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


What, Me Worry?


Good. It’s been quite a ride, but I hope that while you read this book you weren’t vaporized, crushed, irradiated, flung out into deep space, spaghettified, or had any of your protons decay.

We’ve covered a lot of scary ground (not to mention space and time). It seems as if the whole cosmos is trying to snuff us out. In a sense, it is—there’s danger aplenty in the Universe—but we have to take a practical view here. We have to appreciate the vastness of space and time, and our ability to manipulate events around us.

Asteroid impacts provide an excellent example of practical versus theoretical danger. They have done ferocious damage to the Earth in the past, and our relatively fragile economic system could be destroyed by far smaller impacts than the one that did in the dinosaurs. To understand the actual danger, we have to balance the idea that they don’t happen very often with the knowledge that they do in fact sometimes happen.

In your daily life, this may not present much of a problem (except for when you lie awake at night and your brain, unfettered by the common sense available during the day, is wondering if tonight is the night). But in the case of impacts, we can actually prevent them from happening. It would cost a lot of money (hundreds of millions to test various ideas, and hundreds of millions more to implement them), time, and effort. Can we afford to start worrying about this now? Can we afford to wait?

Scientists are asking these questions, because they have to ask Congress for a lot of money to be able to do anything about them. And what Congress decides influences your money (other governments may become involved as well). To make sure we get the right answers, we really have to understand the issues. Astronomers are looking at a theoretical danger and finding a practical action to take against it.

I hope this book has cleared some of that up. I studied gamma-ray bursts for many years, and I personally am not at all worried about them. Nor do I fret over the death of the Sun, the eventual decay of the Universe, or a black hole slipping through the solar system and snacking on the odd planet or two. That’s because I understand the odds of these events actually taking place, and they are vanishingly small. There’s no need to worry about them, whereas an asteroid strike or a particularly nasty solar coronal mass ejection can do a lot of damage. Even then we have it within our power to minimize their impact.

In an effort to make all this a little easier to digest, here is a table that gives the odds, the potential damage, and our ability to prevent the disasters described in this book. What you’ll see is that chapters 1 and 2—asteroid impacts and solar events—cover the only two events we can do anything about. While they may not happen tomorrow, they will happen, and it’s in our best long-run interest to do something about them.

Following the table is a description of how I came to those conclusions. Bear in mind that astronomy is a field of science, and that means that things change as better data and better ideas come along. Don’t assume any of this is written in stone.

Of course, in 1040 years or so, even stone will be long gone.





Asteroid impact

Local for a small rock, global for a big one

1 in 700,000

Almost 100% preventable


Identify potential impactors, then blow them up or push them out of the way

Solar flare/CME

Collapse of power grid, potential ozone depletion


Not preventable, but mitigable


Build robust power grids


Ozone depletion, radiation

1 in 10,000,000

Not preventable

Gamma-ray burst

Ozone depletion, radiation, setting planet on fire

1 in 14,000,000

Not preventable

Black hole

Destruction of Earth

1 in 1,000,000,000,000

Not preventable

Alien attack

Humanity wiped out by aliens; space bugs give us runny noses


Preventable, assuming we colonize the galaxy first; otherwise, forget it

Death of the Sun

Earth cooked to a crisp


Not preventable, but we have a long time to go yet

Galactic doom

Ice ages, radiation, eaten by supermassive black hole


Not preventable, but again, none of these will happen on a human time scale

Death of the Universe

Decay of all matter, collapse of false vacuum


Not preventable, but dwarfs any time scale we can imagine

 Fatilities are very unlikely from a solar event, but they can still cause extensive damage.

 These events all take billions of years (at least!) to unfold, so the chances of their happening during your lifetime are zero, but they are inevitable over longer times.


Of all the woes facing us from space, this is the one that is nearly 100 percent preventable. Scientists and engineers have viable ideas on how to stop big impacts—ones big enough to do significant damage. The real problem is in finding these objects in time, and even that is improving as more surveys of the sky find more objects. However, it’s physically impossible to find every single potential impactor; some come from so far out in the solar system that we simply cannot see them until they are on the way here.

So no matter how many we find, there is still some risk. The American astronomer Alan Harris has composed a table of risks from impacts, and the results are surprising: if you live in the United States, the overall risk of dying from an impact in your lifetime is only 1 in 700,000, somewhat less than being killed in a fireworks accident, but still more probable than being killed on an amusement park ride or by an act of terrorism.

Interestingly, before we started surveying the skies, those odds were calculated to be about 1 in 70,000, which was ten times higher. Why? The old numbers were based on statistics using previous impacts as a basis, and the statistics on those are a bit spotty. Now that we have surveys peering at the skies, most of the big impactors have been found. Since these tend to kill lots of people (mass extinctions, anyone?), eliminating them from the known list of potential impactors dropped the odds of getting killed significantly. Actually, the odds haven’t really changed; we just know how to calculate them better. It’s like the difference between trying to figure out your odds of winning a poker hand before and after the cards are dealt. Once you see those cards, your odds are far easier to calculate.

So how’s that for a return on investment? This makes it very clear why just getting out and finding these things is so important. We spend billions on terrorism, but the risk from an asteroid impact is actually higher.

Mind you, this does not include comets, which can come from deep space and have orbits that are not easy to determine in advance.


These are tricky things to evaluate. For one thing, big solar blasts are rare, making statistics spotty. Also, the effect they have on Earth depends on many factors, including time of year: during the peak summer and winter months the power grid on Earth is heavily loaded with electricity and more susceptible to flares and CMEs, while in the spring and fall months the grids are more robust.

Perhaps the most important issue is that few or no deaths will result directly from a solar event. There are some direct effects like exposure to radiation for long-duration or polar airline flights, but these are difficult to pin down. Astronauts are at risk, but none has ever been killed by a solar event, though this will certainly be a problem for long-duration lunar and Martian voyages.

The large effects are all indirect: loss of power, loss of communication, and so on. These can cause deaths, of course: heatstroke in summer, hypothermia in winter. But the relationship is difficult to determine.

While a whopping big solar event can seriously impair or destroy a nation’s infrastructure and economy, in general it will not directly cause deaths. So we have to rate this a zero for human fatality, but with an asterisk as a nod to the destructive power it has in other ways.


Supernovae happen about once per century in any given galaxy. But galaxies are huge, and the damage from an exploding star is limited by proximity: it needs to go off less than about 25 light-years away to impart significant damage to Earth’s ozone layer. This only happens about once every 700 million years or so. Assuming the event would cause a mass extinction, killing everyone on Earth, the odds of your specifically dying from one over your lifetime are therefore about 1 in 10 million.

Gamma-ray bursts are a little different: they inflict far more damage, and are dangerous from distances of more than 7,000 light-years. But they are also rarer than regular supernovae and also pickier about targets: we have to be in the path of the relatively narrow beam to get hurt. All in all, these effects cancel out, leaving us with just slightly lower odds as being killed by a supernova: for GRBs, the odds are 1 in 14 million.

In both cases, you’re literally more likely to be killed by a shark.


The odds that a black hole will get close enough to the Sun to do any real damage are very low. A normal star passes within about three light-years of the Sun about every 100,000 years, and there are something like 20,000 normal stars in the Milky Way for every black hole (assuming 200 billion stars and 10 million black holes, the latter of which is probably also an overestimate). This means a black hole gets within three light-years of the Sun, statistically speaking, every two billion years, or three times over the solar system’s current lifetime.

From that distance, the black hole can’t do very much to us; remember, from a long way off its gravity is no different from that of a normal object. Even a black hole with ten times the Sun’s mass won’t do much from that distance.

If the black hole is actively eating, then it will emit X-rays that can hurt us from a greater distance. But three light-years is probably a safe distance; from that far away the X-ray emission powered by a black hole is actually far less than the X-rays you’d expect from a solar flare. Even a closer passage by about a light-year would only cause minimal damage to the ozone layer, and that kind of close shave would be rarer still, both because the odds of one getting that close are small and because active black holes are far less common than quiescent ones.

To be as bright as a big solar flare in X-rays, a typical black hole would have to get within about 150 billion miles of the Earth. This is an extremely unlikely event, happening about once every 100 trillion years. Obviously, this is not likely to have ever happened in the history of the solar system, nor is it ever likely to.

So, your lifetime odds of Death by Black Hole are about one in a trillion.


So how do you calculate the odds of being attacked by aliens? We could use the Drake Equation outlined in chapter 6, but as we saw, the output of that equation (the number of advanced civilizations in the galaxy) is not well constrained, as scientists say: it could be one (us), or it could be millions.

Even if the galaxy is buzzing with life, it’s nearly impossible to quantify how many of these aliens would be hostile, and how many would come here specifically to wipe us out. It only takes one race, of course, but what are the odds?

I argue that we can’t know. The best we can do is say that if a race wants to wipe out all life, everywhere, then they must visit every planet not only with life on it but also capable of having life on it. Why take chances?

We have enjoyed a nearly uninterrupted existence of life on Earth for over three billion years. During that time, we have not seen a single alien intelligence sterilize the planet. That puts a lower limit on the odds of alien-induced extinction to one in three billion. Given that a species can explore the entire galaxy in a small amount of time (a few million years) compared to the amount of time life has existed (billions of years), the odds are for all practical purposes zero.

As I pointed out in the chapter, it’s fun to think about, and it makes a great bedtime story, but in the real world I think estimating the probability as equal to zero is close enough.

Also, we cannot really assign odds to a space virus or bacterium turning us into goo anytime soon either. We have not seen a single example of such a beast, though of course we haven’t been looking all that long. Still, because of the lack of data, we have a true unknown here. Personally, using just my guts and hunches, I would put the odds in the range of billions-to-one against, but that is not very scientific. So to be truly skeptical, as any real scientist is, I will have to leave this blank, and hope that advances in astrobiology will allow us to make some safe estimates of the odds sometime soon.


Of all the ways the Cosmic Grim Reaper can pay us a visit, just this one and one other (Death by End of the Universe) have odds of 100 percent. You just have to wait long enough!

The Sun will begin to die in six billion years, and unless we develop technology indistinguishable from magic, there’s nothing we can do about it (short of migrating to another star). But this is a funny case: the odds of it happening in your lifetime are exactly zero, but if you live for six billion years, the odds jump to 100 percent. However, this being an almost entirely predictable event, I’m going to have to go with the lower-limit case.


Just living in the Milky Way Galaxy is dangerous, but how dangerous? Actually, it’s not hard to get some statistics. For example, magnetars—supermagnetized neutron stars—can cause damage from several thousands of light-years away. Like GRBs they are created in supernova explosions, and are also quite rare. It’s reasonable then to assign the same death-inducing stats to them as for GRBs, or about 1 in 10-20 million or so. It’s worth remembering that no magnetars are known close enough to hurt us, and it’s likely that if any were that close we’d see them.

Plowing into a dense dust cloud is a rare event as well, happening on the order of once every billion years. This happens most often when the Sun enters a spiral arm of the galaxy. The Sun is located about 6,400 light-years from the nearest spiral arm, and so even if we were headed straight at it, it would be another 10 million years or so before we hit it. Therefore the odds of hitting a dense dust cloud are extremely low, but go up substantially in a few million years. Statistically speaking then, the odds of dying from entering a dust cloud are essentially zero right now, and will be for quite some time.

The same is true for solar oscillations out of the plane of the Milky Way; we are currently heading up into the danger zone, but won’t be there for at least another 20 million years. So again, currently the chance of getting killed by a flood of intergalactic cosmic rays is zero.

Finally, the collision with the Andromeda galaxy has several avenues for killing us: we could get dropped into the galactic center whereupon we get eaten by the supermassive black hole dwelling there, or we could get close enough that high-energy radiation from said black hole may do us in, or the (possible) vast amounts of new stars born in a cosmic baby boom could produce a nearby supernova (making the “boom” particularly appropriate). While any of these are dangerous, this collision is another event that cannot happen for several billion more years, so the odds of its killing you are currently zero.


Like the Death by Solar Death, this is inevitable. Take your pick: do you care more about the Milky Way colliding with the Andromeda galaxy (which will happen before the Sun dies), or the eventual decay of protons, or the evaporation of black holes, or the total annihilation of the Universe by quantum collapse to the true vacuum state?

But the time scales! Billions of years, nonillions of years, vigintillions 135 of years . . . if you wait long enough, any number of bad things will happen, and any one of them is pretty final. Again, though, like the death of the Sun, most of these events are not random, but happen at a certain pace. Protons almost certainly will decay, but not for more than 1033 years at least. So we’re safe over our lifetimes from these disasters.

There is some finite chance at any one time that we’ll collapse to the true vacuum state, but since this is entirely theoretical, I don’t think any number has ever been assigned to it. The odds are so low that even unlikely events—like getting eaten by a rogue black hole—are fantastically more likely to happen long before the Universe decides to erase all of space and time. I would hang a big zero on this one as well.


The last few events are interesting: they won’t happen in your lifetime, but they will eventually happen. This skews the statistics, since calculating a lifetime risk from such an event doesn’t really work. The death of the Sun won’t kill you—unless you plan to see the ripe old age of six billion—but it’s out there, someday.

This gives us some insight on how much we should worry about these astronomical harbingers of doom. The only two we can do anything about (asteroids and solar events) are things we should do something about. The cost isn’t that big a deal in the long run, and the savings are enormous.

As for the rest, well, you shouldn’t fret too much. They are certainly fun to think about, and by studying them we learn more about the awesome nature of the Universe, its scale and its capabilities. One of the biggest thrills of science, and its ultimate goal, is understanding. Maybe the price to pay for that is a little bit of fear, but I think in this case it’s only a very tiny price. A bargain, in fact.

Sure, the Universe is scary. But it’s also beautiful. A supernova can kill us, but the expanding wave of gas creates one of the most intricate and delicate objects we can see. Colliding galaxies dance to a tune millions of years long, creating graceful and elegant structures for us to ponder. The Sun may evolve into a red giant, but the planetary nebula it may eventually become will shine green, red, and blue, and will grace the skies of countless planets for a dozen millennia.

And there’s more to this story as well. Supernovae create heavy elements and then seed nearby gas clouds with them. These elements are necessary for planets to form and for the rise of life. Colliding galaxies make new stars, new chances for life. Even the death of the Sun means its material is returned to space where it may get used again. Life, death, and life: this is the real story of the Universe.

Most of the Universe is lethal, but our little section of it is pretty cozy. The cosmos takes away, but it also provides. So go outside, enjoy a sunny day or a star-filled night. There is danger to avoid, but also beauty to behold, and you understand a little bit more about that sky now.

And understanding is always good.