Measurements of the X-ray afterglow of long γ-ray bursts largely clarified the origin of these bright flashes of cosmic radiation. Their shorter-lived siblings are now beginning to divulge their secrets, too.
The distance, energy output and source of the mysterious flashes of high-energy radiation known as short γ-ray bursts have so far resisted all attempts at their elucidation. Observations presented in this issue1,2,3,4 establish for the first time the cosmological distance of the bursts and provide solid support for the favoured theoretical model for their origin. In this scheme, short γ-ray bursts are the expression of the enormous energy released when two compact objects — remnants of exploded stars in the form of highly dense neutron stars or black holes — merge.
Scientists have been fascinated by γ-ray bursts (GRBs) for more than 30 years. The most glamorous celebrities of the sky, they appear unpredictably, and — after a brief and intense show, during which their brightness overwhelms that of any other celestial source of γ-rays — they disappear. There are two classes of burst, identified according to their duration: long (longer than two seconds) and short5. The source of long bursts was revealed in 1997, thanks to the precise and fast localization capabilities of the BeppoSAX satellite put into orbit by the Italian and Dutch space agencies. The result was the discovery of the bursts' fainter, long-lived afterglow emission at X-ray6 and optical7 frequencies, lower than the original γ-ray emission. The observations linked these events to the explosive collapse of the cores of young, massive stars in sub-luminous, star-forming galaxies at redshifts of 1–2. (The redshift of cosmological radiation is an indicator of a source's distance; a redshift of z indicates that the Universe has expanded by a factor of 1+z since the radiation was emitted.)
The long-wavelength afterglow of short GRBs has until now evaded detection. The inference — that their afterglow is simply much dimmer than that of the long GRBs — fits with the merger model favoured to explain short GRBs. This model starts with the supernova explosions of two massive progenitor stars in a binary system, which collapse under their own weight to form extremely dense neutron stars or, in the case of more massive progenitors, black holes. The kick imparted to these compact objects by their supernova explosion sends them into less-dense spaces in their galaxy, where they eventually merge, producing the short γ-ray burst (Fig. 1). The afterglow of a burst is produced when material ejected by the explosion at relativistic speeds interacts with the surrounding medium. Therefore, because the merger occurs in less-dense regions of a galaxy, little or no afterglow emission would be expected from a short GRB.
Nevertheless, when the light curves of tens of short GRBs that had been observed independently by BeppoSAX and NASA's BATSE satellite were summed up, they revealed hard, or high-frequency, X-ray tails lasting several tens of seconds8,9. It was an indication, albeit of limited statistical significance, that some long-lived emission was indeed present, and left open the possibility that the X-ray afterglows of short GRBs had been missed not because of their faintness, but simply because none of the available X-ray cameras was pointing the right way. The launch in November 2004 of Swift, a new NASA satellite carrying a wide-field hard-X-ray imager, was intended to eliminate that possibility.
Elsewhere in this issue, the fast and precise localization of two short GRBs is reported in four papers1,2,3,4. Gehrels et al. (page 851)1 report the identification by Swift of a first event (GRB 050509B) on 9 May this year. The satellite was slewed rapidly in the direction of the burst, enabling its narrow-field X-ray telescope to pinpoint to within 10 arcseconds (1/360th of a degree) a source of faint X-ray emission one minute after the burst. This allowed the authors to assign a likely origin for the burst — a luminous, non-star-forming elliptical galaxy at a redshift of 0.225.
Villasenor et al. report (page 855)2 the observation of a second short burst, GRB 050709, pinpointed by NASA's HETE2 satellite on 9 July this year. Fox et al. (page 845)3 and Hjorth et al. (page 859)4 found long-lived X-ray and optical afterglows following this burst, and could thus associate it unambiguously with a star-forming galaxy at a redshift of 0.16. This second event would thus seem at first glance to share some properties — a definite afterglow and an origin in a star-forming galaxy — with long GRBs.
Taking the distance of its source into account, however, the luminosity of the X-ray afterglow of GRB 050709 is, in fact, some three orders of magnitude lower than that typical of a long burst. Short GRBs might also be expected to occur in younger galaxies where stars are still forming, as well as in elliptical galaxies (which tend to consist of older stars), as anything from a few million to several billion years can separate the formation of the compact objects and their merger10. Having said that, two more short bursts localized by Swift in recent months (see ref. 3 and references therein) also seem, like GRB 050509B (ref. 1), to originate in elliptical galaxies — a predominance that seems to indicate that the progenitor stars on average evolved over billions of years.
The energy output of the short GRBs can be easily derived once the distance of the source is established (assuming the emission is isotropic), and is found to be in the range of 1048–1050 erg. Again, this is around 1,000 times lower than that of long GRBs. It is nonetheless large enough to rule out flares from highly magnetized neutron stars, known as soft γ-ray repeaters (SGRs)11, as the explanation for the bursts. It does not, however, preclude the possibility that some short bursts might be the product of an SGR in a nearby external galaxy.
The properties of the long-lived afterglow emission observed for GRB 050709 are consistent with the possibility that the emission arises from low-density surrounding material that is shocked by a collimated (directed) relativistic blast wave — the same process thought to be at work in the long bursts. But the lower energy of the short bursts, together with the fact that they originate closer to Earth in a mixture of galaxy types, suggest that short and long GRBs are in fact separate populations. The observed characteristics1,2,3,4 of the short GRBs are all consistent with models of the merger of two neutron stars, or of a neutron star with a black hole.
Several questions remain. First, if short bursts are preferentially associated with nearby bright elliptical galaxies, why has no such galaxy been found previously in locations where short GRBs have been well defined12? This may be because the population of progenitors had different properties, or because these were distant events originating in faint galaxies that have yet to be identified. Second, why does the afterglow of GRB 050709 show3 evidence of flaring X-ray activity two weeks after the short burst? This is particularly puzzling because it suggests a long-lived energy injection from the central engine driving the burst.
As occurred with their longer cousins, the discovery of long-lived afterglows of short GRBs sets the stage for detailed studies of these exotic cosmic explosions through their emissions of electromagnetic radiation. In the future, there is the exciting prospect of detecting gravitational waves from these events using the second generation of laser interferometry detectors, LIGO and VIRGO.
Gehrels, N. et al. Nature 437, 851–854 (2005).
Villasenor, J. S. et al. Nature 437, 855–858 (2005).
Fox, D. B. et al. Nature 437, 845–850 (2005).
Hjorth, J. et al. Nature 437, 859–861 (2005).
Kouveliotou, C. et al. Astrophys. J. 413, L101–L104 (1993).
Costa, E. et al. Nature 387, 783–785 (1997).
van Paradijs, J. et al. Nature 368, 686–688 (1997).
Lazzati, D., Ramirez-Ruiz, E. & Ghisellini, G. Astron. Astrophys. 379, L39–L43 (2001).
Montanari, E., Frontera, F., Guidorzi, C. & Rapisarda, M. Astrophys. J. 625, L17–L21 (2005).
Fryer, C. L., Woosley, S. E. & Hartmann, D. H. Astrophys. J. 526, 152–177 (1999).
Hurley, K. et al. Nature 434, 1098–1103 (2005).
Hurley, K. et al. Astrophys. J. 567, 447–453 (2002).
About this article
The Astrophysical Journal (2013)
International Journal of Modern Physics D (2011)
Astronomy & Astrophysics (2009)
Monthly Notices of the Royal Astronomical Society (2009)
AMBIO: A Journal of the Human Environment (2007)