Episode 41: How to Train Your Galaxy

with Dr Rob Yates

We are joined by Dr Rob Yates who chats with us about how to recreate the formation of galaxies inside a supercomputer!

Rob is a research fellow working on the chemical evolution of galaxies at the University of Surrey, and formerly at the Max Planck Institute for Astrophysics, Germany.

He explains how he uses semi-analytic simulations to try and understand how various elements are created and dispersed around galaxies. This helps us to understand things like where the oxygen we breath came from and how gold and silver were created.

A simulated portion of the Universe. Robertson, B.E., Banerji, M., Brough, S. et al. Galaxy formation and evolution science in the era of the Large Synoptic Survey Telescope. Nat Rev Phys 1, 450–462 (2019).

This week’s guest

Related Links

Dr Rob Yates website: https://robyatesastro.wixsite.com/robyates


[00:00:00] Jacinta: Welcome to The Cosmic Savannah with Dr. Jacinta Delhaize

[00:00:08] Dan: And Dr. Daniel Cunnama. Each episode, we’ll be giving you a behind the scenes look at world class astronomy and astrophysics happening under the African skies.

[00:00:16] Jacinta: Let us introduce you to the people involved, the technology we use, the exciting work we do, and the fascinating discoveries we make.

[00:00:23] Dan: Sit back and relax as we take you on a safari through the skies.

[00:00:34] Welcome to episode 41.

[00:00:36] Jacinta: Today, we’ll be speaking with Dr. Rob Yates about chemical enrichments and computer simulations

[00:00:42] Dan: Very exciting and right up my alley. We haven’t spoken about simulations for a while.

[00:00:46] Jacinta: It has been a very long time. I think the last time was maybe with Romeel Dave and also with Nicole Thomas back in season one.

[00:00:56] Dan: Wow! Yeah, and we spoke to Rob about something, which is slightly different to what Romeel and Nicole spoke about. They spoke about hydrodynamic simulations and today we’ll be talking about semi analytic simulations.

[00:01:10] Jacinta: Now, if you are like me listeners you’ll have no idea what those two things are. Rob explains it quite nicely and with Dan’s help. But Dan, do you want to just prepare us with a very broad overview? What is a simulation in the context of astronomy and what are these two different types that you’re talking about?

[00:01:28] Dan: Sure. Let me see if I can do it in 10 lines. So basically in trying to describe how galaxies form and evolve we have theories. Such as gravity and various other laws of physics, which we can use to predict how matter, dark matter will move through space and will form into clumps, which form galaxies and that’s where stars form. So we have all of these laws. They are very complex. And when they interact on a scale of a galaxy or on the scale of the universe, it becomes a very big computational problem trying to work out the forces between all of these things at any given time and how they’re going to move is very, very complex.

[00:02:20] So what we do is we build simulations. Where we put essentially particles, which are big, hundreds of thousands of times the mass of our sun, each particle. And we let them interact under the laws of gravity, which we’ve coded into the simulation. So that’s a very basic simulation. And what you can do then is what Romeel and Nicole do is something called hydrodynamic simulations, where in addition to just the dark matter, and interacting through gravity, you add things like gas, stars and the winds that come from stars and supernova and things.

[00:03:01] And you try and follow all of these particles too. So it gets, a whole lot more complicated, but it’s still based in physics and trying to track individual particles. Semi-Analytic models, on the other hand, use the dark matter simulations as a skeleton so you can see where the dense areas are, where the galaxies should form, and then they take a more analytical approach.

[00:03:27] So basically a kind of a simple model of each of the processes, which will happen today. So today we’re going to be speaking to Rob and Rob talks to us about chemical evolution. So basically what he’s doing there is trying to work out, in a large skeleton of a simulation, given certain processes which will happen, how much of each gas would form, you know, if you had X many supernova, you would have this much carbon, for example.

[00:04:02] And then they do that across the entire simulation. How was that?

[00:04:06] Jacinta: Well, it was a bit longer than a few sentences, but I think that made a lot of sense. Thanks Dan! And don’t worry listeners, Rob will go through that all again. So you’ll get another chance to follow that on. But basically all you need to know is that part of astronomy is not just observations with telescopes, but also simulations in supercomputers that do these millions of calculations over time, which would take a human, I don’t know how many lifetimes to actually calculate by hand.

[00:04:35] And that creates a simulated universe. And we compare that with the real universe that we see, and then that helps us to understand what’s going on better. So the observations that we’re taking inform what goes into the simulation and then the results of the simulation then also inform what we choose to look at with telescopes. So you kind of need both of these things in order to really understand what’s going on in our universe. Yeah. And as you said Dan, Rob is using simulations to look at chemical enrichment, which is how all of the elements get where they are in the universe.

[00:05:08] How did all of the elements end up on earth? Like why do we have oxygen and carbon and gold and silver and any element that’s on the Earth. Like, how did it get here and where did it come from? You’ve heard the expression we’re all made of stardust. And so Rob will answer , will explain that a bit more in the chat that we have with him.

[00:05:30] So I guess without further ado, shall we just hear from Rob?

[00:05:33] Dan: Yeah, we should point out that when we spoke to Rob, he was still a post-doc at the Max Planck Institute for Astrophysics and Garching in Germany. He since moved to the University of Surrey, where he is now a Research Fellow.

[00:05:44] Jacinta: With us here now is Dr. Rob Yates from the Max Planck Institute for Astrophysics. And he’s going to tell us where the oxygen we breathe actually originates from. Welcome, Rob.

[00:06:03] Rob: Hi, thanks for having me.

[00:06:04] Jacinta: Rob, can you just start by telling us a little bit about yourself, who you are, where you’re from?

[00:06:08] Rob: Yeah, sure. So I’m a postdoc at the Max Planck Institute, as you said, I come from London, in England. And I’ve been working in Garching at the Max Planck now for quite a while. I did my PhD there and now I’m a post-doc again, working on the chemical enrichment of galaxies and how we model that.

[00:06:25] Jacinta: And what brings you to South Africa?

[00:06:28] Rob: I have some collaborations here with the university UCT. So for example, I’m teaching one of the courses at the University.

[00:06:35] So that allows me to make some, some trips over here throughout the year, which is really nice.

[00:06:40] Dan: And you’re also supervising some students here.

[00:06:42] Rob: That’s right. So I’ll be starting supervising a Master’s project here, which will begin soon. And we’re going to be using the models that I worked on in that project that the student will get to do some galaxy evolution modeling.

[00:06:54] Dan: So what exactly is chemical enrichment?

[00:06:57] Rob: That’s a great question. It mainly involves us trying to understand how the heavier elements, the heavier chemical elements in the Universe were formed and how they were distributed throughout galaxies and how they eventually ended up on the Earth in the air we breathe and in our bodies itself.

[00:07:15] Jacinta: What do we mean by heavier elements?

[00:07:17] Rob: Right. So typically in astrophysics, we consider anything heavier than helium as a heavy element. And the reason for that is that it’s those elements that are formed in stars. So actually it was about a hundred years ago this year. This proposition was first made by Arthur Eddington when he suggested that the energy required for stars to shine so brightly is formed through the fusion of lighter elements into heavier ones. And so we now have this firm theory that the elements heavier than helium, the heavier elements formed in the centers of stars.

[00:07:53] Dan: So we know that most of the hydrogen and helium formed in the Big Bang and the origin of the Universe, and then everything heavier than that had to originate from stars. And this has done to kind of sequentially building up with the fusion process. So if we have a picture of an individual star fusing these light elements into heavier elements. How does this then enrich a galaxy?

[00:08:23] Rob: Alright! So, I mean, if we take for example, oxygen, which is a pretty useful element here on Earth, we need that to survive. For the energy that the cells in our bodies utilize. That particular element is mainly formed in massive stars in the cores of massive stars through what is known as the alpha process. So that is simply the fusion of a helium nucleus with a carbon nucleus. And you need very hot dense environments, like the centers of massive stars for that thing to form. But of course that’s not the whole story because it’s all well and good forming these elements.

[00:08:58] But if they remain locked up in stars, they’ll never get out into the galaxy and eventually into, planets like our Earth. So what happens in these massive stars is that when they finally run out of elements to burn and form fusion, they collapse under their own gravity. These things are very, very massive systems.

[00:09:18] And as they collapse in on themselves, their cores become denser and denser. And then you get this core bounce effect. Where the energy from that collapse and a lot of neutrinos that have been formed during the collapse explode the outer layers of this star and those outer layers contain some of that oxygen that we have today.

[00:09:39] So it spewed out at great speeds, across the galaxy

[00:09:42] Jacinta: Supernova!

[00:09:44] Rob: Supernova, exactly. So in the case of these massive stars is a type II supernova or core collapse supernova. And that’s where most of the oxygen starts its journey from its creation in the centers of those stars, to where it is today.

[00:09:57] Jacinta: So if we didn’t have supernova explosions, so dying stars, we wouldn’t have oxygen.

[00:10:02] That’s right. That’s right. Without the synthesis of those elements in stars, we wouldn’t have any life on Earth today. But of course, that’s just the start of the story and there’s lots of other, there’s more events. There’s other events that have to take place to get it from this exploded supernova into our bodies, into the earth as well.

[00:10:21] Dan: And presumably the, the oxygen that resides here on Earth didn’t all come from one supernova that went off nearby, right? The Earth is four and a half billion years old, and the universe is 14 billion. So, there may be a few generations of stars, which have sort of contributed to the oxygen abundance on earth.

[00:10:41] Rob: That’s right. So that’s where the types of models that I work on come in. Once we understand or think we understand how the oxygen has exploded out in supernovae, we have to then track how more stars form and more supernovae go off over as galaxies evolve. So this is the main part of the models that I work on.

[00:11:01] We look at how galaxies evolve over the last 13 or so billion years. And how, the chemistry of those galaxies changes over time.

[00:11:11] Jacinta: You work on simulating galaxies and galaxy evolution. Is that right? Okay. So can you just give us a rundown of, of what a simulation is and why you use it.

[00:11:22] Rob: Sure. So there’s a wide range of simulations, of course, but they kind of all, there are the objective of all of these is to model a large fraction of the universe over a large amount of time.

[00:11:33] So like I said, typically these things try and model the evolution of galaxies from about 13.5 billion years ago down to today. And we do that using huge computers.

[00:11:45] Dan: So we’ve spoken before about observations with large spectrometers, such as SALT and from this, we can observe distant stars and other galaxies and kind of measure how many, you know, how much oxygen or how much other elements are, are in there.

[00:12:04] In terms of your models, you obviously need to align with the observations and try and build in some sort of evolution across time. So we’re looking back further and further with a big telescope like SALT at early galaxies. Do you see a difference on chemical abundances and early galaxies to current galaxies like the Milky Way?

[00:12:26] Rob: Yes we do. So the understanding that we have is that as galaxies evolve over time, more and more stars will form. And that means there’s more opportunity for supernovae to explode and dump even more oxygen into the galaxy’s gas. So over time you build up the amount of oxygen and other elements that you have in galaxies.

[00:12:46] Jacinta: Okay. So you’ve got the first supernova goes off, it’s created oxygen and its core. So the oxygen atoms and molecules get thrown out into space after that explosion. And then what happens?

[00:12:57] Rob: That’s a great question. So it’s an ongoing area of study, particularly in models to understand how this works after that point.

[00:13:05] So for example, one of the things that simulations have managed to determine is that some of that material that’s exploded out of supernovae actually has to leave the galaxy altogether. The belief is that the supernovae seem to combine together. If many of them go off simultaneously in a similar amount of time, they can drive winds that actually throw material right out of galaxies, into deep space.

[00:13:29] And not only is this observed with observations of galactic winds and galactic outflows, but it’s required in models to get the masses of galaxies right. If we don’t allow some of that material, some of that oxygen enriched material to leave galaxies, then we formed too many intermediate mass galaxies in our models.

[00:13:49] Jacinta: So there’s these big galactic winds in a galaxy and they’re sweeping material out of the galaxy. And you need to account for that in your simulations. You need to tell you stimulations, okay, this galaxy has got some wind and it’s removing mass because otherwise in your simulation, you grow a galaxy that’s too big.

[00:14:07] Rob: That’s right. If the gas doesn’t leave, it’s hanging around and can very easily form stars again. And so you find that you get your galaxies too big. If you don’t have these winds..

[00:14:17] Dan: And there’s a certain percentage of that gas, which is enriched, right? So a certain percentage of the elements you’re forming oxygen and the like are getting expelled from a galaxy altogether.

[00:14:27] Rob: That’s right.

[00:14:28] Dan: And where do they end up?

[00:14:30] Rob: So those elements that leave the galaxy all together can drift out into deep space. And so there are observations to suggest that there are some heavier elements found out there in deep space. And some of them, of course, will return. So there’ll be driven in slightly weaker, galactic outflows, and this material can be accreted back onto a galaxy driven by gravity, and it can then cool again and form into new stars at a much later date.

[00:14:57] Dan: So we’ve spoken a bit about oxygen and how it sort of forms in the core of the star and then gets distributed through a supernova. Do all of the elements form in this way.

[00:15:07] Rob: Not all of them. So we were focusing mainly on massive stars for oxygen, right? There are other elements, lighter elements, such as nitrogen, carbon. Which also have a decent amount of formation taking place in low-mass stars. And then there are some heavier elements, for example, like iron, which are predominantly formed in different type of supernova called a supernova IA. so about 60 to 70% of all the iron in the universe is formed in these types of systems.

[00:15:35] Dan: So in your model you need to account for many different types of stars, which are contributing many different types of elements in order to build a complete picture of how a galaxy is getting enriched, chemically.

[00:15:48] Rob: That’s right. We need to account for all these different mechanisms. And then we can compare the abundance for example, oxygen from type II supernovae from core collapse, supernovae with the amount of iron from type IA supernovae.

[00:16:01] Jacinta: And why does it matter? Why do we need to know the ratios of heavy metals of oxygen or iron in a galaxy?

[00:16:08] Rob: It’s a very important diagnostic to understand if our models are reproducing, what we observe, for example. So elements like oxygen and iron are relatively easy to measure. Of course there’s complications with any observation, but they’re relatively easy to measure in the stars and the gas of galaxies.

[00:16:25] And so we can look at their ratio and compare it between our model and observations to see if we seem to be getting things right.

[00:16:31] Jacinta: And that’ll tell you whether you’ve put the right parameters into your simulation in the first place. Is that right? Whether you’ve got the chemical enrichment process, correct.

[00:16:38] Rob: That’s right. So it helps inform us of parameters and by parameters, we mean efficiencies of certain processes that we don’t fully understand that we have to parameterize, we have to assume are working at a certain rate. We can check if those parameters make sense with these results and can also see if we need to have additional physical processes included in our models. Maybe we’re missing something that needs to be added.

[00:17:03] Dan: So, what are the, some of the particular project you’re working on with regard to chemical enrichment?

[00:17:08] Rob: So for example, one of the results that I’ve I’ve worked on recently is looking at the ratio of this oxygen to iron and different components of galaxies in the stars and the gas and in the hot gas, within galaxy clusters.

[00:17:20] And from this, we can do things like constrained. For example, the type 1-A systems that originally exploded and enriched those galaxies with iron.

[00:17:30] Jacinta: So before you said it was type two, and now we’re talking about type one, what is the difference here?

[00:17:35] Rob: Yeah, so a type two supernova is effectively a type of core collapse, supernova. Like we described with that massive staff that run out of fuel and collapsed. A type 1-A supernova is a slightly different system and we don’t really understand what type of stars led to type 1-A the classifications two and 1-A are purely observational. In fact, they’re not much to do with the physics of these systems and type 1-A is just a supernova that seems to have no hydrogen in its debris. A type two does have hydrogen in it.

[00:18:07] Dan: And this points to a slightly different supernova. The explosion from a slightly different origin story for the star, right?

[00:18:14] Rob: That’s right. So one of the things we can do with my models of galaxies is try to constrain what might be these formation mechanisms of type 1-A supernovae.

[00:18:25] For example, we have found that you need to have a certain percentage of these type 1-A supernovae exploding quite quickly, within a hundred mega years of the stars forming. And if you don’t have about 20, 25% of them going off quickly, you don’t get the right ratios of oxygen to iron in your galaxies compared to observations.

[00:18:45] Jacinta: And what’s the main theory about what’s causing the type 1-A supernova.

[00:18:49] Rob: So there are two main competing theories both or neither of which could end up being true. The first is called the single degenerate scenario. You have a white dwarf next to a more normal star.

[00:19:00] Jacinta: So white dwarf being the core of a dead middle-size star?

[00:19:05] Rob: Right? So that’s an evolved star at the core of an evolved star, normally containing predominantly carbon and oxygen. And the interaction between this and a normal star can lead to the mass of that white dwarf, getting above a critical level where it explodes. So that it detonates.

[00:19:21] Jacinta: Detonates?

[00:19:22] Rob: Exactly like a huge bomb in space.

[00:19:25] That’s the single degenerate scenario, the double degenerate scenario for type one, a supernovae is two white dwarfs merging and causing an explosion. Because again, that takes them above some critical mass.

[00:19:37] Jacinta: Cool!

[00:19:39] Rob: Yeah.

[00:19:41] Jacinta: To bring it home. So you’ve got oxygen forming in stars, getting exploded out in supernova.

[00:19:46] And if there’s lots of supernova going off causing a big wind, which blows it all out of the galaxy, and then it rains back down through gravitational interaction or whatever processes, how does it get to us on Earth? What happens then?

[00:19:58] Rob: Right. So as this oxygen builds up in galaxies over these millions and billions of years, we continue to have stars forming and what people seem to think, stepping back from my particular research, and what people seem to think is that a supernova went off near the birth cloud of our sun and the shockwave from this supernova caused that cloud to collapse and start forming a protostar, our Sun in its very early form and also a proto-planetary disc around that star, which is where our Earth was formed.

[00:20:30] So you have this debris from many generations of supernovae containing that oxygen that slowly accumulates into dust and then pebbles or rocks and then eventually our earth orbiting the sun. And there you have that journey from the center of a massive star, many billions of years ago, all the way back down to the contents of the Earth today.

[00:20:50] Jacinta: So I’ve heard that if you have more of these heavy metals in a, in a gas cloud, that becomes a proto star system that it helps actually to form that star. Is that true?

[00:21:03] The amount of metal in, by metals I guess we should say heavy elements, right? The amount of heavy elements in a star forming cloud does affect how it forms stars. And it can also affect how quickly those clouds form.

[00:21:18] So with the cooling of gas, for example is more efficient when you have heavier elements in that gas.

[00:21:25] So we need a cloud of gas to cool down in order to condense and form a star. Is that right? And so then the presence of the metal somehow helps it to cool.

[00:21:34] Rob: Yeah. So metal atoms are a good method for cooling and astrophysical environments because when they are excited, by the collision of some other atom or by a photon, they can release photons themselves.

[00:21:49] And because these systems are relatively low density compared to the densities of air on Earth, for example, those photons can then leave the gas cloud, and that is taking energy, taking heat away from a gas cloud, and that can allow it to cool more quickly,

[00:22:03] Dan: What are some of the challenges that still exist in understanding the chemical enrichment of galaxies and how astronomers?

[00:22:10] Rob: So there are challenges still in both the theoretical and observational side. So one of the biggest limitations, I would say for the chemical evolution modeling that I do is understanding exactly how much of each of those heavy elements are blown out by supernovae in the first place.

[00:22:28] So we have astronomers doing models specifically on the evolution of stars and they tell us effectively how much oxygen is going to come out of my type two supernova. And these different groups don’t necessarily agree with each other. And of course, different values for the amount of oxygen per supernova will have a big effect on the final oxygen abundance that we get for our galaxies and our simulation.

[00:22:52] So that’s an area of continued research on the theoretical side. On the observational side is actually not as straightforward as it might seem to measure how much oxygen there are in stars and galaxies. We can’t just count the number of oxygen atoms as we would like to do. We have to look at spectra, for example, emission lines from oxygen atoms and the strength of those emission lines depends quite a lot on a number of different processes that we can’t completely constrain.

[00:23:20] So we’re not even sure really how much oxygen we should expect from our models at different points in the evolution of the Universe.

[00:23:26] Dan: And is SALT contributing to this? Are you working with SALT at all?

[00:23:30] Rob: I have done. Yes. So SALT is a, spectrograph a really great spectrograph he had done in South Africa and that allows us to try and measure the heavy element abundances, for example, in the gas of galaxies. And it can allow us to measure it in a more accurate way than is usually possible by allowing us to measure very faint emission lines from heavy elements, like oxygen, which you can’t necessarily do with other types of telescope. And so that’s helping us constraint better, how much oxygen we really expect out in galaxies.

[00:24:00] Jacinta: What about the heaviest elements, even getting out beyond iron. We’ve had some pretty exciting discoveries in the last few years about gravitational waves and things. So how does, how does things like gold and silver and all of that get formed?

[00:24:15] Rob: Yeah, the super impressive elements, gold, platinum. Even some highly radioactive elements are usually not formed in the types of supernovae are in the centers of status that we’ve discussed. They can be formed for example, by the merger of two neutron stars, which is one of the sources of the gravitational wave detections that we’ve heard about in the last few months. These neutron stars themselves, however, actually formed in the collapse of a massive star.

[00:24:40] So when that core runs out of fuel and the core and starts to collapse in on itself, it actually gets so dense that the protons or neutrons inside that core are actually touching each other. There’s no space between them. And that is a neutron star. It’s a super dense, heavy thing. To get a handle on how dense it is, it’s about a hundred trillion times more massive than the sun. If you had one teaspoon of neutron star here on Earth, it would weigh as much as 15 moons. So this is a pretty, a pretty dense little thing.

[00:25:14] Jacinta: A hundred trillion?

[00:25:15] Rob: Yeah

[00:25:16] Dan: That’s a hard number to get your head around.

[00:25:18] Rob: That’s a lot of zeros on the end of that number. And these things are formed in those massive stars. And then later on, they’ll go on perhaps to merge with each other, giving off these gravitational waves. And with that huge amount of energy forming things like gold and platinum.

[00:25:31] Jacinta: So the ring that I’m wearing on my finger right now, it’s got some gold in it. So you’re saying that this is actually neutron star dust.

[00:25:38] That’s one of the mechanisms that we believe could have formed the gold.

[00:25:42] Dan: So at the center of these stars, which then become neutron stars, obviously will be forming all the way up to iron, these elements, but then as it collapses to a neutron star, the protons and electrons and those atoms actually get completely destroyed and form this core of only neutrons. And then only when that explodes through a merger with another one, do these very heavy elements form.

[00:26:07] Rob: That’s right. So you have a huge number of neutrons there available that you can then fuse together into very massive elements. If you have enough.

[00:26:16] Jacinta: So it’s actually the dust of a neutron star explosion, not a neutron star itself, in my ring

[00:26:23] Dan: or of new two neutron stars

[00:26:26] Jacinta: Two neutron stars.

[00:26:27] Dan: I mean, that is definitely a subject for a whole another podcast. we should get into the nuclear physics of gravitational force.

[00:26:35] Jacinta: Getting off track a little bit. Sorry. Right.

[00:26:37] Rob: Interesting stuff.

[00:26:38] Jacinta: Now, I know that what you do is called semi analytic modeling, is that right? Or SAM, and there’s another type called hydrodynamical simulations. And we’ve been talking about those in some previous episodes and with Romeel Dave and other people. And so can you just explain what the difference between those two is, and what’s the benefit of doing semi analytical modeling?

[00:27:03] Rob: Sure. So the name is a bit cryptic, right? Semi analytic. What does that mean? So effectively we are using old school equations, analytic equations to describe how the material in galaxies, the baryonic material in galaxies switches from gas to stars into a black hole out of the galaxy.

[00:27:27] So that’s the analytic part. The semi comes in in that we are for the dark matter that surrounds these galaxies, using numerical simulations to form the dark matter halos in which these galaxies evolve and that’s because it’s not actually possible to calculate how these dark matter halos form purely from written equations, from, from analytics, you need to use numerical solutions for that.

[00:27:52] Jacinta: And so a dark matter halo is like a big clump of dark matter, right?

[00:27:56] Rob: Yeah. So a dark matter halo is something that forms before galaxies form in the universe. And it provides a nice gravitational sink in which the matter that we see and interact with every day can fall in and start to form galaxies.

[00:28:10] Dan: And solving the physics of the gravity of these things and how they’re formed is obviously very complex. As you said, you can’t solve it with a single equation. So what you need to do is set up a field of particles, which then you can allow to interact using the forces of gravity time step of after time, step after time step and see how they move under the influence of gravity and therefore form your underlying simulation.

[00:28:39] Rob: That’s right. And then we take the analytic prescriptions for how normal matter forms and put it on top of that. And one of the advantages of this semi analytic method is that it’s quite efficient. So we can run an evolution of the whole Universe effectively, many billions of years in a few days, compared to simulations, which independently model the visible matter as computational particles as well, that takes a much longer time to do

[00:29:09] Jacinta: So again, what exactly does analytic mean?

[00:29:14] Rob: So by analytic, we mean that we can calculate how the amount of mass and energy that is present in galaxies changes over time using an equation that we can write down. So you might remember back from school, perhaps differential equations, how a property changes over time.

[00:29:34] Dan: I’m not sure how many people remember differential equations. What school did you go to?

[00:29:40] Rob: Yes. So we can use simplified equations that we can write down and solve to understand how the amount of oxygen for example, goes from gas, into stars and elsewhere in galaxies.

[00:29:53] Jacinta: Okay. So that’s different to what you would do in a hydrodynamical simulation?

[00:29:57] Rob: In a hydrodynamic simulation. They have computational particles, these little billiard balls inside the computer, and they calculate how these things move around and interact following our understanding of gravity and also hydrodynamics. So how fluids behave effectively. So each of those little computational particles is independently modeled inside those simulations.

[00:30:20] Dan: So if you were modeling a, an individual galaxy say, in hydrodynamics that galaxy would be made up of say tens or hundreds of thousands of individual gas particles each with their own properties. So they have mass temperature, chemical abundances, and then they interact and sort of share temperature, share chemical abundances.

[00:30:45] And we see how that evolves using physical laws over time. Whereas in a semi analytic model, you’ll look at the model of an individual galaxy and you’ll say that for a galaxy of the size, say the size of our Milky way. We know that over time it will have X amount of oxygen and then that’ll change sort of broadly over it’s evolution. So it’s a sort of simpler explanation for the global properties of the galaxy rather than individual particles inside.

[00:31:18] Rob: That’s right. So it’s like a lower resolution version of a hydrodynamic simulation. So in a hydro sim, each particle will have an amount of oxygen. Okay, we’ll have a temperature and in a semi analytic model, a certain amount of gas within the galaxy.

[00:31:34] So many of those particles will be assumed to be the same. So it’s a lower resolution version effectively, and that’s why it runs much quicker.

[00:31:42] Dan: Than you can do it on a much larger scale, sort of universe wide scale.

[00:31:46] Rob: That’s right. And so you can model over 20 million galaxies, for example, in our simulation over 13.5 billion years, relatively quickly using this method

[00:31:56] Dan: And doing that in a hydrodynamics simulation will take 20 million years , probably.

[00:31:59] Rob: It will take a long time certainly.The equations that govern how the chemicals evolve in these simulations are actually the same, of course. But the difference is that you solve those equations for each individual, computational particle in a hydro simulation, and then you do it for a larger blob of gas for a semi analytic model.

[00:32:19] Dan: I hope that was clear.

[00:32:20] Jacinta: I’m slightly lost. Let this be a lesson to listeners. Don’t put two simulators in the room at the same time.

[00:32:27] Rob: Please send your questions to Dan,

[00:32:33] Jacinta: To Dan, Yeah! and to Rob. Okay. So right. I’m going to go back and ask another basic question. So, okay. I’m imagining little blobs, little particles flooding around in your simulation and you’re using equations to figure out what these blobs are doing. Now in a hydrogen dynamical simulation, each of these blobs is like a certain amount of gas or dust or stars, a star in a galaxy, right.

[00:32:58] And you actually calculate what each individual star, what each individual blob does. And then you figure out exactly, you know, using fluid dynamics and all of these things what’s going on in this galaxy. In semi analytic modelling, each blob is a whole galaxy?

[00:33:13] Rob: That depends on the semi analytic model. So for example, in the one I use the gas in the galaxy, the interstellar medium is split up into different rings.

[00:33:23] So the interstellar medium tends to form in a disc in a kind of a flat plate shape. And we split that up into rings. So our blobs is one ring of this galaxy is gas.

[00:33:35] Jacinta: Okay. And then, so would the resolution of your simulation, we know what a resolution is for observations. It’s how small features you can resolve of an object that you’re looking at, but in a simulation, resolution, what does that mean? Is that if you have higher resolution, your blobs are smaller and smaller parts of the galaxy?

[00:33:56] Rob: That’s right. So if we think about the mass resolution of these simulations, the typical hydrodynamic simulations for a whole evolution of many galaxies would have about a billion times the Sun as a typical particle mass.

[00:34:11] Dan: While we can do slightly better than that now.

[00:34:13] Rob: Okay. That’s great to know improvements always good. The typical hydrodynamical simulation nowadays we’ll have particles of about 10 million solar masses in size. Whereas a semi analytic model, it will depend on how much gas is in that ring. So we will have some rings with about that same amount, but of course we will have some which are considerably bigger than that. And so that means that we can evolve these simulations more quickly.

[00:34:40] Jacinta: Cool. Well, I think we’ll leave it there for today. Rob, thanks so much for coming in and joining us and speaking to us all about oxygen and how we are able to breathe and neutron stars and galaxies and semi-analytical models and all of these things.

[00:34:54] Rob: It’s been great. Thank you very much.

[00:35:03] All right. That was really awesome. So interesting to talk to Rob and to learn about something that I really, really don’t know anything about. So I was just fascinated, like 20 million galaxies over 13.5 billion years is what they can simulate in the semi analytical models. That’s amazing.

[00:35:21] Dan: Yeah, and I think it’s, you know, it’s just getting better and better so that the resolution that we can get to is just going to get better and better. And as we have better telescopes, which we talk about almost every week, we will get more and more data and more and more information on how these galaxies are forming and evolving. And all of that is data, which we can now feed back into these models and sort of link it all together and see how the universe is evolving.

[00:35:48] Jacinta: Yeah. And Rob actually mentioned the SALT telescope. Perhaps Dan, you’d like to just explain to our new listeners what SALT is.

[00:35:56] Dan: Sure. Yeah. It’s salt is the Southern African Large Telescope. It’s based here in South Africa at the South African Astronomical Observatory and it is the largest optical telescope in the Southern hemisphere. It is a 11 meter segmented mirror.

[00:36:13] Jacinta: Awesome. So Rob is involved a little bit with actual observations with telescopes.

[00:36:19] So we’ve gotten this far, Dan, very unusual. We jumped straight into the science without getting distracted, but now I’m going to distract us slightly. What have you been up to lately?

[00:36:31] Dan: So, well, I mean, I hope that listeners don’t turn off now. I don’t know if I’ve mentioned before that there’s a new planetarium, which has recently been built in South Africa in Cofimvaba in the Eastern Cape. So they have built a new science center there and it has a planetarium. So I was recently there to help set that up and we’ll be showing our new planetarium film at the premiere or at the opening of the science center.

[00:37:02] Jacinta: Oh, finally. Yeah. The, how soon is it coming to Cape town after that?

[00:37:07] Dan: Hopefully by November or so.

[00:37:10] Jacinta: All right. Awesome. And then eventually we’ll be distributed around the world.

[00:37:13] Dan: Yeah. So as with Rising Star, we will get it on the ESO platform where any planetarium around the world can download it and show it for free.

[00:37:22] Jacinta: Very exciting.

[00:37:23] Dan: Yeah. I think so. And yourself, what have you been up to just Jacinta? Free from quarantine, right?

[00:37:30] Jacinta: I am, I’m finally home in Western Australia and it feels amazing to be back in my family home with the family dogs and we live in the Bush land. So just chilling out in the Bush land and taking photos of the animals and the birds. And yeah, just really awesome. And in fact, there’s no pandemic here in Western Australia. COVID is not here. So because of all of the very, very, very strict quarantine laws here, so it’s life as normal and wearing no masks is a very, very strange concept. I’m still working from home because I’m just not used to the concept of going into an office to work.

[00:38:13] So I’ll, I’ll get there, but haven’t quite done that yet. Do you know, strangely, the one thing I’m worried about is wearing shoes all day.

[00:38:24] Dan: And is that for spiders or

[00:38:26] Jacinta: no, just like, if I go into an office, I have to wear shoes all day while I’m working. And they haven’t done that a couple of years now or so. Stay tuned to find out the exciting news about whether I wear shoes all day in the office.

[00:38:48] All right. All right. I think we can leave it there quite safely. So that’s it for today and thanks very much for listening and we hope you’ll join us again for the next episode of The Cosmic Savannah,

[00:38:58] Dan: You can visit our website, thecosmicsavannah.com where we will have the transcript links and other stuff related to todays episode.

[00:39:06] Jacinta: You can follow us on Twitter, Facebook, and Instagram @cosmicsavannah that’s Savannah spelled S A V A N N A H.

[00:39:14] Dan: Special thanks today to Dr. Robert Yates for speaking with us.

[00:39:17] Jacinta: Thanks to our social media manager, Sumari Hattingh.

[00:39:21] Dan: And also to Mark Allnut for music production. Jacob Fine for sound editing, Michal Lyzcek for photography Carl Jones for the astrophotography and Susie Caras for graphic design.

[00:39:31] We gratefully acknowledge support from the South African National Research Foundation, the South African Astronomical Observatory and the University of Cape Town Astronomy Department.

[00:39:40] You can subscribe on Apple Podcasts, Spotify, or wherever you get your podcasts.

[00:39:44] And we’d really appreciate it if you could rate and review us or recommend us to a friend.

[00:39:48] Jacinta: And we’ll speak to you next time on the cosmic Savannah.

[00:39:56] Hey, Dan, I actually put on a gold ring today, just so that I can say it’s the dust from a double neutron star explosion.

[00:40:11] Dan: I have a silver one also from a neutron star.

[00:40:14] Rob: I’m ring free today. I only do my heavy elements in simulation within real life.

[00:40:20] Dan: But that gold tooth is pretty.

[00:40:22] Rob: Oh yeah. Thank you. You like that?

[00:40:23] Dan: yeah , I do.

[00:40:23] I can see you from London.