Two sources of highly energetic flares have been discovered in archival X-ray data of 70 nearby galaxies. These flares have an undetermined origin and might represent previously unknown astrophysical phenomena. See Letter p.356
Since our beginning, humanity has observed and chronicled stars, comets and supernova explosions. Nowadays, we have instruments that can cover a wide range of the electromagnetic spectrum, from low-energy radio waves to high-energy X-rays and γ-rays. The high-energy sky is the theatre for many time-varying and energetic phenomena, often involving compact objects such as neutron stars or black holes. On page 356, Irwin et al.1 report what looks like a previously undescribed kind of bright X-ray flaring activity in neighbouring galaxies.
In 2005, an unexplained source of X-ray flares was discovered2 in proximity to the galaxy NGC 4697. Motivated by this discovery, Irwin and colleagues analysed archival X-ray data from the Chandra and XMM-Newton space observatories and found two further X-ray sources that exhibit particularly bright flares. The authors' sources are located in the outskirts of nearby galaxies: one near the Virgo galaxy NGC 4636 and the other near the Centaurus A galaxy (NGC 5128; Fig. 1). The archival data for the latter source revealed five different flaring events. The peak luminosity of the flares (up to 1034 watts) is intriguing, because it is even greater than the luminosity that can be attained by a neutron star under normal circumstances. All the flares had short rise times — they reached their peak luminosities in less than a minute — and lasted for about an hour.
What is the origin of these flares? In general, such events can be identified by their duration and whether or not they recur. Non-recursive events that have durations similar to the observed flares include long-duration γ-ray bursts3 and explosions that signal the deaths of massive stars (energetic supernovae). However, these explanations require a population of young stars, whereas the sources are located in the outskirts of elliptical galaxies, which would indicate an old stellar population.
Repeated bursts of high-energy radiation are observed from the flaring of stars that are less luminous than our Sun. Meanwhile, neutron stars are generous sources of different kinds of burst. Highly magnetic neutron stars called magnetars in our Galaxy can give rise to bright flares that last for less than a second4, but can still influence Earth's ionosphere. By contrast, neutron stars that accrete matter from a companion star can produce 'type-I' X-ray bursts, resulting from runaway thermonuclear combustion of matter accumulated on the neutron star's surface. These bursts repeat, but they cannot explain the observed signals because they last for only a few minutes and are less luminous by a factor of about 100 (ref. 5). Finally, neutron stars can also exhibit super-bursts, which are longer versions of type-I X-ray bursts (lasting for a few hours6), but these are also under-luminous compared with the flares discovered here. Therefore, even within this zoo of different transient phenomena, Irwin and colleagues' flaring sources remain unmatched.
Do we have any other information? Both sources are located in old stellar populations, which probably rules out the presence of young neutron stars such as magnetars. In addition, the sources reside in what seem to be globular clusters or ultracompact dwarf galaxies. The optical spectrum from one of the sources shows that it is at the same distance from Earth as is its host galaxy, which confirms that we are not looking at a star in the Milky Way that is being projected onto a more distant galaxy. Both sources also have a high X-ray luminosity before and after their flares that is larger than can be achieved by a neutron star (even if exceptions exist for young neutron stars7). All of these observations seem to suggest the presence of a black hole.
If the black-hole interpretation is correct, there are at least two viable explanations. One possibility, which is suggested by the authors, is that there exists an intermediate-mass black hole (100–1,000 times the mass of the Sun) at the centre of each source that, for some unknown reason, emits flares that last for about an hour. Alternatively, one could envisage a lower-mass black hole whose X-ray emissions are beamed directly towards Earth. A binary system that has a highly eccentric orbit8 might explain the repeated flares detected from the source near NGC 5128). The flares from such a binary system would be strictly periodic, because sporadic surges of accretion would occur at the point of closest approach.
The mystery of Irwin and collaborators' flares might be solved, as often happens in astrophysics, with more observations. In particular, one could investigate the frequency at which the flares occur. An important aspect of the authors' discovery is that both sources were identified after a careful search through archival data, which provides further testimony for the legacy value of such archives. Now that we know these strange objects are out there, they will remain on the watch list and more examples will be searched for.Footnote 1
Irwin, J. A. et al. Nature 538, 356–358 (2016).
Sivakoff, G. R., Sarazin, C. L. & Jordán, A. Astrophys. J. 624, L17–L20 (2005).
Gehrels, N., Ramirez-Ruiz, E. & Fox, D. B. Annu. Rev. Astron. Astrophys. 47, 567–617 (2009).
Palmer, D. M. et al. Nature 434, 1107–1109 (2005).
Strohmayer, T. & Bildsten, L. in Compact Stellar X-ray Sources (eds Lewin, W. & van der Klis, M.) 113–156 (Cambridge Univ. Press, 2006).
Cornelisse, R., Heise, J., Kuulkers, E., Verbunt, F. & in't Zand, J. J. M. Astron. Astrophys. 357, L21–L24 (2000).
Bachetti, M. et al. Nature 514, 202–204 (2014).
Maccarone, T. J. Mon. Not. R. Astron. Soc. 364, 971–976 (2005).