News & Views | Published:


Supernovae brought to light


The existing catalogue of Galactic supernova remnants contains only a small fraction of the true number of these stellar explosions. A different observational technique is being employed to find the missing ones.

Just as nuclear bombs on Earth leave their legacy of devastation, stellar explosions — supernovae — leave their mark on the cosmos in the form of supernova remnants (SNRs) in the interstellar gas. The rate of supernova explosions in our Galaxy can be inferred from historical records of visual sightings and through comparison with other galaxies. These data, together with theories of SNR expansion and evolution, allow a prediction of the number of SNRs that should currently reside in our Galaxy. The result is tens of times larger than the number of known remnants, which are catalogued almost exclusively from the radiation they emit at radio frequencies1. Writing in The Astrophysical Journal, Koo, Kang and Salter2 use a new technique to identify the more numerous older SNRs, which were missing from the catalogue because they are radio-dim.

The legacies left by supernovae, extending beyond cosmology, are often unappreciated. First, there is the anthropocentric point that we wouldn't be here without the elements, such as iron and carbon, that supernovae have caused to disperse into the interstellar gas. Over Earth's lifetime, several dozen nearby supernovae (within a distance of around 30 light years) would also have seriously modified Earth's environment, each for several hundred years3. The most serious consequences would have been the destruction of the ozone layer and a great increase in the flux of highly energetic cosmic rays. That would have led to direct biological effects through increased radiation, as well as probable climate change that could dwarf today's global warming. It's even conceivable that greatly enhanced radioactivity emanating from a nearby supernova triggered the formation of life on Earth by producing complicated molecules4.

Supernovae have also greatly affected science and scientific culture. A supernova of 1572 landed the astronomer Tycho Brahe a job with King Frederick II of Denmark, an appointment that jump-started the principle of accurate astronomical measurements5. Tycho's protégé, Johannes Kepler, introduced and extended critical thinking and rigorous analysis as the basis for scientific endeavour6. Moving into the twentieth century, in seeking an explanation for the origin of a supernova, Walter Baade and Fritz Zwicky correctly predicted the existence of neutron stars7. More recently still, distant supernovae have been the signposts revealing the increasing acceleration of the Universe and the ‘dark energy’ that is held responsible for this phenomenon8,9.

Supernovae come in two basic types, I and II. Type I supernovae are binary white dwarf systems, in which mass is transferred from an active companion star to a white dwarf (a low-to-medium mass star that has exhausted its nuclear fuel); type II supernovae are massive stars near the end of their lives. In both cases, a very dense object — the white dwarf or the core of the massive star — grows until it exceeds a critical mass, known as the Chandrasekhar limit. At this point, its internal pressure becomes unable to resist the force of its gravity, and it collapses in size by a factor of perhaps 1,000. That process releases a total of about 1046 joules of energy, the lion's share of it in the form of the chargeless, almost massless, particles known as neutrinos.

The 1% of a supernova's energy not emitted as neutrinos produces the spectacular visible explosion, and is deposited as a highly energetic SNR that expands into the surrounding gas, sweeping it into a dense shell. The expansion of this shell slows with increasing radius10 until it merges with the interstellar medium, maintaining that medium's turbulence. This process takes about a million years.

From the eight supernovae visible to the naked eye that have been recorded during the past two millennia11, and from comparison with other galaxies, we infer that our Galaxy produces two or three supernovae per century. Accordingly, there should be thousands of SNRs. But we know of only 265, all through measurements of radio-frequency synchrotron radiation emitted by relativistic electrons produced by the supernova explosion1. This number does not, therefore, include SNRs with no radio emission. Moreover, the observed range of radio luminosity of SNRs covers at least a factor of 100; why some SNRs are bright and others are not is not known. So the bias of the current SNR sample towards the brighter, younger, radio-emitting supernovae is unknown, and the sample's statistical properties are correspondingly unreliable. A dif-ferent technique for identifying supernova remnants is needed.

Enter Koo, Kang and Salter2. When the expansion of an SNR slows down, its shell of swept-up gas turns into neutral atomic hydrogen (H I). The shell's velocity remains anomalous compared with that of the surrounding medium for some time, and, if the shell's angular diameter is large enough, it can be distinguished from ambient gas through a Doppler shift of the characteristic spectral emission line of neutral hydrogen at a wavelength of 21 cm. Using existing low-resolution H i surveys, Koo and colleagues2 identified some 200 new SNR candidates. They mapped one, named FVW190.2+1.1, in detail using the Arecibo telescope on Puerto Rico. This SNR has a radius of some 290 light years, a velocity of expansion of 77 km s−1, and an age of 340,000 years. SNRs as old as this lose all detectable radio emission, so the H i technique opens up an unknown and unsampled set of remnants. Compared with catalogued SNRs, FVW190.2+1.1 is situated unusually far (around 52,000 light years) from the centre of the Galaxy.

An intriguing aspect of this H i technique is its ability to see so-called supershells. Type II supernovae come from massive stars, which are formed in groups localized in space and time. These stars have fairly short, similar lifetimes, so multiple simultaneous supernovae can produce gigantic versions of SNRs. Two local examples of these supershells (or ‘super-bubbles’), the Orion–Eridanus superbubble and the North Polar Spur shell, dominate our sky, with each being tens of degrees in diameter. Several dozen more distant supershells have been catalogued, and the largest known supershell, at a distance from the Galactic Centre of around 13,000 light years, has been mapped12 from its emissions at 21 cm (Fig. 1). The production of this spectacular supershell required upwards of 100 supernovae; fortunately, this is consistent with its location in our Galaxy's star-forming hotbed.

Figure 1: Supernova supershell exposed.

The most spectacular known supershell, produced by 100 or more supernovae, is revealed by this map12 of emissions of the atomic hydrogen line at a wavelength of 21 cm, the tracer used by Koo and colleagues2. The width of the image covers 20° of the sky; brighter yellow corresponds to higher and darker green to lower hydrogen density. The combined explosive energy has lifted the entire Galactic plane (white portion at the bottom) and sent shards of gas flying about around 6,500 light years into the Galactic halo, from which they will eventually return. (Figure courtesy of Y. Pidopryhora.)

Compared with the classical radio SNR identification, the H i technique emphasizes not just supershells, but also the more numerous older SNRs, and thus promises to remove the radio-selected statistical biases. The mapping of the whereabouts of interstellar gas with the 21-cm atomic hydrogen line is undergoing a renaissance as new multifeed instrumentation at the Arecibo telescope13 and the Green Bank Telescope in West Virginia comes on stream. The future is bright for studying radio-dim SNRs.


  1. 1

    Green, D. A. Bull. Astr. Soc. Ind. 32, 335–370 (2004).

  2. 2

    Koo, B. -C., Kang, J. -J. & Salter, C. J. Astrophys. J. 643, L49–L52 (2006).

  3. 3

    Clark, D. H. & Stephenson, F. R. The Historical Supernovae (Pergamon, New York, 1977).

  4. 4

    Shklovsky, J. S. & Sagan, C. Intelligent Life in the Universe (Holden-Day, San Francisco, 1966).

  5. 5

    Gingerich, O. in 1604–2004: Supernovae as Cosmological Lighthouses (eds Turatto, M., Benetti, S., Zampieri, L. & Shea, W.) 3–12 (ASP, San Francisco, 2005).

  6. 6

    Lombardi, A. in 1604–2004: Supernovae as Cosmological Lighthouses (eds Turatto, M., Benetti, S., Zampieri, L. & Shea, W.) 21–29 (ASP, San Francisco, 2005).

  7. 7

    Koenig, T. in 1604–2004: Supernovae as Cosmological Lighthouses (eds Turatto, M., Benetti, S., Zampieri, L. & Shea, W.) 53–62 (ASP, San Francisco, 2005).

  8. 8

    Riess, A. G. et al. Astron. J. 116, 1009–1038 (1998).

  9. 9

    Perlmutter, S. et al. Astrophys. J. 517, 565–586 (1999).

  10. 10

    Cioffi, D. F., McKee, C. F. & Bertschinger, E. Astrophys. J. 504, 761–772 (1988).

  11. 11

    Stephenson, F. R. & Green, D. A. J. Hist. Astron. 36, 217–229 (2005).

  12. 12

    Pidopryhora, Y., Lockman, F. J. & Shields, J. C. Astrophys. J. (submitted).

  13. 13

    Stanimirovic, S. Astrophys. J. (submitted).

Download references

Author information

Rights and permissions

To obtain permission to re-use content from this article visit RightsLink.

About this article

Further reading


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.