Closing In on Gamma-Ray Bursts

Inspiralling neutron stars
This illustration shows two neutron stars spiraling toward each other. When the two compact objects merge, they form a black hole. According to theory, two short-lived jets shoot away from the black hole, generating a short gamma-ray burst.
Courtesy NASA.
Thanks to the Swift and HETE-2 satellites, astronomers have just gotten a lot closer to cracking a long-standing mystery: the origin of "short" gamma-ray bursts (GRBs). Observations of one of these bursts, which flared on July 24th, support the leading theoretical model. In this picture, two compact stellar remnants in a binary system come together in an explosive flash of gamma rays. The binary can consist of either two neutron stars or a black hole and a neutron star.

First discovered in the 1960s, GRBs fall into two broad categories. About 85 percent last anywhere from 2 seconds to several minutes, whereas the remaining 15 percent last less than 2 seconds. In the late 1990s and early 2000s, the Dutch/Italian BeppoSAX satellite quickly pinpointed the location of several "long" GRBs. These quick detections allowed for follow-up observations of afterglows, which showed convincingly that these extraordinarily violent events occur when the cores of massive stars collapse, forming black holes. Magnetic fields channel some of the infalling stellar material into two narrow jets that shoot away from a black hole at near-light speeds. Shock waves within the jets generate the gamma rays.

GRB 050509B location
On May 9th a short gamma-ray burst went off somewhere in the dashed circle, as determined by Swift's X-Ray Telescope. It likely happened in the fringe of the 17th-magnitude giant elliptical galaxy that dominates this visible-light view taken with the 3.5-meter WIYN Telescope. But there is a small chance it occurred in a background galaxy. The July 24th short GRB definitely went off in an elliptical galaxy.
Courtesy Joshua Bloom, et al.

Unfortunately, BeppoSAX could not slew fast enough to pinpoint the precise locations of short GRBs, making follow-up observations all but impossible. But by living up to its name, Swift has changed everything. Swift detected a short GRB on May 9th and then observed a short-lived X-ray afterglow in the outskirts of an old elliptical galaxy with minimal star formation. This fits nicely with the merger theory. It takes 100 million years or longer for two compact objects in a binary system to merge as their mutual orbit slowly decays. This gives a binary plenty of time to wander far from its birthplace. In contrast, long GRBs always have been observed near star-forming regions, because the massive stars that produce them don't live long enough to travel far from their place of origin.

On July 24th Swift pinpointed another short GRB's location. A team led by Edo Berger (Carnegie Observatories) immediately followed up and caught a fading afterglow with the 6.5-meter Magellan Telescope, the 40-inch Swope Telescope, and the Very Large Array network of radio antennas. Unlike the May 9th event, whose location could not be nailed down precisely because its afterglow faded within minutes, the July 24th GRB afterglow lingered for 35 hours. This enabled Berger and 23 colleagues to link the GRB definitively with an ancient elliptical galaxy 3 billion light-years away. As reported in a paper submitted to Nature, Berger's team discovered that the afterglow is 8,000 light-years from the galaxy's center, which is not actively forming stars. Like the May 9th short GRB, the July 24th burst is far from any star-forming region.

Merging neutron stars
Not a place you'd want to be. An artist depicts the violent merging of two neutron stars. According to theory, this cataclysmic collision sets off a chain of events that triggers a short GRB.
Courtesy NASA.

The optical and radio observations of the fading July 24th afterglow suggest that short-lived jets injected energy into the surrounding interstellar material, fulfilling another prediction of the merger model. This theory predicts that colliding compact objects will leave behind a small amount of material that is channeled into two short-lived jets. Moreover, the July 24th short GRB was 10 to 1,000 times less energetic than a typical long GRB, which also agrees with the merger theory.

Swift isn't the only NASA satellite that is catching short GRBs. On July 7th HETE-2 localized a short GRB. But in this case follow-up optical and X-ray observations detected a fading afterglow in a star-forming region of a relatively nearby galaxy. This is not a death knell, however, for the merger scenario. "Galaxies of all types and ages should and do produce compact binary systems that eventually merge," says Joshua Bloom (University of California, Berkeley), who led the studies of the May 9th GRB. "The Milky Way has more than a half-dozen known neutron star binaries that will eventually merge. So it would be surprising to me if short-bursts only occurred in ellipticals."

GRB generator
An artist's concept of high-speed jets shooting away in opposite directions to create a gamma-ray burst.
Courtesy NASA, Dana Berry, and SkyWorks Digital.

While the May 9th and July 24 events provide strong evidence that short GRBs come from mergers, Robert Duncan (University of Texas) points out that an unknown fraction could result from powerful flares on magnetars — neutron stars with stupendously powerful magnetic fields. A giant flare from the Milky Way magnetar SGR 1806-20 was observed on December 27, 2004. The characteristics of this flare had much in common with short GRBs, and Swift could have detected the event had it occurred in a galaxy up to 200 million light-years away.

Bloom emphasizes that despite the great strides made in solving the GRB mystery, the recent observations fall short of "absolute proof" that the short bursts are caused by mergers. "What is clear is that the progenitors of short GRBs must be very different than those of long GRBs," he says.

"Finding an unambiguous counterpart for a short hard bursts has been at the top of almost everyone's list of the most pressing issues in GRB science for years," adds Stan Woosley (University of California, Santa Cruz), a leading GRB theorist. "It is amazing that, after all that time, two independent confirmations would come within weeks of each other."