Berkeleyan: A newspaper for faculty and staff at UC Berkeley


Berkeleyan HomeSearch BerkeleyanBerkeleyan ArchiveUCB NewsUCB Calendar

This Week's Stories:

Regular Features



High Excitement at High-Z

Astounding Findings on an Expanding Universe

by Robert Sanders, Public Affairs
posted December 02, 1998

The Cygnus Loop supernova as observed by campus's Extreme Ultraviolet Explorer satellite.

Alex Filippenko isn't entirely comfortable with the implications of his findings, theorists can't explain them, and everyone's trying to poke holes in the results. But he and his colleagues may have made one of the most important astronomical discoveries of the century.

What they've found is that nearby supernovas - exploding stars - are moving faster than supernovas billions of light years away. That is, the universe is expanding faster today than yesterday - it's accelerating without end.

The implication, Filippenko says, is that "the ultimate fate of the universe is to expand forever, and to become cold and dark."

Not to worry. The final candle won't go out for perhaps a thousand trillion years. But the idea of an accelerating universe has astounded astronomers and flummoxed theorists.

"There was no hint of this when we started the project," says Adam Riess, a postdoc with Filippenko who did the bulk of the analysis for the study. "We expected to see the universe slowing down, but instead all the data fit a universe that is speeding up."

It was only in 1929 that astronomer Edwin Hubble discovered that all galaxies are receding from one another, implying an expanding universe and a Big Bang that set it off some 14 billion years ago. But astronomers bet their money on a universe that would expand ever more slowly with time, stopping only in the infinite future, or one that would actually stop and begin to fall back on itself. The latter would eventually produce a Big Crunch, like a rewound Big Bang.

Then comes Filippenko and his international team of astronomers, calling themselves the high-z supernova search team, who found that the universe is not only expanding without end, but speeding up as time goes on.

At a workshop on supernovas in late October Filippenko presented new data that only solidifies the findings. Plus a competing team of astronomers headed by Saul Perlmutter of Lawrence Berkeley National Laboratory has reached the same conclusion.

"Both teams are getting the same results with an independent set of supernovas, which means we probably aren't making some great blunder," Filippenko says. "Other astronomers and theorists have to take this seriously."

What this means is that some hidden energy is pushing the universe apart, overwhelming the gravitational forces that tend to slow down the expansion. Even when you add in the so far unobserved dark matter, which is assumed to make up 90 percent of the mass of the universe, there is not enough matter to halt the acceleration, let along stop the universe's expansion.

"It appears there is some funny energy in the universe that makes it expand forever, but at an ever increasing rate," he says.

The supernova experiment began about three years ago, when Filippenko, Brian P. Schmidt of the Mount Stromlo and Siding Spring Observatories in Australia, Nicholas Suntzeff of the Cerro Tololo Inter-American Observatory in Chile, and numerous colleagues began searching for Type Ia supernovas at large distances and measuring the speed at which they are receding from us.

Type Ia supernovas are exploding white dwarfs - burnt-out Sun-like stars at the end of their lives - and astronomers assume they are all about the same intrinsic brightness. By measuring the apparent brightness scientists can calculate how far away they are. Their speed is calculated by measuring how much their color is shifted toward the red end of the rainbow, i.e. the redshift (called "z", whence the term high-z in the team's name).

In a universe that is slowing down, distant supernovas - ones that exploded billions of years in the past - should be moving faster than nearer and more recent supernovas.

Filippenko's group and the Berkeley Lab group reported just the opposite at a meeting last February that drew national attention to the findings. The New York Times even quoted UC Berkeley astrophysicist Richard Muller as saying, "This is one of the top astronomy discoveries of the century, certainly of the decade. It's worthy of a Nobel Prize."

When Riess, Filippenko and other members of the supernova search team finally published their findings in the September issue of The Astronomical Journal, they had a good selection of 16 supernovas from as far back in time as half the age of the universe, 6-7 billion years ago. The competing group at Berkeley Lab, which has accumulated data from 42 supernovas, will soon publish essentially the same conclusion.

One of the main criticisms of these results is that the groups may be misjudging the distance of the supernovas because intervening dust is making them appear dimmer and thus apparently farther away. Another is that supernovas long ago were different from those today, and thus not the same brightness.

"Are we really looking at 100 watt light bulbs, or are they actually 90 watt bulbs and we only think they are 100 watts," Filippenko explains.

The recent meeting in Chicago was called to discuss Type Ia supernovas and all the possible ways that the two groups may have misinterpreted their data and gotten the wrong result.

Filippenko was heartened by a singular lack of consensus among the theorists at the meeting. There was no agreement on how the stars could be dimmed by galactic or intergalactic dust, given that the data show no other indications of dimming. And seven-billion-year-old supernovas don't appear to differ in any obvious way from supernovas today.

If the results are correct, cosmologists are stumped.

"If the universe is accelerating, the only way this could happen is if it contains some substance that is very, very strange compared to normal substances," says Marc Davis, a theoretical astrophysicist at UC Berkeley. "Either that, or we are missing something really fundamental about the workings of the universe."

Most explanations to date are "otherworldly and creepy," he says. A negative pressure unknown in the real world could act to speed up the expansion. Einstein may have anticipated this years ago when he inserted a "cosmological constant" in the general relativistic equations describing the universe. Though he later rejected the constant, a non-zero cosmological constant is the only way to explain the results, Riess says.

One possible explanation for the cosmological constant is that the vacuum of space is not really empty, but seethes with subatomic activity that adds up to a non-zero "vacuum energy." The more space, the more energy to push outward.

Unfortunately, theorists haven't been able to calculate what this should be, and a small but non-zero number wasn't what they were expecting.

Another theorist has suggested a fifth force or essence, quintessence (in keeping with Aristotle's four essences, earth, air, fire and water, and also the four known forces, gravity, electromagnetism, the strong and weak forces), that could explain an accelerating universe.

"These explanations are so unintuitive that it's hard to believe them apart from the mathematics," Davis admits. "But unfortunately, that is the way the universe works."

Nevertheless the result is profoundly important, he says, and other experiments are planned that could confirm or disprove the results. He and colleagues are soon to begin observations with the Keck telescope in Hawaii to survey around 25,000 galaxies at a distance of about 8 billion light years to determine the large scale structure of the universe in the distant past. By comparing that structure with today's structure, project DEEP (Deep Extragalactic Evolution Project) could be able to measure the acceleration of the universe.

"Supernovas aren't the only way to measure this," he says. "We have a variety of tests we can perform, independent of the problems with supernovas, that will let us check whether the supernova results are right."

Another astronomy professor, Joseph Silk, is involved with a European satellite project called Planck that also will measure the universe's acceleration. Due to be launched in 2007, the satellite will measure fine fluctuations in the cosmic microwave background, from which can be deduced the accelaration and curvature of the universe.

For the time being, Filippenko and Riess hope to tighten up their results, and if possible find supernovas even further away.

"As we look at greater and greater distances, at some point we will be looking back before cosmic repulsion became significant, and the observed acceleration should decrease," Filippenko says. "If dust is the problem, we won't see this. So by going to big enough z we can distinguish between whether this is an effect of dust or whether it's due to an accelerated expansion."

UCB Home

This site is produced and maintained by the Office of Public Affairs, University of California, Berkeley.
Copyright 1998, The Regents of the University of California.
For comments concerning this web service please e-mail