HILO, Hawai'i — The telescopes on Mauna Kea have given scientists some new clues about "dark energy," the mysterious force that for billions of years has been driving the universe to expand more and more rapidly.

Scientists determined in 1998 the universe is expanding at an accelerating rate, which was a revelation to astronomers who had believed the expansion of the universe was gradually slowing because of gravity.

Theorists attribute that ever-accelerating expansion to "dark energy," a force scientists know only from its effects on the cosmos. The 1998 work triggered a torrent of theoretical work that attempts to explain what the "dark energy" might be, and how it works.

The Supernova Legacy Survey, an international team of researchers in France and Canada, is beginning to provide new data to help whittle down the number of viable theories about dark energy.

"The goal is that we start narrowing in where it is we're looking for what possible explanations could there be for the dark energy," said study co-author Saul Perlmutter, a physics professor at the University of California Berkeley.

The survey team has collaborated with telescope observers in England and the United States to use telescopes on Mauna Kea to study exploding stars for more hints about how dark energy works.

The new research, which will be published in an upcoming issue of the journal Astronomy & Astrophysics, centers on the study of a particular type of supernova stars that consistently explode in very similar patterns.

The astronomers used a 340 million-pixel camera called MegaCam, built by the Canada-France-Hawai'i Telescope and the French atomic energy agency, Commissariat à l'ƒnergie Atomique. The MegaCam images from the 11.8-foot Canada-France-Hawai'i Telescope on Mauna Kea are used to search for distant supernovae.

Light from the exploding stars are then studied with some of the largest telescopes worldwide, including the Frederick C. Gillett Gemini North Telescope and the W.M. Keck Observatory on Mauna Kea, as well as the Gemini South Telescope on the Cerro Pachón mountain in the Chilean Andes, and the European Southern Observatory Very Large Telescopes (VLT) at the Paranal Observatory in Atacama, Chile.

The stars are so bright that they can be seen billions of light years away, which means astronomers are observing how the stars behaved billions of years ago.

Perlmutter said the new research is based on the exploding stars' "redshift," or the amount that light from the distant stars is stretched as it travels to observers on Earth.

"Anything that is not held together in an expanding universe gets stretched, and that includes the wave lengths of light," Perlmutter said. "The amount that we see them stretched is exactly the same amount by which the universe is expanding."

Since some of the exploding stars the astronomers study are much closer than others, astronomers can calculate how much the universe stretched during different periods of its history.

One important finding is that "dark energy" behaves a great deal like the "cosmological constant" that Albert Einstein proposed but later recanted. Einstein proposed the constant in 1917 in an attempt to balance equations of General Relativity. He set the constant at zero — the value it would have to be for a non-expanding universe — because astronomers at the time believed the universe was stable. Later, when it was proved the universe was expanding, scientists noted the constant as a blunder. The new research suggests though his proposed zero constant was an error, Einstein had the right idea.

The new observations reveal that dark energy behaves very consistently like Einstein's cosmological constant idea, to a precision of 10 percent over enormous stretches of time covering billions of years.

That effectively ruled out a small group of theories that predicted the effect of the dark energy would change more dramatically over time as the universe expanded.

While that hardly unlocks the secrets of "dark energy," it is based on only about 20 nights of data from a project that will involve nearly 200 nights of observing time.

"All of this work that we're doing now and in this next five or 10 years ... is aimed at even getting us into the ball park of what kinds of theories are possible," Perlmutter said. "It's a huge job, and this is the first step in that job at a whole new level of precision that we have to reach."

Reach Kevin Dayton at kdayton@honoluluadvertiser.com.