Astronomers have reached a fundamental impasse in their understanding of the universe: they cannot agree how fast it is flying apart. And unless a reasonable explanation can be found for their differing estimates, they may be forced to completely rethink their ideas about time and space. Only new physics can now account for the cosmic conundrum they have uncovered, many believe.
“Five years ago, no one in cosmology was really worried about the question of how fast the universe was expanding. We took it for granted,” says astrophysicist Daniel Mortlock of Imperial College London. “Now we are having to do a great deal of head scratching – and a lot of research.”
This view is backed by the US astrophysicist and Nobel prize winner Adam Riess, of Johns Hopkins University. “I think this issue has become a big deal. We were getting better and better at understanding the universe and yet the closer we have been looking, the more we have found there are all these mysterious components.”
Over the decades, these surprises have included the discovery of dark matter – believed to be made up of as yet undetected particles – whose extra gravitational pull explains why galaxies do not fly apart. In addition, astronomers have also discovered the existence of dark energy, which is accelerating the rate at which the cosmos is expanding.
“Those two discoveries were remarkable enough,” adds Riess who won his Nobel for his involvement in the discovery of dark energy. “But now we are facing the fact there may be a third phenomenon that we had overlooked – though we haven’t really got a clue yet what it might be.”
Scientists first realised the universe was expanding in the 1920s, when the US astronomer Edwin Hubble found that the greater the distance between two galaxies, the faster they are moving apart. It remains one of the most important scientific discoveries ever made.
One way to think about this phenomenon is to imagine a fruit loaf that is being cooked in an oven. As its dough rises and expands, the raisins and sultanas embedded inside the loaf move farther away from each other. The universe is like that fruit loaf. It is expanding and causing the galaxies – the raisins and sultanas embedded in it – to speed away from each other.
But if the universe was getting bigger and bigger, one key question remained: what is the exact rate of this expansion? Just how quickly is the cosmos flying apart? Or to be more precise: what is the exact value of the Hubble constant, as the speed of the universe’s expansion was subsequently termed? It is a highly important, much sought-after value because it will tell us a great deal about the origin, age, evolution, and ultimately the fate, of the cosmos. Hence the effort astronomers have made over the past century to find a precise answer.
These observations have been based on two very different approaches, however. One has focused on the behaviour of galaxies near our own galaxy, the Milky Way, and has involved scientists working out – with increasing precision – just how fast they are speeding away from each other. “That’s the local approach,” says Riess whose own work has concentrated on trying to perfect the measurement of distances between galaxies in our region of the universe. (See “Hubble’s constant and Henrietta Leavitt”, below.)
The other method for establishing the Hubble constant has involved astronomers looking at the rippling pattern of light, called the cosmic microwave background, that formed just after the big bang birth of the cosmos 13.8bn years ago. This background has been surveyed with increasing precision by US and European satellites – most recently by the European Space Agency’s Planck observatory – and these observations have allowed scientists to build a model that takes account of dark energy and dark matter and that shows how the early universe’s growth would probably have produced an expansion that astronomers can measure today.
And until very recently, these two different approaches produced estimates that appeared to be consistent with each other, although there were considerable uncertainties associated with both measurements. “Everyone’s best bet was that the difference between the two estimates was just down to chance, and that the two values would converge as more and more measurements were taken,” says Mortlock. In other words, as the two values were tested with greater and greater precision, their differences would disappear.
Unfortunately for astronomers seeking a simple fix to the problem, this has not happened. “In fact, the opposite has occurred,” says Mortlock. “The discrepancy has become stronger. The estimate of the Hubble constant that had the lower value has got a bit lower over the years and the one that was a bit higher has got even greater.”
Today, those using Planck and cosmic background data to obtain a value for the Hubble constant get a figure of 67.4 plus or minus 0.5. By contrast the local approach gives a figure of 73.5 plus or minus 1.4. These values represent the two different values we have for the expansion of the universe. (See “A matter of metrics”, below.)
The dissimilarity may not sound great but it is significant. They say there is now less than a one in 100,000 chance that this difference can be accounted for by chance. “This is not just two experiments disagreeing,” says Riess. “We are measuring something fundamentally different. One is a measurement of how fast the universe is expanding today as we see it. The other is a prediction based on the physics of the early universe and on measurements of how fast it ought to be expanding. And these measurements have now been independently corroborated by other groups so that the discrepancy does not depend on any one tool or any one team.
“And if these values don’t agree, that means there is a very strong likelihood that we’re missing a factor in the cosmological model that connects the two eras.” In short, something appears to be absent from our understanding of the universe and the Hubble constant has become the focus of a hotly contested battle to discover the nature of this invisible influence.
For a start, the difference between two values has implications for the age of the universe, lopping off more than a billion years of its existence in one case. “Changing the Hubble constant from 67.4 to 73.5 would mean it must have been flying apart faster than previously supposed and so must be younger than its currently accepted age of 13.8bn years,” says Mortlock. “In fact it would cut to 12.7bn years.”
And this does cause problems. There are some very old stars in the universe that have estimated ages of around 12bn years, and this makes a revalued cosmic chronology a bit difficult to stomach. Stars take a long time to form, after all.
However, that is not the real issue, says Mortlock. “The basic problem is that having two different figures for the Hubble constant measured from different perspectives would simply invalidate the cosmological model we made of the universe. So we wouldn’t be able to say what the age of the universe was until we had put our physics right.”
Because of the independent corroborations, Riess has become more confident there must be a fundamental discrepancy involved, one that is not due to methodological flaws or mistakes in observations but caused by a feature of our universe of which scientists have had no previous inkling. “I think that there is something interesting going on,” he says. “And I do not consider measurement errors interesting.”
But if measurement error can no longer be considered a cause of the differences in Hubble constant values, what new concepts could explain this discrepancy? Astronomers have already put forward a number of suggestions.
One idea proposes that the universe contains a new class of subatomic particle that travels close to the speed of light. These entities are called dark radiation and could also include already known particles such as neutrinos. These would affect the speed of the universe’s expansion.
Another idea is there was a special, intense dark-energy episode not long after the big bang, which expanded the universe faster than astronomers had previously appreciated.
And finally there is the possibility that the particles that make up dark matter interact more strongly with normal matter than previously assumed. Again this would have an impact on the Hubble constant.
Not every scientist is over the moon about the prospect that one of these proposals is the answer to their measurement quandary and still hope that in the end it may be possible to reconcile the two values they get for the Hubble constant. This point is stressed by Mortlock. “We already have found that our universe is dominated by dark matter and by dark energy whose effects we can observe but whose basic nature is a mystery. They are two huge question marks that are already hanging over our understanding of the cosmos. Personally, I do not feel the need for a third.”
For his part, Riess takes a slightly more optimistic view. “We are not made of dark matter or dark energy but we have uncovered their existence even though, intuitively, they are not part of our experience of life on Earth. That suggests we are heading in the right direction in understanding the universe – though it just may be that we have at least one other step to take.”
Hubble’s constant and Henrietta Leavitt
Stars known as Cepheid variables have played a critical role in our understanding of the expansion of the universe. These stars, which are relatively common, vary in brightness over periods of days or weeks. In 1908 Henrietta Leavitt discovered there was a relationship between the brightness of a Cepheid variable star and the time it took to go through a full cycle of change in its luminosity.
As a result, by measuring the period of a Cepheid variable, it became possible to calculate its true brightness. Then, by comparing this to its apparent brightness, astronomers could calculate the distance of the star – and the galaxy in which it is found. Hubble used this understanding in his work to calibrate cosmological distances, and Cepheids today continue to provide key calibration for astronomical distances for the local method for calculating the Hubble constant.
A matter of metrics
A Hubble constant of 70 would mean that the universe is expanding at a rate of 70 kilometres per second per megaparsec. To understand what this means, you must first appreciate that a parsec is a measure of astronomical distance and that a megaparsec is the equivalent of a million parsecs. In turn there are 3.3 light years to a parsec, so a megaparsec is the equivalent of 3.3m light years. Thus for each 3.3m light years that a galaxy is distant from us, it will move an extra 70 kilometres (43.5 miles) per second faster from us, as a result of the expansion of the universe.
guardian.co.uk © Guardian News & Media Limited 2010