08:40:08 Okay, so it is 840.
08:40:12 Our next speaker is Dr. You want Lee, who is, as I mentioned a new professor at University of North Texas, and you want, are you there, she's going to talk about observational signatures of turbulence in Halo, and the Halo.
08:40:32 Yes, I'm here.
08:40:32 Okay.
08:40:35 Thanks for joining us, you on. I'm pretty excited about this.
08:40:41 turbulence is always exciting, especially after the conversation yesterday.
08:40:45 Yes.
08:40:53 All right.
08:41:00 Okay, so thank you for giving me this opportunity.
08:41:04 I had never thought that I would give a talk was such a title.
08:41:10 I must theorist.
08:41:13 So am I talking about observational measures.
08:41:16 I also mostly work on clusters. So what do I know about Halo turbulence.
08:41:21 Please allow me to explain.
08:41:26 I think that the CGM, the sun and the ICM are really tightly related.
08:41:34 I think we all agree that the SM and the CGM are connected.
08:41:41 And we have a Slack channel on that. There's lots of interesting discussions there.
08:41:45 And I would also argue that the ICM is basically the CGM of the biggest galaxies in the universe.
08:41:54 And also these biggest galaxies are extremely extended, you probably have heard of the ICL the entire cluster light, which is to extend to the setup of a note of these brightest classroom galaxies.
08:42:08 So technically, the ICM is also is them.
08:42:15 Now, even if you don't agree with my argument I think we can all agree that all these three AMS are multi phase and turbulent. So many of the analysis methods can be shared.
08:42:28 Because the SM study is relatively mature and many brilliant people, including many people that are here have done a lot of excellent work, a lot of really good analysis tools have already been developed there.
08:42:46 And today I'll be talking about one specific tool that I borrowed from yes I'm community, and it's the velocity structure function.
08:42:56 So, it will also structure function is a way to study turbulence.
08:43:01 And it's like I said, widely used and yes I'm community.
08:43:06 Many thanks to panel for a very wonderful presentation yesterday that basically covered the basics of turbulence.
08:43:15 I will just very briefly reiterate the characteristics of a turbulent flow in a turbulent flow, we see Eddie's world swirls twirls whatever he called him of different sizes.
08:43:29 And we think there is a energy cascade.
08:43:33 We're big worlds have little world, which feed on their velocity and little worlds have lesser roles, and so on to Wisconsin.
08:43:43 And 1941, come on or off came up with a famous come up our power spectrum for a steady state incompressible turbulence.
08:43:53 Many of you have seen it many of you know it at some large scale L, or some small k kinetic energy is injected into the system.
08:44:04 And it cascades down to smaller scales all the way down to the dissipation scale.
08:44:10 And between the injections go under dissipation scale is related to pay without power at all.
08:44:27 has stopped 530 minus 531 can derive this with very simple dimensional analysis.
08:44:28 Let's assume that epsilon is the rate of energy injection in a steady state, this should be equal to the cascade rate, and is also the rate of energy dissipation.
08:44:41 So what's the expression for epsilon.
08:44:43 Well since its energy rate.
08:44:47 It should go as the square to buy to buy some time t.
08:45:03 What is T. That's just us Eddie turn over time, which is about L divided by me.
08:45:00 And God gives us epsilon is proportional to me cubed divided by pal, and there should be a factor their lab experiments, tell us that this factor is about our unity.
08:45:12 So that gives us the relation between velocity and size, to me, is proportional to out to the one third.
08:45:22 And from there, one can easily derive the relation between UK and K. Do you have E proportional to cater the minus five thirds. And this is the power spectrum.
08:45:37 And this velocity size relation is basically the policy structure function.
08:45:42 Now we've had.
08:45:45 We have our tool.
08:45:48 Before I show you how to use it.
08:45:51 Let me first review how turbulence was previously measured for the hot ICM.
08:45:57 I put a question mark there. Extra turbulence, because whenever we talk about Terminus there's always a debate about whether or not we're seeing turbulence or bulk flow here for now, I'm calling everything turbulence, any random motion that's turbo Lister.
08:46:15 So on the left hand side, you're seeing resident residual Chandra extra image of the central region of a nearby galaxy cluster Perseus.
08:46:24 You can see that there are lots of fluctuations. For example, you see these bubbles that are inflated by the agent jets.
08:46:36 And you were in as Ramallah and collaborators developed a method called a doubt, it's a modified Delta variance method to try to extract information about turbulence from these images and adult have parents method, interestingly was also originally developed
08:46:54 to study as intervals.
08:46:57 So what Irina does is she derives the density power spectrum from the surface brightness fluctuations here.
08:47:06 And then she translates that that's the power spectrum to I lost the spectrum.
08:47:13 Assuming that data VRB is proportional to data Roll over roll.
08:47:20 And then she could get basically the kinetic power spectrum.
08:47:25 And what she found is for to nearby Cassie closers Perseus and wearable inferred level of turbulence.
08:47:33 What give you enough dissipation to balance rated cooling.
08:47:40 That's pretty neat.
08:47:42 But it's in disagreement with numerical simulations of class raging and feedback, including my own.
08:47:49 So in general, we find that Shockwave sound waves they debated processes and mixing are all more important than turbulent dissipation numerical simulations.
08:48:00 So was the renowned wrong.
08:48:03 Not really.
08:48:05 Now there is another way to measure turbulence in the high Zm.
08:48:18 And it's, I think in spirits very similar to the absorption line studies in the CGM that we've heard a lot about. So you can look at the lines coming from the hot gas.
08:48:23 And this was done by had told me, he told me it was an extra telescope was very high spectral resolution
08:48:33 and made one observation before it tragically died. that one observation was staring at the central region of the Perseus cluster.
08:48:42 And from these x ray emission lines, you can measure it brought me.
08:48:47 And you say okay there's some thermal branding and normal branding. So if you say the non thermal branding comes from turbulence, you can infer a line of sight velocity dispersion of about 100 something commenters per second, which is consistent was the
08:49:01 Irina's analysis.
08:49:05 Does that mean theories are wrong.
08:49:09 The set shoes that I found out is that when theories are wrong nobody really cares, except the theorists themselves.
08:49:18 Um, so I had to save myself and my fellow theorists have. I wanted to bring on her back to the theorists.
08:49:28 And the only way to do that is to prove that the previous observations were biased or limited.
08:49:37 Luckily for me, the centers of galaxy classrooms are often modified face.
08:49:44 We can observe the cool filaments using ground based optical telescopes, with a fuse like news and CFHT, etc.
08:49:56 We can even see some of the core components with Alma.
08:50:01 And all of these instruments have wonderful spatial and spectral resolution.
08:50:06 So what you're looking at here is the line of sight velocity map of the hf of elements in the central region of Perseus over plotted on the extra image.
08:50:20 And usually when people see a map like this, they'll say, Oh look, the motions of the fundaments appear, kind of random.
08:50:30 Not always seeing turbulence there.
08:50:33 Well there's one way to find out.
08:50:47 To obtain this first order velocity structure function. I simply look through the map for every pair of pixels. I record their separation and their velocity difference.
08:51:04 Then I just find the relation between pair separation, and the meaning velocity difference at the separation.
08:51:05 And that's shown on the right hand side. As you can see, all small scales, there is a beautiful, powerful, indicating a cascade of big world, turning into little worlds.
08:51:20 So we are seeing terrible is more interestingly, the Structural Function also reviews.
08:51:27 The scale where the energy is being injected, that's where it turns round.
08:51:36 And that scale is about 10 kilo per sec. That's exactly the size of these extra bubbles on the left hand side highlighted was the remote control.
08:51:44 So we're not only seeing turbulence, we're seeing turbo is driven by the feedback from the supermassive black hole.
08:51:52 Now in case you're not convinced. I've also divided the sentiments into enter filaments an outer feminists by simply draw a circle.
08:52:01 The filaments the inner filaments in Percy as follows structure function shown it red.
08:52:07 And you see that bump, a technical person.
08:52:11 The outer filaments in Perseus follows structure and function shown it green, and they don't seem to care about the intervals, which was a lot of sense.
08:52:22 Instead, the outer filaments review or driving skill and about 20 to 30 kill person, which is exactly the size of this otter Bible.
08:52:33 So it's likely that the outer feminists and Perseus are falling turbulence driven by the previous agent outburst.
08:52:42 And then we also look at another nearby galaxy cluster about 2597. When we saw this exact same story, the inner filaments follow turbulence, driven by the inner bubbles and the outer feminists follow turbulence, driven by the outer bubbles.
08:53:02 And we've also got all my data on this cluster.
08:53:06 We've looked at turtle is traced by the molecular guys, that's shown in bra.
08:53:13 And it overlaps perfectly was Terminus traced by the, I guess, everything is moving together.
08:53:21 And lastly, we also looked at the central got to have the burger classroom and seven.
08:53:29 We only have data on the very central region, unfortunately, but still kind of look at the structural function of the central filaments in Virgo. And we see again it's very beautiful power, small scales.
08:53:43 And then the driving so here is about 1.5 kilo per sec, which is exactly the size of the bubbles here, and the size of that cute little chip there that's currently inflating these bubbles.
08:53:59 How cool is that. So in all three classes I've looked at, I see turbulence and I see that the turbulence is driven by the activities of the supermassive black hole.
08:54:12 Now we can also think about what this means for the coupling between a CT heart phases, which many people here care about.
08:54:21 What do I mean by that.
08:54:23 masking do coke blobs move around like apples, falling off a tree or they move around like the leaves, which fly with the wind.
08:54:37 I've looked at my own sure hydrodynamic simulations. unfortunately, structure function of my focus is to flat.
08:54:46 And I think it's because my coat clouds are too ballistic and char and long had a paper out recently where they looked at structural functions avian flu vaccine relations with magnetic fields.
08:54:59 And it seems that was magnetic fields, cold and hot faces are much better coupled, and the Structural Function agrees really well with the observation.
08:55:11 To not that we can use the coal, gas as a tracer of the Hawkeyes.
08:55:17 We can talk about trouble anticipation.
08:55:20 So first let's look at how our measurements compare with the extra measurements.
08:55:26 So to toe horizontal lines represent the Hitomi results for the inner and outer regions of Perseus which more or less correspond to the inner and outer filaments, as I said earlier, had told me you had excellent spectral resolution, but it doesn't have
08:55:41 good spatial resolution just sort of results have to be represented was just horizontal lines.
08:55:49 Nonetheless, at the driving scale, they are really well with our optical measurements.
08:55:56 The Purple Line is the Chandra extra fluctuation analysis that you run as Robert I did for a region that matches the outer region in the optical analysis.
08:56:08 And again, Grace really well, was the turbulence trust by the ionized guys showing green.
08:56:15 So there's excellent agreements, near the driving scale.
08:56:21 However, the dissipation rate of small scales was previously estimated, assuming that turbulence with cascade falling Comoros relation.
08:56:35 What we see here is that the structure functions steeper than a classical coma growth turbulence.
08:56:44 So the actual amount of kinetic energy that gets cascade down to smaller scales is much smaller than the previous estimation.
08:56:56 So there's actually less trouble anticipation than you would assume, from what you see on larger skills.
08:57:05 So that means the previous observations were limited. And I saved the theorists.
08:57:13 But why do we see a steeper slope.
08:57:17 I don't know, I don't think anybody knows. At this point, we had a bunch of hand wavy possibilities in the paper, my current favorite theory is that black holes are just terrible drivers and turbulence, because everything changes on very short time skills.
08:57:39 For example, we can estimate the ages of different generations, the bubbles.
08:57:48 And they're separated by only 10 to 20 million years. But the return of our time alarm skills is on the order of 200 years.
08:57:57 So, we just can't ever establish a classic Common Core off turbulence from the driving scale.
08:58:05 But Mark wood has a different new idea. And I encourage all of you to look at his new videos about gravity waves.
08:58:17 And now that we can use coal, gas as a tracer of the hawk us, we can have some plasma flow.
08:58:26 So, earlier I said, turbulence cascades, all the way down to dissipation scale. where is this dissipation scale.
08:58:35 If dissipation is due to speech or viscosity, we would expect as a patient to happen, above the mean for a pass at the so called come up or off scale.
08:58:47 and the expectation is shown as this dashed line.
08:58:53 What you see here is we see turbulence below where you don't expect turbulence.
08:59:02 If it's dissipated.
08:59:05 With as a topic that's constantly, which means.
08:59:09 So traffic because these suppressed.
08:59:13 And even though particles are no longer directly interacting with each other on small scales. They're still effectively interacting with each other so the plasma is still effectively conditional, and it's probably because of particle may feuds interaction.
08:59:32 And the previous constraints the best constraints from previous does come from your Ashraful was surface brightness fluctuations. I just want to point out that that was obtained was 12 day launch hunger observations of Kumar cluster.
08:59:57 cheap, a poor person's mindset. I'm very happy to see that our analysis is much cheaper. And we can push down to a smaller scales to constrain the viscosity even better.
09:00:05 Now that I've been converted to an observer, I want to look at more systems. I want more data. So a student of mine is currently analyzing more cluster of data compiled by malaria, all the various, and hopefully I'll have some update for you soon.
09:00:28 And lastly, I just want to briefly mention some of the other correlation functions that people have used in CGM studies for a while, I thought that I was the first person who did too bossy structure function and analysis and CGM observations.
09:00:45 But, factories, he pointed out to me that, oh no I was wrong, up to about two decades ago, people had already done it.
09:00:56 So, and Raj at all, they, they looked at the bosses director function in the CGM using absorption lines.
09:01:06 Now I do know that Cassie Lucas used boss charter function to analyze turbulence in the CGM simulations number drama fielding.
09:01:18 In fact, my first version of the script was based on categories script and just work has looked at density correlations of ions in the Milky Way, and very very recently, and she at all.
09:01:49 with Joe Brockman has looked at both the density correlations and velocity correlations for absorption lines in the Milky Way. And I'm very much looking forward to the discussion.
09:01:54 Next week, hope that they will talk about these interesting results.
09:01:58 So, to summarize the CGM of the biggest galaxy. Biggest galaxies. Okay. The ICM is turbulent and different faces up here Walt capote, and the turbulence is mainly driven by the supermassive black hole feedback at least in the central region of the galaxy
09:02:19 closers, and in turbulent dissipation seems to be a sub dominant heating source.
09:02:25 So traffic viscosity is surprised.
09:02:28 And lastly, on to point out the correlation functions are a lot of fun and easy to do. So they said in the perfect position of that pain diagram that he showed us yesterday, fun and easy.
09:02:44 and they're also very useful.
09:02:46 And I also want to say that it's very important for us to learn from each other. That's why I'm so in love with this workshop, I'm learning so much and we can really learn a lot from the SM community.
09:02:59 Thank you very much.
09:03:02 Thank you very much, you on that was that was a great tutorial, and really broke down some of the things certainly that, that I haven't understood in the past so hopefully other audience members got something out of it too because I think it was really
09:03:14 excellent.
09:03:16 There are a number of questions. I encourage people to raise their hands will take questions for a little bit and then move everything to the breakout room and the next like 1015 minutes.
09:03:25 I'm going to take moderators privilege and ask my question first because I'm pretty excited about this.
09:03:32 So, the stuff you were showing with the the ICM and the inform and you looked at the, the, the function of the drop off.
09:03:42 In, in where the motions were.
09:03:45 What do you expect. Do you expect a similar power law for or scaling for involved of galaxies into a halo and then the stripping of gas behind those galaxies, so you know you have your, your main Halo, you have a bunch of installing satellites that go
09:04:04 in, they depart they get rammed pressure stripped or tightly strip their their debris, so you'd also expect some sort of power loss, to, to define the behavior of the gas, does it.
09:04:19 I mean, I don't know.
09:04:19 I don't know what that would give but presumably that's different than the coma garage turbulent scaling.
09:04:26 I think you're thinking about turbulence on top of folk motion.
09:04:31 Yes, like be a cut off at small scales because this scale size of the individual streams that are coming off that I don't know, it just seems.
09:04:40 I wish I could tell you, but I have no idea how to remove Bob motion.
09:04:48 And I'm working on it.
09:04:50 I've been thinking about it
09:04:53 for the cases I was looking at I was lucky that I could argue does not hold out of fog motion right you can see it from the just, we'll all see maps, but there might be a little bit of rotation there.
09:05:07 I think I just sort of wave my hands and said you know whatever bulk motion, there is a true cascade. I just goes into the driver of turbulence.
09:05:26 I agree there could be bulk motion, that's just bulk motion.
09:05:27 Okay, yeah, if anybody has good ideas on how to remove bulk motion, let me know.
09:05:35 Alright the first person who had his hand up was Frank Vanden Bosch Would you like to ask a question.
09:05:41 I want excellent talk thanks. I'm just being in an internal critic skeptical let's put it that way. I was wondering, the results that you showed with this power law scaling relation, based on the velocity measurements.
09:05:56 Of course, there's two issues that I would worry about, and one is projection of facts we're seeing this in projection, for example if you know, if for some reason it took this will be largely confined playing or then you know depending on the inclination
09:06:08 and global see different slope.
09:06:10 In my inferred coma or a spectrum. So what is the issue there and the Secondly is is seeing you have a finite resolution element and that will always of course at small scale will drive your delta V down and so what's going to extend to several fluid
09:06:24 half max of OKBC so so how do you, how are you sure you're not just measuring a seeing function.
09:06:31 Excellent, excellent questions. Those are the two questions that can be lots and lots of headaches and
09:06:40 I, I'll just say, okay, for the first one the protection. In fact, I'm here, I totally was you were looking at everything, and the projected plane of the sky, but all of these structures are presumably 3d structures.
09:06:57 And you can work out sort of analytical and other things so y'all She recently did some work where she showed the analytics of how the projected structure function should be related to the 3d structure function.
09:07:10 If we assume that this thing is reasonably sin than the protected structure function should be the same as the 3d structure function.
09:07:19 If it's not sin.
09:07:22 The protected one would be flatter than the real one. And in this case, because we're already seeing steeper than Comoros soaps. In projection, the real soaps can only be even Steve PR.
09:07:38 So I won't change the conclusion aa Well, it might change the exact soap.
09:07:44 But I don't think that in any time soon we can actually remove this project in fact just because we don't really, we, we can't figure out the 3d structure of these things.
09:07:55 Now the second point is another excellent point about seeing a, and it's so funny that comes from your theorist, but all.
09:08:08 At first I was like, you know, I don't, I mean I don't think seeing was doing anything because seeing what affects smaller scales and I didn't see any change of soap at small scales, but actually our referee asked about seeing again.
09:08:26 and that's fine.
09:08:43 So I did an experiment where basically I asked the observer co authors to run the data through like a not a seeing simulation to just blow it up again, and compare the structural function of the further blurred up map, and the one, The original one.
09:08:51 And
09:08:51 actually seeing surprises the overall amplitude by a little bit, just a teeny little bit. Um, without changing the slope.
09:09:02 So that was a little bit unexpected I realized oh ok so seeing doesn't work like step function, it's like, I've got ocean thing so everything gets smoothed out a little bit.
09:09:11 Now if you ask me.
09:09:13 How does this compare with the original pure date off photos without anything in fact I don't know because I don't have a space telescope that could give me.
09:09:27 It'd be, it'd be great if if there's a tool to generate of like the real spectrum, the real have your observation, out of a simulated cluster, then we can really try to better understand the same effect.
09:09:40 That's a good question. Thank you.
09:09:45 Excellent. So our next question is from Christoph from.
09:09:50 Yeah, thanks. Excellent talk and graver, I really love this work that he does.
09:09:57 Now regarding Of course interpretation that's the sort of really exciting part and, and I think they're the jury's still out but I think recently john one one has written a very nice paper.
09:10:12 I'm a great student of materials Kosky that offered actually an explanation of the steeper slope, and his explanation was that the code phase velocity structure function is driven by is this due to turbulent motions primarily driven due to the gravitational
09:10:26 attract acceleration of ballistic clouds and this interestingly happens to accept you give you the observed slow. Now in this picture, sort of, then the hga would sort of, sort of, excited or are sort of triggered because of dictation so that's why you
09:10:50 kale. And then once you sort of have to code clouds that up, then ballistic be moved in in class the potential drive turbulence. And of course the big question is, how much in this picture do couple of hot pretty cold face, what is your view on this sort
09:10:58 sort of somewhat different picture than you propose.
09:11:03 Um, yeah, I actually, I thought it was the opposite so I thought that my coat flops were to ballistic, and the gravitational acceleration would work against having a steeper slope.
09:11:17 So my thought is.
09:11:21 And I'm so sorry for you feel full on writing off that paper, but my thought is that when you have gravitational acceleration on these co blobs and making them going ballistic, you're going to have basically blobs passing each other at very high velocities,
09:11:35 without knowing about the existence of each other. So then when, when you run the structure function, you would have very large was the difference at small separations.
09:11:45 And that's just because they're not connected within Eddie's.
09:11:49 So I actually thought sort of the opposite.
09:11:58 I mean it's.
09:12:00 I don't think anybody knows our.
09:12:04 At this point, nobody has really convinced anybody else.
09:12:09 Yeah, and you should look at Mark White's new idea that's also really interesting and warnings is very convincing.
09:12:20 Okay.
09:12:23 Uh, we'll take one more question, I think, and then move to the breakout room with this discussion, Marcus.
09:12:31 Hey, yeah.
09:12:32 So, who said, The other cool thing is I have a couple.
09:12:37 So what's your, what's the evidence for this didn't quite get this, and then you, then you refer to me see simulation starts going show this. What are the parameters within which staff actually works.
09:12:54 Good question. It so indirect evidence right the direct evidence will come from, hopefully, far future X ray observations where they actually measure the spectrum of the hot gas.
09:13:09 Right now it's so indirect I mean, at the driving scale or right below the driving school where you really not controlled.
09:13:16 There's excellent agreements between the two faces. And then at smaller scales. My argument is that.
09:13:26 And in the case of poor coupling like my own have a pure hydro simulations I see a flatter so then column world.
09:13:35 But then the observed So, and in better capote simulations like tower us. I mentioned you simulations you see a more steep steep slope. And so it's indirect.
09:13:50 I know I can't show directly that they're there well couple of all the way down to dissipation skill.
09:13:58 So, fair enough. Cool.
09:14:01 Um, I think we have time for one more question, particularly since Mateo she's about to give his tutorial and he's next in line, I'll let him ask the question since he won't be able to join the discussion afterwards.
09:14:13 But there's a boy, a young white time to prophesied.
09:14:28 Who said I will just comment on this in a moment. Anyway, so the perfect. Okay.
09:14:22 divert up. Hi, you are excellent talk and that was a that is an excellent work. I this is less of a question more like a verification of what I see in my simulations.
09:14:35 And I see that steepening of slow in my cluster simulations.
09:14:40 But the stiffening of slope depends like where it is depends depends on like at what stage the Asian outburst is, like, if the Asian is full blown as deepening is like more like a 10 to 15 k pieces care while like the at, you know, the stage where the
09:14:54 agents have gone out and it's more like a cooling phase now, then this deepening is much more closer.
09:15:06 So, I do see this deepening of slope, but the scale at which this deepening happens from Colorado grow.
09:15:14 It changes with the state of the Asian feedback, interesting, I would say that I agree that was my bad driver theory.
09:15:27 Okay, so we will take a five minute rest you on and anyone who is interested in continuing to discuss turbulence and turbulence signatures in the in the halo are welcome to join the appropriate breakout room where discussion will will continue.
09:15:46 And, yeah, we'll have a short rest, and then in five minutes Mateo will get started with his tutorial on AGN non thermal processes.