© SCIENCE SOURCE/SPLThis multi-wavelength image of Abell 520 shows the aftermath of a complicated collision of galaxy clusters, some of the most massive objects in the universe. In this image, the hot gas as detected by Chandra is coloured red.
For more than a decade, astrophysicists have grappled with evidence of a baffling force that seems to be pushing the universe apart at an ever-increasing rate. Exactly what constitutes the dark energy responsible for this cosmic speed-up is unknown, says Michael Turner at the University of Chicago. "The simplest question we can ask is 'does the dark energy change with time?'"
So far, the evidence has suggested that dark energy is constant, though its effect on the universe has become stronger as the universe has expanded and the gravitational force between objects weakens with distance.
Now an analysis of supernovae suggests dark energy may actually be on the wane. In a paper on the physics preprint website, a team led by Arman Shafieloo at the University of Oxford examined a newly released catalogue of supernova explosions, including a number of relatively recent blasts nearby. They found that the new data made the best fit with a universe in which dark energy is losing strength. "It seems acceleration is slowing down," says Shafieloo.
The first evidence of dark energy emerged in 1998, when two teams of astronomers spotted distant supernova explosions that appeared dimmer than expected, and so further away. The find suggested the exploding stars were receding from Earth faster than anticipated, and therefore so was the rest of the universe. "Dark energy" was invoked to explain the apparent anomaly. Since then more supernovae have been catalogued to help build up a picture of how the universe has expanded over time.
The biggest set of supernova data was released earlier this year by the Harvard-Smithsonian Center for Astrophysics in Cambridge, Massachusetts. It includes data on 147 supernovae that exploded in the last billion years, more than half of them newly discovered. The Harvard team analysed the new supernovae assuming that dark energy has remained unchanged.
Shafieloo, however, dropped the requirement that dark energy be constant over the universe's history. Together with Varun Sahni of the Inter-University Centre for Astronomy and Astrophysics in Pune, India, and Alexei Starobinsky of the Landau Institute for Theoretical Physics in Chernogolovka, Russia, Shafieloo used an approach he says is particularly sensitive to rapid changes in the universe's rate of expansion.
Beginning with factors like red shift - a measure of how much the expansion of space has stretched the light from each explosion - they calculated a representative number for the epoch in which each supernova occurred. After plotting all of these numbers, they found that the best fit was a scenario in which dark energy has weakened over the last 2 billion years, causing cosmic acceleration to slow down. Shafieloo cautions that their result is preliminary, but adds that it could be time to begin revisiting other models of dark energy.
"Their approach is reasonable," though the effect is slight, says cosmologist Dragan Huterer of the University of Michigan, Ann Arbor. "If that is really the case it would be a tremendous discovery."
Indeed, it would change our ideas about the source of dark energy. Until now, all signs have pointed to the cosmological constant as the simplest explanation for the accelerating expansion of the universe. This constant is an unchanging energy that arises from quantum fluctuations in the vacuum of space. "The cosmological constant is the only thing that makes any sense to particle physicists right now," says Huterer.
Yet if dark energy is changing, the cosmological constant could not be the driver. Instead, it would suggest far more exotic physics at work. It might even mean dark energy does not exist at all. One example of an exotic origin is "quintessence", a theoretical quantum field that permeates space like the as-yet-unidentified field thought to have driven inflation right after the big bang. This field could be dissipating and losing energy, eventually causing the universe to decelerate and collapse back on itself.
A more likely explanation for the team's result is a slight bias in the new supernova data, Huterer says. Robert Kirshner, a member of the Harvard team, agrees. "I think these are serious people whose analysis should be taken seriously, but there can be more than one cause for the apparent effect," he says.
For example, a potential bias could have been introduced thanks to dimmer objects being easier to see if they are nearby. It is possible that the Harvard team happened to catalogue a disproportionate number of nearby supernovae that were faint or obscured by dust. Astronomers must correct for the dimming effect of dust and other subtleties in order to estimate a supernova's true peak brightness. But the team may have overcompensated in this correction, producing a catalogue of nearby supernovae that are slightly too bright for their distance. That would create the illusion that the universe's acceleration has been slowing.
New observations from other groups need to be examined to look for the same effect, Kirshner says, though determining whether dark energy really is changing could take a while. The fine details of so many supernovae have been recorded that the so-called "systematic floor" has been hit - a scenario in which everything from subtle differences between supernova explosions to the warp of a telescope mirror can skew results, Huterer says.
Upcoming "precision projects" like the Dark Energy Survey, which will mount a supersensitive 500-megapixel camera on a 4-metre telescope at the Cerro Tololo Inter-American Observatory in Chile, aim to reduce some of the sources of uncertainty. One of the project's aims is to measure some of the universe's most recent history, by recording about 2000 supernovae that have exploded in the last 7 billion years.
Other probes that will push the limit in sensitivity are still in early planning, including two space probes - the US's Joint Dark Energy Mission and Europe's Euclid. Some astronomers suspect a partnership will be forged between these missions to send up a single international probe instead.
It is practically impossible to definitively discover if dark energy is constant. "There isn't a target to shoot for," says cosmologist Sean Carroll of Caltech. "As we narrow down the error bars and get closer and closer to perfectly constant, there's no point at which you say 'OK. We're done. Dark energy is constant.'"
However, the next burst of effort could reveal in glowing detail if dark energy has been changing. "It would be a surprise if we found that dark energy were varying with time," says Carroll, "but it would be so hugely important that it's still worth looking."
We don't need the stuff
Some theories claim to explain the universe's accelerating expansion without resorting to dark energy.
Some have it that cosmic acceleration is the result of the breakdown of general relativity in which space is forced apart at large scales. If that were the case, discrepancies should show up between measurements of the universe's expansion and the number of galaxy clusters, which is used to gauge how easily the universe can grow large structures. Such evidence has been lacking.
What's more, most such theories have a hard time explaining the acceleration of the cosmos without introducing other effects, such as instabilities that could preclude the formation of stars and galaxies. "None of the ideas make everything snap into place," says Sean Carroll of Caltech.
Another possibility is that matter is not uniformly distributed over large scales, and that the Earth is inside a vast bubble of space that is relatively devoid of matter. Gravity would have less pull in such a bubble, so it would expand rapidly. This expansion would affect light as it reached us from supernovae, meaning that they only appear to be moving away from us increasingly quickly, and that dark energy need not be invoked.
This scenario seems unlikely, says Alexei Starobinsky of the Landau Institute for Theoretical Physics in Chernogolovka, Russia. But the opposite - a smaller region with a slight excess of matter - could create a tug that would look like dark energy weakening of late. "That could explain this difference," he says.
Reader Comments
to our Newsletter