Could an asteroid wipe out human civilisation like it may have eliminated the dinosaurs? Big asteroids come along extremely rarely and our monitoring systems are effective and well funded.
On Friday the 15th, February 2013, there was an eerie coincidence. NASA had been tracking a 45-metre-wide space rock, Asteroid 2012 DA14, for almost exactly a year.
2012 DA14 had been spotted when it passed moderately close to the Earth in the previous February – moderately close being about two million kilometres away, roughly seven times the distance from the Earth to the Moon. But it had a very similar orbital period to the Earth’s own, 368 days to our 365, which meant that it would come fairly close every year.
NASA was able to predict that on its next pass it would come much closer – about 28,000 kilometres away, barely twice the diameter of the Earth, and within the orbits of geosynchronous satellites. It would be travelling, relative to Earth, at 7.8km/s, or about 28,000 kilometres an hour. That pass was due on 15 February 2013 and would be the closest ever recorded pass of an asteroid that large.
Newtonian physics being as reliable as they are, 2012 DA14 did indeed pass close to the Earth that day. But something else happened. At 3:20am British time – 9:20am locally – there was a huge explosion in the sky over Chelyabinsk, in central southern Russia. A thousand people were injured, mainly by flying glass from broken windows. Dashboard cameras and phones captured hundreds of startling videos.
The Chelyabinsk meteor was established to be a 20-metre-wide rock, estimated to weigh about 11,000 tonnes and travelling even faster, relative to the Earth, than 2012 DA14 – about 19km/s, or 60,000km/h. It hit the Earth’s atmosphere at a shallow angle; as it did so, it decelerated suddenly, compressing the gas in front of it and causing rapid heating. Thirteen seconds after entering the atmosphere it exploded with the force of about 30 Hiroshima bombs.
Chelyabinsk and 2012 DA14 were entirely unrelated. The Chelyabinsk meteor had travelled undetected, partly because it was smaller and partly because it was coming from the direction of the Sun. But the coincidence called much more attention to the risk of “near-Earth objects” colliding with the Earth. Later that year, the US House of Representatives Committee on Science, Space and Technology met to review efforts to “track and mitigate asteroids and meteors”.
But how likely is a major, devastating impact? And what can we do, and what are we doing, about it?
Different sizes of asteroid
Chelyabinsk was the largest object to collide with Earth in more than a century, since a large object, perhaps a comet, had exploded in 1908. That, too, had been in luckless Russia, in the remote region of Tunguska. Estimates of its size vary widely, from 60 metres in diameter to 1,000 metres.
Had the Tunguska object hit a city, it would have been devastating: it exploded with a force of between 3 and 20 megatons, equivalent to a large nuclear bomb. Similarly, NASA estimates that 2012 DA14, had it hit, would have had an explosive force of around 2.4 megatons. If it had exploded in the air above Charing Cross, central London, it would have damaged buildings in St Albans, caused third-degree burns to people as far away as Essex and Surrey and (depending on the height at which it exploded) vaporised everything from Buckingham Palace to Vauxhall, probably killing many millions.
But DA14 was relatively small, as asteroids go. In Earth’s history, much larger rocks have hit us. The most obvious example is the Chicxulub object, which crashed into the shallow seas in what is now Mexico, 66 million years ago. It was about 10 kilometres across, with an explosion of about 100 million megatons; the dust clouds it pushed into the atmosphere killed plant life around the world and led to the extinction of all dinosaurs except the ancestors of birds. The discovery of the Chicxulub crater prompted Congress to direct NASA to find and track at least 90% of near-Earth objects more than one kilometre across.
The destructive energy of an asteroid is proportional to the cube of its diameter. A 100-metre-wide asteroid will explode with a force 1,000 times greater than that of a 10-metre one, assuming that they are made of the same materials and travelling at the same speed and angle.
So a 10-metre-wide rock might cause damage roughly comparable to that of Chelyabinsk: smashed windows, injuries from flying glass. A 100-metre one might be more like a nuclear strike; a kilometre-wide rock would make a crater 15 kilometres across and would essentially remove a city from existence, should it hit one. Buildings 150 kilometres away would be damaged by the pressure wave and earthquakes; people at that distance would suffer extensive third-degree burns.
But a 10-kilometre-wide asteroid would be different. In his book The Precipice, about existential risks, Dr Toby Ord, a researcher at the Future of Humanity Institute, writes that asteroids of that size and above “threaten mass extinction. It is possible that humans would survive the cataclysm, but there is clearly a serious risk of our extinction. Last time all land-based vertebrates weighing more than five kilograms were killed.”
How likely are we to face each of these types of impact in a given century?
The chances of an asteroid strike
We can be relatively sure that major, civilisation-ending asteroid impacts are rare, simply by virtue of the fact that we have a civilisation. If the chance of such a strike was even 1% per century, then the chance of Homo sapiens surviving the 2,000 centuries that it has would be around 0.0000002%. In fact, we can use equations like that to put some plausible upper limits on the rate of all natural existential threats, including not just asteroids but supervolcanoes and gamma-ray bursts and anything else that could take us out without our being involved at all.
But smaller impacts are much more common: analysis of the size of craters on the moon shows that the frequency of impacts decreases with the cube of the diameter of the asteroid. The Earth Impact Effects Program, set up by Imperial College London and Purdue University, says that the number of near-Earth objects above a given length, L, in kilometres can be calculated with the equation N(>L) ≈ 1,148L-2.354, and that for any given near-Earth object, there’s a ~1.6 × 10-9 chance of hitting the Earth in any given year.
To go back to our 10-metre, 100-metre, 1-kilometre and 10-kilometre space rocks earlier, using that equation we would expect to, on average, be hit by a 10-metre asteroid every 11 years or so; a 100-metre one about once every century and a half; and a kilometre-wide one about every two millennia. A 10-kilometre space rock, on the other hand, should arrive only once every million years. (All these are, of course, averages; the impacts do not come at regular intervals but randomly. A better way of saying it would be that you’d expect about nine 10-metre rocks per century.)
| Asteroid Size (diameter) | Estimated Direct Impact | Estimated Freqeuncy (years) |
| 10 metres | Smashed windows, injuries from flying glass | 11 |
| 100 metres | Similar to a nuclear strike | 150 |
| 1 kilometre | Destruction of an entire city | 2,000 |
| 10 kilometres | Threat of mass extinction | 1,000,000 |
These numbers overstate the chance of a disaster caused by an asteroid impact. Large asteroids, those several kilometres across, could cause huge devastation wherever they land – a Chicxulub-scale object landing in the middle of the ocean might not cause huge dust clouds to block out the sun, but it would create enormous tsunamis that would wreak havoc along coastlines thousands of miles away. But smaller ones will only be of real danger if they land on or very near a city.
John P Holdren, director of the Office of Science and Technology Policy under the Obama administration, told the 2013 House committee looking into asteroid risks that while events as large as Tunguska might happen about once in every thousand years, the chance of widespread loss of life is much smaller “because land covers only 30% of the area of the Earth, and urbanised areas cover only two to three percent of the land area”.
What are we doing to prevent it?
It’s worth noting that all the above numbers are the likelihood in any given century or year. So in a randomly selected century, there’s about a 2% chance of a devastating kilometre-wide asteroid impact, and about a one in 10,000 chance of a Chicxulub-event disaster.
But this isn’t a randomly selected century. The risk of a devastating asteroid impact has been written about in science fiction for at least 100 years; since 1981, and the discovery of the Chicxulub crater, it has moved from the realm of plausible fiction and into that of actual Earth history. Accordingly, people have been doing something about it.
Since 1994, NASA and other space agencies have been actively looking for large near-Earth objects, with the goal of discovering at least 90% of all of those greater than one kilometre in diameter. They achieved this goal in 2011, and now believe they have discovered at least 95% of asteroids of that size that approach Earth – 888 of them have been detected, and those that remain will disproportionately be from the smaller end of that scale, since they are both more numerous and harder to detect. NASA has since been looking for asteroids larger than 140 metres with the goal of finding more than 90% of them. So far, around 10,000 objects between 100 and 1,000 metres in diameter have been detected.
Taking that into account, the chance of a Chicxulub-level event in the next 100 years is vanishingly tiny – about one in 150 million, according to Ord. Even a smaller but still devastating impact of an asteroid between one and 10 kilometres across has around a one in 120,000 chance per century. Since the discovery of the Chicxulub crater, humanity – notably the US government – has made great strides in addressing the risk and identifying dangerous asteroids. There are several asteroid-spotting organisations, including the UN-backed International Asteroid Warning Network and the Spaceguard Centre. NASA’s “planetary defence budget” for spotting asteroids has gone up from $4 million in 2004 to $150 million in 2019.
Should we do more?
Probably not.
Our asteroid-spotting capabilities are excellent and well-funded, and the risk – already small – is now vanishingly so. The only question is whether we would be able to divert an asteroid if one should arise.
Ord says that there are several methods which could be used to do so, if the situation arises. Kinetic impacts – that is, firing solid objects into the approaching rock – and nuclear weapons are two of the examples he gives.
But it’s not clear that it’s a good idea to get those systems ready in advance. Major impacts are unlikely. And any technology that could be used to direct an asteroid away from Earth could equally be used to direct one towards Earth; it is very unlikely, but then so are natural impacts, and it may be that the chance of a deliberate impact, in war or terrorism, is more probable than a natural one. This “deflection dilemma” is an ongoing debate. Certainly, developing the technologies would be a large expense for an at best uncertain gain.
That said, technologies for diverting asteroids may become useful in their own right, if we start to mine asteroids for rare earth metals in the coming decades – but developing them specifically for planetary defence seems unnecessary and possibly even counterproductive with no real threats on the horizon.
Chelyabinsk-scale events are harder to predict: smaller asteroids are harder to spot. But while Chelyabinsk itself was terrifying, it was not overly ruinous: a thousand people suffered (mostly minor) injuries, and it caused property damage worth around $33 million. If asteroids of that scale are spotted ahead of impact, it may be worth mitigating the damage caused with evacuation or shelter-in-place orders, rather than attempting to divert the course of the asteroid.
If we detect a large and dangerous asteroid on a probable collision course with Earth, it is likely that we will do so with years to spare, since the Earth’s early-warning system is now excellent. Preparing a mission to divert it – with robot-delivered nuclear weapons or kinetic railguns or whatever – should not be beyond our abilities. Even 2012 DA14, which was relatively small by the standards of asteroids, was known about a year before it arrived. Unlike several other problems, humanity has the asteroid problem largely under control.