08:05:07 Good morning, everyone. Welcome to our second keynote, for the week on outflows. 08:05:18 I chose for the image today, an observational image of Centaurus A known for its relativistic jets that are being flung out by AGM feedback. Perhaps we'll hear something about that in our keynote. 08:05:35 This is just our schedule of eight themed weeks. 08:05:40 So we're finishing up outflows this week and next week we'll move on to inflows. 08:05:46 Just a reminder that we have a YouTube page which features new results videos that are contributed by members of the community for minute long videos there's a video on the thing for making a video. 08:06:00 It's a four minute long videos there's a video on the thing for making a video. We've been a little bit light this week so I just maybe people are out of new results. 08:06:06 I hope that's not true. But we, we encourage people to submit videos and we'll, we'll post them on the channel. We've got like 80 subscribers now everyone's watching these things so it is getting a pretty good coverage by the community and hopefully your 08:06:23 results will be seen by others. 08:06:27 On that same YouTube page we're featuring Junior scientists so if you're a junior scientist who's potentially on the job market. You can go to the halo 21 on the market Slack channel and submit a video about yourself, this. 08:06:41 We just had a submission from running to Jonah from the Ramon Institute in India. 08:06:46 So check that out and we're also encouraging people to post job listings if you're hiring, feel free to post in that same Slack channel. I know we've had a couple of jobs posted so so definitely a place you want to check out if, especially if you're a 08:07:01 junior scientists potentially looking for a job now or in the next couple of years. 08:08:26 So today, we will have a keynote presentation by Tim Heckman from Johns Hopkins, that will take us through the hour. 08:08:36 And then that will have a short break of about five minutes or so, and then follow it up with a panel discussion, featuring dr Cassie lock us from Space Telescope, Professor, Professor crystal Martin from UC Santa Barbara, Professor Evan Schneider from 08:08:50 University of Pittsburgh and Professor Chuck style from Caltech during Tim's presentation, and during the discussion panel we encourage rather than typing things into the zoom channel for, you know, questions or comments, we encourage people to do so 08:09:09 in the Slack channel Hello 21 week five outflows that will allow it to be more preserved after the zoom channel just closes and allow other people to comment, and in a threaded way so it's a bit more organized and you can get some pretty organic discussion 08:09:25 taking place there so and. And what's more, during the panel discussion which I'll be moderating, I'll be selecting questions and comments from the Slack channel, to ask of our panelists, and so on, so forth. 08:09:38 Another thing that you can do is by clicking on the little emojis next to a comment that you like, or perhaps dislike. That's effectively a way of uploading it so that if I see there's a comment that has 12 up votes on it, I'm more inclined to ask that 08:09:54 of our panelists. 08:09:57 So, so that's one way you can kind of informally vote in the slack. 08:10:02 Okay. 08:10:05 So our keynote today is given by Professor Tim, Tim Heckman. 08:10:09 He needs little introduction he's one of the most senior members of our community. And as a chair at Johns Hopkins, as well as a member of the Academy, the National Academy of Sciences and the American Academy of Arts and Sciences. 08:10:25 Tim has done you know he's an observer by trade, he's worked in a variety of different wavelength bands including optical UV radio with surveys like manga Sloane fuse galaxies, much of his work is that is is extremely seminal and has defined the direction 08:10:46 of our field in, in terms of wins and outflows and AGN and Starburst driven wins. And so I'm really excited to hear what he has to say. Today on the topic of of outflows for us. 08:11:01 So, Thank you for joining us, Tim. Thank you. 08:11:56 Okay. 08:11:54 This is great. Thanks for inviting me I've been really enjoying the workshop so far. 08:12:00 And I wanted to spend some time walking you through galactic winds and how they might impact the circle galactic medium. 08:12:07 So was I tried to put this together I realized I can't cover everything. 08:12:13 So I decided to have some points of emphasis. 08:12:17 I will talk about whole range of red shifts but the bulk of the talk will be focusing more on the low redshift universe and this is because, as I'll show you is outflows are multi phase. 08:12:32 And most of these phases can only be looked at in the local universe. And also, it turns out that you learn a lot of interesting things by spatially resolving the flows, which is also a lot easier to do than the nearby universe. 08:12:44 I'm going to focus on feedback from stars and not talking about Ag and feedback and massive galaxies I think this was really while incredibly important it's a sort of a separate talk. 08:12:55 And I would invite everyone to join Mark Floyd's after party where we're going to talk about this at great length I suspect. 08:13:02 And I'm also going to even though the outflows are multi phase I'm going to emphasize primarily the warm that is kind of the 4k up to say 10 to the seven k gas. 08:13:13 And this is basically just being practical This is where most of the data 08:13:20 pertain to. And I think they're the phases of the outflow of that probably have the most relevant to the certain galactic medium. 08:13:27 So this is just an outline, I wanted to just start by giving you a brief motivation and some background. I wanted to walk you through MIT to which is the best study galactic when you can see all the different phases and intricacy of these outflows, and 08:13:43 then talk about how mad to in fact is not unique it's typical. And I want to emphasize sort of spatially resolved observations of the hot and warm phases of wins in the local universe, then talk a bit about how we measure systematic properties of wins 08:13:59 things like outflow velocities mass flow rates and so on. And the scaling relations that resolved and then conclude by summarizing the observational evidence that wins impact the circle galactic medium. 08:14:18 So, motivation and the biggest picture of course as we all believe that feedback regulates the efficiency of turning accreted gas into stars and I think one really interesting way of looking at this, is to consider this was from a paper. 08:14:29 Last year from guru zl, which basically says that the global efficiency of converting Barrick created baryons into stars has been roughly constant with redshift. 08:14:40 And it's a border 10%. 08:14:42 So the data or just the normal Medallia plot, showing the star formation history of the universe. 08:14:48 And then the red and blue lines are basically a model in which effectively. 08:14:54 There's just a constant efficiency of conversion of the creative baryons into stars as a function of time. 08:15:00 So that's actually pretty simple right so now plot is often called mysterious and you need feedback and so on. It really is primarily just tracing the Christian history of dark matter and baryons into halos. 08:15:15 So why is this interesting Well, this is one of the clues as to what's going on here. Why is it 10% Where does the 10% number come from, is that there's almost no evolution in the Omega, each one. 08:15:29 So if you were to plot on the left this is. 08:15:32 Most of the color data points there are just the accumulation of stellar mass and galaxies. 08:15:38 And the hollow red lines are from a compilation by readouts of omega h1 measured in different ways. And what you see is from Russia for to zero. 08:15:49 The mass and stars has grown by a factor of 30. 08:15:53 The master. Each one has stayed almost exactly the same. 08:15:56 So for that to be true, it says that basically the creation rate of h1 through cooling has to be equal to the destruction rate of h1, and that destruction right as effectively going to be some combination of outflows and star formation so the instruction 08:16:12 rate of each one is one plus data at times of star formation rate where a does this famous mass loading factor that we discussed at some length yesterday afternoon within Evans, nice session. 08:16:25 What we'll see later in the talk is that at the relevant scale the h1 disc and the stars, this mass loading factors order unity for the stellar mass for the LCR galaxies are really dominate the stellar mass budget. 08:16:43 So this essentially implies that the creation rate of each one is. 08:16:52 Sorry I'm 08:16:55 think I need to minimize. 08:16:57 I've got my thumbnail covering my screen. Yeah, so, implies the completion rate of each one has to be lot less than the berry on accretion rate, and that suggests that when just 10% factor is not really being set by anything happening inside the galaxy, 08:17:13 it's determined by the effectiveness of converting warm CGM baryons into cold is. 08:17:21 So why do we expect feedback would couple especially exceptionally well to the stricken galactic medium. 08:17:28 It's hard to couple feedback to the disk, right if you have feedback in the disc. 08:17:44 It has a high column density which means when you push on it it doesn't move very much. And it has a very unfavorable geometry, so if you release the energy in the disk, it finds a path of least resistance and just blows out into the Halo. 08:17:55 In contrast, that's tripping galactic medium has low volume densities low column densities and a sort of quasi spherical geometry so it's going to capture, whatever energy or momentum is dumped into it. 08:18:05 So I think these sorts of arguments, really explain why I think the circle collecting medium is really the battleground were a Christian is fighting with feedback. 08:18:16 So let's then talk a little bit about how hot winds are created by Starburst so round numbers for every solar mass with stars forum, you get about kind of the 49 hours of kinetic energy from supernovae and from stellar wins that's about 10 to the minus 08:18:32 five c squared physics units. 08:18:44 And that doesn't sound very impressive, but the potential well depth of the galaxy like the Milky Way is only roughly 10 to the minus six c squared. In this conventional picture that was first proposed by sheer volume and Clegg. 08:18:48 These stellar objectives are thermal shocks and produce hot gas. The temperature is of order tend to the eight Kelvin modular these factors alpha and beta, two alpha is the thermal ization efficiency. 08:19:01 if radiative losses are negligible, then alpha is one, and beta. I'm going to call the poisoning factor to make a distinct from the mass loading factor. 08:19:11 So, beta is one for pure stellar ejecta. And as you mix stuff in and beta can be larger than one pressure this gas is much greater than the surroundings, it's going to expand along the path of least resistance and blow out into the Halo and form of sort 08:19:27 of wide angle bipolar wind. 08:19:30 If You just have conservation of energy of idiomatic expansion of this hot gas, and the terminal velocity of the wind is going to be something like 3000 kilometers per second. 08:19:40 with these alpha and beta factors. 08:19:43 The kinetic energy flux carried by the wind is about 10 to the 42 times of star formation rate if radiative losses are negligible. And the momentum flux of the wind is about five times 10 to the 33 kind of star formation rate again for alpha beta one. 08:20:01 And because this material in principle is could be pure stellar ejecta, it could have a very high metal history so the metal is the alpha elements produced by core collapse supernovae is 1.5 over beta time solar. 08:20:18 So, here's 20 years of simulations on the left is a simulation by Strickland Stephens in 2000, showing a bipolar wind, and on the right is Evans, most recent work from last year. 08:20:33 And, yes, we've made progress but I think conceptually. 08:20:38 I think the picture of proposed by Strickland Stevens is still I think conceptually accurate. That is, we have most of the volume being occupied by this. 08:20:47 Kenny was energetic, when fluid. 08:20:51 And what we actually observe and admission or absorption comes from denser material that is either interacting with, or possibly was created from the wind fluid. 08:21:00 And it has small volume filling factors. 08:21:03 So now let's take a quick guided tour through the multi-phase wind and M 82 so we see warm gas on the left. In a Java. 08:21:11 We see hot gas tend to the seven k gas and x rays in the center panel, and on the right we actually see dust scattering FEV light from the central Starburst, and you can see the structures are remarkably similar, and all three bands the scale here these 08:21:27 inserts are about 20 kilo parsecs in height. 08:21:33 Let's talk about the, the driver of the wind is when fluid we actually observe it directly and the answer is yes we see it. Me too. And about a dozen other Starburst directly through the hydrogen and helium wipe iron lines at 6.9 and 6.7 KV. 08:21:52 These are signposts of extremely hot gas in the case of MIT to the temperature is about 60 million degrees. If you will allow this gas to not suffer any rate of losses and just expand a diabetic Lee would reach terminal velocity of about 2000 kilometers 08:22:08 per second. 08:22:10 So it's detected as a sort of ghostly blue glow in the central region of the panel on the right, which is just color coded by energy bands and the x rays. 08:22:20 It's only detected in the central region, and that's because a diabetic expansion and cooling causes the gas and this hot gas to effectively disappear becomes very low intensity and at radically cools. 08:22:35 Now what we commonly think of when we think of X rays from the wind is not the wind fluid necessarily it's gas that's roughly order of magnitude cooler than what we see in the very center of the galaxy. 08:22:47 So its temperature is something like four to 10 million degrees. 08:22:53 And, at least in the soft band below sort of half a kilovolt charges change becomes important tweets about 20% of the total X ray luminosity and this is X ray Michigan produced by an interaction between highly ionized and mutual gas. 08:23:09 You can see from the picture that gas at 10 of the seven K is clumpy it's not volume filling their structure their filaments and and structures that actually correspond to some of the structures we see in the alpha image right in the central Starburst, 08:23:26 the elemental abundances are super solar about three times solar in silicon and silver. 08:23:32 But if you go further out into the outflow abundance, roughly solar. So in the outflow. This is not consistent with this gas being pure stellar ejecta. 08:23:45 Probably the most familiar phase is what we see in H alpha mission. 08:23:50 These are just a couple of pictures of the inner and outer hf of filaments and I'm 82. And you can see just from looking at this at this is very clumpy through elementary material accelerated by some kind of momentum transfer from the wind fluid, and 08:24:04 it has a very small volume feeling factor, something like 10 to the minus three to 10 to the minus four, depending on the radius or the geometry that we have. 08:24:15 Now, one important point that I think doesn't get talked about enough is that it's possible to actually measure the momentum flux of the wind fluid. 08:24:24 And that's possible because using optical data we can directly measure the electron densities and electron temperatures, and we can actually measure the pressure in the flow as a function of radius. 08:24:36 And in this plot on the left hand side the black solid dots are measurements of the pressure in the warm ionized gas and MH is a function of radius. This is log are on the x axis so we're going from the central star wars to little beyond the thousand 08:24:53 perfect. 08:24:56 We can also measure the pressure is using x rays. This is a little bit tricky because we know don't know the volume filling factor, and the pressure is going to depend on the inverse square root of the volume filling factors so the minimum pressures of 08:25:14 these hollow dots are the X ray pressures and I've adjusted them, assuming a volume filling factor of 20%. And that was done a bit of a cheat to make the pressures of the X rays sort of line up with the emission line gas if the volume filling factor with 08:25:29 was 100%. These would be lower by a factor of about two. 08:25:33 But I think it's fairly obvious from the X ray image that the volume filling factor of the hot gas can't be 100%. 08:25:40 So the tiny little dots are basically just doing a stupid as possible thing which is just take the standard Chevallier and click model scale to the star formation rate of 82 and the radius of the Starburst led to. 08:25:54 And you can see that the model is an excellent agreement with the pressure that we measure. 08:26:00 Now, in this model in the central region interior to a few hundred parsecs This is the thermal pressure because the, there is no wind there. It's just hot gas. 08:26:11 Sonic point of the wind is about the Starburst radius and so that really I larger than a few hundred parsecs ram pressure takes over and you get this one of our r squared pressure dependence in the model. 08:26:24 And one of the things that's been puzzling me I guess is, in this emerging popular picture in which the kind of the 40 k gas is somehow cooling out of mixing layers associated with the hot gas, would you expect this guest to be at the same pressure is 08:26:40 the ram pressure of the wind, rather than just the thermal pressure of the wind fluid which is going to be considerably lower because the wind is supersonic. 08:26:52 In any case, you can use this to measure the momentum flux. When you have momentum per unit area, momentum flux per unit areas and same thing as ram pressure. 08:27:03 If you just multiply that by the surface area, you get the momentum flux and in this case, the momentum flux in the wind fluid matches what we expect the input from the Starburst. 08:27:14 Now what about the structure and kinematics with this warm gas. If you do a detailed study of the kinematics as Fabry perot data cube 3d mapping of the outflow by Chuck Belen bland Hawthorne, what you see is that the kinematics and structure of the gas 08:27:33 implies that the gas is flowing out along the surfaces of hollow by cones. 08:27:37 So this is projected velocity as a function of distance from the Starburst, and the lines are double peak, because you're seeing emission from the front side of the colon in the backside of the cone along each line of sight. 08:27:52 You can also see that the gas rapidly accelerates the radiance of the Starburst is about 300 parsecs. So this has reached its maximum velocity, very quickly as it exits the Starburst it reaches the maximum observe velocity of about 300 kilometers per 08:28:07 second. 08:28:08 In the modeling that they did you deep project us to get the true outflow speed would be about 600 kilometers per second. 08:28:18 Now this is just Overlay and the warm gas, which is shown in contours, with the hot gas which is children x rays from looking at this it's a little bit unclear what's going on but let me go to the next slide. 08:28:31 This is actually taking a ratio of H alpha and soft x rays and in this picture. the orange and yellow regions are regions in which he alpha is bright and the soft x rays are relatively faint, and the green and dark blue regions are areas where the X rays 08:28:55 are bright, relative to h alpha. And what you see is although it's quite messy. It really looks like the warm gas at 10 of the 4k is sort of in closing the X ray gas the extra gases interior to or upstream of the warm gas so there's clearly some kind 08:29:12 macroscopic interface between the hot and warm gas in the system. We also see a magnetized relativistic phase which was traced by synchrotron permission from cosmic ray electrons and the magnetic field. 08:29:20 This has been known for some time. This is the most recent paper. I could find on radio properties of 82 and what they find in this case is that the field geometry of the magnetic field is radio in the region of the outflow. 08:29:35 Assuming Aqua partition the magnetic pressure be squared over eight pies about 10% of the thermal pressure in the wind. 08:29:43 And in this paper the authors and basically are interpreting this as the magnetic field is being affected by the wind fluid and carried out with the field lines getting stretched and a radial direction. 08:29:55 Just to complete the picture. There's also molecular and atomic phase, this is sort of a cartoon, put together by Leroy Adele, in which the cooler material the molecular and atomic phases are in a sort of fountain flow, roughly 100 kilometers per second 08:30:12 that surrounds the region of hot and warm gas. 08:30:17 Okay, so now what I want to do is basically try to convince you that me to is not some weird. A typical object as a typical of galaxies today but it's typical of Star Wars today. 08:30:28 And this is a cartoon put together a paper that David Strickland led some time ago, and it's just a schematic, which I think is a cartoon representation to what I was saying before with me to that we have sort of Central cavity filled by this wind fluid 08:30:45 which is not emitting very much, and then we have the sort of limb Brighton structures that are visible in H alpha and soft x rays around the boundaries of this cavity. 08:30:57 So I'm going to just show you some examples. 08:30:59 This is NGC 3079 in the central region. This is a killer parsecs skill bubble, we can see these filaments kind of a film entry bubbles structure. 08:31:20 or the soft x rays. If you measure the kinematics of the gas, put a slit along this you see the same thing is to get me to that there are split lines from the mission from the front and back of the bubble with implied velocities of orders 600 kilometers 08:31:33 per second. 08:31:35 This is going into the outer part of NGC 3079 is Rhl fulfilling filaments extending out about 10 kilo parsecs on the left hand side. 08:31:45 And then on the right hand side is from the recent paper by Hodges Gluck Adele, showing the soft x rays, and probably just scattered FEV emission on scales of 10s of killer parsecs. 08:31:58 And again, These are all sort of Islam brightened supplementary structures. 08:32:04 This is the core region killer prospect scale region of NGC, 253 in soft x rays on the left hand side, you see again is limb Brighton structure to the lower left of the nucleus. 08:32:19 On the right hand side is just said as a musical cut the surface brightness in H alpha x rays showing limb brightening and both h alpha and x rays. 08:32:29 This is the outer halo of NGC 253 on scales of 10s of killer parsecs. Again, soft x rays, on the left hand side and he off on the right here is scaling up by a factor of 30 and star formation to you legs merger or up to 20 again soft x rays on the left, 08:32:48 he off on the right these three limb Brighton double bubble structures. 08:32:52 Same thing in the US, or in history 6242 Elementary structures and he alpha and x rays that align with one another. 08:33:00 And then, most spectacular is a recent paper by rookie Adele. 08:33:06 This is something that's probably, you know a factor of three to 10 even more powerful than up to 20. So redshift have about a half. And this shows ionized limb Brighton bubbles stretching out roughly 50 kilowatts section radius from the central Starburst. 08:33:24 Okay, so I want to just shift gears while I'm talking about measuring spatially resolved structure because one of the other ways of doing it so far we've been talking about. 08:33:36 You can also use resonance scattering and associated fluorescent admission to study wins. This is just showing a picture on the right hand side. This is the ground state and it's fine structure levels in silicon two photon is absorbed from the ground 08:33:59 state. It can go to one of the excited states and then it can either decay back to the ground state, which is resonant scattering, or it can decay to the fine structure level, and this is called for essence. 08:34:13 So, as for every photon that's absorbed in the outflow. That photon is going to be readmitted either as a residence line or associated for us in line. 08:34:23 And once you can immediately grasp is that the relative strength of the absorption lines and the emission lines are going to depend on the relative size of the outflow compared to the aperture of the spectrograph used to measure the, the spectrum. 08:34:36 If the outflow is large compared to the aperture size as a shown in the right hand, left hand side there. Then you get all the absorption. Because you see the full Starburst. 08:34:49 So the absorption line is strong, but most of the emission is outside your aperture. And so you get a strong absorption line with a week admission line, and that means, in principle, one can learn something about the size and structure about flows. 08:35:01 By comparing the absorption lines and the emission lines. 08:35:06 This is just some examples. This is an ongoing project with a graduate student Benji one. 08:35:12 We are using cos UV spectra of linemen break analog galaxies, these are low wretchedness Starburst that were chosen to have property similar to Lyman break galaxies. 08:35:23 And here the black. 08:35:26 Sorry, the data for this galaxy and orange This is silicon to 1260 has an associated fluorescent line at 12 64.7, and the data are shown in orange. 08:35:39 If you have a starburst wind, and you have an aperture large enough to capture the entire wind region, you would get a spectrum like what's shown in black, so very strong fluorescent addition line. 08:35:54 You can see the data fluorescent light is much weaker. And so we're trying to make models to sort of use these to infer what kind of structure the wind would have to have to be consistent with the weakness of these fluorescent emission lines. 08:36:08 And what we're finding is that you cannot fit the specter of these galaxies, using a sort of traditional density law, like art of the minus two, that's too centrally peaked too much emission ends up in the aperture, you have to have a radial density profile 08:36:23 is considerable the shallower than that. And that has very interesting implications for the structure of the wind, 08:36:31 even better is the case where you can spatially resolve the, the resonance emission and this was from a recent paper by Rich Adele, where they've used Casey wi to map magnesium to wind emission from a Star Wars, performing galaxy to wrench into point 08:36:48 detect magnesium to admission out to radio about 20 kilo parsecs, and again in the modeling, they find that they have to have a velocity which is falling with radius, and a density profile is considerably shallower than art of the minus two. 08:37:04 So this is consistent with what I were inferring from these analysis of the spectrum. 08:37:11 Okay, so now I want to switch gears and talk a bit about the systematic properties of galaxies, what are the important parameters that can be derived from the data. 08:37:21 And what are some of the assumptions we have to make. 08:37:23 This was something that I haven't covered very nicely in her talk yesterday, so I hope those many of you probably heard it so I can go through the introduction to this relatively quickly. 08:37:36 I'm going to primarily focus on the large data sets which are only really available for these sort of down the barrel absorption lines but I wanted to start out, not using absorption lines but going back and talking about this business of measuring radio 08:37:51 pressure profiles that we that I showed you and me too, so we can do this and more than Me too, me too. We have the best data, but we can actually go through the same exercise and try to estimate radial pressure profiles and use those to infer a momentum 08:38:08 flux. So we're just basically measuring the momentum flux per unit area is being the pressure, and then integrating over the area we can get the momentum flux and what's shown here is the momentum flux we estimate using this technique on the y axis, versus 08:38:25 the injected momentum flux gorilla Starburst on the x axis so you can see that the blind through the middle of just the one to one line so measured momentum flux in this hot when fluid is comparable to the amount of momentum injected by the Starburst 08:38:42 Okay, so now I'm going to focus on working lines. 08:38:46 And there have been quite a few studies john Chisholm wrong on borderline among many others crystal Martin, the people have used this technique to try to infer properties of outflows there are many papers on this subject, I'm going to focus on a particular 08:39:01 sample. I worked on with since birth a car. 08:39:09 And that, mainly because I'm most familiar with the assumptions that I made, but also I think this is a very nice sample in the sense that most of these are live and break analog so what we learned from studying these i think is relevant to galaxies at 08:39:25 higher redshift. We restricted the sample to only cases where the aperture cover the entire Starburst we're talking about global properties. 08:39:34 And the goal of this was to kind of figure out what the scaling relations look like and confront the models and simulations with them. 08:39:41 This is just showing profiles of the time that we work with here, this is showing five different ions you see range and velocities pile up strong absorption in their vehicle zero and then the absorption becoming weaker as we go to higher velocities. 08:40:00 Okay, so 08:40:04 how do we do this with there are a number of problems that. Then, let me briefly walk you through one problem is, how do we define the velocity right we want to measure an outflow rate, and to measure out flow rate we need to have a velocity. 08:40:20 You can see that there is no well defined velocity here. 08:40:25 Sometimes people. 08:40:27 Select a maximum velocity, which in this case would be sort of eight or 900 kilometers per second, that's kind of a bill to find quantity in some cases in the paper that I'm going to summarize the results we actually use the centroid. 08:40:42 So it's sort of a 08:40:45 non parametric measure of just the centroid of the line that has some problems as well but so most straightforward thing to do. 08:40:53 We also need to know the column density and problem was getting the column density is that these lines are often optically thick so the strong lines that are easiest to measure have significant capacity and that means that the, the, basically equivalent 08:41:11 width of the line is relatively insensitive to column density. So what we did in this paper was to actually use week lines which are unsaturated and measure the calm densities. 08:41:32 These are measuring lines from typically things like sulfur. Three and sulfur for, magnesium, or iron too. And then, basically, we spend it enough, we spend a large enough range and ionization state that we could do a refined ization correction. 08:41:45 And we then converted these columns in metals into hydrogen or total columns assuming that the middle of the city and the outflow and gas was similar to what was derived from Nebula emission lines for the Starburst itself. 08:42:02 So what are the results of this exercise. 08:42:06 We find fairly strong correlations between the velocity of the outflow. And both the stellar mass or circular velocity that shown in the pot on the left hand side. 08:42:19 And with the star formation rage shown on the right hand side. 08:42:24 This has been, this is not surprising many people have found qualitatively similar results. 08:42:32 We thought it was maybe more physically instructed to look at a sort of normalized outflow velocity. In other words, the velocity normalized to the rotation speed of the galaxy. 08:42:41 That's what's plotted on the left hand axis here. 08:42:45 And we wanted to also have then something that was a more normalized star formation rate on the x axis so this is star formation rate per unit area. 08:42:55 And what you see is a steep rise in velocity, this normalized velocity is a function of star formation rate per unit area. And then it sort of saturates once you get above the star formation rate per unit area about 100 solar masses per year. 08:43:23 Calabar sir. 08:43:14 The scale of these outflows what's clouded here is this V max the maximum velocity and the velocity in the strong outflows at the top upper right hand side of this plot are typically five to 10 times circular velocity. 08:43:28 If you use the centroid velocity instead they're still about three to five times the circular velocity so there for the strong outflows, the velocity is that we observe in the warm gas are significantly larger than the escape velocity. 08:43:43 Okay, I see I had my slides out of order this is basically a slide that explains why walk you through before you take a centralized velocity. 08:43:56 We take week unsaturated lines. And then finally there's this question of what you assume about the radius. 08:44:04 So the get our flow rate you have to adopt value for the radius radius is something that's poorly constrained. 08:44:14 Basically we just adopted a scaling where the radius is assumed to be some dimension list, quantity beta times the radius a starburst in beta. We chose to be a more one. 08:44:26 Okay. 08:44:28 So doing that. 08:44:30 We can now measure the outflow rate compared to the star formation right and this was a clock that haven't showed yesterday. And within the uncertainties basically the answer is the star formation rate is of order. 08:44:44 I'm sorry the mass outflow rate is a ward of the star formation right. 08:44:48 We can also measure the momentum flux. 08:44:51 We have a mass flux you multiply that by V you get a momentum flux. 08:44:56 And what we find is they're basically two regimes. There are these data points of the upper right, where the momentum flux in the warm gas. Now this is different from what I started out for the beginning I was talking about the momentum flux in the wind 08:45:09 fluid. This is the momentum flux and attended the 4k gas. 08:45:15 And there are a family of these objects in which the momentum flux in the warm gas is comparable to the injected momentum flux from the Starburst. So there's been a very effective coupling of momentum from hot wind into the warm gas on the other hand, 08:45:30 there are these data points in the lower left part of the plot where only about 10% of the momentum available to the gas has been transferred to it. So the momentum flux transfer has been ineffective in these so called week wins. 08:45:46 So what do we think the physics going on. I think the simplest model or way of thinking about this I think it's just to try to gauge the relative importance of the outward force acting on these this material, due to the momentum flux in the wind and the 08:46:05 force of gravity, the inward direction. 08:46:10 And it's possible if you do this, it's almost like an Eddington limit kind of argument or similar to it, you can define a critical momentum flux from the Starburst, in which the momentum flux from the at the location of this cloud will produce a force 08:46:27 on it that exceeds the inward force of gravity. And this mode critical momentum is written down below the figure their beta is this sort of additional radius radius of the Starburst, the column density. 08:46:41 The main master particle and the circular velocity squared of the galaxy. 08:46:45 We know all of these things except for beta, and if we just take beta to be roughly or unity. 08:46:52 We can make the plot above where we compare the momentum flux delivered by the star burst on the y axis to the critical momentum flux needed overcome gravity on the x axis. 08:47:05 And what you can see is in the majority of our sample. 08:47:09 The momentum flux from the Starburst is roughly an order of magnitude larger than what's required to overcome gravity, so it's not surprising that we see an outflow. 08:47:23 And in fact in the strong Star Wars, the ratio of momentum available, compared to the amount of momentum need to overcome gravity is, you know, much larger than one. 08:47:36 And in fact, there's a good correspondence and these two plots. 08:47:41 The cases where there's been an effective transfer of momentum from the wind fluid to the gas points in the upper right. 08:47:50 Oops, 08:47:53 are the same objects in which the momentum flux from the Starburst is much larger than what's needed to overcome gravity. 08:48:02 Now, in this picture, is there a natural explanation then for why wins evolve with cosmic time why they're not so important now and we're very important early times and that's because you can rewrite this ratio of outward inward forces as being proportional 08:48:19 to the star formation rate divided by RV squared. 08:48:23 At six mass star formation rate goes like one plus the two the 2.5. 08:48:30 That's a specific star formation right in other words is evolving, like one plus either the 2.5 and effectively by definition, at fixed mass RV squared is independent redshift because RV squared is just mass. 08:48:44 So this means that as you go from Richard zero. 08:48:48 You're in this regime where there are no wins or only very weak wins. 08:48:52 As you go to higher redshift because of this one plus leader the 2.5 this ratio in typical star form and galaxies grows, and by the time you have a redshift larger than one you're in this regime of strong winds. 08:49:06 Okay. 08:49:08 So I want to spend the last 10 minutes talking about how winds might impact the circle galactic medium. 08:49:15 I'm going to start at redshift one because that's where the, we have the most data, and it's in this regime, as I mentioned on the previous slide where you don't have to look at Starbucks to see wins typical galaxies on the star formation main sequence 08:49:29 are driving outflows, the left panel this, these two plots are taken from a recent paper by land and Mo. These are based on stacking sightlines of quasars societally with either luminous read galaxies in the Sloan survey or admission line galaxies that 08:49:48 were used in the E boss survey from Sloan. 08:49:51 So it's a very large sample with lots of stacks available. And this enables you to reach the equivalent with limits of few hundreds of Amsterdam, and these stacks. 08:50:02 And there are two points to take away from this one is that in the inner CGM interior to about a impact parameter of 100 kilo parsecs. 08:50:13 There's an excess of magnesium to absorption in the mission line galaxy is compared to the luminous read galaxies. 08:50:20 This is comparing things with in killer parsecs this would be even more impressive. If you re scale the, the x axis and the left hand side to be in units of zero radius because the red galaxies are quite a bit more massive than the blue galaxies. 08:50:36 On the right hand side is showing for iron and magnesium to in the stacks. 08:50:40 The fact that the magnesium two lines within this region, interior to about 100 kilo parsecs are stronger along the minor access than along the major axis. 08:50:53 Now this is a very nice paper that I think is pretty much consistent and goes further than the land and mo paper in terms of what they're able to model. 08:51:02 This is from shredder now based on Muse data redshift about 1.1 1.5. 08:51:09 On the left hand side, what they find is a by modal distribution of magnesium to absorption. 08:51:15 There's a peak near zero, that would be in the plane of the galaxy. 08:51:21 And then there's another peak. 08:51:24 Roughly from 50 to 90 degrees which would be along the minor access the interpretation of this is that the peak closed within you know sort of 10 degrees of the major axis represents accretion and the plane of the disc and peak. 08:51:44 The broad peak from 50 to 90 degrees would represent an outflow what they're able to do is with a model of the outflow sorted by conical flow is to try to make some inferences about velocities and mass loading factors. 08:51:54 The middle panel shows the velocity in units of the escape velocity on the y axis, and then the stellar mass of the galaxy on the x axis. And you can see that it's only when you get down to below about few times 10 to the nine solar masses that the inferred 08:52:13 velocities from these models will exceed the escape velocity. 08:52:17 These are on scales of roughly 100 kilo parsecs. 08:52:21 On the right hand side is the estimate of the mass loading factor you can see there's a lot of scatter the data points are shown and then the colored lines are various theoretical model predictions on the scales of 100 kilo parsecs the mass loading factors 08:52:35 are measured to be of order a few, there's a wide range you can see, I would buy the median looks like maybe, you know, two or three. And it seems to decline slowly with radius. 08:52:49 Or sorry with mass I'm sorry. 08:52:53 What about redshift two to three. This is when every galaxy affects every star forming galaxy effectively is driving a strong when this is some work seminal work by Chuck from 2010. 08:53:12 The actual stack spectra shown on the left hand side and the right hand side is basically showing the strength of the Georgian lines equivalent with a various ions as a function of impact parameter. 08:53:25 And what you can see is that you detect, not only blind in alpha and beta to detect how the elements out to a radius of about 100 kilo parsecs, and then it looks like there's some sort of cut off beyond 100 kilo parsecs where the lines get rapidly weaker 08:53:44 and check and talk about this of course. 08:53:48 Soon, but I think the interpretation that this is sort of representing the zone of influence of the wind, to this region interior to about 100 kilo parsecs is where these heavy elements have been are flowing out into the certain galactic media 08:54:06 at redshift zero data are quite limited. And that's because at redshift zero only relatively rare galaxies dr wins and if you look at cost halos or cost gas. 08:54:16 They're not very large surveys to begin with, and they're not large enough to really include Starburst galaxies. So we had to actually do data mining took a million galaxies and Sloan from them we selected 17 Starburst galaxies that had a quasar that 08:54:33 was bright enough within an impact parameter of about 200 kilo parsecs to observe with HST, the parameters for the star wars were derived from the SPSS spectra these are mad to like Starburst something like kind of the 59 herbs and total energy injected 08:54:51 over a few hundred million years. 08:54:53 Roughly 20% of the mass of the galaxy has been formed in this burst. 08:54:58 And this is all described in a paper by myself and others, a few years ago. 08:55:03 Well what did we see we're using cost burst and costs, so we're using costs gas and costs halos is our control sample, and selecting star forming galaxies. 08:55:14 And those are shown in blue and these two plots so this is the equivalent with of silicon three on the left hand side, carbon for on the right hand side, as a function of normalized impact parameters, this is input parameter and units to the viewer radius. 08:55:29 And once you can see is that the Starburst in red, have significantly stronger metal lines on large scale so we're going all the way up to the bureau radius, you see an enhancement by a factor of have several to 10 in the strength of these metal absorption 08:55:47 lines with a covering faction of about 50% in the outer CGM. 08:55:53 We also see evidence for superhero velocities. So on the left hand side or the width of the lines in the control sample. The bottom panel their typical velocities are the lines or maybe 100 hundred and 50 kilometers per second wide. 08:56:15 In the cost per sample there are many much broader lines and on the right hand side is just a stacking the linemen alpha absorption lines, and the cost per sample that shown in black, and what's shown in red is what you would expect if these were just 08:56:23 clouds orbiting in the gravitational potential of the Halo. So there's evidence that not only are the lines stronger in the halos of Starburst but they are kinematic Lee distinct showing evidence that something other than gravity's involved in their emotions. 08:56:41 Now Canada's really work. We just did a incredibly simple minded model we just took the classic windblown bubble, dating back to Weaver Adele and 1977 and retooled it to the circle galactic medium. 08:56:57 So this is just something that either be driven by an energy or by momentum. 08:57:02 You can we took our main Starburst age of 300 million years. 08:57:07 The main energy injection rate is 10 to the 43 or x per second. And then we use similarity solutions from Dyson 1989, assuming that there's a sort of volume feeling X ray Halo, whose properties are given a total mass of about 10 to the 10 solar masses, 08:57:20 with a density drops like one of our are sort of loosely based on the Milky Way's Halo. And what you find is that the outer radius of this windblown bubble is roughly 200 kilo parsecs in the case of an energy driven bubble. 08:57:42 And even if you have radiative losses in the wind fluid is driven by momentum. You still get out to nearly the same scale. So it is plausible that servers could be affecting the circle galactic medium all the way up the rural radius and these systems. 08:57:57 Well, 08:58:05 my next last slide I wanted to advertise something that I think is very exciting in terms of the future prospects for learning about the interaction between outflows and circle galactic medium. 08:58:11 Subaru prime focus spectrograph project is going to deploy roughly 2400 fiber is over. 1.25 Square degree field of view at the prime focus of the Subaru telescope, we will have spectra with resolution of about 3000, covering 3800 extra ones you 1.26 microns. 08:58:30 We have 360 nights for the survey over six years starting two years from now, and we're going to be doing many, many things with this but one of the things that's the most relevant to this talk is that we can use combinations of quasars and background 08:58:45 galaxies. 08:58:46 We estimate will be able to prob about 60,000 sightlines through the circle galactic medium of galaxies, with impact parameters less than 200 kilo parsecs over a redshift range of point 622 point four. 08:59:01 And by stacking these stacks of 100. So, we will be able to reach limiting equivalent width of a few tenths of an extra. So I think we'll be able to really expand on the kind of work that I showed you with the land and mo approach. 08:59:17 But now, over a wide redshift ranch with more ions and greater diversity of galaxies. 08:59:24 Okay, so let me just leave my conclusions here. You can read them, and I'm happy to take questions or I guess we're going to take our little coffee break and then come back for the panel discussion. 08:59:39 Excellent, thank you very much Tim that was, that was great good overview. 08:59:44 and a lot of points for discussion later. 08:59:47 Yeah, so it's nine o'clock now Pacific. We will reconvene in 08:59:55 five minutes 10 minutes will say five minutes we'll reconvene at 905. 09:00:02 So go ahead and get some coffee or do whatever you need to do, and I encourage people to look at the Slack channel week five outflows healer 21 week five outflows for some of the points that have been made by members of the community, respond to them 09:00:25 vote them, add your own comments and discussion points, and I will be selecting those. And I encourage that to be the point where you you mark things as opposed to just raising your hand or blurting out in the discussion. 09:00:32 So, we'll reconvene in in five minutes.