The evidence in the afterglow

The X-ray spectrum of an afterglow from a γ-ray burst reveals a smoking gun. It links the γ-rays to the expanding fireball that occurs after a supernova explosion.

Our understanding of γ-ray bursts (GRBs) becomes clearer with the new observations by Reeves et al.1 on page 512 of this issue. GRBs are thought to result from the collapse of an extremely massive star to a black hole (prompting a supernova2) or when a neutron star falls into a black hole3. Reeves et al. found that the X-rays emitted in the wake of a GRB come from extremely hot gas expanding outward from the source of the γ-rays and that this gas is highly enriched with the by-products of a supernova explosion. The new results strengthen the link between GRBs and supernovae, and favour models where the GRB occurs within a few days after a supernova.

For some time following their discovery in the late 1960s, bursts of γ-rays were detected by satellites or deep-space probes throughout the Solar System, but their sources were unknown. The directions to most GRBs were very uncertain. Except for a small subclass, called soft γ-ray repeaters, no two bursts with precise locations were ever seen in the same direction in the sky, or identified with any particular class of astronomical object, deepening the mystery of their origins.

During the 1990s, however, a much clearer picture emerged. Today, it is generally accepted that GRBs originate in distant galaxies, rather than in our Galaxy, and are probably the results of the most powerful explosions in the Universe. Most of the alternative models of the birthplace of GRBs were knocked out by a formidable one–two punch. The first blow came from NASA's Burst and Transient Source Experiment on the Compton Gamma Ray Observatory4, which showed that GRBs are distributed evenly across the sky rather than being associated with the Milky Way. The second was in the follow-up observations by the BeppoSAX mission, which indicated that GRBs occur in galaxies at cosmological distances5,6.

The follow-up observations have also found X-ray, optical and sometimes radio 'afterglows' of GRBs that fade over several days. The X-ray afterglows could be evidence of a spherically expanding, cooling fireball following a supernova. But another possibility that has gained favour is that they come from a jet of plasma, moving at relativistic speed — close to that of light — and oriented towards Earth in a conical geometry.

A further puzzle has been that, according to the fireball model, emission lines should be observed, because the elements produced in the supernova are hot enough to excite electronic transitions in the X-ray band, much like sodium atoms in a sodium vapour lamp emit lines in visible (yellow) light. Such lines could provide key diagnostic data about the environments that give rise to GRBs or the physical state of the material that emits γ-rays. These lines have not yet been detected in γ-ray spectra themselves, however. As to the reasons why none have been detected, one can only speculate that the γ-rays result from a nonthermal process after a supernova explosion, or that the emission lines are very weak and the spectral resolution of γ-ray instruments is insufficient to discern them.

Meanwhile, the most powerful X-ray telescopes — NASA's Chandra X-ray Observatory and the European Space Agency's XMM-Newton — have recently obtained evidence of emission lines from iron in the X-ray afterglows7,8. The Doppler widths of the lines (a measure of velocity) give a value of about 10% of the speed of light, giving credence to the expanding-fireball hypothesis.

But these reports have been criticized because the X-ray spectra are very weak and the lines are on the border of detectability. Furthermore, if the lines are identified with iron, there is a mystery as to where the iron originates because it doesn't form in a supernova until after the decay of radioactive 56Ni via 56Co; and the latter has a half-life of 78 days. So most of the interpretations of the X-ray afterglow spectra have invoked environmental material that was presumably ejected before the explosion. Relativistic jets could also irradiate the ejecta, causing the iron line emission.

Reeves et al.1 have been able to go further in analysing afterglows. Using data gathered by the X-ray spectrometers on the XMM-Newton satellite, they have identified several emission lines in the X-ray spectrum of the afterglow from a GRB that was first observed on 11 December 2001 by BeppoSAX (information about this burst is summarized in ref. 9). The X-ray spectrum of the GRB appears to be different from that of other GRBs and may provide a new context for interpreting the previous X-ray lines. The strongest emission lines Reeves et al. detected were from silicon, sulphur and argon; there were other, weaker detections of magnesium and calcium — but no iron was detected. With several different emission lines, especially at lower X-ray energies, one can be much more confident of the identity of these elements.

Reeves et al. take full advantage of the additional information to be quarried from these emission lines, including the speed of the gas outflow, its temperature and composition, and the time delay since the supernova explosion. First is the measurement of a Doppler blueshift: the line-emitting gas is moving at about 26,000 km s−1 towards the observer, suggesting that the X-rays are being emitted at the nearest portion of an expanding shell of gas that is illuminated by a narrow cone of γ-rays from the GRB. Next, the lines they detected are highly ionized, indicating that the gas has a temperature of about 5 × 107 K, which is a reasonable value for hot ejecta from a supernova-like explosion.

The data indicate that the material is enriched in these elements, as one might expect in such an explosion where most elements are created and then ejected into the interstellar medium. Reeves et al. also put a figure on the size of the expanding shell, estimating it as 1015 cm. Given the shell's expansion rate, this gives a time delay from the initial supernova explosion of 10–100 hours, which is consistent with the absence of iron and provides another crucial link to the supernova origin of GRBs.

As with any ground-breaking observation, this one raises questions even as others are answered. For example, why aren't the putative iron emission lines in other GRB afterglows blueshifted? Why don't lines from lower ionization states appear as the afterglow cools, as would be expected? Also, why don't some of the other GRB afterglows show a similar profusion of elements? The optical and X-ray properties of GRB afterglows can be quite varied7,8. Some are not detected optically, earning them the monikers 'dark GRB' or 'γ-ray bursts hiding an optical source-transient' (GHOST). It may turn out that some of this variation can be explained by the diversity of environments in which the fireballs detonate.

There is evidently a great deal yet to be discovered about GRBs, and two further NASA missions are dedicated to studying them. One (Swift) will be launched next year. The other, the High Energy Transient Explorer, was launched in 2000 and provides accurate GRB positions within minutes to hours so that other instruments can be trained on the source to catch the afterglows. Continued observations of various types will help to elucidate the origins of GRBs. But it is clear that high-resolution spectroscopy, as exploited by Reeves et al., is fast becoming an especially valuable tool in the endeavour to understand these hugely energetic phenomena.


  1. 1

    Reeves, J. N. et al. Nature 416, 512–515 (2002).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Meszaros, P. & Rees, M. Astrophys. J. 397, 570–575 (1992).

    ADS  CAS  Article  Google Scholar 

  3. 3

    MacFadyen, A. & Woosley, S. E. Astrophys. J. 524, 262–289 (1999).

    ADS  CAS  Article  Google Scholar 

  4. 4

    Meegan, C. et al. Nature 355, 143–145 (1992).

    ADS  Article  Google Scholar 

  5. 5

    Costa, E. et al. Nature 387, 783–785 (1997).

    ADS  CAS  Article  Google Scholar 

  6. 6

    Metzger, M. et al. Nature 387, 878–880 (1997).

    ADS  CAS  Article  Google Scholar 

  7. 7

    Antonelli, L. A. et al. Astrophys. J. 545, L39–L42 (2000).

    ADS  CAS  Article  Google Scholar 

  8. 8

    Piro, L. et al. Science 290, 955–958 (2000).

    ADS  CAS  Article  Google Scholar 

  9. 9

Download references

Author information



Corresponding author

Correspondence to Herman L. Marshall.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Marshall, H. The evidence in the afterglow. Nature 416, 484–485 (2002).

Download citation

Further reading

  • Astrophysics in 2002

    • Virginia Trimble
    •  & Markus J. Aschwanden

    Publications of the Astronomical Society of the Pacific (2003)


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Sign up for the Nature Briefing newsletter for a daily update on COVID-19 science.
Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing