ON 1 May 1006, the southern sky was pierced by a star of such intensity that observers from Europe to Japan reported it lighting up an otherwise dark and moonless night. The spectacle was bright enough to cast shadows and afforded enough light to read a manuscript. “Its light illuminated the horizon and it twinkled very much,” wrote the Egyptian astrologer Ali bin Ridwan. “The magnitude of its brightness was a little more than one quarter the brightness of the full moon.”
Bin Ridwan was lucky enough to have seen a supernova, a type of catastrophic stellar explosion that ranks among the rarest and most intriguing of celestial phenomena. Modern astronomers consider the supernova of 1006 to be the brightest in recorded history. Since then only a handful of others have occurred in our corner of the Milky Way, all before the invention of the telescope. Today researchers reading historic accounts can reflect on the unfairness of it all: had the galactic lottery turned out differently, perhaps we could have viewed some of those supernovae with state-of-the-art optical hardware. As it stands, whatever precious data their light contained is lost forever.
Or is it? One man thinks we can still observe them – and he may have already succeeded. Armin Rest, from the Cerro Tololo Inter-American Observatory in Chile, is part of a team trying to study the great supernovae of the past by spying their faint reflections in the present. The light from those ancient cataclysms is still out there, the team reasons, spreading like ripples in a galactic pond. All it takes is some interstellar dust in the right place to bounce a “light echo” in our direction, and this time Rest and his colleagues will be watching. “That would be really wonderful,” he says. “To see the same light that was seen long ago, and to know what it is.”
Advertisement
Tracking down light echoes from a long-gone supernova would be an astronomical tour de force, opening a trove of scientific riches. For the first time we could compare measurements of the explosion with the glowing remnants left behind centuries later. This would reveal new details of how supernovae work and how they interact with their surroundings, which could shake up our view of galactic history, planet formation and even the fate of the universe.
The modern study of supernovae got off to a quiet start in 1885, when Ernst Hartwig at Dorpat Observatory in Estonia noticed the appearance of a new star in the direction of the Andromeda galaxy. Decades later, when astronomers realised how far away Andromeda was, it became clear that the “star” must have been billions of times brighter than our sun. In 1934, Walter Baade and Fritz Zwicky of the Mount Wilson Observatory in California coined the term “supernova” to describe these remarkable events. One by one, the outbursts of the past millennium were tied to supernova remnants still visible in the night sky.
The discovery of supernova light echoes is more recent. In February 1987, the brightest supernova of modern times appeared in the Large Magellanic Cloud (LMC), a satellite galaxy that orbits the Milky Way. Despite being 150,000 light years away, supernova 1987A was visible to the naked eye for weeks. A year later, Arlin Crotts of the McDonald Observatory in Texas reported the appearance of two faint semicircular rings around the supernova remnant. His analysis revealed that the rings were light echoes: the initial glare from the supernova had hit two regions of dust and reflected towards Earth, reaching us after a year-long detour.
“Supernova light echoes could shake up our view of the fate of the universe”
Crotts’s finding was significant because it offered a first glimpse of the dusty environment around a supernova. What was not immediately clear was that the same principle could yield information about supernovae from the distant past. That discovery would fall to Rest, while he and his colleagues were looking for something else.
Ghost hunters
Between 2001 and 2005, the researchers were scanning the LMC for objects that might help account for dark matter, the invisible mass thought to make up 23 per cent of the universe. In particular they were searching the space between the LMC and the Milky Way for “massive compact halo objects”, or MACHOs. That includes brown dwarfs, black holes, wayward planets or any other substantial objects too dim to see at great distances.
Hoping to spot a rare instance of a MACHO passing in front of an LMC star, the team took images captured on separate occasions and digitally subtracted one from the other, leaving behind only those features that had changed. There is something eerie about looking at a galaxy when most of its stars and nebulae have been removed; it is a bit like peering through the walls of a house and seeing ghosts moving around inside.
Not surprisingly, among the ghosts the MACHO hunters saw drifting through the LMC were the ring-shaped light echoes of supernova 1987A. Echoes give the appearance of motion because as the supernova’s light travels outward it illuminates successive sections of dust, creating an ever-expanding pattern. As the survey entered its third year, however, other features started turning up that baffled Rest and his colleagues.
These were not stars, but elongated patches of light that tended to appear side by side. At first the researchers thought they were seeing internal reflections in their telescope, but the features persisted, says team member and astronomer Doug Welch of McMaster University in Hamilton, Ontario. “They kept marching in the same direction.”
Rest wondered if the motions of the mysterious objects might provide a clue to their origins. On a scrap of paper, he marked out the locations of the objects within the LMC and added arrows that showed which way they were moving. They were arranged in expanding circular patterns, and each could be traced back to one of three points of origin. The circles were reminiscent of the supernova 1987A light echoes but much larger.
Next, Rest checked a list of supernova remnants in the LMC, and in each case he found one at the centre of a circle – except these remnants were far older than 1987A. Rest knew then that he was seeing echoes of three supernovae from centuries ago. “It was the best single moment I’ve ever had as a scientist,” he says.
The team published its discovery in December 2005 (Nature, vol 438, p 1132), but within a week of Rest’s revelation they were already considering a more exciting possibility. Could they use the same technique to find light echoes from historical supernovae in our own galaxy? Compared with the LMC, the Milky Way presents a much more challenging hunting ground. Although light echoes would be closer to us, that also means they should appear larger and more diffuse. Furthermore, they could turn up just about anywhere in the sky – a vast search area. “We have to be selective,” says Rest.
The researchers acquired telescope time in both the northern and southern hemispheres, and drew up a search strategy based on the six known Milky Way supernovae of the past 1000 years (see “Supernovae in history”). To help with this, they turned to a bit of basic geometry. Since light travels with a constant speed, all echoes arriving at the same time from a given explosion must have travelled the same distance.FIG-mg26021202.jpg
That matches the geometry of an ellipsoid, the surface of a 3D ellipse: the combined distance from two points, called foci, to any point on an ellipsoid is always the same. With the explosion at one focus and Earth at the other, an ellipsoid represents all possible points from which a reflection of a supernova might reach us now (see Diagram). You need the date of the explosion and its location to calculate the shape of the ellipsoid.FIG-mg26021201.jpg
Rest and his colleagues reckoned the best places to search were where these ellipsoids are likely to intersect with dust that could scatter light our way. That means the mid-plane of our galaxy, where most interstellar dust is confined. The angle at which the light hits the dust is also important. Even with these restrictions, the search area for each supernova remains large, and can only be covered by taking hundreds of images. Since light echoes form irregular patterns that cannot easily be recognised by software, the team must eyeball each “difference image” made by subtracting two regular images, and look for anything unusual. “It’s hard slogging,” says Welch, “but the rewards…”
If they do succeed, the first reward would be to confirm what kinds of supernovae our ancestors saw (see “Meet the supers”). There are various types, and each produces a characteristic light curve – showing how its brightness changes over time – and displays a distinct pattern of lines in its spectrum that reveal what elements were present. “Looking at a spectrum is the unambiguous way to determine what type of supernova you’ve got,” says Frank Winkler of Middlebury College in Vermont. “Of course, we don’t have that from hundreds of years ago.”
Forged in a fireball
Fortunately, the spectrum of a light echo would preserve the details of the direct light, revealing the signatures of atoms forged in the supernova’s fireball. Also, the width of the spectral lines would reveal the speed at which the explosion unfolded. These details would allow astronomers to compare a supernova’s remnants with the original explosion. “Information from these two time periods can really help you to understand the physics between the supernova explosion and what it does to its surrounding medium,” says Rest.
Supernovae are key players in the interstellar ecology that determines how solar systems come to be. In fact, all elements heavier than iron are produced exclusively by supernovae, and heavy elements seem to be abundant in stars that have planets. Stellar explosions may even make our galaxy a more likely environment for life by generating the kinds of elements that make up Earth-like worlds, including radioactive ones that help keep planets hot and geologically active. But the relative abundances of elements left behind in remnants of historical supernovae are not always as expected; recapturing the light from those supernovae could reveal why, and lead to new models of what happened.
Another breakthrough might come from light echoes of type Ia supernovae, the brightest and most spectacular variety. The peak brightness of these explosions is remarkably consistent, making them “standard candles” that provide an independent measure of the distance to remote galaxies. By combining this with the speeds at which such galaxies are receding, astronomers a decade ago reconstructed the expansion history of the universe. To their surprise they found the expansion was accelerating, and a new entity known as dark energy has been proposed to explain this (żěè¶ĚĘÓƵ, 17 February, p 28).
A better understanding of how type Ia supernovae explode could improve how accurately we can quantify dark energy and its impact on the universe. Until now astronomers could only observe supernovae from one direction – head-on from Earth. In using type Ia supernovae as a distance yardstick, astronomers assume that they look the same from all angles. However, recent evidence from their spectra suggests they can be asymmetrical, which would mean their brightness depends on their orientation (Astrophysical Journal Letters, vol 652, p 101).
That could affect our estimates of dark energy’s influence on the universe. “Our understanding of dark energy is now limited by our understanding of type Ia supernovae,” says Lifan Wang of Texas A&M University in College Station. If these explosions do not brighten equally in all directions, it could explain why the data that support dark energy have an uncertainty of about 15 per cent. That in turn could make the difference between a universe that will expand with a constant acceleration, or one that will be eventually torn apart in a scenario known as the big rip.
Of the six historic supernovae that Rest’s team is searching for, three are suspected to be type Ia, including the supernova of 1006. If any yields multiple light echoes – each reflecting a different view of the supernova – astronomers will for the first time be able to test whether the viewing angle makes any difference. “To me this is the great piece of new science that is possible with light echoes,” says Welch. “You can suddenly talk about observational constraints on symmetry and supernova explosions.” Other researchers think the plan is promising but could be difficult. “It would be a challenging thing to do,” says J. Craig Wheeler of the University of Texas in Austin. “But the potential is there.”
Already there is encouraging news: going back to one of their LMC echoes, Rest’s team recently measured the first spectrum of a centuries-old supernova. They confirmed that it was a special variety of type Ia – information otherwise impossible to obtain. The researchers have yet to confirm the discovery of any light echoes from the historic supernovae in our own galaxy, but they have seen promising candidates from more than one explosion. As żěè¶ĚĘÓƵ went to press, they were preparing to do spectroscopic work to verify their findings.
Since echoes grow dimmer with time, the easiest to spot should come from more recent supernovae. That includes those seen by Tycho Brahe in 1572 and Johannes Kepler in 1604, as well as the most recent of all: dubbed Cas A after the constellation Cassiopeia, where it occurred, the remnant of this supernova was discovered when early radio telescopes picked up the signal of its energised debris. Astronomers have since calculated that Cas A should have appeared in Earthly skies during the mid-17th century. That it is not mentioned in records suggests interstellar dust blocked it from view, but the same may not be true of its light echoes. “Now that would be really interesting,” says Winkler.
Still earlier supernovae could have generated light echoes that remain visible. This should be true for the 1006 supernova, says Welch, because its position high above the galactic plane affords very favourable scattering angles. If so, we may yet witness the brightest supernova on record. When bin Ridwan recorded it, he could scarcely have imagined that his words might be compared with observations of the same event, more than a thousand years later. “It’s like seeing it again for the first time,” says Welch. “The cosmos has given us another chance.”
Meet the supers
Observations of distant galaxies reveal that supernovae come in two main types. Those designated type II are the products of massive stars undergoing “core collapse” when they run out of fuel. The collapse gives birth to a neutron star or a black hole in the star’s interior, while at the same time releasing enough energy to rip the outer portions of the star to bits.
Even more spectacular is a type Ia supernova, produced when a white dwarf star draws vast quantities of gas off a nearby orbiting companion. As the gas piles up, the white dwarf becomes critically unstable and suddenly erupts in a thermonuclear flash. Lastly, types Ib and Ic are special cases of core collapse in which a dying star has already lost its outer shell of hydrogen.
