Published online 10 May 2005 | Nature | doi:10.1038/news050509-4


Swift satellite spies cosmic crash

Massive energy burst supports theory of neutron star collisons.

Crash bang: a computer simulation of two neutron stars colliding.Click here:/news/2005/050509/multimedia/050509-4-m1.html to see the video.Crash bang: a computer simulation of two neutron stars colliding."Click here":/news/2005/050509/multimedia/050509-4-m1.html to see the video.© D. Bock/D. Swesty/A. Calder/E. Wang/NCSA

A satellite that NASA sent to investigate mysterious energy bursts has succeeded in glimpsing its quarry. The data it is beaming to Earth support the leading theory that the bursts are generated by the dramatic collision of two neutron stars as they form a black hole.

NASA's Swift satellite spotted the energy burst at 4:00 GMT on 9 May. The event, named GRB050509b, threw out a very short blast of gamma-rays (γ-rays): 90% of the event's total energy was released in just 30 milliseconds.

The blast is thought to come from two neutron stars colliding in a galaxy about 3 billion light years away. "All of the evidence we have so far points towards this being a neutron star merger," says Paul O'Brien, an X-ray astronomer from the University of Leicester, UK, and a member of the Swift team.

Astronomers think the neutron-star pair originally formed when two orbiting stars exploded in supernovae. The dense remnants of these stars twirled around each other in ever-decreasing circles until they eventually collided and formed a black hole, explains Andrew Levan, a research student on Leicester's X-ray team.

Short and strange

Astronomers think there are several sources of γ-ray bursts in the Universe. Blasts that last more than a few seconds are thought to come from the death of supermassive stars as they collapse in a violent supernova explosion. These are quite well understood, simply because the length of the bursts makes them easier to study.

But short γ-ray bursts have been very mysterious. Some astronomers argue that they are generated when highly magnetic neutron stars, called magnetars, throw out plumes of material.

Swift's observation goes against this theory. O'Brien says that GRB050509b is at least ten times farther away than the most distant magnetar eruptions that Swift would be able to spy. It is more likely, he says, to be a neutron star collision, because that is a much brighter source.

"From simulations of neutron-star mergers, the timescales of the predicted γ-ray burst match up pretty well with these observations," adds Kim Page, another researcher on the Leicester team.

Quick swinging

Swift was carefully designed to catch short bursts of γ-rays and swing around fast enough to record the aftermath of whatever caused them. Just 56 seconds after detecting this burst, Swift got into position to look for the afterglow of X-rays and visible light.

It then sent out e-mail alerts to astronomers around the world, who used ground-based telescopes to look at the area. No one has spotted anything so far, says Page, but this may be because the distant glow is too faint.

This isn't the first microsecond burst that Swift has seen. It spotted one back in February, but couldn't swing into place to observe its source because the Sun was in the way and would have blinded its sensors. "This is the first one we've been able to follow up on," says Page.

Spotting similar short γ-ray bursts in the future will help astronomers to work out how common twinned neutron stars are throughout the Universe, and how many black holes neutron-star collisions account for, says Levan. There are only four pairs of neutron stars known in our Galaxy.

"This short burst is the first of many we hope Swift will detect," says Levan.