Simulation Computed at NERSC Matches Historic Gamma-Ray Burst

BERKELEY, CA — After three decades of scientific head-scratching, the origins of at least some gamma-ray bursts (GRBs) are being revealed, thanks to a new generation of orbiting detectors, fast responses from ground-based robotic telescopes, and a new generation of computers and astrophysics software. A GRB detected on March 29, 2003 has provided enough information to eliminate all but one of the theoretical explanations of its origin. Computational simulations based on that model were already being developed at the National Energy Research Scientific Computing (NERSC) Center at the U.S. Department of Energy's Lawrence Berkeley National Laboratory when the discovery was made.

Gamma radiation from outer space is blocked by the Earth's atmosphere. But if it were not, and if we could see gamma rays with our eyes, about once a day, somewhere in the world, people would see a spectacular flash, a hundred million times brighter than a supernova. These gamma-ray bursts, first discovered in the 1960s by satellites looking for violations of the Nuclear Test Ban Treaty, are the most energetic events in the Universe, but they are also among the most elusive. GRBs appear randomly from every direction. They do not repeat themselves. And they last from only a few milliseconds to a few minutes — astronomers consider a GRB long if it lasts longer than 2 seconds.

With data so hard to pin down, scientists for a long time could only speculate about what causes GRBs. Speculations ranged from the intriguing (comet/anti-comet annihilation) to the far-fetched (interstellar warfare). By 1993, 135 different theories on the origin of GRBs had been published in scientific journals. But in recent years, three theoretical models emerged as frontrunners — one involving colliding neutron stars, and the other two involving supernovas, the "supranova" and "collapsar" models.

In the collapsar model (introduced in 1993 by Stan Woosley, professor and chair of astronomy and astrophysics at the University of California, Santa Cruz), the iron core of an aging star runs out of fuel for nuclear fusion and collapses of its own weight, creating a black hole or a dense neutron star. Material trying to fall onto this object forms a hot swirling disk and a narrow jet, which shoots out of the star in less than ten seconds at nearly the speed of light. When this jet erupts into interstellar space, it creates a "fireball" of gamma rays — the highest energy, shortest wavelength form of electromagnetic radiation. The rest of the star explodes as a supernova, but that event is eclipsed by the brighter GRB.

In the late 1990s, GRB research took a great leap forward when a new generation of orbiting detectors were launched, including the Italian-Dutch satellite BeppoSAX and NASA's High-Energy Transient Explorer (HETE). These satellites were designed to provide quick notification of GRB discoveries for immediate follow-up observations by ground-based instruments. In 1997 astronomers discovered that long GRBs leave an afterglow of lower-energy light, such as X-rays or visible light, that may linger for months, allowing researchers to pinpoint where the GRB originated and providing important clues about the event that produced it. The presence of iron in the afterglow light strongly suggested star explosions. In 1998, the supernova theories got an additional boost when a GRB and a supernova appeared in the same vicinity at roughly the same time; but the data was inconclusive and some scientists remained skeptical about the connection. Several similar suggestive but inconclusive events in subsequent years divided astronomers into two camps on the issue of associating GRBs with supernovas.

Figure 1. This image from a computer simulation of the beginning of a gamma-ray burst shows the jet 9 seconds after its creation at the center of a Wolf-Rayet star by the newly formed, accreting black hole within. The jet is now just erupting through the surface of the star, which has a radius comparable to that of the sun. Blue represents regions of low mass concentration, red is denser, and yellow denser still. Note the blue and red striations behind the head of the jet. These are bounded by internal shocks.

The skepticism was dispelled on March 29, 2003, when HETE detected an unusually bright and close GRB — only 2.6 billion light years from Earth instead of the typical 10 billion. The discovery triggered a swarm of observations that found the unmistakable spectral signature of a supernova in the afterglow. This event, named GRB030329 after its detection date, was dubbed the "Rosetta stone" of GRBs in a NASA news release, because it conclusively established that at least some long GRBs come from supernovas, and it confirmed the collapsar model as the only theory that matched the data. The March 29 burst's afterglow was so bright that it allowed astronomers to study the event in unprecedented detail; they even joked about its casting shadows. The Hubble Space Telescope will be able to study the afterglow for another year, and radio telescopes may be able to track it longer.The first 3D computational simulations of jet formation and breakout in the collapsar model (Figures 1 and 2) were already being conducted on NERSC's "Seaborg" IBM supercomputer by Woosley and Weiqun Zhang, a Ph.D. candidate at UC Santa Cruz, under the sponsorship of the DOE Office of Science's SciDAC Supernova Science Center, one of two SciDAC (Scientific Discovery through Advanced Computing) programs focusing on developing new computational methods for understanding supernovas. The goal of the Supernova Science Center is to discover the explosion mechanism of supernovas through numerical simulation.

Figure 2. This image from a computer simulation shows the distribution of relativistic particles (moving near light speed) in the jet as it breaks out of the star. Yellow and orange are very high energy and will ultimately make a gamma-ray burst, but only for an observer looking along the jet (± about 5 degrees). Note also the presence of some small amounts of energy in mildly relativistic matter (blue) at larger angles off the jet. These will produce X-ray flashes that may be seen much more frequently.

Zhang, Woosley, and colleagues had been modeling jet formation after core collapse in both two and three dimensions for a few years when observations from GRB030329 validated their work and made it newsworthy. "We weren't utterly surprised," Woosley said, "because the evidence associating GRBs with supernovas had been accumulating for several years. But we weren't totally confident, either. When the light spectrum from the March 29 burst confirmed that it came from a supernova, that felt good."

Woosley emphasized, however, that "the data is still way ahead of the theory." The computational simulations are still incomplete, and the resolution needs to be improved. Upcoming work will include higher-resolution 3D studies of jet stability — which will help scientists interpret the differences in GRB observations — as well as simulating the full star explosion that accompanies the GRB and the core collapse that precedes it. The project will be challenging, even on Seaborg, currently America's largest computer for unclassified research.

"This does not mean that the gamma-ray burst mystery is solved," Woosley added. "We are confident that long bursts involve a core collapse, probably creating a black hole. We have convinced most skeptics. We cannot reach any conclusion yet, however, on what causes short gamma-ray bursts."

Aside from explaining the mysterious long GRBs, the study of supernovas will fill an essential gap in our knowledge of the Universe, because supernovas are currently thought to be the source of all the elements heavier than iron. Because supernovas cannot be recreated in a laboratory, numerical simulation is the only tool available for interpreting observational data and for developing a detailed understanding of the physical processes involved. NERSC's IBM supercomputer is currently being used by several research groups studying supernovas.

The first analyses of GRB030329 were published in three papers and a commentary in the June 19, 2003 issue of Nature. Additional images and animations are available on the NASA Goddard Space Flight Center Web site.