Supernova 1987A:
20 years old and still growing stronger!

The news of a new supernova in the Large Magellanic Cloud quickly swept around the world after its discovery by Canadian astronomer Ian Shelton at Las Campanas Observatory, Chile, on 24 February 1987. A neutrino burst, which signified the event as a type II core-collapse supernova (SN), was retrospectively detected by Kamiokande II and other detectors at UT 07:35 on 23 February.

Figure 1: Compact Array observations of SN1987A at 8 GHz between 1992 and 2006. The images shown have been super-resolved to a resolution of 0.5 arcsec and were taken at multiple frequencies in the 3-cm band and typically include two complementary 6-km configurations. The inset photo is a high-resolution optical image from the Hubble Space Telescope's Advanced Camera for Surveys (ACS), courtesy NASA, P. Challis, R. Kirshner and B. Sugerman (click on image for larger version).

Being the brightest supernova for 400 years, and the closest supernova since the invention of the telescope, it was an amazing event. Many astronomers not previously noted for their expertise in supernovae rapidly became inducted if they happened to be sitting at a suitably southern telescope at the time. For radio astronomers, it was unfortunate that the Compact Array was still a few years from being operational. However, the Sydney University Molonglo Observatory Synthesis Telescope and the Fleurs Synthesis Telescope were operational and, within a few days, the Australian Very Long Baseline Interferometry (VLBI) network also swung into action. Early detections, summarised by Turtle et al. (1987) and Storey & Manchester (1987), were extremely important in understanding some of the blast wave astrophysics, including the density of the immediate circumstellar environment. Even the VLBI non-detection (Jauncey et al. 1988) was important in establishing that the initial shock velocity was in excess of 19,000 km/s, consistent with contemporaneous optical spectroscopy.

Figure 2: Number of refereed astronomy and physics papers mentioning supernova "1987A" in their title or abstract published each year since 1987. Data from NASA's Astrophysical Data System (click on image for larger version).

Being the brightest supernova for 400 years, and the closest supernova since the invention of the telescope, it was an amazing event. Many astronomers not previously noted for their expertise in supernovae rapidly became inducted if they happened to be sitting at a suitably southern telescope at the time. For radio astronomers, it was unfortunate that the Compact Array was still a few years from being operational. However, the Sydney University Molonglo Observatory Synthesis Telescope and the Fleurs Synthesis Telescope were operational and, within a few days, the Australian Very Long Baseline Interferometry (VLBI) network also swung into action. Early detections, summarised by Turtle et al. (1987) and Storey & Manchester (1987), were extremely important in understanding some of the blast wave astrophysics, including the density of the immediate circumstellar environment. Even the VLBI non-detection (Jauncey . Figure 2 shows that 1,522 refereed astronomy and physics papers, with 32,277 citations to them, have been published on SN1987A, according to the NASA Astrophysical Data System! Interestingly, the publication rate has hovered around the 20-30 mark for the last decade. Moreover, two major conferences at the Aspen Center for Physics in Colorado and in Hawai'i will shortly be marking the 20^th anniversary of the explosion, each with a wide-ranging set of talks on the physics of supernovae and Gamma-ray bursts (GRBs). Why the continued interest? The answer lies in the high-density environment around the expanding shock wave which continues to create a firework display in all wavebands from the radio to the X-ray regime. Fourteen years of Compact Array imaging data, shows that, rather than fading after a few weeks as has been seen in all other radio supernovae, SN1987A has considerably brightened in the intervening years, as can be seen in the various panels of the front cover image.

The reason for the brightening in the radio and other regimes relates to the extremely high densities in the pre-existing circumstellar medium. This medium is most visible in the Hubble image shown in the inset of the front cover image. Unlike the radio image, the Hubble image mainly shows the pre-existing medium excited by the rapidly engulfing shock wave. However, detailed understanding of the formation of this ring-like structure is still absent, with recent work by Nathan Smith on similar structures around Luminous Blue Variable stars in our own Galaxy calling into question the usual evolutionary scenario which invokes a transition of the progenitor of SN1987A from a red supergiant to a blue supergiant in the recent past.

The unusual brightening of SN1987A and the fact that the radio morphology is no longer due to interaction with an inverse square-law stellar wind, means that it is often regarded as a supernova remnant (SNR) in its early phase, rather than a radio supernova. The gradual flattening of the radio spectral index from an initial value of -1.0 (S∝ν^α) to its current value of -0.7, which is closer to that of canonical SNRs, confirms this.

Figure 3: Seventeen years of Compact Array monitoring of SN1987A up to day 7231
(11 December 2006) showing that the flux density is not only increasing at all
frequencies, but that the rate of change of flux density is also increasing
(click on image for larger version).

Figure 3 shows that the flux density of SN1987A has now exceeded 300 mJy at 1.4 GHz, and, moreover, that the rate of change of its flux density is also increasing, but not quite at the same rate as Chandra observations of the 1 keV X-ray luminosity have suggested in recent years (Park et al. 2006). Future predictions of the flux evolution are fraught with uncertainty — partly due to poor detailed knowledge of the cosmic ray acceleration mechanism and magnetic field evolution — but mainly due to the poorly understood circumstellar density profile. The continuing interaction of the shock front with the circumstellar medium provides the power source for the generation of the synchrotron radiation responsible for the radio brightness.

Recent observations have resulted in images of even higher resolution compared to the mosaic shown on the front cover (Manchester et al. 2005) and have resulted in the first detection at 3 mm using the upgraded Compact Array. What does the future hold? Low frequency observations with the more sensitive eVLBI array will be useful in establishing the variation of spatial structure across an order of magnitude in frequency, which will help us better understand the acceleration mechanisms. The detection of a pulsar would be extremely important on a number of fronts — stellar evolution theories are divided between a neutron star and a black hole end-point for the remnant of SN1987A; and pulsar birth periods currently remain poorly understood. Finally, will SN1987A continue to brighten in the radio regime and perhaps one day rival that of the neighbouring giant HII region, 30 Doradus? Probably unlikely at this stage, but SN1987A has produced many surprises in the last 20 years, and more can be expected!

References

Jauncey, D., et al. 1988, Nature, 334, 412

Manchester, R., et al. 2005, ApJ, 628, L131

Park, S., et al. 2006, ApJ, 646, 1001

Storey, M., Manchester, R. 1987, Nature, 329, 421

Turtle, A. J., et al. 1987, Nature, 327, 38

Lister Staveley-Smith (UWA), Lewis Ball (ATNF), Bryan Gaensler (USyd), Mike Kesteven (ATNF), Dick Manchester (ATNF) and Tasso Tzioumis (ATNF)
(lister.staveley-smith@uwa.edu.au)