Portrait of a doomed star

Some stars explode in thermonuclear supernovae, but understanding of why this occurs comes mainly from indirect clues. Now, the progenitor of a member of a strange class of such explosions may have been detected directly. See Letter p.54

Explaining the nature of supernovae is one of the classical problems in astronomy. Supernovae are not only enticing mysteries in which explosions of awesome power and brilliance are perpetrated by well-hidden culprits, they are also of broad importance to astrophysics — so our lack of certainty about the progenitors of some supernovae is embarrassing. On page 54 of this issue, McCully et al.1 report that a combination of good fortune and careful analysis has pointed them to the probable system behind a particularly puzzling type of supernova.

In ancient times, people interpreted new lights in the sky as heavenly signs of earthly fates. Now, astronomers use a form of supernova known as type Ia to infer the history and future of the Universe. Those ancient new lights appeared to be new stars only because the stars that produced the celestial displays had previously been too faint to see. We now have a similar problem with supernovae. Although supernova explosions can be seen across more than half the age of the Universe, detecting pre-explosion stars is difficult even when they are massive and luminous. For such supernovae we still only have a small number of definitive pre-explosion detections. The progenitors of type Ia supernovae have proved even more elusive. This type of supernova is thought to occur when a star called a white dwarf undergoes runaway nuclear fusion, but we have yet to directly see what circumstances cause this to happen. One such system may have been detected in X-rays2, but perhaps the clearest trace of these progenitors has been in material lingering close to the explosion3,4,5. However, theorists have been more successful in making multiple models consistent with the presence of that material than in convincing everyone to agree on one interpretation.

We are therefore lucky that two supernovae separated by only a decade have been detected in the spiral galaxy NGC 1309 (Fig. 1). After the first explosion, detected in 2002, a team led by one of McCully's co-authors used the Hubble Space Telescope to stare at the host galaxy6. The astronomers did this so that they could compare the distance to NGC 1309 inferred from two methods — one using the 2002 supernova and the other involving a type of star for which the method has been well-calibrated with nearby examples — and improve the precision with which we measure even larger cosmological distances. So when the second explosion occurred, in 2012, there was a wealth of data to examine for signs of the supernova progenitor.

Figure 1: A double bill.

W. Li & A. V. Filippenko/UC, Berkeley; G. H. Marion, S. W. Jha & R. P. Kirshner

The spiral galaxy NGC 1309, which is about 30 million parsecs from Earth, hosted a type Ia supernova in 2002, SN 2002fk, and then a type Iax explosion in 2012, SN 2012Z. McCully et al.1 have detected the probable progenitor of SN 2012Z in pre-explosion images obtained by the Hubble Space Telescope (see Fig. 1 of the paper1).

At first, the 2012 supernova was classified as a type Ia (ref. 7), although one with very unusual features. However, it and similar supernovae have become known as type Iax, following the name of a peculiar prototype event called SN 2002cx (refs 8,9). Astronomers sometimes worry too much about classification: dividing objects that show continuous variation in physical properties such as luminosity and velocity into discrete categories can become unhelpful. However, this new name reflects the realization that the 2002cx-like explosions may be even less like normal type Ia supernovae than first thought. Superficially similar qualities might have previously misled us into doing something akin to classifying a platypus as a type of duck.

McCully and colleagues have found a source in the pre-explosion images from the Hubble Space Telescope whose location closely matches that of the 2012 supernova. They also show that this alignment has only a roughly 1% probability of being a coincidence, which leads them to conclude that this source is very likely to have produced the supernova. Although apparently exciting coincidences can happen by chance, there have been very few pre-supernova observations that could have identified a progenitor that resembles this one. So this result is not due to astronomers taking snapshot after snapshot and waiting for an interesting-looking but meaningless alignment to randomly occur.

Assuming that the observed source is the progenitor, the first question in interpreting the results is whether it was emitting mainly simple starlight. Because we might also expect light to be generated by matter falling onto the white dwarf before it explodes, it is possible that McCully et al. have detected a signal from this process. However, the authors argue that it is more likely that the source they have identified is a star made of helium, which has been transferring matter onto an unseen white dwarf. This option is tempting because it matches some models for type Iax supernovae in which the white dwarf accretes a surface helium layer that eventually ignites in a thermonuclear runaway, in turn triggering fusion in the underlying star.

If this explanation is true, then the source should still be present when the light from the supernova has faded sufficiently to allow the researchers to look for it again. However, the supernova might well have heated the helium star, stripped it of some surface layers or polluted it. It would be frustrating if the star appears exactly as it did before the explosion, and in that case we would have to worry more about whether the true progenitor was too faint to have been detected. It is to be hoped that future observations of this source will help us to understand not just type Iax events, but also the impact of supernovae on nearby stars. Another possibility noted by McCully and colleagues is that this supernova was the death of a massive star, in which case the source will have disappeared. This conclusion would show that Iax events truly are produced by qualitatively different systems from type Ia supernovae.

If McCully et al. have identified the progenitor, then the observation will be one of the most memorable signposts on the road to understanding supernovae. And because it may improve our knowledge of what can happen after a layer of helium ignites on the surface of another potentially explosive collection of fusion fuel, it could also help to explain the progenitors of other recently discovered events for which variations of this model have been proposed10,11,12,13. The 2012 supernova in NGC 1309 has not yet yielded all its secrets, but this discovery might help to solve mysteries both old and new.


  1. 1

    McCully, C. et al. Nature 512, 54–56 (2014).

    CAS  Article  ADS  Google Scholar 

  2. 2

    Voss, R. & Nelemans, G. Nature 451, 802–804 (2008).

    CAS  Article  ADS  Google Scholar 

  3. 3

    Patat, F. et al. Science 317, 924–926 (2007).

    CAS  Article  ADS  Google Scholar 

  4. 4

    Sternberg, A. et al. Science 333, 856–859 (2011).

    CAS  Article  ADS  Google Scholar 

  5. 5

    Dilday, B. et al. Science 337, 942–945 (2012).

    CAS  Article  ADS  Google Scholar 

  6. 6

    Riess, A. G. et al. Astrophys. J. Suppl. Ser. 183, 109–141 (2009).

    Article  ADS  Google Scholar 

  7. 7

    Cenko, S. B. et al. Cent. Bur. Electron. Telegrams 3014, 1 (2012).

    ADS  Google Scholar 

  8. 8

    Li, W. et al. Publ. Astron. Soc. Pac. 115, 453–473 (2003).

    Article  ADS  Google Scholar 

  9. 9

    Foley, R. J. et al. Astrophys. J. 767, 57 (2013).

    Article  ADS  Google Scholar 

  10. 10

    Perets, H. B. et al. Nature 465, 322–325 (2010).

    CAS  Article  ADS  Google Scholar 

  11. 11

    Kasliwal, M. M. et al. Astrophys. J. 755, 161 (2012).

    Article  ADS  Google Scholar 

  12. 12

    Poznanski, D. et al. Science 327, 58–60 (2010).

    CAS  Article  ADS  Google Scholar 

  13. 13

    Kasliwal, M. M. et al. Astrophys. J. Lett. 723, L98–L102 (2010).

    Article  ADS  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Stephen Justham.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Justham, S. Portrait of a doomed star. Nature 512, 34–35 (2014).

Download citation


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.


Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing