SCIENCE AT THE SHINE DOME canberra 5 - 7 may 2004
Symposium: A celebration of Australian science
Friday, 7 May 2004
Professor Brian Schmidt
Research School of Astronomy and Astrophysics, Mount Stromlo and Siding Spring Observatories, Australian National University
Brian Schmidt is an Australian Research Council
Professor at the Australian National University's Research School of Astronomy
and Astrophysics. In 1994 he formed the High-Z SN Search team, a group
of twenty astronomers on five continents who used distant exploding stars
to trace the expansion of the universe back in time. This group's discovery
of an accelerating universe was named Science Magazine's Breakthrough
of the Year for 1998. Brian was awarded the inaugural Malcolm McIntosh
Prize for Physical Scientist of the year in 2000. He is continuing his
work on the expansion of the universe, as well as studying exploding stars,
searching for clues to the nature of dark matter, and looking for mini-planets
beyond Pluto. He received undergraduate degrees in physics and astronomy
from the University of Arizona in 1989, a Master's degree in astronomy
from Harvard in 1992, and his PhD in astronomy from Harvard in 1993. He
joined the staff of the Australian National University in 1995.
The universe from beginning to end
Cosmology is the study of the universe as a whole, and this started, in my opinion, in 1916 when Vesto Melvin Slipher went out and started measuring the distances to galaxies. Well, he didn't measure the distances to the galaxies; he actually measured their velocities, by looking at their light. Their spectra, made up of the various element lines, such as iron and sodium, were stretched.
Doppler Shift gives velocity of galaxy
(Click on image for a larger version)
And as he looked at galaxies – he looked at about 50 of them – he noticed that every single one of them, bar about one, was stretched to the red. It was moving away from us. This was a big mystery in 1916. Why would all the galaxies be moving away from us?
Measuring distances with standard light bulbs
(Click on image for a larger version)
We used physics, in terms of 'standard light bulbs', to measure distances to solve this problem. The idea is that if you have a light bulb, for example, at 1 metre, and you move that to half the distance, it appears four times brighter. We call this the inverse square law, and it is a physical law that we can use to measure distances in the universe.
Edwin Hubble, using the biggest telescope on Earth at the time, in 1929, used the brightest stars as his standard light bulbs in various galaxies, and he found that the larger the redshift of the galaxy – the higher its velocity – the fainter the stars. In other words, the further the object was away, the faster it was moving away from us. He concluded in 1929, at the US Academy of Sciences, that the universe is expanding.
Hubble's data
(Click on image for a larger version)
To understand that, let's just look at his data (and marvel at how ugly it actually looks): distance/velocity. With this data Hubble made one of the greatest discoveries in astronomy.
A toy universe
(Click on image for a larger version)
To understand this, let's look at a toy version of the universe. I take my model of the universe, I expand it – through the magic of PowerPoint – by 5 per cent and then I overlay it. I use my reference point right up here. You can see that every object from this point has moved away. Furthermore, the objects down here [at far lower right], a long ways away, when I have expanded my toy universe have moved further than the objects nearby. So this is exactly what Hubble saw. If you expand something, you will see something where the further away you are, the faster it is moving. And that's why we call it the 'expanding universe'.
Time
(Click on image for a larger version)
As the universe expands, we can wind it up and go backwards into time.
The Big Bang
(Click on image for a larger version)
And the inevitable conclusion is that everything gets piled on top of itself. That's the Big Bang.
If you know how fast the universe is expanding now, then you can figure out how long it has been since everything was piled up on top of itself and the Big Bang happened. So you can measure the age of the universe by how fast the universe is expanding now. If we know that, we can measure the age of the universe, but all the Galaxies in the universe have gravity, gravity pulls on the universe as it expands and that will slow the universe down in time. So there is a complication.
The distance between two galaxies
(Click on image for a larger version)
To illustrate this, imagine I plot in this diagram the separation of two Galaxies in time. So here is where they are now; they have a certain distance. And if the universe is just expanding without slowing down or speeding up, they project back and this [at x-axis] is the time of the Big Bang. Of course, if the universe is slowing down because of gravity, they take a different trajectory and the Big Bang didn't happen as long ago in the past.
Looking towards the future
(Click on image for a larger version)
We can use this to project forward in the universe as well. So take the same situation and let's project forward. If the universe is not speeding up or slowing down, then the objects are just going to get further and further apart in time.
This is the universe. If a universe is empty and it is expanding, it is going to expand forever and it is going to go off to infinity. On the other hand, if the universe has lots of 'stuff' in it so that it is slowing down, you can imagine it slowing down completely if you have enough 'stuff' in it. So where we have the Big Bang here [at left of graph], we get on the other direction the gnaB giB, which is the Big Bang in reverse. So we can predict the future by measuring things now or looking into the past.
The slowing down or speeding up of the universe affects how old we think the universe from the Hubble constant. The Hubble constant is how fast the universe is expanding now. (We name it after Hubble for his great discovery.) It tells us the ultimate fate of the universe, whether or not it is going to exist forever or whether or not it is going to end in the gnaB giB, or, finally, it also turns out it tells us the shape and therefore, essentially, the weight or the density of the universe.
Cosmic geometry – curvature and density
(Click on image for a larger version)
To illustrate my point, here is a closed universe, space that wraps onto itself – a finite universe. That is a heavy universe that has a lot of 'stuff' in it. It will end in the gnaB giB.
This is a light universe. This is a universe that keeps going on forever, and it is open.
Finally, we have a just-right universe, a universe which is flat, which also goes on forever, which is precariously balanced between having enough 'stuff' in it to keep the universe not being open and not enough to close it as well. So 'just-right' is the correct title for this, because theorists who try to understand the Big Bang have long said that this must be the geometry of the universe.
Measure universe's past
(Click on image for a larger version)
So you can imagine an experiment where you go through and you look into the universe's past to see what it did long ago, so that you can predict the future and how much 'stuff' is in the universe. That is something that I have worked on for the last decade.
The idea is this: you measure the expansion rate – that is, the Hubble constant – now, and then you look at very distant objects, and by looking along to very far distances you are looking back in time because the light takes time to reach us. So imagine a universe which is not changing its rate of expansion over time. Well, in this diagram, it will have no change. It will be a constant. The expansion rate will be the same over time.
On the other hand, imagine a universe which is slowing down, so it was expanding faster in the past. This trajectory [curving down from original point on x-axis to a point on y-axis labelled 'Long ago'] gives you a universe which is going to end in the gnaB giB, the 'gravity wins' situation. Gravity will eventually halt the expansion of the universe.
If, however, the trajectory of the universe lies above this line [ie, the previous trajectory], then gravity loses. The universe is infinite and expands forever.
Finally, if the universe is above this line, that means that gravity isn't really, in the traditional sense, playing a role in what the universe is doing but rather something else would be accelerating the universe.
Type 1a Supernovae
(Click on image for a larger version)
So we used objects called Type 1a Supernovae – here is one observed with the Hubble Space Telescope – as standard candles.
What is a Type 1a Supernova? Here is the Sun Earth. This is the life of the Sun, in five seconds. The Sun is eventually going to become a red giant and swallow the Earth, and at the end of this it is going to collapse down to this tiny little thing called a white dwarf.
If the Sun had a friend, a sibling, a binary star, the larger of the two stars would become a red giant. Stars can survive that. This little white dwarf would still remain and eventually this star would become a red giant, a big star, spill material onto that other star, and make it grow in mass.
White dwarfs are interesting stars, in that if you make one to be 1.3 times the mass of the Sun, it explodes. When it explodes it becomes, basically, brighter than about five billion suns. And they have this wonderful uniformity of about 8 or 9 per cent in their brightness. They take about 20 days to rise from being faint to bright, and they fade off into oblivion.
So at this point these things can be used as standard light bulbs to a uniformity of about 7 per cent.
So we can go out and use a telescope such as the Cerra Tololo four-metre in Chile, which is the sibling telescope to the Anglo-Australian Telescope up at Coonabarabran, watch the sky go by and take lots of pictures.
We can then uncover objects in data that looks like this. We take about one of these every 60 seconds when we are observing, and there are roughly 5000 Galaxies. We can find the supernovae in here, using some image processing techniques. It happens to be right there. How do I know? Well, we don't take just one picture; we take two. We take one now but we compare it with an image taken roughly three or four weeks earlier, and so you can see the object has popped up here in the intervening three weeks. This is a supernova 4.5 billion light-years in distance. So this supernova exploded at about the time the Earth was being formed.
Time (billions of years)
(Click on image for a larger version)
We can now look at what the data looks like on my little diagram. It shows the cosmic expansion rate, and you can notice – putting our trajectory that I showed you earlier – this is the trajectory of a universe which ends in the gnaB giB. So these points lie sort of uniformly half in there and half not, but these [other] points clearly do not lie on a trajectory which is down here.
We can put in the 'gravity loses' scenario. So this is a universe which is infinite, but gravity is slowing it down a little bit but not enough to end the universe. These points don't really lie on average there as well. Instead they lie above this line, and the universe appears from this data to be accelerating.
A similar result was found by Perlmutter et al.
(Click on image for a larger version)
We had competition during our work, and the competition was by a group at Berkeley led by Saul Perlmutter. He had similar data, gathered at the same time, and he at the same time came up with an identical conclusion. And I should say this conclusion was not something people expected. It was indeed something that made me very concerned about my future career when we published, because it was so crazy. But it has resonated because it solves many problems in cosmology, it turns out.
So we have an accelerating universe. To summarise, the Type 1a Supernovae measurements indicate that the universe is about a 30:70 mix of normal gravitating matter – that is, matter which can slow the universe down – and material which can push on the universe, accelerating material. We tend to call that material Dark Energy, where we call this [other] material Gravitating Matter.
This is true if you believe our results. So we have got to look at this problem from another perspective. One thing to do is to do other physics experiments and see what is in the universe.
What is in the universe? – take II
(Click on image for a larger version)
So let's look at what is in the universe in maybe a more direct fashion. As one of the things we can do – and I will show equations just to show that we do use very simple equations sometimes (actually, we use much more complicated equations most of the time) – let's say we wanted to weigh the Solar System. You can go through and measure the mass of a system with spherical symmetry by measuring an object's velocity in orbit and its distance from the centre, where it is orbiting, and you have the gravitational constant. So if you use Earth and you use its orbital parameters, you can go through and use this equation [at bottom of slide] and measure the mass of the solar system: 1.989x1030 kilograms. That is how much our Sun weighs.
Weigh the Solar System with Neptune instead!
(Click on image for a larger version)
And for consistency's sake you could go out and do it with Neptune, and use its orbital parameters. You get a number which differs in the last decimal place by 0.1; that's actually the mass of Jupiter and Uranus and Saturn. This is a very precise tool for measuring mass. So let's go out and apply it to a Galaxy.
A galaxy
(Click on image for a larger version)
Here is a nearby spiral, M33, the Triangulum spiral, and here is what it looks like in radio wavelengths. You can go through and measure the velocity – the orbital velocity, in this case – of hydrogen. If we plot it out, you can see that the material, the hydrogen, is moving at about 150 km per second. But the interesting thing when we do this measurement is that for the hydrogen, which we can see where there no stars, the velocity is still constant even though the distance from the centre of the Galaxy is increasing. So, because this v [in the equation M=v2R÷G] is more or less constant, and this is getting bigger, the mass as we go further and further out is getting bigger and bigger in this Galaxy, even though there are no stars. We can add up how much gas there is, and it turns out there is not very much gas.
A generic thing that we have in all Galaxies is something we call Dark Matter. In and around Galaxies there is roughly 10 times more Dark Matter than matter that we see in the form of stars and gas – atomic matter. This has long been a puzzle to astronomers: what is this Dark Matter? Does it really exist?
2dF Redshift Survey: A census of the 1-billion light year
neighbourhood
(Click on image for a larger version)
So let's go and measure the universe in a different way. Matthew Colless, who just recently left Mt Stromlo Observatory to go run the Anglo Australian Observatory, conducted what we call the 2dF Redshift Survey, which was a census of the 1 billion-light-year neighbourhood around our own solar system. Each one of these points is a Galaxy, and instead of measuring distances with the 2dF facility on the Anglo-Australian Observatory, they go through and just measure velocities. It turns out that within this neighbourhood, this equation – Hubble's Law, as we call it, this being the Hubble constant – is that if you measure the velocity you get the distance. And that is good to, on average, about 1 per cent over these distances. So Matthew went and did this with his colleagues to 240,000 objects, and there is one interesting feature this which makes it a useful tool.
If you imagine that you have distance measured by velocity, what happens if there is gravity which slows or attracts the Galaxies against and relative to the expanding universe? So if you have a bunch of Galaxies which are randomly put on the sky, but there is a big mass here, they all have a velocity, they fall in towards that. And so when you measure their distance you make a mistake. So instead of the objects which are really here in physical space, when you measure their velocity they look like this: they look a little collapsed. The more matter, the squishier this diagram gets.
Measuring gravity
(Click on image for a larger version)
And so by comparing the distance of Galaxies in redshift direction versus the physical direction that you see on the sky, you can measure the strength of gravity. This is what Matthew and his team have done in this diagram. This is basically showing the average distance in redshift versus the average distance of the sky of all 240,000 Galaxies in his survey. You can see that the contours here are squished, and that gives him a precise measurement of what is in the universe.
So what they have done in this measurement is gone through and done the physics and calculated that 28 per cent of the density necessary to be flat is in the universe. It is not 35 per cent, it is not 15 per cent, it is 28 per cent. It is a very precise measurement. This is a great piece of work. And so 28 per cent of the universe is gravitating mass.
The cosmic microwave background 13.7 billion years ago, Uv
3000°C
(Click on image for a larger version)
Let's look at another experiment, which has just recently within the last year come out: cosmic microwave background. This is a picture of the universe when it was roughly 100,000 years old – 13.7 billion years ago. The universe is roughly 3,000°C here; it is glowing like the Sun. And it is smooth to one part in 100,000. The little bumps and wiggles you see here are places where the universe is a little hotter or a little cooler than other places.
Gravitational collapse
(Click on image for a larger version)
The reason it is hotter and cooler is that as you have little bumps and wiggles you get gravitational collapse. The gravity causes the universe to start collapsing there. But as the universe collapses, it heats up, and that provides pressure which halts the collapse.
This is a very useful tool, because the physics here is very simple and we have the interesting physical point that the scale where the pressure can halt the collapsing universe is equal to the speed of sound times the age of the universe. It is a physical ruler which we understand very, very well.
Physics how big (in metres)
(Click on image for a larger version)
So if you go out and you can make a measurement – for example, in physics, the physics that we understand tells us how big the blobs in the early universe are in metres. We can then go out with measurement and see how big they appear now. And the ratio of how big they are and how big they appear tells us the geometry of space; it tells us the shape of space. Why is this?
Well, imagine a universe which is curved and, in this case, closed – in blue – and a universe which is flat – in red. An observer here, up at the Earth, sees on these two lines exactly the same size on the sky, but because the lines in the curved universe are bent, this big object has the same apparent size as this little object in the curved space. That is, the size of objects, if you really know how big they are, tells you the curvature of space.
Geometry of the universe
(Click on image for a larger version)
And so the geometry of the universe was determined by the WMAP satellite last year, with exquisite accuracy, by doing this experiment. They have modelled the physics of what the cosmic microwave background, this early snapshot of the universe, should look like in an open, a flat and a closed universe. You can see how these [closed] are big, these [flat] are medium and these [open] are small. So the size of those blobs tells you.
Open universe / closed universe (first of two slides)
(Click on image for a larger version)
What they have measured – in this diagram actually showing you the size of the biggest blobs – is that these are about 1° in size, and that corresponds to almost exactly borderline between the open and the closed universe. The universe has 102 per cent, plus or minus 2 per cent, of the matter necessary to be flat. Now, you would say, 'But you just told me Matthew said 28 per cent.' Well, it turns out that Matthew Colless measured the matter that has gravity, any type of matter that has gravity as you and I would have. This measurement is sensitive to any type of matter, whether it gravitates in the normal way or not. So this inconsistency is not as bad as you might think.
Open universe / closed universe (second slide)
(Click on image for a larger version)
The other thing, which I don't have time to explain, is that the position and height of this bump [towards right-hand side of graph, at about 2000] tells you how much atomic matter there is in the universe. It turns out the measurement says that the universe has 4 per cent of the atomic matter necessary to be flat.
In synthesis, the universe is, from supernovae, 30 per cent gravitating matter, 70 per cent accelerating matter; the universe from the 2dF Redshift Survey is 28 per cent gravitating matter to be flat; from the cosmic microwave background, we have it that the universe has 102 per cent of the necessary matter to be flat; and from Galaxies we know that there is 10 times more Dark Matter than stars and dust. And, finally, the universe is 4 per cent atomic matter, and that is also from the cosmic microwave background.
The messy universe
(Click on image for a larger version)
This all leads to a consistent but very messy explanation for the universe. The universe is 4 per cent atomic matter – 4 per cent made up by the stuff that the Earth and you and I are made up of; 24 per cent of it is Dark Matter, which has gravity but we really don't understand what it is – we think it most likely is some sort of elementary particle that has not yet been discovered, things called axions being one possibility; and then 72 per cent of the universe is Dark Energy. It is something which is pushing the universe apart but which we know nothing else about.
All right, so what's the future of the universe?
Well, it's really a battle for domination of the universe between Dark Energy and Dark Matter. Dark Matter slows the universe down, Dark Energy speeds it up.
As the universe expands, Dark Matter slows it down, but the universe is still expanding.
And as the universe gets bigger and bigger, the Dark Energy becomes more and more dominant, and eventually it is able to take the universe and accelerate it, and make the universe expand faster and faster over time.
Graph
(Click on image for a larger version)
So, going with the Hubble Space Telescope, we can look back now 11 billion years. This is the most distant object every discovered, a supernova 11 billion light-years in distance. And here is current data that we have compiled over the last 5 years. If you want the universe to end in a gnaB giB – that's the fit to the data – you can see the data and that line doesn't fit very well. But we have two other possibilities: the Big Rip or the Empty Oblivion. Those are two models of Dark Energy which fit the data.
So as I finish off I will tell you about the Final Act of the universe.
The Empty Oblivion: Dark Energy has basically won the battle of the universe – that is not in doubt – and will continue to accelerate the universe for the rest of the age of the universe. Eventually the whole universe, bar a few Galaxies which are gravitationally bound to us now and which will form some super-Galaxy, will be looking out at a universe which is completely empty. All the other things in the universe will be pushed out beyond our ability to see it.
The Big Rip option: It turns out that if the Dark Energy has a property that it gets created faster than space – and I should say that this is a long shot, but it is certainly within the observations we have made so far – then every piece of the universe will be eventually pushed away from every other piece, even to the subatomic level, leaving nothing, or at least, I would say, a very good definition of nothing. Every thing would be in complete isolation in the universe.
So, in summary, I think we can say, 'So this is how the universe ends, not with a gnaB but a whimper of endless oblivion – or just possibly in a scream during the Big Rip.'
Question: You talked about Dark Matter. What is an axion?
An axion is a little particle that people have made up. It has a certain energy and it would also have a certain mass. It doesn't necessarily do much other than it can fit easily into the laws of physics as we understand them. As I say, it is a made-up particle which has some nuances of how it fits into physics but would provide the mass of the Dark Energy. It turns out there is so much of it that you could actually have the chance of putting very sensitive detectors in the bottom of mines and letting these things pass through them, and they will excite an atom and you can measure that. So people are trying to do it, and they may have a chance of measuring it in the next decade or so.
Question: Does calling it Dark Matter mean that it actually absorbs light coming from beyond it?
No. Dark Matter means that it does not emit light. Astronomers see things that emit light. So you and I are possibly Dark Matter. The Earth is Dark Matter, because it doesn't emit light, it only reflects it. Our Sun is luminous matter. However, because of shortage of time: the 50-inch telescope did a very famous experiment last decade called the MACHO experiment. It went out and looked for the gravitational magnification of little Earths and things going by in our Galaxy in front of a nearby Galaxy. That experiment, while it did detect a little bit, did not detect anywhere enough to explain the Dark Matter as being in stars, planets or small black holes, and so it indicated it has to be something else.
Question: Did I understand you properly to know that the amount of Dark Energy is constant, or is it changing over time?
Dark Energy is often called the cosmological constant, because it is part of space itself. As you create space, this stuff is attached to the volume of space. So if you double the volume, you double the amount of Dark Energy. But if I double the volume of space around you, I don't double your energy. You are still you. So the density of Dark Energy stays constant but the effect on the universe becomes much stronger over time because its density remains constant, where the density of everything else in the universe becomes smaller and smaller over time.
Question: Is Dark Matter all the same stuff, or are there hundreds, potentially, of different types of Dark Matter and every couple of years you will get up and tell us about another 2 per cent of the Dark Matter?
Well, astronomers live by something we call Occam's Razor – it is throughout physics, math and, hopefully, biology as well – where you are always looking for the simplest solution. So the addition of Dark Energy, the fact that you have this 30:70 mix of two things, is already violating that. That is not the simplest thing. The universe should be made out of one thing. As you point out, I have just come up with three things. I got Dark Energy, Dark Matter and atomic matter.
Now, the atomic matter you will believe because you are here. But everything else is stuff I am telling you we don't understand very well. So it is my sincere hope that the Dark Matter is one thing, but I can't guarantee it. As you divide it up, it becomes – despite fitting all the observations – less and less plausible that you understand everything that is going on. It's a very good question.
Question: A year or two ago, I recall, there was great excitement that neutrinos, or some species of neutrino, were supposed to account for Dark Matter. Now you are telling us that they are axions rather than neutrinos?
The mass of neutrinos has more or less been measured or it has been limited, and it turns out that they are at least a factor of 100-1000 times less than the amount of Dark Matter in the universe. Physics has gone through and done the measurement. The axions are literally made up. We do not know that they exist; that is a guess. So my best guess is that either there has to be some particle out there which is lying waiting to be discovered – that is a prediction – but it is extraordinarily difficult to define them.
Question: Looking at the future, what do you think are the experiments, the measurements, that should be done to throw more light on Dark Matter and Dark Energy? Does it need new equipment, something special in a satellite to measure things?
To measure Dark Energy turns out as extraordinarily difficult, so let's talk about Dark Matter. I think the correct way to measure Dark Matter is that we should be able to detect it. There are literally millions of pieces of it, probably, going through us right now, and there should be some way to measure it in the lab. People are doing that. Especially, at the University of California at Berkeley there is a great group doing that.
On Dark Energy, one of the things that people are doing here in Australia is that Matthew Colless again, Brian Boyle and a number of other people are going through, with our colleagues throughout the world, to do experiments which will better quantify how the material behaves as the universe expands. We are doing that with supernovae; my colleague in the United States, Saul Perlmutter – of whom I guess I am a loose affiliate – has lined up a $1.4 billion satellite to go out and to measure supernovae with precision, I am told. I have to admit that I am somewhat reluctant to be spending that amount of money on an experiment that I am not sure will be able to find anything, but I think we need some good ideas. We have no observations; it is time for theory to come in with some new ideas to settle things down.
So I am waiting for things to settle down a little bit so we get some better ideas, because I think we are actually very short on good ideas of how to go forth right now.
Question (cont'd): Are you saying that perhaps more money should be spent on theory?
Yes, I would actually say that within the Australian physics community in general, I think theory is generally underfunded relative to – and I say this not as a theorist – the experimental and observational side.
Question: What is the evidence or reason for Dark Energy to have a constant density as the universe expands?
As we measure the trajectory of what the universe is doing, back in the past, that is sensitive to the behaviour of Dark Energy. If it is being created faster than space is created, then it pushes on the universe more recently than if it were doing that slightly more slowly. So we can quantify the physical characteristics, and we measure its equation of state, in the same way as you do for a gas. We have done that with the supernovae, and Matthew Colless plus the cosmic microwave background can do it as well. We get the same answer. It is very close, very similar to its being perfectly aligned with space, but the uncertainty is still 30 per cent in that equation-of-state property.
