Astronomers know little about γ-ray bursts other than that they are the most energetic explosions in the Universe. The latest observations indicate that large-scale magnetism contributes to their power.
Gamma-ray bursts (GRBs) are truly amazing astrophysical events. In a matter of seconds, they release more energy than the Sun will do in its lifetime of 10 billion years. These powerful stellar explosions eject material at highly relativistic velocities — differing from the velocity of light by as little as one part in a million. Because they are so bright, GRBs can be seen almost to the edge of the observable Universe, occurring, on average, about twice a day. However, their origin remains a mystery. Their outflows could be gas-dynamic phenomena, driven by gases near their central engine with a high kinetic pressure (to some extent, similar to exhaust jets in aeroplanes)1. Alternatively, and somewhat unexpectedly, magnetic fields, which on Earth have a negligible dynamic role, have been proposed as the dominant mechanism for both driving the bulk of the outflow and accelerating particles within it2,3. The coupling of magnetic fields to gravity — which occurs, for example, in the close vicinity of the event horizon (a boundary beyond which nothing can escape) of a rotating black hole through the Blandford–Znajek mechanism4 — can indeed produce powerful relativistic outflows such as those seen in GRBs.
But magnetic fields in GRBs are notoriously hard to observe — just imagine the difficulties faced by an observer trying to prove the existence of Earth's magnetic field without the use of a magnetic compass. On page 767 of this issue, Steele et al.5 report a possible detection of magnetic fields in a GRB, GRB 090102, through observations of polarization in its optical (visible-light) emission.
GRBs emit light across a broad band of frequencies, ranging from the radio to the high-energy end of the γ-ray part of the electro-magnetic spectrum. As the name suggests, they are detected, at least initially, through their γ-ray radiation. Because Earth's atmosphere effectively blocks γ-rays, thereby protecting the planet from this highly damaging radiation, astronomers rely on space telescopes that operate at high frequencies, such as Swift, to observe them. Once a GRB is detected by a space telescope, its sky coordinates are transmitted to ground-based observatories, which then carry out follow-up observations.
In the optical waveband, GRBs tend to produce a dim flash that lasts for only a few dozen seconds. This makes ground-based follow-up observations challenging. First, the telescope must turn automatically towards a GRB, because any delay may mean that the quickly fading signal is not detected. Second, because the telescope sometimes needs to be reoriented by a large angle during this 'slewing' manoeuvre, modern, large telescopes are often too massive, and therefore too slow, to be useful.
In their study, Steele and colleagues5 used the medium (2-metre) RINGO Liverpool Telescope at La Palma in the Canary Islands. The key advantage of the RINGO telescope is that it is equipped with a detector that can measure the polarization of incoming radiation. The authors were therefore able to measure a considerable degree of polarization — of the order of 10% — in the optical emission from GRB 090102. The fraction of polarization in astrophysical sources rarely exceeds several per cent.
Polarization is a property of electromagnetic waves that describes the preferred direction of their electric-field oscillations. A non-zero polarization indicates that the process that produced the waves is, in some sense, non-isotropic: it is more sensitive to one particular direction than others. The most logical — but not the only — explanation for the high degree of polarization obtained by Steele and colleagues is that the emitting source of GRB 090102 is permeated by large-scale, ordered magnetic fields, and that the emission process is non-thermal synchrotron emission from leptons (such as electrons) as they spiral around the magnetic fields at high (relativistic) speeds in the GRB outflow's frame of reference.
Magnetic fields are necessary to produce synchrotron emission. But if a magnetic field is tangled on small scales, as the favoured model of GRB emission posits, the polarization produced by different emitting regions will average out to nearly zero. A large amount of polarization would therefore be tied to large-scale, non-tangled magnetic fields. If this interpretation is correct, Steele and colleagues' result will lend support to models in which large-scale magnetic fields have an important role in launching and collimating GRB outflows. In addition, large-scale magnetic fields can directly accelerate the emission of particles through the process of magnetic reconnection — a well-established hypothesis to explain solar flares.
Steele and colleagues' detection5 of polarized light from GRB 090102 is likely to contribute to the heated debate about the nature of GRBs. Previously, claims6 of high polarization in GRBs were inconclusive7 because polarization is difficult to measure at high frequencies, giving large uncertainties. At visible-light frequencies, polarization can be measured with much higher certainty. A conservative interpretation of the results of Steele et al. is that the magnetic forces in GRBs are at least as important as the gas-pressure forces. A more exciting possibility is that magnetic fields completely dominate the outflow dynamics, so that the nature of GRBs is mostly electromagnetic and not gas-dynamic8.
Piran, T. Rev. Mod. Phys. 76, 1143–1210 (2004).
Lyutikov, M. New J. Phys. 8, 119 (2006).
Usov, V. V. Nature 357, 472–474 (1992).
Blandford, R. D. & Znajek, R. L. Mon. Not. R. Astron. Soc. 179, 433–456 (1977).
Steele, I. A., Mundell, C. G., Smith, R. J., Kobayashi, S. & Guidorzi, C. Nature 462, 767–769 (2009).
Coburn, W. & Boggs, S. E. Nature 423, 415–417 (2003).
Rutledge, R. E. & Fox, D. B. Mon. Not. R. Astron. Soc. 350, 1288–1300 (2004).
Blandford, R. D. in Lighthouses of the Universe: The Most Luminous Celestial Objects and Their Use for Cosmology (eds Gilfanov, M., Sunyeav, R. & Churazov, E.) 381–404 (Springer, 2002).