Farewell to Reality (21 page)

As much as 90 per cent of the mass required to explain the size of the gravitational effects appeared to be ‘missing', or invisible. Zwicky inferred that there must be some invisible form of matter which makes up the difference. He called it ‘missing mass'.

Missing mass was acknowledged as a problem but lay relatively dormant for another forty years. In 1975, the same problem was identified in the context of the motions of individual galaxies by young American astronomer Vera Rubin and her colleague Kent Ford. They reported the results of meticulous measurements of the speeds of stars at the edges of spiral galaxies. What they discovered was that stars at the edge of a galaxy move much faster than predicted based on estimates of the mass of the galaxy derived from its visible stars.

Simple Newtonian gravity predicts that the further the star is from the centre of the galaxy where most of the stellar mass is concentrated, the weaker the force of gravity that drags it around. Consequently, the speeds of stars being dragged around the centre would be expected to fall the further they are from the centre. Rubin and Ford found that the speeds of the stars actually level off the further they are from the centre.

Although Rubin's results were initially greeted with scepticism, they were eventually accepted as correct. In fact, Peebles and his
Princeton colleague Jeremiah Ostriker had earlier concluded that the observed motions of galaxies (including our own Milky Way) could not be modelled theoretically unless it was assumed that each galaxy possesses a large halo of invisible matter, effectively doubling the mass. In 1973 they had written: ‘… the halo masses of our Galaxy and of other spiral galaxies
exterior
to the observed disks may be extremely large'.
¹⁴
The following year they suggested that the masses of galaxies might have been underestimated by a factor often or more.

The missing matter was now ‘dark matter'.

It is now thought possible to account for a small proportion of dark matter using exotic astronomical objects composed of ordinary matter which emit little or no radiation (and are therefore ‘dark'). These are called Massive Astrophysical Compact Halo Objects, or MACHOs. Candidates include black holes or neutron stars as well as brown dwarf stars or unassociated planets.

However, the vast majority of the dark matter is thought to be so-called ‘non—baryonic' matter, i.e. matter that involves not protons or neutrons, but most likely particles not currently known to the standard model of particle physics. Such particles are called Weakly Interacting Massive Particles, or WIMPs.
^*
They have many of the properties of neutrinos, but are required to be far more massive and therefore move much more slowly.

Dark energy and the return of the cosmological constant

In 1992, the COBE satellite revealed the imprint of quantum fluctuations on the early universe and confirmed, with some confidence, that the universe is flat. Cosmic inflation could explain why this must be so.

But a density parameter Ω equal to 1
demands
that the universe should contain the critical density of mass-energy. The problem was that, no matter how hard they looked, astronomers couldn't find it. Even dark matter didn't help. In the 1990s, the best estimate for Ω based on the observed (and implied) mass-energy of the universe was
of the order of 0.2. If this were really the case, the universe should be ‘open'. With insufficient mass-energy to halt expansion, it would be destined to expand for ever.

Some astrophysicists began to mutter darkly that Einstein's infamous fudge factor — the cosmological constant — might after all have a place in the equations of the universe. Several earlier attempts had been made to resurrect it, but it had stayed dead. Now some were beginning to appreciate that a mass-energy density of 0.2 and a cosmological constant contributing another 0.8 might be the only way to explain how Ω can be equal to 1 and why the universe is flat.

To predict the ultimate fate of the universe, astronomers first have to learn as much as possible about its past. To learn about the past, it is necessary to study the most distant objects visible. Light from distant objects shows us how these objects were behaving at the time the light was emitted (for example, light from the sun shows us how the sun looked about eight minutes ago; light from the Andromeda galaxy shows us how it was 2.5 million years ago).

Because the speed of light is finite, it takes some time for this light to reach us here on earth. So when we study the light from objects at the edges of the universe, we take a journey into the distant past.

The attentions of astronomers turned to supernovae, first discovered by Zwicky in the 1920s. These occur when a star exhausts its nuclear fuel and can no longer support itself against the force of gravity. The star collapses. This collapse may trigger a sudden release of gravitational energy, blowing the star's outer layers away in a massive stellar explosion. Or, under the right conditions, a so-called white dwarf star consisting mostly of carbon and oxygen may gain sufficient mass to trigger runaway carbon fusion reactions, resulting in a similar cataclysmic explosion.

For a brief time, a supernova may outshine an entire galaxy, before fading away over several weeks or months.

Astronomers classify supernovae based on the different chemical elements they are observed to contain from their absorption spectra. Supernovae involving stars that have exhausted their hydrogen are called Type I. These are further subcategorized according to the presence of spectral lines from other atoms or ions. Type Ia supernovae produce no hydrogen absorption lines but exhibit a strong line due to ionized silicon. They are produced when a white dwarf star accretes
mass (perhaps from a neighbouring star) sufficient to trigger carbon fusion reactions.

Type Ia supernovae are of interest because their relatively predictable luminosity and light curve (the evolution of their luminosity with time) means they can be used as ‘standard candles'. In essence, finding a Type Ia supernova and determining its peak brightness provides a measure of the distance of its host galaxy.

Galaxies that would otherwise be too dim to perceive at the edges of the visible universe are lit up briefly by the flare of a supernova explosion.

In early 1998, two independent groups of astronomers reported the results of measurements of the redshifts and hence speeds of distant galaxies that had contained Type la supernovae. These were the Supernova Cosmology Project (SCP), based at the Lawrence Berkeley National Laboratory near San Francisco, California, headed by American astrophysicist Saul Perlmutter; and the High-z (meaning high-redshift) Supernova Search Team formed by Australian Brian Schmidt and American Nicholas Suntzeff at the Cerro Tololo Inter-American Observatory in Chile.

The groups were rivals in the pursuit of data that would allow us to figure out the ultimate fate of the universe. Not surprisingly, their strongly competitive relationship had been fraught, characterized by bickering and disagreement. But in February 1998 they came to agree with each other — violently.

It had always been assumed that, following the big bang and a period of rapid inflation, the universe must have settled into a phase of more gentle evolution, either continuing to expand at some steady rate or winding down, with the rate of expansion slowing. However, observations of Type Ia supernovae now suggested that the expansion of the universe is actually
accelerating.

‘Our teams, especially in the US, were known for sort of squabbling a bit,' Schmidt explained at a press conference some years later. ‘The accelerating universe was the first thing that our teams ever agreed on.'
¹⁵

Adam Reiss, a member of the High-z team, subsequently found a very distant Type la supernova that had been serendipitously photographed by the Hubble Space Telescope during the commissioning of a sensitive infrared camera. It had a redshift consistent with a distance of 11 billion light years, but it was about twice as bright as it had any
right to be. It was the first glimpse of a supernova that had been triggered in a period when the expansion of the universe had been decelerating.

The result suggested that about five billion years ago, the expansion had ‘flipped'. As expected, gravity had slowed the rate of expansion of the post-big-bang universe, until it reached a point at which it had begun to accelerate again. And there was really only one thing that could do this.

The cosmological constant was back.

One supernova does not a summer make, but in 2002, astronauts from the Space Shuttle
Columbia
installed the Advanced Camera for Surveys (ACS) on the Hubble Space Telescope. Reiss, now heading the Higher-z Supernova Search Team, used the ACS to observe a further 23 distant Type Ia supernovae. The results were unambiguous. The accelerating expansion has since been confirmed by other astronomical observations and further measurements of the CMB radiation.

There are other potential explanations, but a consensus has formed around a cosmological constant with a value of 0.73. Its reintroduction into the gravitational field equations is thought to belie the existence of a form of vacuum energy — the energy of ‘empty' spacetime — which acts to push spacetime apart, just as Einstein had originally intended.

We have no idea what this energy is, and in a grand failure of imagination, it is called simply ‘dark energy'.

The standard model of big bang cosmology

The ΛCDM model is based on six parameters. Three of these are the density of dark energy, which is related to the size of the cosmological constant, Λ; the density of cold dark matter; and the density of baryonic matter, the stuff of gas clouds, stars, planets and us.
^*

When applied to the seven-year WMAP results, the agreement between theory and observation is remarkable. Figure 4 shows the ‘power spectrum' derived from the temperature of the CMB radiation mapped by WMAP. This is a complex graph, but it is enough to know that the oscillations in this spectrum are due to the physics of the plasma that prevailed at the time of recombination. In essence, as gravity tried to pull the atomic nuclei in the plasma together, radiation pressure pushed them apart, and this competition resulted in so-called acoustic oscillations.

Figure 4
The 7-year temperature power spectrum from WMAP. The curve is the ACDM model best fit to the 7-year WMAP data. NASA/WMAP Science Team. See D. Larson et al.,
The Astrophyskal Journal Supplement Series, 192 (February 2011), 16.

The positions of these oscillations and their damping with angular scale determine the parameters of the ΛCDM model with some precision. In Figure 4, the points indicate the WMAP power spectrum data with associated error bars, and the continuous line is the ‘best-fit' prediction of the ΛCDM model obtained by adjusting the six parameters. In this case, the best-fit suggests that dark energy accounts for about 73.8 per cent of the universe and dark matter 22.2 per cent.

What we used to think of as ‘the universe' accounts for just 4.5 per cent. This means that the evolution of the universe to date has been determined by the push-and-pull between the antigravity of dark energy and the gravity of (mostly) dark matter.

It seems that visible matter, of the kind we tend to care rather more about, has been largely carried along for the ride.

^*
This sounds a bit like the Copernican Principle described in Chapter 1, but the Copernican Principle goes a lot further. The cosmological principle is in essence a kind of statistical assumption — our perspective from earth-bound or satellite-borne telescopes is representative of the entire universe. The Copernican Principle subsumes the cosmological principle, but goes on to insist that this perspective is due to the fact that the universe was not designed specifically with human beings in mind.

^*
This is a Doppler shift, caused by the speed of motion of the light source, not to be confused with a gravitational shift due to the curvature of spacetime.

^*
A megaparsec is equivalent to 3.26 million light-years (a light-year is the distance that light travels in a year), or 30.9 billion billion kilometres.

^*
Although some heavier elements were formed in the big bang, much of the synthesis responsible for the distribution of elements in the universe is now understood to take place in the interiors of stars and during cataclysmic stellar events, such as supernovae. Hoyle played a significant role in working out the mechanisms of stellar nucleosynthesis. There's more on this in Chapter 11.

^*
And, indeed, of several other, similar, predictions that Gamow, Alpher and Herman had published in the intervening years.

^*
Pope Pius XII had pronounced in 1951 that the big bang model was consistent with the official doctrine of the Roman Catholic church.

^*
More recent evaluation puts this energy somewhere in the region of 200,000 billion GeV.