January 1994 (vol. 10, #2) 1601 N Tucson Blvd #9, Tucson AZ 85716 c 1994 Physicians for Civil Defense


This week's issue of Nature brings a detailed analysis of the risk of a cosmic impact severe enough to cause global catastrophe (Chapman and Morrison, 1/6/94). Although a subject of science fiction (e.g. Lucifer's Hammer by Larry Niven and Jerry Pournelle and The Hammer of God by Arthur Clarke), the threat is not merely hypothetical. Over 140 terrestrial impact scars have been identified. In 1908, an impact with a comet or asteroid caused an area of devastation of 1000 sq km in the Tunguska River region of Siberia. A prehistoric collision with a 10-km asteroid may have caused mass extinctions. A bolide (explosion of a meteor) with the energy of the Hiroshima bomb occurs about once a year, generally at an altitude so high that the shock wave does not reach the ground. The events have been observed by surveillance satellites.

The explosion of a meteor that penetrates to within 25 km of the earth's surface can cause blast damage analogous to that of a nuclear weapon. The Tunguska event was the equivalent of a 10-20 MT blast. Civil defense studies show that the area of destruction scales as the explosive yield to the 2/3 power (Glasstone and Dolan, The Effects of Nuclear Weapons 1977).

Global climatic effects could result if the colliding object is larger than the ``global threshold.'' The event would have to inject into the stratosphere about 100 times as much aerosol as any of the major volcanic eruptions of the past two centuries. (It was indeed calculations of an ``impact winter'' that formed the basis of the ``nuclear winter'' hypothesis.) Such an environmental shock would curtail agricultural production during one growing season. ``The repercussions would be severe, as few nations store one year's worth of food,'' state the authors.

There are large uncertainties in the calculation of the energy required to produce this effect. Chapman and Morrison calculate a range from 1.5 x 104 to 107 MT, produced by objects 0.6 to 5 km in diameter.

About 163 asteroids with orbits crossing that of the Earth have been catalogued; the catalogue is probably less than 5% complete for objects less than 1 km in diameter.

The probability that a locally devastating, 1000-MT event will occur somewhere on Earth during a person's lifetime is approximately 1%. Surprisingly, the average American has a much higher risk of dying from an asteroid impact than from many other highly publicized hazards. However, Tunguska-class impacts are 100 times less likely to occur in a heavily populated area than other natural disasters. Between 1900 and 1985, there were 10 natural disasters that killed between 100,000 and 2,000,000 people each, including 4 earthquakes, 3 floods, 2 droughts, and 1 cyclone.

Chances of dying from selected causes (US)*:

*after Chapman and Morrison, Nature 367:39

The lower-limit risk for an impact is based on the assumption that 1.5 billion persons are killed by a 1.5 x 104 MT event, which occurs, on the average, every 70,000 years. The upper-limit risk assumes that a 107 MT event is required to cause this number of casualties. The probable frequency of such an event is 1 in 6 million years. The average expected number of local casualties ranges from 3 million up to 30 million if the effects of tsunamis (from ocean impacts) are included. The vast majority of the assumed casualties are due to climatic effects and synergistic impacts of social and ``ecosystem'' disruptions.

An extraterrestrial impact with global consequences is in the category of a ``rare/high consequence'' event it is overwhelmingly probable that zero persons will die of a catastrophic impact in the foreseeable future. It is instructive to compare societal response to this threat for which risk estimates, at least of impacts, are based on actual observations with the Chicken Little style response to purely hypothetical models and extrapolations such as Nuclear Winter and Global Warming and the Ozone Hole. [We have to try to prop up the sky in case it might be falling because if we wait for data, it might be too late.]

Chapman and Morrison argue for developing and deploying a system such as the ``Spaceguard Survey'' proposed by the NASA Near-Earth Object Detection Workshop. The estimated cost is $50 million for capital construction and $10 million per year for operation. The system could give a warning time of decades for collision with an asteroid. For long-period comets, the warning time might be only a few months.

A mitigation system, involving launch vehicles and explosive devices, would be far more costly. In addition, it would be controversial because nuclear explosives would be required. The authors argue for developing defenses only after a hazard is detected and only against impacts that might cause global catastrophe, even though impacts causing regional devastation are far more probable. The mitigation system itself would be subject to accident. Natural hazards are far more likely. Furthermore, ``it would be extremely difficult to discover and inventory all of the countless potential colliding objects down to tens of metres in size.''

An immediate issue raised by the study of impacts is the possibility that a high-altitude explosion of a meteor might trigger an inappropriate nuclear response from one of the world's increasingly numerous nuclear powers.

The authors conclude: ``By choosing whether or not to do something about this threat from the skies, society may establish a standard against which its responses to other hazards are measured.''

Questions that might occur to the reader: Should the EPA budget for monitoring and giving ``educational'' seminars on minute quantities of TCE (trichloroethylene) in water be diverted to the Planetary Science Institute of Science Applications International Corporation and the Space Science Division at NASA Ames Research Center, where Chapman and Morrison work? What about other hazards from the sky, say ballistic missiles? Their warheads pack a megatonnage much less than that of the Tunguska asteroid, but they are better targeted and more numerous.

Even today, asteroids may be better surveyed and more thoroughly understood than political and economic trends far more likely to have a global catastrophic impact (see p. 2).