NASA's Swift Observatory, launched into orbit last November 20th, was designed to detect gamma-ray bursts coming from billions of light-years away and to chase them down to their points of origin in a matter of seconds (Sky & Telescope: March 2004, page 33, and March 2005, page 27). It's now doing exactly that, with a vengeance. And on May 9th Swift scored a historic first. It recorded a short burst lasting just 0.03 second, swiveled around, and imaged a weak, fast-fading X-ray afterglow that revealed the event's location to within about 8 arcseconds accuracy. Never before has a mysterious short burst been located so precisely. A closer look at its position with other telescopes is giving hints of what short bursts may actually be.
Gamma-ray bursts (GRBs) come in two distinct types, and astronomers have long suspected that they result from two different causes. The more common long bursts, which last from two seconds to several minutes, have been pretty well solved. In the current picture these happen when the core of a hot, very massive Wolf-Rayet star collapses to form a black hole. An extremely energetic disk of superdense material forms right around the hole and emits narrow jets from its poles at barely under the speed of light; the jets punch clean out of the star (as illustrated at right). If a jet is pointed directly at us, we see it as a gamma burst. The burst is often followed by a lingering afterglow, at wavelengths from X-rays to visible light to radio. The afterglow arises as the jet plows through surrounding interstellar matter in the following hours, days, and weeks.
Short bursts, by contrast, can be as short as several milliseconds. They have shown a striking absence of afterglows and it is by afterglows that bursts can be accurately located. So astronomers have had little chance to follow up and see what short bursts actually are. A leading theory is that they result from two neutron stars (or perhaps a neutron star and a black hole) spiraling together, fragmenting into a jet-emitting disk, and collapsing into a single black hole. Such an event would be over very fast, and it would create little or no afterglow.
That theory is getting a boost from Swift's May 9th observations of a short burst in Coma Berenices designated GRB 050509B. Swift slewed its X-ray and ultraviolet/optical telescopes onto the right area within 53 seconds, in time to record the barest wisp of an X-ray afterglow that faded out in 5 minutes. There was no trace of visible or ultraviolet light.
Ground-based telescopes started imaging the site within seconds and minutes, and follow-up observations are still continuing. No visible afterglow has been seen, even by 8- and 10-meter VLT and Keck telescopes (early reports of an optical afterglow proved false). But the biggest telescopes do reveal that the location is on the fringes of a giant elliptical galaxy of old, low-mass stars located 2.7 billion light-years away (at redshift 0.226) a galaxy where no young stars are forming.
This is important. Long bursts are seen to occur only in galaxies full of very young stars, since massive stars have very short lives and can't wander far from their birthplaces. But neutron stars spiraling together would be much older, typically billions of years old giving the host galaxy time to age, starbirth to cease, and the star pair to be kicked some distance out of the galaxy by the explosions that long ago gave the neutron stars birth.
However, the case is far from closed. Several extremely faint, distant star-forming galaxies do show up in the background within the burst's position-error circle. This leaves open the possibility that the burst happened in one of them, among a population of massive stars at high redshift.
Then again, the chance that the burst would appear so close to a foreground galaxy as bright as the giant elliptical just by luck is estimated to be only one in 1,000 or so.
Observations are continuing; some models of short bursts, involving a single collapsing white dwarf or neutron star, predict a visible supernova brightening into view 20 or 30 days after the burst which in this case means at the end of May or in early June.
Swift could repeat its performance by catching another short burst at any time. Meanwhile, it is recording garden-variety long bursts at a rate of about two a week on average. This is nearly the rate that astronomers had hoped for, and it is enabling much more data to be gathered on more bursts than ever before.
For instance, last December 19th Swift caught a long burst that had a complex light curve. Swift relayed the news worldwide fast enough that the 1.3-meter PAIRITEL robotic telescope in Arizona swung to the spot (in Cassiopeia) and began imaging within two minutes of the start of the burst. It was still happening and PAIRITEL caught a source of infrared light fluctuating in time with the gamma rays. This was a first. Only one other burst has been seen in visible or infrared light while the gamma-ray event was still in progress, and in that case, the light was out of sync with the gamma-ray variations; apparently the light came from interactions of the jet with surrounding gas, afterglow-style. In the December 19th case, however, the light apparently provided a look into the guts of the incredibly powerful jet itself.