Following recent events in/over Russia, I thought this might be of interest.
It is estimated that 2000 asteroids each more than a kilometre across are in Earth-crossing orbits, accompanied by a million objects between 0.1 km and 1 km across and at least 500 million smaller objects (down to 10 metres across).
So the threat is certainly real. But how do you measure such threats? How do you decide what response is cost effective? The best analysis, so far, of the risks involved has been carried out by Clark Chapman, of the Planetary Science Institute in Tucson, Arizona, and David Morrison, of the NASA Ames Research Center at Moffett Field, in California. It was published in Nature on 6 January 1994 (volume 367, page 33), and has since provided the basis for all attempts at deciding how best to combat the threat from space.
Chapman and Morrison drew on several studies carried out during the 1980s and early 1990s, most of which had been inspired, directly or indirectly, by the realization that the Earth had suffered a major cosmic impact at the end of the Cretaceous. The bottom line of their calculation is that there is a 1 in 10000 chance that an object bigger than 2 km across will strike the Earth in the next century, causing major disruption of the world’s ecosystem and killing a large proportion of the human population of the planet. The risk is very small, but the number of people potentially affected is very large, and in statistical terms that puts the risk to any individual human being surprisingly high up the scale of things to worry about, using the actuarial approach of life insurance companies.
For any present-day citizen of the USA, the chance of being killed by the after effects of a cosmic impact are about 1 in 20000. This is bigger than the risk of being killed in any other kind of natural disaster, and is exactly equal to the risk of being killed in an airplane accident on a commercial passenger flight, something that very many people take seriously as a risk to worry about. It compares favourably with the chance of being killed in a motor vehicle accident of 1 in 100, and with the danger of accidental electrocution of 1 in 5000, but unfavourably with the chance of being killed by a tornado (1 in 60000), and with the risk of death by food poisoning (just 1 in 3 million). THe risk of death from cosmic impact is greater than the combined risk of death from tornadoes, hurricanes, earthquakes, forest fires and volcanic eruptions.
The point is, as Chapman and Morrison put it, that “impacts are an extreme case of a low-probability/high-consequence hazard”. In his book Exploring Planetary Worlds, Morrison spells out how the odds are calculated. If an impact big enough to cause a global crop failure happens once every 375,000 years, and you have a 1 in 4 chance of dying if such an impact occurs in your lifetime, then the annual risk of death from this cause is 1 in (3 x 375,000), which is 1 in 1.5 million. If the average human lifetime is 75 years, the overall chance of being killed in this way is 75 times greater — 75 in 1.5 million, or 1 in 20000.
The chance of anyone at all being affected in this way is indeed small; but if one person is affected by a major cosmic impact then literally billions of other people will suffer as well. This changes the way you should think about the risk. For example, many cautious people buy cheap “one off” life insurance policies every time they fly, so that if that 1 in 20000 risk comes up, their dependents will, at least, have financial security. But there is no point in taking out comparable life insurance against the risk of death in a major cosmic impact, because if that event happens you probably won’t leave any surviving dependents, and in any case there will be no insurance company left to pay up. The only kind of “life insurance” against a major cosmic impact is to take global precautions, and the cost of those precautions should be seen in exactly those terms — as life insurance for billions of people. Divide the cost of the precautions by the number of people in the world (or even in the USA and Europe) and you have a sensible figure for the cost of the insurance in personal terms. This makes current proposals put forward by astronomers to keep a proper watch out in our neighbourhood of space look very cost effective — certainly better value for money than those one-off airline flight insurance premiums.
The chance of any one person being killed by a major comet impact in any one year is only about 1 in 2 million, but that still means that on average 2,700 people are killed by large comets each year, 390 of them in the developed world alone. Putting this in a slightly different perspective, the cost of road safety measures in Britain in the mid-1990s works out at just over ￡800,000 for each life that is saved. If we cared as much about the comet and asteroid hazard as we did about road safety, that would imply that the developed world ought to be spending nearly $500 million a year to prevent cosmic impacts. The fact that nobody is killed by such an impact for many years, and then many people are killed at once, does not alter the argument. There are long stretches of road where nobody is killed for many years, and no big accidents occur; but that is not a good argument for saving costs by not bothering to maintain safety barriers and decent lighting on those stretches of road on a year by year basis.
Adapted from Fire on Earth, by John Gribbin & Mary Gribbin. Now sadly out of print.