Light reflected off a dust cloud in the vicinity of the relic of Tycho Brahe's supernova, whose light first swept past Earth more than four centuries ago, literally sheds light on the nature of this cosmic explosion.
The occurrence of a supernova in our Galaxy is a rare event — only six such supernovae were registered during the past millennium1. Although their remnants have been extensively studied, the nature of the original explosions is the subject of debate. Supernovae are classified into different types mainly on the basis of their observed spectral properties, and occasionally through the analysis of well-sampled light curves (plots that show the evolution of their brightness over a period of time). The classification of historical supernovae — those whose light was first detected on Earth centuries ago — is more problematic than that of events detected recently, because the sparse visual observations reported in historical records are often difficult to calibrate, and the resulting light curves are rather uncertain. In addition, the study of their relics does not always give unequivocal answers as to which category they fall into.
On page 617 of this issue, Krause et al.2 offer convincing evidence that the supernova first observed by Tycho Brahe in 1572 (SN1572) was a type Ia supernova — one originated by the thermonuclear explosion of a white dwarf star in a binary system. The authors' investigation was triggered by a recent paper3 that announced the discovery of light echoes from two ancient supernovae, one of which was SN1572. The luminous echo from Tycho Brahe's supernova was visible in recent images of its remnant as an extended light source which, according to astrometric observations that determine the precise positions of the source in the sky, seemed to shift with time through a dust cloud in the remnant's vicinity2,3. Krause and colleagues classified SN1572 through a spectrum of its echo, a method with proven credentials in classifying ancient supernovae. For example, the youngest supernova remnant in our Galaxy, Cassiopeia A (ref. 4), was classified in this way as a type IIb (ref. 5) event (a supernova generated by the gravitational collapse of the core of a massive, moderately hydrogen-poor star), and the relic SNR0509−675 in the Large Magellanic Cloud was classified as a luminous type Ia supernova6.
During the 1960s, Iosif Shklovski7 and Sidney van den Bergh8 independently showed that the detection and subsequent follow-up spectroscopic analysis of scattered-light echoes in the surroundings of supernova sites could be used to get information on the explosions that generated them. But this technique remained only a brilliant idea for many decades because of the technical challenges involved in observing such low-surface-brightness echoes. Now, after 40 years, and thanks to the enormous technological progress and the larger dimensions of modern telescopes, Shklovski and van den Bergh's prediction has finally proved to be correct9,10.
But what is a supernova light echo? A supernova produces a flash of optical radiation that propagates in all directions. If the blast occurred in a vacuum, the observer would detect the event through photons that travel unperturbed along the line of sight. But interstellar space is not empty, and thus photons propagating away from the line of sight are likely to encounter dust clouds that scatter their light in all directions. Therefore, a small fraction of the incoming radiation is expected to be seen by the observer as a faint, local re-brightening in the proximity of the explosion site. Because of the finite speed of light, scattered photons will reach Earth after some delay compared with those travelling along a straight, unperturbed path (Fig. 1). Because this phenomenon is similar to that well known for acoustic waves, it is labelled a light echo.
Interestingly, as the flash propagates through the cloud, the re-brightening shifts in space (Fig. 1), giving the impression that material is moving at velocities that can actually exceed the speed of light. In reality, the effect is purely geometrical and nothing physically moves, but variations of brightness with time allow astronomers to detect light echoes through reiterated observations of selected areas in the sky. A time-evolving re-brightening is exactly what Krause et al.2 observed in a particular (and limited) region of the remnant of SN1572.
The light echo thus carries a sort of fossil imprint of the original supernova, and so analysis of the reflected light can unveil characteristics of it even centuries after it was first detected on Earth. Krause et al. obtained a spectrum of the bright-light echo of SN1572 with the Subaru 8.2-metre telescope on Mauna Kea (Hawaii) and showed that prominent silicon, sulphur and iron lines — which are typical of a normal type Ia supernova around its maximum — were present in the spectrum.
Type Ia supernovae are thought to be generated by the explosion of carbon–oxygen white dwarf stars in binary systems. As the white dwarf accretes mass from its companion star and eventually reaches the critical Chandrasekhar mass (1.4 solar masses), a thermonuclear explosion disrupts the star, releasing materials that are the product of the stellar nuclear fusion (mostly silicon and elements of the iron group). Nickel (56Ni) in particular is ejected in great amounts — up to 50% of the whole white dwarf's mass. This radioactive isotope decays initially into radioactive cobalt (56Co), and then into stable iron (56Fe). Such a decay chain provides enough energy to keep the material ejected by the explosion hot for several months, and supports the supernova luminosity in the early phase of the transition towards the remnant stage. During this transformation, some — but not all — information about the explosion itself is lost. This implies that, in principle, the type of original supernova cannot be uncovered by studying the chemical abundances in the remnant alone.
Nevertheless, Tycho Brahe's supernova was suspected to be of type Ia well before the observations reported by Krause et al.2. The historical light curve11, the studies of the remnant in the radio12 and X-ray13 wavelengths, and the discovery of a surviving candidate G-type companion star14 (a star just like our Sun) all suggested a thermonuclear supernova rather than a core-collapse event. The fundamental contribution of Krause and colleagues' work is to transform these clues into definite proof — we are now fully confident that one of the most popular supernova remnants detected in our Galaxy was produced by an ordinary type Ia supernova that was first detected more than 400 years ago.
The technique of observing light echoes from supernovae is a remarkable observational tool with which to pigeonhole the type of supernova. It will allow astrophysicists to characterize other supernova remnants in our Galaxy and in nearby galaxies. This will hopefully clarify the relationship between supernova relics and their explosion mechanisms. Finally, it is likely that precise information about the frequency of the different supernova types in our Galaxy and its surroundings will shed light on the star-formation history and chemical evolution of the Local Group of galaxies.
Green, D. A. & Stephenson, F. R. in Lecture Notes in Physics Vol. 598 (ed. Weiler, K.) 7–19 (Springer, 2003).
Krause, O. et al. Nature 456, 617–619 (2008).
Rest, A. et al. Astrophys. J. 680, 1137–1148 (2008).
Thorstensen, J. R., Fesen, R. A. & van den Bergh, S. Astron. J. 122, 297–307 (2001).
Krause, O. et al. Science 320, 1195–1197 (2008).
Rest, A. et al. Astrophys. J. Lett. 681, L81–L84 (2008).
Shklovski, I. S. Astron. Circ. (USSR) 306, 2–4 (1964).
van den Bergh, S. Publ. Astron. Soc. Pacif. 77, 269–271 (1965).
Rest, A. et al. Nature 438, 1132–1134 (2005).
Krause, O. et al. Science 308, 1604–1606 (2005).
Baade, W. B. Astrophys. J. 102, 309–317 (1945).
Baldwin, J. E. & Edge, D. O. Observatory 77, 139–143 (1957).
Hughes, J. P. et al. Astrophys. J. 444, L81–L84 (1995).
Ruiz-Lapuente, P. et al. Nature 431, 1069–1072 (2004).