Death from the Skies!: These Are the Ways the World Will End... - Philip Plait (2008)

Chapter 4. Cosmic Blowtorches: Gamma-Ray Bursts


There could be no warning: the wave front was moving at the velocity of light, the ultimate speed limit in the Universe. Nothing can move faster, so the wave of death brought its own announcement.

Across the Earth’s southern hemisphere, people were having a normal day: shopping, working, playing, walking, hunting. When the beam reached Earth, that all changed instantly. The sky looked perfectly normal for one second, and then literally in the next it suddenly lit up, like a switch had been flipped. An intensely bright spot flashed in the sky, so bright that anyone looking at it instinctively looked away, eyes watering from the onslaught.

The new star in the sky was so fantastically bright it could outshine the full Moon, but didn’t last long. It started to fade after less than half a minute, and was bearable to the eye after a few minutes. People stood in the streets, in the deserts, in the Antarctic plains, on ships at sea in the South Pacific and Indian oceans, and boggled at the incredibly bright but rapidly dimming new star in the sky.

But their amazement soon faded, and they began to go about the daily business of their lives.

Most people had already put the event out of their minds when, hours later, a flood of subatomic particles from the fading star slammed into the Earth’s atmosphere. Invisibly, these particles rained down out of the sky, covering the Earth from the south pole to 30 degrees north of the equator. Australia, New Zealand, South America, essentially all of Africa and India, and half of China were blanketed with a lethal dose of radiation. It didn’t matter if people were inside their houses, or outside under a clear sky: all of them were exposed.

Across two-thirds of the Earth, people started dying.

North America, Europe, and much of Asia were spared the immediate effects, but it hardly mattered. With most of the human population dying, the impact on the globe was overwhelming. And those who weren’t killed outright by the burst of radiation were doomed anyway: the Earth’s ozone layer disintegrated in the onslaught, dropping to half its usual strength. Ultraviolet light from the Sun was able to penetrate almost freely to the Earth’s surface, killing off the base of the food chain.

The final blow was yet to come. Spawned by the wave of subatomic particles, a thick layer of smog began to form in the air, and within days the sky was a dank reddish-brown color over the whole planet. Any hardy plants that had managed to stay alive thus far suddenly found the sunlight and temperature dropping . . . which was bad enough, until the acid rain began.

And that was short-lived as well. Within weeks, the Earth’s temperature had dropped enough that a new ice age was triggered. It wasn’t long before the glaciers started their march from both poles.

The people who had survived the initial months of the event learned that they had witnessed the death of the supermassive star Eta Carinae, but that knowledge didn’t help them. The mass extinction the star triggered would be the worst the Earth had ever seen, and when it was finally over, there were no humans left to wonder at how a single star trillions of miles away could destroy all of history in less than a minute.


By the 1960s, the situation between the United States and the Soviet Union was grim. The USSR had put a base in Cuba, less than a hundred miles off the Florida coast. A failed invasion by the United States hadn’t helped. Both superpowers were testing nuclear weapons on, beneath, and above the surface of the Earth. The USSR had exploded the largest thermonuclear bomb in history, equivalent to the detonation of 50 million tons of TNT.28

Needless to say, people on both sides were nervous. The end of the world by our own hand was a very real possibility.

So, in August of 1963, the United States, the United Kingdom, and the USSR signed the historic Nuclear Test Ban Treaty, limiting testing of such weapons. The very first article of the treaty states:

Each of the Parties to this Treaty undertakes to prohibit, to prevent, and not to carry out any nuclear weapon test explosion, or any other nuclear explosion, at any place under its jurisdiction or control [ . . . ] in the atmosphere; beyond its limits, including outer space; or under water, including territorial waters or high seas.

This was a serious restriction. Even after more than a decade, the results of nuclear testing were often surprising. Weapons were tested not just to increase the explosive yield and improve other engineering issues, but also to see what their effects were on the environment. Just the year before the treaty was signed, in 1962, the United States had exploded a device called “Starfish Prime” 250 miles above a remote location in the Pacific Ocean. This height is essentially in space; the Earth’s atmosphere is extremely tenuous that far above the surface. Starfish Prime had the relatively small yield of 1.4 megatons (that is, equivalent to 1.4 million tons of TNT), yet the effects were profound. A vast pulse of gamma rays, extremely high-energy photons of light, was created in the blast. This wave of gamma rays slammed into the Earth’s atmosphere, blasting electrons off their atoms. Moving charged particles create magnetic fields, and the sudden surge of rapidly moving electrons generated a huge electromagnetic pulse of energy, or EMP. This surge blew out streetlights in Hawaii, fused power lines, and overloaded TVs and radios—all from over 900 miles away.

Testing in space was dangerous, and the long-term effects were still not understood at the time. It became more and more clear that fallout and other effects made atmospheric and near-space nuclear tests extremely unwise. The Test Ban Treaty was hailed as a major step toward world peace.

Of course, the United States trusted the Soviet Union completely, knowing they wouldn’t dream of violating the treaty . . . yeah, sure. While the treaty was an excellent start, no one trusted anyone else at all, and each side was very suspicious of the other. In fact, American scientists pointed out that the USSR could blow up bombs on the far side of the Moon and these would be difficult to detect. The Soviets could break the treaty and the United States would never know. What to do?

Nothing feeds engineering progress like fear. The Americans quickly found a way to check up on those scheming Soviets.

While a bomb blown up behind the Moon might be hard to detect visually, its expanding debris cloud would generate quite a bit of radioactive material in space that could be detected. One such radioactive by-product would be gamma rays. Detection technology for gamma rays was relatively new in the 1960s, but it was sufficient to sniff out any of that radiation from translunar explosions. There was one catch: gamma rays from space cannot penetrate the Earth’s atmosphere, so the detectors would have to be launched on a satellite.

Besides the usual problems involved with lofting detectors into space, there was also the issue of accounting for gamma rays emitted by astronomical sources, and not from Soviet nukes. The Sun emits gamma rays, and high-energy particles from solar flares can be mistaken for them as well. A satellite might see a sudden jump in gamma rays, only to have been fooled by a solar eruption or a random particle hit.

The obvious solution was to launch gamma-ray satellites in pairs. A random particle hitting one satellite would not be seen by the other, providing a check against false detections. The data from each satellite could be compared, and if both saw an event, scientists could assume it potentially came from a noncosmic source. Other, existing satellites tracked solar flares, so those could be consulted as well.

The pairs of satellites were quickly constructed and launched. Named Vela—“watch,” in Spanish—the first set was launched just days after the Test Ban Treaty was signed. They were initially crude, only able to positively detect gamma rays after taking an “exposure” of 32 seconds. But things progressed swiftly, and by 1967 the fourth pair had been launched, with a fifth—highly advanced compared to the earlier missions—ready to go.

Two scientists, Roy Olson and Ray Klebesadel, were assigned the laborious task of comparing the observations of one satellite with those of its mate. As they checked, signal after signal turned out to be negative. But in 1969, they found their first hit. The Vela 4 satellites both registered a gamma-ray event from July 2, 1967, shortly after they were initially launched. A quick look at solar flare data revealed no activity that day. Later, they found that the still-flying Vela 3 satellite pair saw the event as well.

There was one problem—whatever caused the gamma-ray event didn’t look like a nuclear blast. The amount of gamma radiation and how it fades with time are very distinctive for a nuclear weapon, and the July 2 event looked completely different. There was a strong, sharp peak of emission lasting less than a second, followed by a longer, weaker pulse lasting for several more seconds.

What could this be? Unfortunately, the Vela 4 satellites couldn’t tell from what direction the radiation came, so there was no way of determining the source. It may have come from behind the Moon, as feared for a nuclear test, or it may have come from some other spot in the sky entirely. Also, the event began and ended so quickly that there was no prayer of using an optical telescope to find it.

However, the Vela 5 and 6 satellites were more powerful—they were more sensitive to gamma rays, and had better time resolution. If the July 2 event repeated, or something else like it occurred, Velas 5 and 6 had a much better shot at figuring out what was going on. Deciding that discretion was the better part of valor, the scientists waited to release the July 2 event data.

It was a good choice. Over the next few years, several more of these mysterious bursts were detected. Plus, there was an added benefit to having more satellites flying: since they were separated by thousands of miles, a crude direction could be determined for each flash. Even at the speed of light, it takes a finite amount of time for a pulse of radiation to get from one satellite to the next. That time delay, together with the known positions and separations of the satellites, could be used to triangulate on the direction of the event.

As the data built up, the scientists were astonished: the gamma-ray flares were originating from random spots in space! None appeared to come from the Sun or the Moon. It became clear that what Olson and Klebesadel were seeing was some totally unknown but extremely powerful astronomical event that no one had any previous clue about. It seemed ridiculous—how could the Universe hide such a thing from the prying eyes of astronomers?—yet there they were.

By 1973, Klebesadel and Olson had accumulated enough data to go public with the news. Together with another scientist named Ian Strong they presented the results at a meeting of astronomers in Ohio, and published a paper titled “Observations of Gamma-Ray Bursts of Cosmic Origin” in the prestigious Astrophysical Journal. The paper outlined the sixteen bursts they had seen up to that time (by 1979, when the Vela missions finally ended, over seventy gamma-ray bursts, or GRBs, had been detected by the satellites).

It should be noted that several other astronomers had found weird gamma-ray emissions in their detectors on various satellites as well, but couldn’t be sure what they were. It took the accumulated high-quality data from the Vela satellites to be able to determine that these events were coming from deep space, or at least from outside the Earth-Moon system.

Not that the scientists had a clue what these things actually were. GRBs are confusing today as well as in those early days. When Klebe-sadel’s team released their results, the origin of GRBs was a complete mystery. Gamma rays can only be generated by high-energy events like exploding stars, solar flares, or nuclear weapons. But they had established that none were from the Sun, and none were associated with any supernovae. And they clearly weren’t nuclear tests—the Vela satellites did detect several atmospheric weapons tests (from other countries), but the signals for those were unambiguous.29

What could the bursts be? To make matters more confusing, the distances to sources of GRBs were completely unknown. It was hard to imagine they were really close by (say, inside the solar system), because it didn’t seem like any object or event could generate gamma rays that we wouldn’t already know about. And again, the data didn’t link the bursts to any observed astronomical events farther away.

As mundane explanations fell by the wayside, odder ideas were proposed. Maybe the bursts were from comets hitting the surfaces of super-dense neutron stars, or maybe they were from some other equally exotic event. No one knew. But one thing most astronomers at the time agreed upon was that GRBs were not very far away—that is, from outside the galaxy. The farther away a source is, the brighter it must be for us to detect it. For a GRB to be outside our galaxy meant it had to generate literally unbelievable amounts of energy.

But this didn’t help much. There were still too many unknowns.

There were two fundamental problems with determining the origins of GRBs: the lack of real-time information, and the lack of directional information.

The former was a significant problem. The time it took for the information from the satellites to be beamed to Earth, recorded, and then interpreted could be measured in days, or even weeks (or, in the case of the first one, two years). The GRBs, however, faded away in mere seconds! By the time the burst was confirmed, it was long gone. There was hope that perhaps GRBs emitted light in other wavelengths—X-rays, or optical light—and that this glow would persist long enough to be seen by other telescopes. Assuming GRBs were some sort of explosion, it would make sense that there would be an afterglow, giving astronomers time to find it. But that leads to the second problem: where to look?

The gamma-ray detectors of the time had poor eyesight: early missions simply couldn’t see the direction from which the gamma rays came.

Optical light—the kind we see—has a relatively low energy. Carefully aligned lenses or mirrors inside a telescope bend or reflect the light, bringing it to a focus. This can be used to very accurately measure the position of a source of optical light. Gamma rays, however, are more like bullets zipping around. Changing their paths is much harder, and even today focusing them is beyond our technology.

What this means is that while a gamma ray can be detected and counted, getting a direction from whence it came is very difficult. Only the crudest of directions could be obtained by the Vela satellites (it wasn’t much better than “somewhere over there”30). But the direction is critical to understanding the object. If the gamma-ray source’s position is known, other telescopes can be trained at that spot on the sky to see what’s what. Then any visible source seen there can be compared to known sources like galaxies or stars listed in existing catalogs. But some degree of precision is required: if the position of the burst can only be nailed down to, say, an area on the sky the same size as the full Moon, there are still thousands or even millions of objects detectable by a big optical telescope.

Eventually, technology started catching up to the problem. In 1991, NASA launched the Compton Gamma Ray Observatory satellite, which had GRB detectors on it. Compton’s ability to get the positions of GRBs was still not great—it could only nail them down to an area on the sky the size of a quarter held at arm’s length—but it was a definite improvement. Over the course of the mission, it detected over 2,700 GRBs. And while the directions were not precise, just getting that sheer number of observations was a huge advance; after enough bursts were detected, patterns began to emerge.

For one thing, that large collection of bursts allowed scientists to determine that there appeared to be two kinds of GRBs: short ones, lasting in general less than two seconds; and long ones, which lasted more than two seconds. Some bursts were even found to emit gamma rays for several minutes. As more GRBs were observed, it was found that the shorter bursts tended to give off higher-energy (“harder”) gamma rays, and the longer bursts had lower-energy (“softer”) gamma rays. While it wasn’t understood why this might be, it was an important clue to their origins.

But the big scientific result from Compton’s observations was perhaps far more important in solving the riddle: it saw GRBs spread out evenly across the entire sky. At first glance this may not seem to help, but in fact it eliminates many possibilities for their origins.

Imagine standing in a field, and insects are buzzing around. If you’re in the center of the field, then you’d expect, on average, to see the same number of insects no matter what direction you look. But if you’re close to the eastern edge of the field, you will see far more insects to the west (looking out across the length of the field) than to the east (looking out over the edge). The number of bugs you see in a given direction tells you something about your placement in the bug swarm (assuming the swarm is relatively random and symmetric).

So the information from Compton—that GRBs were spread randomly across the sky—instantly tells us an important fact: we are in the center of the GRB distribution in space.

If GRBs were inside our solar system, we’d expect to see more in one direction than another, because we are not in the center of the solar system—the Sun is. We’re offset from the center by nearly a hundred million miles, and you’d expect to see that reflected in the distribution of GRBs. But there is no offset, so they are not coming from objects in our solar system.

But this also means that GRBs are not coming from sources spread around inside our Milky Way Galaxy. Since the Earth is halfway to the edge of the galaxy, GRBs in that case would be seen preferentially toward the center of the galaxy as viewed from Earth. They aren’t, so they are not galactic in origin either.

That doesn’t leave too many options. They could come from stars very near the Sun, like only a few light-years away, but not from farther stars, say, more than a few hundred light-years away, because then we’d start seeing more toward the galactic center. The other choice is that GRBs are very, very far away, from well outside the galaxy, millions of light-years distant.


If you are in the middle of a field of fireflies (left), you see equal numbers of bugs in every direction you look. But if you are off-center in the cloud of bugs (right), you see more in one direction than in another. This information can be used to determine the shape of the cloud of bugs—or, more practically (to an astronomer), the distribution of GRBs in the Universe.


Neither of these options is terribly palatable either. Stars shouldn’t be able to make such high-energy bursts, and if they were really far away, the intrinsic energy emitted would be ridiculously high.

Still, astronomers staked their claims on either side of this issue, publishing papers furiously and arguing—sometimes also furiously—over it. They even staged a famous debate about it between two accomplished scientists who took different sides of the debate: one defending the idea that they were from nearby stars, the other saying they were coming from the distant reaches of the Universe. But even by the time the debate was held, preparations were under way to get the real answers.


The joint Dutch-Italian satellite BeppoSAX was launched in 1996. While it was not designed specifically to hunt for GRBs, it had that capability. More important, it had on board a revolution waiting to happen: detectors that could actually get a good direction for incoming X-rays (which, like their higher-energy brethren, gamma rays, are difficult to pin down). It also had a wide field of view, which increased the odds of detecting a randomly placed burst, even if the position was not well known at first.

In February 1997, a long GRB was detected by the BeppoSAX monitor. It also happened to lie within the field of view of the X-ray detectors. Observations were made, and then repeated a few days later. Breakthrough! The results were clear—a bright source of X-rays had faded considerably in the interval. Astronomers knew that must be from the fading afterglow of the burst. And better yet, the X-ray detectors were able to get a reasonably good position for the burst, now called GRB 970228 (for the gamma-ray burst seen in 1997 on February 28).

Within a month, the Hubble Space Telescope was pointed at the location of the GRB and the breakthrough got more momentum: a fading glow in visible light was detected, and it appeared to be right next to a dim, distant galaxy. This was too close to be a coincidence.

Then, finally, the clincher. In May of that same year, the mammoth ten-meter Keck telescope in Hawaii obtained spectra31 of a GRB afterglow. This allowed astronomers to determine an accurate distance to GRB 970228, and they were astonished to see that it was located a numbing nine billion light-years away. That’s more than halfway across the Universe!

Finally, after thirty years, thousands of burst observations, and countless arguments, a major question was answered: bursts were not only far away, they were very far away. After this, no one doubted the vast distances to gamma-ray bursts. They were coming from well outside our Milky Way, and in fact close to the visible edge of the Universe.

But that left one problem, vast in its own right: what event could possibly generate such incredible energies?


No matter how you slice it, GRBs are, for a short time, the most luminous objects in the Universe, the best bangs since the Big One.

This is no small problem. Imagine a source of light in space: the light it emits will expand as a sphere with the source at the center. As the sphere grows, the light gets spread out, and will appear dimmer to an observer (that’s why lights get fainter with distance). When the distance to the object doubles, the area over which the light spreads out goes up by four times,32 so the brightness will dim by four times. If you increase your distance to 10 times farther away, the light will be only one-hundredth (1 percent) as bright, and so on. The brightness of an object therefore decreases very rapidly with distance. This presented a serious problem for GRB researchers: from a distance of billions of light-years, the explosion that formed the GRB must be huge to be able to be detected at all from Earth. When the numbers were crunched, it didn’t make sense. Even converting an entire star into energy using Einstein’s E = mc 2(see chapter 2) wouldn’t provide enough energy to fuel the burst, and that is literally the most energy you can get from a star (ignoring the inconvenient fact that there’s no known way to convert an entire star into energy, and certainly not in the span of a few seconds).

But there was still an out. What if the blast wasn’t symmetric, expanding equally in all directions? What if it was beamed?

If you take a small lightbulb and turn it on, it emits light in all directions, and its apparent brightness fades rapidly with distance. But if you put it in a flashlight, which collects its light and focuses it into a beam, the light appears brighter from farther away.

Astronomers could almost taste the answer to this piece of the GRB puzzle. Instead of a nearly impossibly energetic blast at a colossal distance expanding spherically and fading rapidly, maybe the explosion was less energetic, but focused into beams. Beaming would mean only a tiny fraction of the energy would be needed compared to a spherical blast.

The energy of the detonation would still have to be frighteningly huge to be seen clear across the Universe, but not impossibly so. In fact, the energy involved would be similar to that of a supernova. This gave astronomers hope that they could find the Holy Grail of GRB science: the engine that drove this phenomenon.

And of all the objects in the cosmic zoo that astronomers knew of, only one could possibly generate those kinds of forces.


Black holes are notorious for sucking down matter and energy, not spewing them out, so it might seem paradoxical that they could be at the heart of gamma-ray bursts, the brightest objects in the Universe.

But the key to this is gravity. And the key to that is how black holes form, so let’s take a step back (a good idea when dealing with black holes) and take a look at this singular event.

In chapter 3, we saw that massive stars explode when they run out of fuel to fuse in their core. The incredibly strong gravity of the core makes it collapse, which sets off a series of events that blows up the star. That description focused mostly on what happens to the outer layers in a supernova, but not what happens to the core itself. But it’s there that the power of the GRB lies.

As the iron core of the incipient supernova collapses, the electrons are rammed into the protons, making neutrons (and emitting neutrinos, the major trigger of the supernova explosion). In a flash the entire core of the star becomes a sea of neutrons with almost no normal matter left. What was once a ball of iron thousands of miles across is now an ultradense neutron star, perhaps ten miles across. It has a mass equal to the Sun, but a density magnified beyond belief: a spoonful of neutron star matter would weigh a billion tons! That is somewhat more than the combined mass of every single car in the United States—imagine 200 million cars crushed down to the size of a sugarcube and you’ll start to get an idea of how extreme neutron star matter is.

The neutron star’s incredible mass is supported by a weird quantum mechanical effect called degeneracy (see chapter 3). It is similar to electrostatic repulsion—the idea that like charges repel—but instead it’s a property of certain subatomic particles where they resist being squeezed too tightly together. Degeneracy will occur if you try to pack too many electrons together, but it also affects neutral particles like neutrons. It’s an astonishingly strong force, and is able to keep the vast bulk of the core from collapsing further. The collapsing core of the star slams to a halt, and a neutron star is born . . .

. . . most of the time. It turns out that if the mass of the collapsing stellar core exceeds about 2.8 times the Sun’s mass, even neutron degeneracy cannot hold it up. The core’s gravity is too strong, and the core collapse continues. This time there is no force in the Universe strong enough to stop it.

What happens next is so bizarre that it stretches the human mind to its limit to understand. As an object gets smaller, but retains its mass, its gravity gets stronger. As an easy example, if you were to somehow shrink the Earth to half its current diameter while still keeping its mass, the gravity you feel (and therefore your weight) would increase. The smaller the Earth gets, the stronger its gravity.

If you wanted to launch a rocket to the Moon from this newly shrunken globe, you’d have to give it a lot more power to overcome the Earth’s gravity. If you shrank the Earth more, the rocket would need even more power, and so on. Eventually, the Earth would shrink so much that its gravity would be literally impossible to overcome.

You might think you just need to add more thrust to the rocket, but when matter gets this dense, Einstein has something to say about the situation. He postulated that gravity is really just a manifestation of bent space. What you feel as a force downward, toward the center of the Earth, is actually a bending of space, like the way the surface of a mattress would bend if you plunked a bowling ball down on it. Roll a marble across the bed, and the path of the marble bends, just the same way an asteroid’s path bends because of gravity when it passes near the Earth.

This is more than just a model, more than mere speculation. Its consequences are very real: if too much matter is packed into too small a volume, the bending of space can become so severe that it literally becomes an infinitely deep pit. You can fall in, but you can never climb back out.

An object like this is like a hole in space. Nothing can escape it, not even light. Since it cannot emit light, this hole would be black. What would you name such a thing?

And so it goes in the core of the exploding star. If the core is too massive to form a stable neutron star, it collapses. All the way down. It shrinks to a mathematical point, space gets bent to the breaking point, and a black hole is born.

The gravity of the hole is intense. Any matter close by will be drawn inexorably into it. But there’s a hitch. Stars spin, and so do their cores. As the core collapses on its way to forming a black hole, that rotation increases, the same way an ice skater can increase her spin by drawing her arms in. Once the black hole is created, it will be spinning very rapidly, and any matter falling in will also revolve around it, like water going down a drain. The closer to the black hole it gets, the faster that matter will swirl around it.

So matter falling into a black hole doesn’t just fall straight in—plonk!—and disappear; it spirals in. The matter just outside the black hole begins to pile up, and it forms a flattened disk called an accretion disk (accretion is the process of accumulating matter). This will happen for any star that is spinning before it collapses, but models have shown that GRB progenitors may be spinning even faster than normal. These rapid rotators form an accretion disk much more readily than a slowly rotating star. And once the disk forms, the ferocious gravity of the black hole will get the inner part of the disk moving very close to the speed of light, and even matter farther out from The Edge will still be moving incredibly rapidly.

When a black hole forms, spin and gravity are not the only things to get amplified. Stars also have magnetic fields, like giant bar magnets (see chapter 2). Just as gravity increases as the star shrinks, so does the magnetic field. A typical star may have a magnetic field not much stronger than the Earth’s: just enough to make a needle in a compass move. But if you take a star a few million miles across and squeeze it into a ball just a few miles across, the magnetism increases hugely as well, getting billions and even trillions of times stronger.

Any matter trying to fall into the black hole is therefore under the influence of a witches’ brew of forces. Gravity tries to draw it in, but its angular momentum counteracts that, forming the disk. The magnetic fields also get twisted up like a tornado as the matter spins around the disk. And on top of it all, there’s just plain old heat, created, oddly, by something familiar amid all these exotic forces: friction. As the matter in the disk swirls madly around under the force of the black hole’s gravity, the particles in it slam together at incredibly high speeds, which generates immense quantities of friction. This heats the disk to millions of degrees.

The sheer heat tends to drive particles away from the black hole. If a particle tries to move outward in the plane of the disk, it slams into other particles and cannot escape. But if it goes up, out, it’s free to travel—there’s less material in that direction. Moreover, the monstrously amplified magnetic fields can also accelerate the particles up and out. The heat and magnetism combine to focus a pair of tight beams, like two ultra-mega-superflashlights glued together at their bases. These twin beams shoot out from directly above and below the black hole, firing outward, away from the black hole in directions perpendicular to the disk.

What happens next is a vision of hell so apocalyptic that it’s difficult to exaggerate. Moments after the black hole is created and the disk forms around it, all that energy—a billion billion times the Sun’s output—is focused into twin beams of unmitigated fury. So much energy is packed so tightly into the beams that they blast outward in opposite directions, eating their way through the star at the speed of light. Within seconds, the beams have chewed their way out to the surface and are free. Any matter in their way is torn apart, heated to billions of degrees, rendered into its constituent subatomic particles, and accelerated to within a hairbreadth of the speed of light. Ironically, by the time they punch their way through the star, perhaps only a few hundred Earth-masses of matter are in the beam, which is huge on a human scale, but tiny on a cosmic one. But that also is a key to their power: since the total amount of matter in the beams is relatively small, it can be accelerated to incredible speeds.

Clouds of gas still surround the doomed star, echoes of past eruptions before the final explosion. The beams of energy and matter slam into this material, creating vast shock waves, sonic booms in the material, but on a mind-numbing scale.

There are also shock waves generated inside the jet itself as parts of it move faster than others. When these collide, the awesome energy of the jet churns up the matter inside them, creating unimaginable turbulence, which in turn adds greatly to the energy emitted. The ensuing conflagration emits gamma rays, huge amounts of them, as the magnetic fields and raw energy of the beams bombard the matter.


When a very massive star’s core collapses, twin beams of matter and energy can be focused by the incredible forces in the star’s center. The beams may last only a few seconds, but contain as much energy as the Sun will emit in its entire lifetime, or more.


A gamma-ray burst is born.

The beams continue on. Behind them, the rest of the star finishes its collapse, forming what would otherwise be a normal supernova. Before the discovery of GRBs, a supernova was considered the most violent, the most energetic single event in the Universe. But a decent GRB can dwarf the energy of even a supernova. Because of this, astronomers coined a new word to describe the event: hypernova.

Once the beams pass through the gas, they continue on, leaving behind superheated matter that begins to cool. As it does, it emits light for some time after the beams have moved on. This is the source of the afterglow sought so dearly by scientists on Earth. The matter can get extremely bright—one GRB in 2008 was nearly 8 billion light-years away, but was visible to the naked eye! But the afterglows fade rapidly, dropping in brightness by factors of thousands in just a few minutes. That’s why the optical afterglow was initially so difficult to detect. Even the titanic energy of a GRB is mitigated by raw distance.

But we now know that GRBs are created in a hypernova, when a massive star explodes . . . and we see massive stars in our own galaxy. Sure, all the GRBs we have ever seen have been at terribly remote distances, billions of light-years removed from Earth.

But what happens if one goes off that’s not far away? What if a nearby star becomes a GRB?


An object that finds itself in the path of the beams of a nearby GRB will have bad things happen to it.

Very bad things.

But before I scare the pants off you, remember that if you are far enough away they are no danger at all. The only reason we can see GRBs at all is because we are in the path of the beams: since all the light of the GRB is focused into those beams, if they miss us we don’t see anything. So if they are far enough away you just see a faint blip, and it’s gone. But if you’re too close . . .

The effects from a GRB are very similar to those of a supernova, which isn’t surprising. They are related phenomena, with GRBs being produced in supernovae, and they both emit huge amounts of energy in the form of gamma rays, X-rays, and optical light.

Where they differ is how well they sow their destruction over different distances. With a supernova, which emits radiation and matter in all directions, the effects die down rapidly with distance. As we saw in chapter 3, they appear to be mostly harmless from a distance of more than 25 to 50 light-years or so.

But GRBs are beamed. Their luminosity does not decrease as rapidly with distance, and this makes them dangerous from farther away. Much farther away.

Every GRB is different, making prognostication difficult. But enough have been observed to do a little averaging and get the effects from a typical GRB, whatever “typical” means when you’re dealing with Armageddon focused into a death ray.

Let’s set the scene.

Why fool around? Let’s say a GRB went off really close: 100 light-years away. Even from that very short distance, the beam of a GRB would be huge, 50 trillion miles across. This means that the whole Earth, the whole solar system, would be engulfed in the beam’s maw like a sand flea in a tsunami.

GRBs, mercifully, are relatively short-lived, so the beam would impact us for anywhere from less than a second to a few minutes. The average burst lasts for about ten seconds.

This is short compared with the rotation of the Earth, so only one hemisphere would get slammed by the beam. The other hemisphere would be relatively unaffected . . . for a while, at least. The effects would be worst for locations directly under the GRB (where the burst would appear to be straight up, at the sky’s zenith), and minimized where the burst was on the horizon. Still, as we’ll see, no place on Earth would be entirely safe.

The raw energy that would be dumped onto the Earth is staggering, well beyond the sweatiest of cold war nightmares: it would be like blowing up a one-megaton nuclear bomb over every square mile of the planet facing the GRB.It’s (probably) not enough to boil the oceans or strip away the Earth’s atmosphere, but the devastation would be beyond comprehension.

Mind you, this is all from an object that is 600 trillion miles away.

Anyone looking up at the sky at the moment of the burst might be blinded, although it would probably take several seconds to reach peak optical brightness, enough time to flinch and look away. Not that that would help much.

Those caught outside at that moment would be in a lot of trouble. If the heat didn’t roast them—and it would—the huge influx of ultraviolet radiation would instantly give them a lethal sunburn. The ozone layer would be destroyed literally in a flash, allowing all the UV from both the GRB and the Sun down to the Earth’s surface unimpeded. This would sterilize the surface of the Earth and even the oceans down to a depth of several yards.

And that’s just from the UV and the heat. It seems cruel to even mention the far, far worse effects of gamma rays and X-rays.

Instead, let’s take a step back. GRBs are incredibly rare phenomena. Although they probably happen several times a day somewhere in the Universe, it’s a really big Universe. The odds of one happening 100 light-years away are currently zero. Zip. Nada. There are no stars close to us that are anywhere near the capability of becoming a burst. The nearest supernova candidate is farther away than that, and GRBs are far rarer than supernovae.

Feel better? Good. So let’s try to be more realistic. What is the nearest GRB candidate?

In the southern sky is a star that looks unremarkable to the naked eye. Called Eta Carinae,33 or just Eta for short, it’s a faint star in a crowded field of brighter ones. However, its faintness belies its fury. It’s actually about 7,500 light-years away, and is in fact the most distant star that can be seen with the unaided eye.

The star itself34 is a monster: it may have 100 or more times the mass of the Sun, and it emits 5 million times the energy of the Sun—in one second, it gives off as much light as the Sun does in two months. Eta suffers periodic spasms, blowing off huge amounts of matter. In 1843, it underwent such a violent episode that it became the second brightest star in the sky, even at its vast distance. It expelled huge quantities of matter, more than 10 times the mass of the Sun, at over a million miles per hour. Today, we see the aftereffects of that explosion as two huge lobes of expanding matter, each looking like the blast from a cosmic cannon. The energy of the event was almost as powerful as a supernova itself.


Eta Carinae is the Milky Way’s scariest star. It may be a binary, with one star 100 times more massive than the Sun. In 1843, it underwent a titanic convulsion, almost as powerful as a supernova, that ejected the two lobes of matter on either side. When Eta finally blows, it may explode as a hypernova and a GRB.


Eta has all the markings of a GRB in the making. While it will certainly explode as a supernova, it isn’t known if it will be a hypernova-type GRB or not. It should also be noted that if it does explode and becomes a GRB, the orientation of the system is such that the beam will almost certainly miss the Earth. We can tell this from the geometry of the gas ejected in the 1843 paroxysm: the lobes of expanding gas are tilted with respect to us by about 45 degrees, and any GRB beaming would be along that axis. To make this more clear: we are in no danger from a GRB, Eta or otherwise, in the near or even mid-term future.

But still, it’s fun to speculate. What if Eta were aimed at us, and it went hypernova? What would happen?

Again, bad things. While it wouldn’t get anywhere near as bright as the Sun, it would certainly be as bright or possibly even ten times brighter than the full Moon. While this is bright enough to make you squint, it would only last a few seconds or minutes, and so would probably do no long-term damage to any flora’s or fauna’s life cycles.

The levels of influx of ultraviolet light would be intense, but brief. People outside would be mildly sunburned, but in all likelihood there would not be a statistically measurable increase in skin cancer rates down the line.

But the situation is very different when you look at gamma rays and X-rays. These would be absorbed by the Earth’s atmosphere, and the effects would be far worse than for a nearby supernova.

The most immediate effect would be a strong electromagnetic pulse, far stronger than the one experienced in Hawaii during the Starfish Prime nuclear test. In this case, the EMP would immediately wipe out every unshielded electronic device on the hemisphere of the Earth facing the burst. Computers, phones, airplanes, cars, anything with electronic circuitry would be fried. This also includes power grids; a huge current would be induced in the transmission lines, overloading them. People would be without power, and without any means of long-distance communication (satellites would have all been fried by the gamma rays anyway). This would be more than an inconvenience, as it means hospitals, fire stations, and other emergency personnel would be without power as well.

But as you’ll see in a moment, we may not have much use for emergency services . . .

The effects on the Earth’s atmosphere would be severe. This situation has been extensively studied by scientists. Using the models described in chapter 3, and assuming a GRB at Eta’s distance, they have determined the effects. They are not pretty.

The ozone layer would take a huge hit. The gamma rays from the GRB would blast apart ozone molecules wholesale. Globally, the ozone layer would be reduced by an average of 35 percent, with smaller localized regions being depleted by more than 50 percent. This all by itself is incredibly damaging—mind you, the ozone troubles we have currently are due to a relatively slight dip of just 3 percent or so.

The effects of this are very long-lasting, and could persist for years—even five years later there could be as much as a 10 percent ozone depletion. During that time, ultraviolet light from the Sun would be more intense on the Earth’s surface. Microorganisms that form the base of the food chain are highly susceptible to UV radiation, and would be killed in vast numbers, leading to a possible extinction-level event that would work its way up that chain.

To make matters worse, the amount of reddish-brown nitrogen dioxide formed (see chapters 2 and 3) in an Eta Carinae GRB event would actually reduce the amount of sunlight reaching the Earth by a significant amount.

The exact effect of this is hard to determine, but it seems likely that even a few percent drop in sunlight over the entire Earth (the nitrogen dioxide would spread all over the atmosphere) would cool the Earth considerably, and could conceivably start an ice age.

On top of this, enough nitric acid would be generated in the chemical mix that there would be acid rain, which would also potentially have devastating environmental effects.

Then there’s the issue of subatomic particles (cosmic rays) from the burst. It’s unclear just how damaging these would be from a GRB. But, as discussed in chapters 2 and 3, high-energy particles may have all sorts of effects on the Earth. A GRB from 7,500 light-years away would inject a vast number of subatomic particles into our atmosphere, and they would be moving at just a shade under the speed of light. Within hours of the initial burst they would slam into the air, creating a shower of muons. We see muons coming from the sky all the time, but in small amounts. From a nearby GRB, the number of muons generated would be huge. One team of astronomers calculated that as many as 300 billion per square inch could hit the Earth’s surface all over the hemisphere facing the blast.35 If that sounds like a lot, well, it is. These particles would cascade down from the sky and be absorbed by anything out in the open. Given how well human flesh can absorb muons, the astronomers who did the calculation found that the energy absorbed by an unprotected human would be ten times the lethal dose. Hiding won’t help much; muons can penetrate water to depths of more than a mile and also go right into rock down to depths of half a mile! This would therefore affect nearly all life on Earth.

So in reality, ozone depletion wouldn’t be that big a deal. By the time that really became a problem, most of the animals and plants on Earth would be long dead anyway.

That is the nightmare scenario depicted at the beginning of this chapter. However, before you panic, remember: Eta Carinae is almost certainly pointing in the wrong direction. But while we’re on the topic, there is another possible GRB progenitor to consider. Called WR 104, it’s coincidentally about the same distance from us as Eta. It’s a binary star, and one of the stars is a bloated, massive beast near the end of its life. It may blow up as a GRB, and it may be pointed more or less at us, but those are both pretty iffy. The odds are that we’re safe from this monster as well, but it’s worth mentioning.


So we seem to be pretty safe at the moment, which is a good thing. The odds of a nearby GRB at any given time are extremely low . . . but the Earth is old. Is it possible that we got zapped by a GRB in the past?

Statistically speaking, it’s actually quite probable that the Earth was hit by a relatively close GRB beam at some point in the past. While supernovae are common enough, they have to be close to hurt us. GRBs are far rarer, but are damaging at much greater distances. Some studies have shown that one should be near enough to do some ecological damage to the Earth every few hundred million years or so.

It turns out that there may even be evidence for one such event in the Earth’s past. The end of the dinosaurs may be the most famous mass extinction event in history, but it was not the largest. The Ordovician era ended about 440 million years ago, when as much as half of all genera of life on Earth were wiped out. It happened rapidly, and appears to have had two separate extinction events separated by perhaps a million years. The cause has mystified scientists for many years.

Could a GRB have pulled the trigger on this extinction event? There are many tantalizing clues. In a GRB event, you’d expect incoming UV radiation would more profoundly affect animals and plants that lived near the surface of the oceans than it would affect deep-sea creatures, and there is evidence in the fossil record that that is what happened. Trilobites, those curious crablike animals that dominated the oceans of the time, had a larval stage. It appears that at the time of the extinction event, larvae that lived near the surface of the water were more affected than those that lived in deeper water, indicating that whatever caused the sudden die-off may have come from above, from the sky. Moreover, animals that had longer larval periods in their life cycle were also more likely to go extinct than those with shorter larval stages. These are both consistent with a sudden increase in UV radiation that could affect shallow-water regions, but not deep-water. Animals with longer larval stages would absorb more dangerous UV radiation, which would preferentially kill them off.

Interestingly, such trends are not seen in other mass extinctions, indicating that the Ordovician extinction had an unusual cause. GRBs are many things, but “unusual” would be high on that list.

The second Ordovician extinction event has been associated with rapid cooling of the Earth, followed by glaciation. This is also consistent with the effects of a nearby GRB; the cosmic-ray shower and subsequent increase in atmospheric nitrogen dioxide would contribute to a possible global cooling. In fact, some researchers have found that at this time on Earth, a global glaciation could not have occurred without some sort of “forcing event”—that is, some outside mechanism to kick-start it. Perhaps that force came from a GRB.

This evidence is interesting, perhaps even persuasive, but it is not conclusive. More research, as usual, is needed. But it does give one pause to think that an event that occurred thousands of years earlier and trillions of miles away could so profoundly affect life on Earth.


Are GRBs worth worrying about?

One answer is no, because if one goes off there’s nothing we can do about it. And since gamma rays travel at the speed of light—they are light—we will get literally no warning if one is headed our way. So why worry?

On the other hand, it’s quite possible that there is nothing to worry about anyway.

Almost every gamma-ray burst ever seen has come from an incredibly distant galaxy. But in astronomy, distance is the same thing as time: the farther away you look, the farther back in time you are seeing. When we see a GRB explode in a galaxy nine billion light-years away, we’re seeing that galaxy as it was nine billion years ago. GRBs were common in the past, and became less frequent as the Universe aged.

This is significant because galaxies change over time. Early in their lives, they had fewer heavy elements like calcium, iron, and oxygen in them; these elements are created and distributed into the galaxies by supernovae, and that takes time. It turns out that it’s easier for stars with fewer heavy elements to turn into GRBs when they die. Since most massive stars currently being formed have lots of heavy elements in them, thanks to previous generations of supernovae, they are less likely to go GRB.

Furthermore, stars that explode as GRBs need to be rotating rapidly before they collapse, or else the accretion disk that feeds the beams may not form. It turns out that stars with higher abundances of heavy elements tend not to spin so quickly. It’s not because the elements are more massive, however! Heavier elements are better than lighter ones at absorbing the light coming up from the star’s interior. This makes a star with lots of heavy elements in its gas hotter and brighter than a star with fewer heavy elements in it. Because of this, particles on the surface of the star are more easily blown away in a stellar wind—the equivalent of the solar wind, but from a star other than the Sun.

As the particles leave the star, they are swept up by the rotating magnetic field of the star. This acts like a parachute, slowing the star’s spin in turn: imagine holding a plastic bag open and spinning around; as the bag catches the air, your spinning would slow because of the drag. The same thing happens to stars; their spin slows over time as their magnetic field drags through the stellar wind. In fact, this is why the Sun rotates only once a month. It probably spun much faster when it was young, but over billions of years the solar wind dragging through the magnetic field has slowed its rotation.

So stars that have more heavy elements have a stronger stellar wind, and tend to spin more slowly. The converse—stars with fewer heavier elements tend to rotate more rapidly—means that stars that were born earlier in the life of the Universe will make more GRBs than stars born more recently. The upshot of all this is that GRBs from hypernovae—from massive stars exploding—will be more rare today than they were in the distant past.

In other words, you really don’t have to worry too much about them.


So are we safe from this form of destruction, sitting comfortably in our twelve-billion-year-old galaxy with its heavy elements and slowly spinning massive stars?

Maybe. But maybe not. If you recall, there appear to be two different kinds of GRBs, ones that last longer than two seconds, and ones that are shorter. The kind generated in the collapse of a massive star’s core is the long kind of GRB. But what of the short ones?

Two NASA satellites were critical for understanding the short bursts. The High-Energy Transient Explorer-2 (HETE-2) and Swift missions detected dozens of short GRBs. Using these observations, astronomers were able to craft the idea that a short GRB can occur when two dense neutron stars merge. A neutron star forms when the core of a star going supernova is not quite massive enough to form a black hole. In many cases, massive stars form in pairs, with the two stars orbiting one another, and many such high-mass star pairs are seen in our galaxy. Over time, the more massive star will explode, leaving behind a neutron star. Some time later, the other star explodes, also leaving behind a neutron star.

Through many forces, over billions of years, the orbits of the two stars will shrink. The two ultradense objects spiral closer and closer together . . . and then, finally, they will get so close that they literally merge. Their combined mass may be enough to form a black hole, and if enough matter is left over it will form an accretion disk around the hole. At this point, events are similar to what happens in the core of the massive star when it explodes: the accretion disk, tremendous magnetic fields, and powerful gravity of the black hole focus twin beams that explode outward.

Models of these events indicate that the burst of gamma rays would be much shorter in duration than the massive star type of GRB, and would produce higher-energy gamma rays. Both of these predictions fit the observations. There are other models that also fit the observations (such as a black hole-neutron star binary, with similar results), but this is the leading theory.

One major difference between the merging neutron star GRBs and the massive star hypernova GRBs is the time it takes before one can go off: while in modern times we expect to see few if any massive star GRBs, we expect to see plenty of neutron star mergers. It takes billions of years for the orbits of the two neutron stars to decay and cause the stars to merge, and so they should be able to occur today. This may very well be true, but in raw numbers they are less common than their more massive counterparts. This may be due to their uncommon origin—there are plenty more single massive stars that can explode than there are binary massive stars—so it’s difficult to get a handle on how many potential short GRBs there are in our galaxy. There are many neutron star binaries known, all of which could become short, hard GRBs . . . in a few more billion years. None are known that could go off in a century or millennium, or even in the next million years. But unlike massive stars, which are incredibly bright and obvious, binary neutron stars give off very little light and are difficult to detect.


Two neutron stars finally succumb to their mutual gravity after billions of years of orbiting each other. Torn apart, they merge and collapse into a black hole, which announces its birth with a GRB.


It’s unlikely in the extreme that there are any close enough to do us any harm. But it’s not possible to entirely rule them out either.


What we need, as always, are more observations. As the biggest explosions we know of—and probably the biggest explosions the Universe can make—GRBs are of great scientific interest. They tell us so much about how matter and energy act at the extreme limits of physics, how black holes are born and behave, and also about the environment around them. There’s still a lot we don’t understand about GRBs, of course. We’ve come a long way since the Vela satellites; in 2004, NASA launched the Swift satellite—so critical in understanding the origin of the short, hard bursts—which has observed hundreds of GRBs, including the most distant one ever seen at 12.8 billion light-years away. Swift’s observations have allowed keen insight into both long and short bursts, adding much-needed data to the theoretical models.

As we learn more about GRBs, we’ll be better able to assess the danger from them, including how they may have affected life on Earth in the past. While there’s probably nothing we can do if one goes off—and it’s incredibly unlikely it will happen at all—it’s always better to have a good handle on the situation.

So should you worry? I’m asked this all the time, and I have a simple answer: I have known people who’ve been killed in all sorts of unlikely ways, including car crashes and one who was hit by lightning. How many people do you know who have been killed by a gamma-ray burst?