On Earth we seem insulated from the violence of the cosmos. Though there is
drama a-plenty in the lives of stars and galaxies, the timescales are so long
that from here they seem peaceful and unchanging. But there are some cosmic
happenings that, should you see them, would take your breath away. About once a
day, for no more than a few seconds, the sky is blasted with a huge burst of
radiation. 鈥淚 call them God鈥檚 firecrackers,鈥� says leading theorist Stan Woosely
of the University of California at Santa Cruz. Invisible to the naked eye, these
gamma ray bursts (GRBs) come from incredibly rapid explosions of almost
unimaginable violence. By some estimates, they radiate more energy in a few
seconds than the Sun does in 10 billion years.
Telescopes and satellites have been picking up these dramatic fireworks for
25 years. And yet in all that time their nature and origin have remained a
mystery. Some astronomers believe they come from within our Galaxy while others
hold that they are from the farthest reaches of the Universe. Now, at last, we
should find out. On 28 February, an Italian-Dutch satellite called BeppoSAX
detected a burst that took the astronomical community by storm. And just three
weeks ago, another burst happened which could finally crack the problem.
Observers around the world are scrambling for their telescopes and theorists are
hastily revising their models to try to explain the data. 鈥淚t鈥檚 terrifically
exciting,鈥� says Woosely. 鈥淭his seems to be the breakthrough that we鈥檝e been
waiting for.鈥�
GRBs were first spotted in the 1960s by American satellites on the lookout
for clandestine nuclear explosions in space. The bursts could last from a few
milliseconds to several minutes. But it was not until 1973, after painstaking
analysis, that Ray Klebesadel, Ian Strong and Roy Olson of Los Alamos National
Laboratory finally proved that these bursts were cosmic, not communist. Their
discovery sent a wave of excitement round the world, and GRBs have since spawned
more than a hundred theories about their origin鈥攕ome of them distinctly
bizarre. At a conference in 1975, astrophysicist Mal Ruderman from Columbia
University commented wryly that he was surprised no one had suggested that GRBs
came from comets made of antimatter falling into white holes. The American
magazine National Enquirereven claimed that GRBs were the aftermath of a
battle between alien civilisations.
Advertisement
However, most astronomers came to believe that GRBs originated from neutron
stars somewhere in the flat disc of the Milky Way. Neutron stars are born,
phoenix-like, from the ashes of a giant star. When these huge stars reach the
end of their lives, their central nuclear power source fails and they collapse
in on themselves. This triggers a violent explosion鈥攁 supernova鈥攊n
which much of the star鈥檚 outer material is blasted off into space. A collapsed
core made entirely of neutrons鈥攁 neutron star鈥攊s left behind.
Though neutron stars are typically just a few kilometres across, they are
incredibly dense and are packed with energy left over from the supernova. So
it鈥檚 easy to picture ways to release some of this energy and generate a GRB.
Perhaps an asteroid could hit the surface of the neutron star and release its
gravitational energy as a burst of gamma rays. Or perhaps the crust of the star
could fracture in a giant 鈥渟tarquake鈥�.
These ideas persisted until 1991, when NASA鈥檚 Compton Gamma Ray Observatory
was launched, carrying the Burst and Transient Source Experiment. BATSE was
designed to test the disc model by checking the distribution of GRBs around the
sky. The Earth sits about two-thirds of the way out from the centre of the
galactic disc. So astronomers reasoned that if the GRBs came from within the
disc, we should see more of them in the direction of the centre of the Galaxy.
They were wrong. 鈥淭he distribution was completely even,鈥� recalls Jerry
Fishman, the principal investigator on BATSE, based at NASA鈥檚 Marshall Space
Flight Center in Huntsville, Alabama.
What鈥檚 more, BATSE had trouble picking up fainter GRBs even though it should
have been sensitive enough to see them. Presumably the faintest bursts are the
ones farthest away, suggesting that there is a cut-off in the number of bursts
that BATSE could see at some distance from the Earth. 鈥淲e were seeing the edge,鈥�
says Fishman.
Outer edges
The conclusion was inescapable. If the number of GRBs is the same in every
direction, the Earth must be sitting in the middle of an extremely uniform
sphere of GRB sources. Since our Galaxy is a flat disc, how could the sources be
spread out into a sphere? And the sources clearly have an outer edge. What could
be defining that edge? Many astronomers revised their thinking dramatically. The
natural explanation, they decided, was that GRB sources stretch out to the
farthest reaches of the Universe. In other words, they are not galactic but
cosmic. The same number of sources lie in every direction because the Universe
is vast and uniform. And the outer edge is where the expanding Universe has
redshifted the GRBs out of the energy bands that the satellites can see.
It鈥檚 a shocking notion, because the farther away the GRBs are, the more
energy they must be giving out. 鈥淚f they鈥檙e at cosmological distances, they鈥檙e
the biggest explosions in the Universe,鈥� says Woosely. It is no longer enough to
drop an asteroid onto a neutron star. You need 10 billion times as much energy
as you would if the explosions were in the Milky Way. It takes events that are
truly catastrophic鈥攖he most energetic since the big bang created the
Universe.
But how can you make that much energy in just a few seconds? One way is to
crash two neutron stars together. 鈥淵ou need a way to make a lot of energy and
this would do it,鈥� says Woosely. 鈥淒rop an apple on a neutron star and you get a
megatonne. Drop a neutron star on a neutron star and you get a lot of
尘别驳补迟辞苍苍别蝉.鈥�
This makes sense because there are plenty of star systems containing two
neutron stars orbiting each other. Every galaxy has several thousand of them,
according to Ramesh Narayan of Harvard University in Cambridge, Massachusetts.
The neutron stars radiate gravitational waves and gradually lose energy,
spiralling in towards each other. Eventually they collide with a spectacular
burst of energy. Because neutron stars are so small and compact, these
collisions last only a few seconds鈥攋ust right for making GRBs.
Narayan, working with Tsvi Piran and Amodz Shemi from the Hebrew University
in Jerusalem estimated the rate at which bursts should reach the Earth if they
are caused by colliding neutron stars. Sure enough, the researchers discovered
that we should see around one a day.
Escaping a black hole
There are problems with this model, however. The biggest is that when two
neutron stars collide they form a black hole, which could swallow up all the
material from both stars and allow nothing to escape, not even radiation.
However, it is possible that some debris from the explosion is not swallowed up
by the black hole but remains trapped in an orbit around it, says Martin Rees,
Britain鈥檚 Astronomer Royal. The orbiting material would be spinning very rapidly
in a flat disc, and the energy from this spinning could cause the GRB. It works
like this. The spinning material would be threaded through with a magnetic field
left over from the neutron stars. But because different parts of the disc would
be spinning at different speeds, the field lines would become twisted. 鈥淵ou
might start with a nice ordered field,鈥� says Woosely, 鈥渂ut you twist it like
rolling thread around a spool.鈥� This would quickly become unstable, and release
a sudden burst of energy, he says.
In fact, says Woosely, this scenario is not limited to colliding neutron
stars. Anything that produced energetic material spinning round a black hole
would do the trick. His favourite model is what he calls a 鈥渃ollapsar鈥�. It鈥檚 a
bit like a supernova but for massive stars made of helium, between 10 and 15
times the size of the Sun. When these implode at the end of their life, the
gravitational pull from the core might be so strong that it would stop the outer
material from being blasted away. The centre of the star would form a black
hole, instead of a neutron star, and the rest of the stellar material would be
trapped in orbit around it. This scenario could produce GRBs in just the same
way as the colliding neutron stars, but with much greater energy.
But while theorists are working on these and many other ideas for
cosmological explosions, proponents of the galactic theory have not given up
hope. Don Lamb from the University of Chicago, one of the champions of a
galactic origin for GRBs, speculates that there could be a spherical 鈥渃orona鈥� of
neutron stars surrounding our Galaxy. They would need to extend way out beyond
the edges of the galactic disc鈥攆ar enough to make the distance between the
Earth and the centre of the Galaxy negligible by comparison. Otherwise we would
still see slightly more bursts towards the Galactic centre.
It sounds odd, but it is possible. Lamb points out that when a neutron star
is born in a supernova, it can often receive a 鈥渒ick鈥� from the explosion that
can send it heading out of the Galaxy in any direction. Many such stars would
form a corona around the Galaxy. The problem is how to account for the large
number of GRBs: theory suggests that every neutron star would have to generate
up to a thousand bursts during its lifetime. Lamb counters that events such as a
colliding asteroid or a starquake could happen repeatedly to the same star. In
fact, last year there was a hint that several bursts were coming from the same
part of the sky. Lamb points out that if two neutron stars collided, as in the
rival theory, that would definitely be a one-off.
On the other hand, says Woosely, the cosmological idea could still hold even
if the GRBs repeat. He imagines a scenario in which a black hole slowly captures
another star. On the first pass, the black hole could break the star into
pieces, just as the planet Jupiter did to Comet Shoemaker-Levy 9 in 1995. On the
second pass, the pieces could be drawn inwards. One by one, the pieces would
form discs around the black hole. The magnetic field lines within the disc would
then twist and each disc would release its energy in a GRB, just as for the
colliding neutron star model. This way, there would be a series of bursts from
the same source.
What has prevented astronomers from sorting out the origins of GRBs is a lack
of data. GRBs are spectacularly difficult to study because they occur randomly
in time and position and are over so quickly. By the time astronomers have
managed to work out exactly where they are, and swing their telescopes round to
look at them, there is nothing left to see.
Clues in the afterglow
One thing everyone has been hoping to spot is an 鈥渁fterglow鈥� of X-rays or
even visible light. Rees points out that whether the explosions causing the
bursts are galactic or cosmological, they should send a fireball of ionised gas
tearing into space at very close to the speed of light. The fireball would slam
into the molecules and atoms that exist throughout space and decelerate,
emitting another, more sustained burst of energy at lower frequencies than the
initial gamma rays. This afterglow should last longer than the rapid GRB.
Studying it should provide vital clues about the origin and nature of the
explosions.
But to study an afterglow you still have to move quickly. You also need to
know exactly where to look. BATSE can spot a burst and pass the information
rapidly to telescopes round the world. But unfortunately it is not very good at
pinpointing the exact location of the burst鈥攊ts best guess is within a
huge 2 degrees of the sky, or four times the diameter of the full Moon.
There are ways to get more accurate information, but at the cost of speed. A
whole interplanetary network of spacecraft, which are carrying out other
missions, can also detect GRBs. Comparing the time that a GRB arrives at the
different spacecraft gives the position of the burst very accurately. But it can
take as long as a month before all this information reaches Earth.
Speedy precision
BeppoSAX has changed all that. Launched last year to study high-energy
transient phenomena of all kinds, it has the capacity to pinpoint GRBs with
great accuracy and send the information quickly back to the ground. Oddly, it
seemed to have escaped the notice of many astronomers. But on 28 February it
took everybody by surprise. In the early hours of the morning, the spacecraft
spotted a burst and recorded its position. Within eight hours, the BeppoSAX
team, headed by Luigi Piro of the Institute of Astrophysics in Frascati, had
discovered a fading afterglow of X-rays using instruments on the same
satellite.
Meanwhile, astronomer Jan van Paradijs of the University of Amsterdam was in
Huntsville, Alabama, when he heard about the burst from a graduate student in
Holland. Quickly he alerted colleagues at the British William Herschel Telescope
on La Palma in the Canary Islands, where he had some observing time booked. The
team swung the telescope round to where the GRB had been and found what everyone
had been waiting for: a definite glow of visible light. A week later they
checked again, by which time the light had begun to fade. At last they had found
the afterglow.
When news of the X-ray and optical afterglow broke, astronomers went wild.
Within days, the Internet was flooded with papers detailing revised models and
follow-up observations. On 8 March, a team led by Paul Groot used the European
Southern Observatory鈥檚 New Technology Telescope in Chile to view the afterglow.
To their excitement, they spotted a faint fuzz of light surrounding the point of
light left behind by the GRB. Was it the glow from a faint, distant galaxy,
proving beyond doubt that the burst was cosmological?
A team from Caltech in Pasadena, headed by Mark Metzger, used the 10-metre
Keck telescope to view the region, and they too picked up the 鈥渇uzz鈥�. Van
Paradijs persuaded Robert Williams, the director of the Space Telescope Science
Institute in Baltimore, to use some of his precious discretionary time on the
Hubble Space Telescope to view the afterglow. On 26 March, it also spotted the
faint fuzz of light surrounding the point where the GRB had appeared
But in the next few weeks the picture became increasingly confusing. The
Hubble team looked again on 7 April and saw no signs that the fuzz was
fading鈥攋ust what you would expect if the light came from a galaxy. But the
Keck team looked again and the fuzz had vanished below their detection limit.
Claim followed counter-claim. The Hubble team reported that the fuzz was not
moving, suggesting that it is very distant. On 17 April, Nature
published a paper by van Paradijs and his many collaborators making a strong
claim that the fuzz is a distant galaxy.
But on the very same day, Patrizia Caraveo from the Institute of Cosmic
Physics in Milan and several collaborators in Europe and the US reported that
they had reanalysed the Hubble data and found that the point object was moving
so quickly across the sky that it had to be coming from within our Galaxy.
Hanging around
The story became still more complicated when Rees and Ralph Wijers from the
University of Cambridge issued a paper with their collaborator Peter Meszaros of
Pennsylvania State University, at University Park, Pennsylvania, pointing out
that the afterglow had hung around too long for the GRB to be galactic. If it
were in our Galaxy, they said, it should have disappeared after about a
day鈥攂ut it was still visible a month after it was first detected. 鈥淚f it鈥檚
local, the fireball would have less energy and slow down more quickly,鈥� says
Rees.
鈥淚t鈥檚 been an incredible roller coaster for all of us,鈥� says Lamb. Almost
every day new evidence seemed to swing the argument one way or the other. 鈥淲hen
we鈥檝e been down they鈥檝e been up, and when we鈥檝e been up they鈥檝e been down.鈥�
And in the last few weeks the picture has channged yet again. On 8 May,
BeppoSAX spotted another burst lasting about 15 seconds. Although GRBs happen
every day, the satellite has to pick them up on two different instruments to get
an accurate fix, which will probably happen only about five or ten times a year.
The next day Howard Bond of the Space Telescope Science Institute in Baltimore
spotted a starlike object that seemed to be changing brightness rapidly just
where the burst had been. The Caltech team were quick to respond. After two
nights of observing on the 10-metre Keck II telescope, they had the
prize鈥攁 spectrum of the object. The spectrum held few clues about the
object itself, but to their excitement, the team noticed that some of the light
was being absorbed by an object lying between the light source and the Earth.
The intervening object, they calculated, is at least 7 billion light years
away鈥攎ore than half way to the edge of the Universe. If the light really
was coming from the burst, it had to be cosmological.
鈥淣ow we have a clear case for a cosmological distance scale,鈥� says Bohdan
Paczynski of Princeton University, in New Jersey.
Lamb is staying quiet about the new finding until more information emerges,
but even if it proves correct it might not be curtains for the galactic theory.
After all, it remains to be seen whether this burst is typical. 鈥淧eople can be
very stubborn,鈥� says Paczynski. 鈥淚t鈥檚 very important to see a pattern. If you
always see the object move, or if you always see a galaxy, that should convince
people one way or another.鈥�
While the different camps wrangle, everyone is eager to sort out exactly how
far away the GRB sources really are. 鈥淲hatever they are, they will tell us
something completely new and unexpected about the Universe,鈥� says Lamb. And if
BepposSAX continues to work as well as it has up to now, by the end of the year
the secret should finally be out. 鈥淣ature has put up a marathon chess match over
the past quarter century,鈥� says Lamb, 鈥渂ut she has just lost her queen. Within a
few months, I think it鈥檚 going to be checkmate.鈥�