BBC Future

The Real Risk

How likely am I to be hit by an asteroid?

About the author

David Spiegelhalter is Winton Professor of the Public Understanding of Risk at the University of Cambridge, UK. His professional background is in medical statistics. He is a frequent contributor to the media, and has been an advisor on risk issues as diverse as breast implants, volcanic ash, clinical trials, surgical mortality, and the inquiry into mass murderer Harold Shipman. He is a Fellow of the Royal Society and was awarded an OBE in 2006. You can see more of David’s explorations of risk at the Understanding Uncertainty website, or follow him on Twitter at @undunc

(Copyright: Science Photo Library)

(Copyright: Science Photo Library)

A report suggests that there should be 91 deaths every year from asteroid strikes, but what are the chances of that actually happening?

Buy insurance. Tick. Health check. Tick. Drive sensibly. Tick. As a general rule, we humans like to control our lives. But let’s face it, all of this caution is a complete waste of time if a huge rock from space has your name on it.

Take the recent ‘near-miss’ by the poetically-named asteroid 2012 BX34, which was only discovered two days before it sailed past within 40,000 miles (60,000km) of Earth. What if it had been heading straight for us?

A wonderful report from the US National Research Council (NRC) says that on average there should be 91 deaths per year from asteroid strikes - a remarkably precise figure and one that deserves some digging.

Try to think of when you last heard about an asteroid striking the earth. There really aren’t that many of them, or at least that many that are noticed or reported in newspapers.

One of the last significant impacts occurred on 30 June 1908, when an asteroid or comet exploded 6.2 miles (10km) above a secluded forest in Tunguska, Siberia, flattening trees over an area of 625 sq miles (1600 sq km), which surprisingly few people cared about at the time due to the remoteness of the region and the fact that there seem to have been no casualties.

Calculations suggest that if it had landed 4 hours and 47 minutes later, it would have hit St Petersburg(1), in which case people might have cared a lot, particularly as it was rather a delicate time in Russian history. According to estimates, such an airburst occurring over New York would cost $1.19tn to insurers in property damage, not to mention causing approximately 3.2 million fatalities and 3.76 million injuries.

But that has not occurred. In fact, there are surprisingly few reports of fatalities involving asteroids. A few cars in the United States have been damaged and there was a case of a cow being killed in Valera, Venezuela in 1972 – the unfortunate animal was duly eaten and bits of the meteorite were later sold to collectors. A home outside Paris was also recently hit by an egg-shaped meteorite, but the appropriately-named Comette family were away at the time.

So how can the NRC be so precise? Well, to understand we need to understand how astronomers and statisticians think about these risks.

As these threats come in all shapes and sizes astronomers have established a classification system to help gauge the potential level of risk. If asteroids or comets come within one-third of the distance from the Earth to the Sun – just over 30 million miles (48 million km) – they are labelled as Near Earth Objects (NEOs). ‘Near’ is clearly a relative term, though this does show how comparatively close asteroid 2012 BX34 came. Fortunately, there are observatories watching over us and Nasa’s Near Object Program is keeping count of what is passing through our local area. By December 2011, over 8,500 NEOs had been found and named, with around 500 being added to the list every year.

If an object is found that is at least 480ft (150m) across, and which will pass closer than 20 times the distance between the moon and the Earth (around five million miles, or eight million km), it earns itself the status of Potentially Hazardous Asteroid (PHA). If anything this big were to hit the Earth, the consequences would be serious. So far 1,271 of these have been found, of which 151 are of more than 0.6 miles (1km) in diameter, a size that could be globally catastrophic. 

Working out the chances of an Earth-asteroid collision and the damage it would cause is not like an insurer dealing with a collision between two cars: there is almost no direct historical data, so astronomers create equations relating to the size of an asteroid, how many there are around, how often they might hit the Earth and what the explosive force of any impact would be. These estimates are continually being revised and are subject to some esoteric disputes.

Asteroids in the 16-32 feet (5-10m) diameter range find their way to Earth around once per year, releasing energy equivalent to approximately 15,000 tons of TNT when they explode: about the same as the bomb detonated over Hiroshima. Most go unseen and unrecorded. An airburst of an 80ft (25m) asteroid would release energy equivalent to around one million tons of TNT, or one megaton (Mt), equivalent to around 65 Hiroshima bombs. That same NRC report estimates an average of 200 years between such impacts.

The Tunguska devastation in Siberia is thought to have been caused by an asteroid of around 160ft (50m) diameter exploding 6 miles (10km) up, releasing 10Mt of energy. The NRC estimates an average interval of 2,000 years between events of this size, although other studies have suggested that Tunguska-like events could be caused by objects as small as 100ft (30m), which would equate to nearly a 50:50 chance of such an event occurring in a given lifetime(2).

Bigger asteroids start being considered as ‘continental-scale events’, although it is tricky to predict what damage such a major impact would cause. Due to the differing coverage of the Earth’s surface, there is a 70% chance that any such impact would hit the ocean, and models have suggested a 1,300ft (400m) asteroid could cause a tsunami 650ft (200m) high(3), although with regards to possible consequences, there is great uncertainty about whether such a wave would break on the continental shelf, whether the population could evacuate, and so on.

Mass extinctions

Nasa estimate that there are currently around 900 Near Earth Objects of over 0.6 miles (1km) across, and if one hit us it would release around 100,000Mt of energy. Such an impact would almost certainly be globally catastrophic. Thankfully, such strikes should only be expected every 700,000 years, but even bigger collisions have occurred. Considerably more than just one cow perished when a 6 mile (10km) wide, 100,000,000Mt lump hit the Yucatan peninsula in modern-day Mexico 65 million years ago: this impact is credited with changing the climate and wiping out the dinosaurs. But such mass extinction events are only estimated to come along every 100 million years or so. So not much need to worry there, then.

In terms of future threats, the current catalogue indicates no serious risk from asteroids that we know of. The biggest known danger Nasa can point to is a 1 in 600 chance of a collision with the 460ft (140m) 2011-AG5 asteroid sometime in the 2040s. But most asteroids less than 1,640ft (500m) across remain undiscovered, and although these would be unlikely to cause a global catastrophe, they could certainly take us by surprise. 2008 TC3 was 6-16ft (2-5m) across and weighed around 80 tons when it exploded over the Sudanese desert on 7 October 2008. It was the first asteroid to ever be detected before impact, but it was only picked up 19 hours prior to its destruction, and one can imagine the crisis if its predicted path had passed over a big city. It exploded with a force of around 2,000 tons of TNT and 10kg of fragments were picked up afterwards – though thankfully nobody was hurt.

Is it worth getting up?

So, once again, back to that figure of 91 deaths.

This now begins to make more sense when you realise that it is an average taken over millennia, in which almost all years show no deaths, but a few isolated events cause massive casualties. In fact, the figure of 91 deaths is almost evenly balanced between more frequent, smaller-scale impacts, and statistically very infrequent globally catastrophic impacts.

So, I know what you are now thinking: what is the risk of being killed by an asteroid in my bed – or anywhere else for that matter – in my lifetime?

As I said in my first column, statisticians like to define such risk events in terms of micromorts: that is, a one-in-a-million chance of dying.  Since there are 7 billion people on Earth and 91 people are expected to die every year from an asteroid impact, this works out at 1/77 micromorts per person per year – about the same risk as a 3 mile (5 km) car journey, but much less risky than riding a motorcycle for 60 miles (97km) in the UK (around 10 micromorts).

Assuming a lifetime of around 77 years, this comes to the delightfully round number of one micromort per lifetime from asteroids.  Which, as risks go, is hardly the end of the world.

References 

1. Woo G. Calculating Catastrophe. 1st ed. Imperial College Press; 2011.  368 p.

2. Boslough MBE, Crawford DA. Low-altitude airbursts and the impact threat. International Journal of Impact Engineering. (2008) 35: 1441–1448.

3. Ward SN, Asphaug E. Asteroid Impact Tsunami: A Probabilistic Hazard Assessment. Icarus. (2000) 145:64–78.