Leonardo Silva Reviewer's Good evening, everyone, and welcome to this lecture. We are going to do a journey into space, and we have to start from the place we live in the cosmos, which is our wonderful planet. Colorful, full of different things, and very different from the empty space. Indeed, space is basically empty. The universe is dark, and ordinary matter has a very low density of the order of one atom of hydrogen per cubic centimeter.
In spite of this darkness and emptiness, when large telescopes such as the James Webb Space Telescope, presently my favorite telescope, launched and started to stare into the empty space, look at the wonders that come out. All of those patches of light represent galaxies at various distances in cosmic time. This is called the James Webb Deep Field, and deep means that by looking at these images, we actually look at the light that was sent by these galaxies shortly after the Big Bang. So, as you know from the title of this talk, I will show you how, in modern science, we are able to decode that light, to model the light of galaxies across the universe, which allows us to do astrophysics.
As I will tell you, the dark component may be fascinating in themselves, but they don't allow you to do much, exactly, because we don't see them. So, let's start from our home, the Milky Way, in one of these fancy pictures. that I've taken from the southern hemisphere, where we have actually privileged view into the highest density of stars, which is the Milky Way belt. So what is a galaxy, after all? Let us remind us, and this is a prototype galaxy, a real observation taken from Australia, the galaxy of the kind, the same where we live.
A galaxy generically contain loads of stars. It may have a structure such as this one. where the component line up into a disk, a central density, high density of starch, that we call the bulge, and right in the middle of it, a nucleus.
And if there is a black hole, it is exactly in the middle of it. All of this, and we will explain this in a second, appears to be embedded in a halo of a matter that we don't know what it is, and we call it dark matter. That indicates also the approximate position of our Sun. So, after all, a galaxy is like an industry which continues to convert its content of gas and dust. And I put for your eyes and pleasure these beautiful images of the Orion Nebula with the James Webb Space Telescope.
If you could compare, that's the second one there. What the Hubble Space Telescope saw, the resolution, the gain is incredible. It's full of stars. Stars create chemical elements, including those that actually make up our bodies.
Gas and dust are the fuel for star formation. Black holes are lurking around, not only in the very middle of galaxies, but also from those stars that are so big they leave them out. And dark matter.
which, again, I will stress, we need them, and I will explain why, but we don't know what it is. So, is there one type of galaxy? Actually, there are so many in different shapes, although we recognise three main components, that this gave origin to a naming that was actually conjetured in Oxford, the Galaxy Zoo, where the people, the normal... public and not necessarily with a degree in astrophysics are called to actually try to classify the myriad of galaxies. Do you want to have a good number of how many galaxies we think are in the universe?
Hundreds of billions of galaxies. That's actually a number that seems fancy in itself, but it's not too incorrect. You see that there are galaxies that are perfectly spherical. We call them ellipticals.
Actually, those are the ones that are most interesting to me, because they seem to be the ones that actually formed first, very shortly after the Big Bang. There are galaxies that we call spirals or disk galaxies, where actually you see these spiral arms like in our own. There are tiny galaxies that gain the naming of dwarfs, where the actual content of stars is very little, and yet we need to conclude that they are galaxies because they are expanded. sustained by this invisible matter. And also, there are galaxies that are actually in interaction with each other that we call, in jargon, mergers.
So... How do they come about and how can we actually study them across cosmic time? At this point, we need to start from the start.
The model that is the best at the moment that we have was suggested by George Lemaître and is called the Big Bang model. Besides the fact that it is actually a valid model for the universe, it also satisfies some very... strong observational constraints on the abundance of elements and also the expansion of the universe, as I will immediately tell you. So right at the beginning you need to imagine, that's very relevant for the formation of galaxies for the reason that I will tell you now. You start, we think, to start from a singularity, that's normal practicing in physics, but we need then to justify the huge universe that we need, we live in.
And also, this has to happen quite fast in time. And so you have this expansion that we call inflation, in which, from this point, the universe very soon reaches incredibly large dimensions. At the beginning, because of the high temperature of the Big Bang, we call the universe actually a cosmic hot soup, in which all these elementary particles, Those that we know, but maybe others that we don't yet know, all jump around freely, like in the soup that you see. In the meantime, the universe is expanding further. And, as you know, if you have ever cooked a risotto, which I do often because I'm Italian, but you will see the analogy between astronomy and gastronomy over and over in this talk.
So you see that when the result is very hot, in order to decrease the heat, you actually distribute it on a larger surface, and so does the universe. But that's the key in order to form the first thing that I told you before, the hydrogen atom. Because if the temperature is far too high, the particles are all moving at very high velocity.
You do not form anything stable. So it is the expansion of the universe that allows actually to create... matters. So that's an excellent idea, but there are problems with that, that I will come in a second to explain. So in this view graph that you can find anywhere, I will depict a brief history of the universe, which is a history of expansion.
And as we know, since 1998, even more dramatic, the expansion is not, let's say, constant, but is actually in acceleration. But besides the start... point in which we don't have light, and then we form hydrogen and helium and all the rest, it's all about tracing by galaxies. And that's why study galaxies is actually key to understand the universe on the larger scale. At some point in the late universe, we astrophysicists used to say, the sun got formed 4.5 billion years ago, as many other stars, and the planet.
our planet with it. Now, how do we think we can form galaxies? Surely, we need gravitation, and here you know we need to attribute to Newton.
That's excellent, but do we have enough matter to actually start to explain that this primordial matter gets together in something that will become huge as a galaxy? We actually don't. If we start...
to count all the normal matter that we call baryonic, but you can ignore this term, that we know of, in an expanding universe working against gravity, which tend to condense things, you would not form any galaxy. Therefore, we would not be here, which obviously is an uncomfortable situation. In order to cope with this, scientists hypothesized that actually there is more matter than the one that we know interacts with light. And because of this, we call it dark matter.
So as you will see, we require the presence of this extra matter not only at the level of galaxies, maybe this is something new that I tell you, but also at the beginning to start making the process of galaxy formation. So it is a crucial ingredient in our modeling. Another analogy with food is then, what is the content of the universe? This is something that we call the cosmic pie, which a satellite like Planck determined quite precisely.
So if I show this to my students, they start to say, hmm, are we actually in a serious physics course? Because 95% of the universe is in an unknown form. It is actually the case.
Quite a sizable amount of dark matter that needs to start to form structure in the first place and since 98, magically the year I got my PhD, an even larger amount of something that pushes the universe into acceleration that then we call dark energy. At the present time, we actually don't know much about both of them and because of this, there are plenty of satellites and investment in Europe, in the UK, in the United States mostly, in order to probe these two strange things. However, Atoms, galaxies, gas, and so on.
It's 5%, but they count a lot, because actually they convey to us all the information. So, first of all, let me show you who has actually demonstrated at the level of galaxies, so what we want to explain, the reason why we need to invoke this extra matter, Professor Vera Rubin, who I was lucky to meet in 2009 in Canada. You will see her here.
In the... Systematic work on all the spiral galaxies that were known since the late 50s up to the 60s, Vera determined, as a function of the distance from the center, the velocity of the particle on these huge disks. And you can apply a usual Keplerian law, and you would expect, which is the blue line, that the more I get farther from the barycenter, from the center of matter, the velocity decreased for the simple reason that particles don't need to run that fast in order not to fall into the center, in a way.
Instead, the observations showed the orange line. So the orange line means that the velocity remains high even in regions where I see no stars. And therefore...
the most logical assumption is that there is matter that actually sustains these velocities. So when we model galaxies, we need actually to do it in this larger context. Nowadays, and the UK has been a leader on that, we can actually model this dark matter with an accuracy which is surprisingly high, considering that we don't know what we are talking about. But the thing is, we know that this matter obeys only to the law of gravity.
What I'm showing here in this graph is the so-called millennium simulation, which is a very large supercomputer type of effort of m-body simulation of these dark matter containers performed in conjunction with Munich in Germany and Durham in the United Kingdom. The number of particles has been very fancy. This simulation was running for months, so one would hope that we would start the simulation without making any effort to start with, because otherwise you have to wait several months. But nowadays, we have an exquisite resolution of the substance we don't know.
I'm always stressing that. Sometimes, somewhat it's easier. Why we need that? Because on this, we start to model the actual luminous component. So, it is fair to say that in the dark cosmic web, we grow galaxies pretty much as we make pizza.
And here I have to come to this second analogy. For making a pizza, you essentially start to put the dough all together and to round them up. So this is what happened in this large container of dark matter where cosmic...
gas flows into due to gravity, gain angular momentum, and start to be shaped in a round structure. And that's why we think in the cosmos, the central part of galaxies are those forming first. And the more you condense, similar to how stars form, the more you condense the central part, the more they will gain gravity and they will attract more.
At some point in this container, however, you... happen to form a disk by rotating the structure. And this is how actually the James Webb is seeing primordial disks in galaxies very close to the Big Bang. Disks are a fragile structure, on the other hand, contrary to bulges that are held together by this strong gravity. And therefore, in the long history of the universe, disks can actually get destroyed, reinforcing the bulge.
On the other hand, a galaxy like the Milky Way happened to have its disk since at least 8 billion years, because this is what the observation seems to show. So, in a sense, the universe is very efficient after we gave them a bit more gravity to form this huge structure. Our solar system happened to be, in fact, in an outside arm, which is quite protective.
from a cosmic catastrophic event, such as explosion on very massive stars that destroy necessarily any planetary system that happened to be forming. Planetary system always formed together with stars. But our location is actually quite good. We also do not have a companion star.
So there are many things that we call cosmic coincidence. But let's go on. So the way we see galaxies or our telescopes collect and notice that there are galaxies and we invest a lot of money on these huge satellites such as James Webb is because they send us light.
And this light comes from thousands, billions of stars like the Sun or different from the Sun. So the problem is, are we able to actually make a theoretical modeling of this light. And this is something we're going to try to show now. So first of all, how does this light look like?
The galaxy you see there is called the M87, Messier 87. And this is the galaxy where people have actually pictured the central black hole. It's the most massive galaxy that we have in our local universe. The most massive.
There is nothing more massive. And it's packed with stars. Do you want to show exactly how the light looks like?
Like this. What I'm showing there is an actual spectrum collected by a previous student of us, Tania Parikh, of M87 with the Sloan telescope, which is an optical telescope, so I'm showing the optical spectrum. So stay with me.
On the y-axis, I have the emission of flux in certain units. And on the x-axis, there are the wavelengths at which this light is emitted. This, if you like, is called spectrum, and we can have spectra in many branches of physics, but this is an astrophysical spectrum. So we see that this is a line that increases towards the visible wavelengths and stays rather flat, proceeding towards the near-infrared. And there are a lot of lines, which we call absorption feature.
Of good elements, let me show you. We have iron. Iron lines are everywhere in the spectra of galaxies.
Magnesium, which besides helping our muscle, is actually what we call a cosmic clock because it's going to tell us, and this would require a larger explanation, the ratio between types of supernovae that are exploding in galaxies. We do have sodium, which obviously is familiar to you. And calcium, the two lines of calcium that you see there, are actually powerful for age in this galaxy.
So, after all, a distant galaxy like M87 is full of chemical elements of the kind that we are used to in our planet. And that's because the generation of stars that are born and die in that galaxy have actually filled the galaxy with those elements, and we see the light. and I always thought that this is fantastic. It's my favorite application of chemistry. It's actually in the extra-relative realm.
So people started to say, okay, we get so much detail. Because what you see, those wiggles are not measurement error. They're actually feature. So people thought, are we actually able to make a theoretical spectrum in order to perform astrophysics? Let's see.
Guess what? But we need to start with Einstein for good reason. Because the first thing we need to understand is where does the energy of a galaxy come from?
Obviously from the stars. How do stars shine? Actually very simply following the Einstein equation. They keep converting hydrogen into energy with an efficient, with a coefficient of efficiency, apologies that I didn't quote there.
but these are all the order of 10 to the minus 3, 10 to the minus 4 for the nuclear reaction that happen in stars. So the first thing is to have this equation in mind. And this is what any star does. This is a beautiful image of the sun composed by 70% of hydrogen, 20% of helium, and just 2%, only 2% of those elements, but they mean everything to us because we actually see the features.
and we can actually determine the cosmic evolution of the chemical elements. So, if we walk into the Sun, which we will never be able to do, but virtually we can, we would notice that it is actually in the very center, that the Sun is self-consuming itself at a nice, steady pace, and so do other stars, with very different time scales, however. It's that... energy that comes out of the Sun and if you integrate over all stars of a galaxy, that's what the light of a galaxy is. So, as we know, not all stars are identical to the Sun, although the Sun is a very common star.
So the question is, what type of stars will be present in these very distant galaxies? Depends, but in principle, I will have small stars like the Sun. They will live on timescales of 10 billions of years, become giants during their life, and then... re-collapse to live as a white dwarf. This is what the Sun will do.
There are the stars that are much bigger than the Sun. On the other hand, they will live much shorter, of the order of one million years, so different scales enormously. However, they are far more luminous.
They will become super giants and implode into a black hole or a neutron star. So when you do a model, For a galaxy, you need to take into account all these species, in fact, of stars. So, very simply, and I'm sure you are with me, if I am able to model the energy emission of one single star, by summing all of them, I should be able to make something similar to a galaxy.
How are we able to calculate? This problem is called the population sink. the synthesis of a population in external galaxies. Let me have an analogy with normal life. Suppose you have this audience where you see a variety, let us call ourselves species, spread in age, spread in other characteristics, and we are very near and we can actually see all this in detail.
If we zoom out, This would be, of course, a different situation, like a stadium. The only thing that you actually notice is that there are many people, but you won't be able to recognize the characteristics of each of them, but probably you don't need that. Then you still zoom out, and this you see from Europe and the UK. I see a lot of light in certain places and not many in others.
You don't have the detail that you have in the room or nearby, but you can still see. that you can measure the emission of energy. This is an analogy, but not too wild, believe me, because the detail room is actually a stellar field in the Milky Way.
This is a real one from the Hubble Space Telescope. The stadium is a star cluster, the smallest condensation of stars that we have in our own Milky Way, and the continent is actually a real galaxy. So our task as a theoretical astrophysicist is to know in detail the physics of the species in the room and to construct a model that we can apply to the first galaxy born after the Big Bang. So please let me give a tribute to Beatrice Tinsley, a British-born scientist who was actually active in Harvard, who in 1972 literally invented the field.
And she called it evolutionary population synthesis, and that was the theme of my PhD. And I'm very happy I did that because it's a lot of fun. And what I'm doing is really needed in every part of astrophysics.
So Beatrice, in a series of papers before she passed away in 1981, very young unfortunately, wrote that the contribution in any far, far away galaxy, a reminder of Star Wars, the contribution from different stars to the light of a galaxy must be proportional to the stellar energy and lifetime. We have equations for that. According to nuclear physics, because we convert hydrogen and helium in the cores, and the stellar evolution theory.
And so the field was born now 51 years ago. So it's relatively young for a branch of physics, but it ramped up in parallel with the observation. So let me show you, before I show you the actual...
model that we make in the computer that really resembles a galaxy As we said, in this galaxy there are 100 billion of stars. So how do their individual spectra look like? What I'm showing here are actual Milky Way spectra that we collected as part of a collaboration led by Young just recently. And now it involves the largest collection ever of Milky Way stellar spectra available in the literature.
It was published in 2019. it includes 60,000 spectra. Our students in Portland actually can have their master project on actually these things. Now, what am I showing there?
Again, the spectra, the flux emitted as a function of wavelengths. The very high up lines correspond to very hot stars. How do you know this? Because the energy distribution tend to peak in the left part, which is the ultraviolet.
So going there. And the line is almost featureless. The stars are very hot, and they effectively behave as almost black bodies. When the temperature decreases, instead, you see those huge bumps, and their stars are showing us molecules of actually water, carbon oxygen, carbon dioxide, all these complexities, and even more, because the temperature is low, and therefore matter can actually aggregate into... a complex molecule that doesn't happen in the North Star.
So remember that the light that I get of a galaxy will be a combination of all of those. So summarizing before I'm going to show you an actual model spectra, the component of a good model for a galaxy will be the energy emissions per stellar mass, the time scale, How long do these stars live? I told you that there are huge variations.
The Sun shines for 10 billion years, but the 30 solar masses are done in 30 million years. There are a lot of stars that live much shorter than the period in which the dinosaurs were on the Earth. Actually, 65 million years ago, several stars in the Milky Way would be dead. Then I also need the distribution, which is then similar to the analogy in this room.
How many stars get born with a certain mass? It's like, how many people get born with blue eyes or black hair? You know, this is one of the toughest problems in astrophysics.
It's called the initial mass function, and it is actually not known by first principle. So what we do, we use a function that looks sensible, that has been calibrated on our own galaxy, but effectively, We are not able to predict yet, we work for that, how a mass of gas fractalizes into the component star. It's by far not a trivial problem, but I'm going to show, to mention it because it's a part, of course, of our model.
The other thing that we need, guess what, is a supercomputer. And there I'm actually underestimating the number of cores that we have in Portsmouth now. Ten years ago, we were part of the Cosmos Supercomputer of Cambridge, but now we have, of course, our own.
And then you do your integral and let it go. And usually that doesn't take too long before you have this beauty. Now, here you see not an observation, but a model spectrum. That is a spectrum of a galaxy, of an hypothetical galaxy, that I have calculated. And they are all available, by the way, at our website.
What do I show you there? Again, the flux emitted as a function of wavelengths, apology for using the logarithm, that's typical of astrophysicists, because we need to stretch orders of magnitude of scales in which the variable changes. This spectrum shines from the ultraviolet to the infrared, and it should represent a galaxy that is just 0.5 billion years old. So...
relatively young. As many as those, you can actually calculate thousands of spectra at an arbitrary resolution in time. It's very satisfactory, I need to say, because I will show you how they compare with the real data.
So in that plot, I show you the evolution of a galaxy, when it formed one million years, since the beginning, up to when it aged. to 10 giga year galaxy and stars are like people. They get born and then they start to age. And what changes their energy emission, which I show on the y-axis.
Young is up there with a lot of emitted flux. Young galaxies are super energetic, which explains why we can see galaxies near to the Big Bang. If they would be less, we would not be able to see them, not even with the James Webb Space Telescope. Instead, they have an order of magnitude more energy than an older galaxy that you see on the other part of the diagram.
When they are young, they emit a lot of light in the ultraviolet, which is on the left. on my x-axis, when they age, they actually tend to emit more in the near infrared. Now, this is a movie that we made at the University of Portsmouth and shows how the energy emission of a galaxy changes with age, where age is the little number that you see there in billions of years. Now, the galaxy is aging up to...
15 billion years. Do you want to know why I stopped at 15 billion years? I could actually calculate the model further, because in principle, stars that are smaller than the sun live for the order of 25 billion years.
I stopped there because cosmology tells us that the universe should be about 15 billion years old. But in principle, this model being fully theoretical, they can actually predict the future of the universe. So they can even get older.
The different colors are the different chemical composition of these models. So the lowest color indicates an extremely enriched composition more than the sun, so more than what we have in our planets. This is typically found in this. elliptical galaxies, and in our Milky Way, it is actually found in the central region. There are a lot of chemical elements there.
The blue one instead, that is the highest up, that shows what we call metal-deficient galaxies that are the first ones to form shortly after the Big Bang, when, as we know, there was only hydrogen and helium. But this is just to show you the variation of... appearance and the quantitatively energetic that this beautiful model gave us.
How do they compare with observations? Sorry, we put also the reference there. They do an egregious job, because in this plot, in physics what we do, the art is to compare a theoretical prediction, the blue line in that case, with something that the telescope observed, which is the black line. And then you calculate at each wavelength how far the model to the data is. And that's a metric called chi-square.
And essentially it's going to tell you how good your model is. And every time we have this, we see, oh, are they good or not? And this is, of course, excellent for astrophysical eyes. I hope you agree with that.
I can tell you what are the black lines at the edges. These are in the data. the observation at the edges of the detector of the telescope where we have more noise than signal. But in the central part, it's a fantastic comparison.
And there I put the... This is an actual plot that we published in a paper, so stay with me. By this comparison, you determine the age of the galaxy completely, the chemical composition, in this case a bit more than the solar chemical composition, and the mass in stars. which I quoted there as 1.5 in 10 to the 8th, the mass of the sun.
So the comparison with this model gives you all the astrophysics that then you see, you hear, you read in books. It's all from decoding the light. Nothing comes from the black, from the dark side, but the light is going to give us actually robust physical quantity. And so people are using these models.
across the universe with different detectors and so on and so forth in different wavelengths range. But this is exactly how we proceed. So I spoke so far about the shining living stars, but I also told you that stars like us live and die. When they die, it means that they stop shining. They actually leave us what we call dead.
remnants, which sounds a bit creepy, but this is exactly how we call it in the field. The remnants that they leave us with depend on their initial mass. Stars that are smaller than eight times the sun.
all leaves us white dwarfs. Between 8 and 25 times the mass of the Sun, we get neutron stars. And above 25 times the mass of the Sun, we get black holes. So, are we able to actually model the graveyard in a galaxy? That's the other task in the model.
So far, I modeled the light, because the light is really giving me chemical composition, formation, epoch, all the cosmological quantities. But the darkness is actually giving me the structure of the dynamical potential of a galaxy. What I show you again is the stellar field sampled by the Hubble Space Telescope, because to me, this is so fantastic.
The white spots are living stars. The little The little dots encircled are those that are already dead in the same field. And this is very near to us, and so they, with Hubble, they can actually pinpoint the white dwarf.
You can't see the black holes, so we have to model them for you. I can tell you that there can be, in the Milky Way, there are going to be at least a million floating stellar mass black holes. There's nothing to do with the central, just due to stellar evolution. We calculated that. But you can see that this is the case.
And so in our school, we also started to model the graveyards of galaxies. And in fact, people love this title, but it is actually fair. And I'm going to show you some maps. This we published quite recently, and it's a new concept. So in the first column, you have real galaxies in the local group.
Local group means in the local universe, essentially, with different... morphological types, a spiral, more compact, a double galaxy. To the right, you have the resolved map, the time model of the remnants, white dwarfs, which are the pink ones. That is the reason why we picked that color. Neutron stars, the orange, and, of course, black holes are the black.
That is where, according to this prediction, actually this reservoir of compact objects should be in any galaxy. The thing is, you don't see them. But they participate to the total mass of the galaxy, including the dark one, and also they explain the large motion of stars in a galaxy.
So I just showed you that we can model the light and we can model the darkness out of stellar death. So now you want to know the future of our galaxy, which I'm going to... tell you just because it sounds catastrophically but effectively.
So I told you the light of galaxies is just due to the secular life cycle of stars. If there is new stellar formation like in our Milky Way, and the disk and new stars keep being formed, they will shine. But ultimately, galaxies will run out of gas, which is the fuel for star formation. So what we think will happen in the future?
Galaxy will consume the gas, most stars will be dead, galaxies will dim. If you combine that with the accelerated expansion of the universe, there will be very little light. It's very hard to consume all the gas. It is more an asymptotic behavior than a zero, in fairness, because when a star dies, it also expels part of the envelope. So there is always a little bit of residual gas.
However, if you... allow enough time to go, at some point there will be nothing, and also the expansion will counteract interaction. But in 3.5 billion years, the Milky Way will collide with Andromeda. And that's because, thanks to the dark matter, we are actually part of the local group, which, where actually gravitation between these two giant galaxies work against the expansion of the universe. And so the velocities that they display actually tell us that the galaxy will become bigger, just the sum of them, because of their internal motion.
I'm getting to the end, so please stay with me, because now what I'm going to show you is something more fancy than that. We used to say that the universe is a time machine, and the reason is that by using the light, the light with a well-defined finite speed, we can actually go back in time and, as I told you, see galaxies in different snapshots of their epochs. And here you see a very small me a few decades ago.
And I wish I could go back, but I can't. So I tried to use the universe to do that. And so what we have done...
is to use all our model quantities for all the galaxies in certain surveys that are 1.3 million overall, and to associate their properties. And all of this has been put in an app that is called the Cosmic Stroll. The website is there, www.cosmicstroll.com, that works on both iPhones and Android.
It is free to download. And using this app that now works with the phone but also on a desktop, you can actually dive into the dark spaces of the universe and then see these real galaxies farther away and then approach them and see how old they are, how big they are, all according to these models. So feel free to get...
to this app, I thought it's pretty good. We are still developing it to a final stage, but I thought it looks already good now. Thank you very much.