of white light image in, and you start to see a whole bunch of this complex 3D structure as your, as your spacecraft is like kind of in and between all of this, all of this fun stuff. And the good thing about simulations is you can take your Parker Solar Probe trajectory and just kind of move it around and sample different parts of your simulation. So here's, here's a version where we run

Right into part of that CM at the bottom here. So this is kind of a flanking counter. You can see the white light signal just running into the ejecta up here. And this is the kind of in-situ, um, time series that, you know, my fake Parker would measure flying through this simulation. Um, here's a here's kind of an even more impactful, shorter duration trajectory where basically you go right through the middle of that strong field region. And you can kind of see there's a bunch of field lines drawn basically from the spacecraft position. And you can kind of see the difference in connectivity right when you're in the middle of the ejecta; those field lines start, and they jump, and they kind of flap around, jump back.

Um, so for anyone who's like interested in looking at CMEs and spacecraft data, they're always this complicated, and in fact, even more complicated than this, but I can at least understand what happened here because I made the simulation. So that's progress, and that's I don't know, there's some more stuff here. We're we're we're it's noon. Okay, so I I will finish in 30 seconds by saying, once you've gotten all of your flux ropes into space, you get to like figure out what else happened. So between the Sun and where you measure, there can be things like rotation, deflection, CMEs running into each other. Um, it's a mess, but that's why you guys are here to solve all of these problems. So there's some stuff here, and there's some stuff here. Basically, let's let's check. I believe I uploaded them correctly, so yes.

Why do we need a box? I wanted to ask for the um, sympathetic CME simulation. I think the first one you showed, um, how in the simulation do you inject the energy to sort of initiate it? Yeah, so so I I do it the same way I do in all of my simulations, is is basically just apply shear to the lower boundary. And so in the get back or not, but okay, so like, for example, here in 3D, um, you would inject shear by basically putting a a sense of rotation in one direction on one side of the polarity inversion line in the other. So you're basically doing this in a circle. The previous, uh, axisymmetric simulation in Cartesian is basically infinitely long out of the page, and so it's done the same way, but it's so two and a half FD is like we maintain all of the components of every vector, but they only vary in two of the three directions. And so, you know, I can kind of create a B Fe component, which is then my shear, which is then what the system tries to get rid of.

Um, but you don't have to do it that, and in fact, um, a bunch of the stuff I showed is more generic and universal than is associated with that particular form of energization because you could introduce the energy a number of different ways; it would all still look like this. Um, so whether you brought it up from the interior slowly or, you know, uh, did a some combination of shearing and reconnecting, you know, flux cancellation or, you know, kind of you name it, but sort of any CME model for introducing energy will have um, slightly different details, but generically more or less the same; you know, you're inflating the arcade via some mechanism. Okay.

Um, so where do you think things are going wrong in terms of the M? Do you think it's a lack of observation, so we can't constrain the models, or is it just a physical knowledge is not there yet? Um, it's it's I think it's a little bit all of the above. I mean, there's a there's a way in which the models are much better than they have any right to be, um, which means we're doing well, um, but it's not I mean, the Sun has infinite resolution, and we don't. And that's kind of the, you know, the Sun has no problem being stuck in one regime, and we do. So it can figure out how to switch between fluid and kinetics and where to dump all the energy into particles and waves and kind of, and we have a limited version of that. Despite all of those limitations, we got a bunch of the large-scale stuff right in a in a way that's pretty impressive. I mean, so I I I once had a a colleague who was writing her first of a proposal on solar flares, and and she started out the proposal, we've studied solar flares for 50 years, and there's still so much stuff we don't understand. We kind of don't understand anything. I'm like, you have to rewrite that first sentence. You cannot say we've studied this problem for 50 years and didn't make any progress, which is why you need to give me some more money to study it. Say, no, no, no, we've made a ton of progress; we just haven't solved it yet. So here are the outstanding issues. And that's it's kind of both. I mean, it's it's both; we have made a lot of progress, and there's a lot of a lot of work to be done.

Hey, um, can you tell me how do you produce the EUV emission in that simulation? Yeah, um, okay, so there's there's two there's basically two aspects of it. Um, one is okay, so so in general, both the EUV and the white light are are line-of-sight integrals where you add up contribution at a number of points along your line of sight. Um, for white light, there's a very convenient sort of scattering formalism that was worked out 50 years ago or whatever, um, and so your your your white light intensity from scattered electrons or white scattering off electrons, um, has a well-known geometric kind of configuration, and it means that your signal at the end of the day is proportional to density, line-of-sight integrated density. For EUV is proportional to line-of-sight integrated density squared because EUV is emission line stuff. So you've got basically as much material as emitting, and then you need kind of another factor of density for collisions and the atomic physics to make. Now, in real EUV observations, those emission those emission lines are like not only density dependent but temperature dependent, and then all of a sudden you've got all your like instrument response on the basis of wavelength and all that kind of stuff. So I don't do any of that; I just do density squared because I don't have enough physics in the simulation to do the, you know, calculate the emission from a thetical iron 12 or, you know, like my my coronal heating is so basic that it's not realistic enough to do um, the kind of keeping track of the temperature dependence of that emission. Now, there are plenty of people that do have that much physics, and so the the emission from like, say, you know, that is generated from simulations by like, say, Michigan or like predictive science or, you know, like like if you've got a better coronal, uh, heating model, then all of a sudden things like EUV emission, um, you do have to start folding in the real dependence on temperature, and you do have to start paying attention to the geometric factors if you want to create basically data number per second for direct comparisons. I don't do any of that stuff; I'm just density squared, and it's still looks like there's a ton of structure. So if I put real heating into this and you actually had all of this, uh, you know, material here sort of heated from its participation in the reconnection region, then I think there's a ton more structure, different lines you could then look at.

I mean, there's a lot of of room for improvement here, uh, considering the models for the single flux, so such as European all, but uh, we know that the from the famous, uh, PSP observation that we are not looking for the single flux, so there are multiple flux slopes which are coming towards us. Rather, if we see the storms on May 21st, it consists of rather five to seven there. So yeah, what's our modeling capability are there for simulating these type LS multiple flux LS? Um, it's in in some ways we have all the stuff that we need to do that; it's just a very difficult calculation. So, you know, for kind of all of the time I spent analyzing my one eruption, I could create a scenario where I keep energizing it or I have two of these right next to each other or three of these right next to each other, and then I just have to run the whole thing. And okay, that's um, people are working on it, and it is challenging because it's not for the single eruption at the same time; multiple eruptions at the same time. There are some things there where fast CMEs get de-accelerated, slow CMEs get accelerated, the scene that we have observed on July last July, that miscalculation between modeling and actual measurements. Right, right. Yeah, yeah, no, exactly. That was kind of the stuff that I skipped over at the end, but but even after you've gotten your set of eruptions out of the corona, you then have to figure out what's happening for that entire duration, and a lot of stuff can happen. I mean, that's that's where you do get those kind of false, you know, we're expecting to see CME at 12:00 on, you know, Thursday, and it never showed up. Well, it's because the, you know, there was some high-speed stream that knocked it out of the way or some other thing ran into it. I mean, it it's there's a lot of stuff to work out there, but we've made a lot of progress, and there's more work to be there. Is there is any last questions for Ben? Great. Thank you, Ben. All right. Thank you very much, guys.