This article is reproduced with the permission of New Scientist for exclusive use by Nova users.

Dark energy: Was Einstein right all along?
03 December 2005
Magazine issue 2528 of New Scientist magazine
by Stephen Battersby

THE world's most powerful telescopes are being turned on distant supernovae to close in on dark energy, the mysterious stuff that is thought to be pulling the universe apart. And the results so far suggest that dark energy resembles the "cosmological constant", which Einstein proposed before anyone even knew that the universe was expanding.

Astronomers first discovered dark energy in the 1990s, while studying type 1a supernovae - exploding stars that act as useful markers for time and distance in the universe. It's known that these supernovae always shine with about the same peak brightness, so by measuring how bright one appears from Earth, you can work out how far away it was and how much time has passed since the star exploded.

Because the supernova's light has travelled through an expanding universe, its wavelength has stretched and shifted towards the red end of the spectrum - so the greater the red shift, the more the expansion since the supernova explosion. Put the ages of different supernovae and their red shifts together, and you know how the expansion rate of the universe has changed over time.

From observations of tens of type 1a supernovae, astronomers concluded in 1998 that the expansion of the universe is accelerating and they gave the unknown cause a name: dark energy. What it is and how it works is a complete mystery, but there are a few rival theories.

To narrow them down, the Supernova Legacy Survey (SNLS) team aims to study hundreds of type 1a supernovae and use them to plot as precisely as possible the history of the universe's expansion. The international team is using the 3.6-metre Canada-France-Hawaii telescope (CFHT), atop the 4200-metre Hawaiian mountain Mauna Kea, to monitor large chunks of the sky at a time. The survey has already seen about 200 supernovae, says team member Isobel Hook of the University of Oxford.

Once a supernova is picked up by the CFHT and its brightness measured, an even larger telescope swings into action to record the red shift of the faint supernova. The project is using most of the world's biggest telescopes - the four 8.2-metre instruments of the Very Large Telescope in Chile, the two 8.1-metre Gemini telescopes in Hawaii and Chile, and the two 10-metre Keck telescopes in Hawaii.

The team's preliminary conclusion, based on an analysis of 70 supernovae, fits the most conservative theory of dark energy: that space itself has some inherent energy. Einstein showed that if the vacuum of space has a fixed energy - which he called the cosmological constant - it would produce a force that would counter gravity. The latest SNLS findings show that the expansion rate of the universe is changing with time in just the way you would expect if there is a cosmological constant. More precisely, the observations show that dark energy's repulsive force has not changed by more than about 20 per cent since 8 billion years ago, when the universe was half its current size.

This finding seems to rule out some alternative theories. For instance, one idea is that the repulsion comes from fractures in the fabric of space-time that formed as the universe cooled down after the big bang. Another arises out of a version of supergravity, which is an attempt to describe gravity as a quantum force carried by particles known as gravitons and gravitinos. According to both these theories, the density of dark energy should have faded faster with time than the new observations suggest it has.

But not everyone is convinced that the SNLS findings constrain the nature of dark energy any better than other projects so far. "Their pace of discovery is faster than anyone else's," says cosmologist Adam Riess of the Space Telescope Science Institute in Baltimore, Maryland. "But I could point to half a dozen papers that have the same constraint."

However, the studies he refers to have all relied on combining several different data sets. The SNLS data is uniform. "That gives us improved confidence in the result," says SNLS team member Saul Perlmutter of the Lawrence Berkeley Laboratory in California.

For the latest from New Scientiist visit www.newscientist.com