In fact, astronomy and astrology were closely intertwined as recently as four centuries ago, when Tycho Brahe laid the foundations of modern astronomy while simultaneously maintaining a lucrative business in personal horoscopes. Modern astronomers generally scoff at such superstitious beliefs, so it is somewhat ironic that science has in the past few decades uncovered compelling evidence for celestial interference in terrestrial matters.
It is now clear that asteroids occasionally wander from the main belt beyond Mars because of chaotic instabilities caused by Jupiter. Some of these errant asteroids strike the Earth with terrible consequences. On page 165 of this issue, Rabinowitz et al.1 report that the number of threatening near-Earth objects (NEOs) larger than 1 km in diameter is only half the previous estimates. But we still have no effective means of detecting them all, and no form of self-defence.
The Earth bears the scars of previous encounters with NEOs. Hundreds of impact craters, some the size of small American states, have been discovered on the surface of our planet. Each was produced by a devastating explosion that must have been fatal to life in the surrounding areas on scales from local to global (Fig. 1). The Cretaceous-Tertiary mass extinction of 65 million years ago seems to have been triggered by the impact of an asteroid 10 km in diameter2. Ten thousand people killed by 'falling stones' in Shanxi Province, China, in 1490 were possibly the victims of a much smaller and thoroughly fragmented projectile. Still more recently, on 30 June 1908, 1,000 square kilometres of Siberian pine forest in Tunguska were blown flat by a 10-megaton atmospheric blast caused by a 70-metre asteroid.
The gradual acceptance of the evidence for impacts by asteroids (and comets) has led naturally to questions about the magnitude of the threat posed by NEOs to life on Earth3,4. Rabinowitz and colleagues1 provide the most recent and best controlled estimate of the number of large, potentially Earth-threatening NEOs.They report that there are nearly 1,000 NEOs larger than 1 km in diameter and that, given the present rate of discovery, it will take 20 years for 90% of these objects to be found. Should we worry?
The answer depends on the number of fatalities to be expected, but also on personal assessments of risk. The number of NEOs found by Rabinowitz et al. is within a factor of two of previous estimates based on less-controlled samples, so published estimates of impact mortality are essentially unchanged. Considering events of all energies there is about 1 chance in 20,000 of being killed by an impact during the course of a human lifetime4, similar to the likelihood of being killed in an airplane accident. The perception of risk from impacts is smaller than for being killed in a plane crash because planes crash at a steady rate with (relatively) few deaths per event, whereas lethal impacts are rare but kill a lot of people. At the very least, the potential consequences of impact are large enough to cause concern.
In the past decade, thanks to several reported near-miss encounters with small objects, the impact threat has become a subject of intense interest to the general public (spawning the popular movies Deep Impact and Armageddon). In 1994, the United States House Committee on Science and Technology went so far as to order the US space agency NASA to "catalogue within 10 years the orbital characteristics of all (Earth-orbit-crossing) comets and asteroids that are greater than 1 km in diameter". This particular cut-off diameter was picked in part because 1-km NEOs are thought to be the smallest objects capable of wreaking global havoc (for example, by disrupting the climate and shutting down photosynthesis). Smaller objects cause regional damage but would be unlikely to precipitate a major extinction like the Cretaceous-Tertiary event.
Last summer, astronomers devised a new risk-assessment scale, similar to the Richter scale used for earthquakes, to help the public understand the hazard posed by a given NEO. The so-called Torino scale ranges from zero (no chance of a collision) to 10 (certain collision causing global devastation). No known NEO has yet had a Torino number greater than one. This is just as well because we presently have no coherent plan of action should a real threat arise. The simplest option -- massive evacuation of the impact site -- would be impractical because of the positional uncertainties and large numbers of people involved, and would be ineffective because the damage from large NEOs will be global. One option that has been discussed is the thermonuclear destruction of the incoming NEO (a bad idea because the shower of debris produced by the exploding NEO might be as damaging as the initial object, and would be radioactive). Given enough time, the NEO might be deflected from an Earth-intersecting path by a series of smaller explosions, or by attaching rockets or solar sails that use radiation pressure from the Sun.
The focus on NEOs larger than 1 km ignores the threat from smaller but much more numerous objects. The Earth's atmosphere offers little protection against objects larger than 100 metres in diameter4. These smaller objects outnumber NEOs larger than 1 km by a factor of 100, so they are much more likely to strike in our lifetimes. There is a 1% chance that the Earth will be struck by a 300-metre NEO in the next century4. Such an impact would deliver a withering 1,000-megaton explosion and cause perhaps 100,000 deaths. If the impact occurred in or near a densely populated region -- the eastern seaboard of the United States, for instance, or Western Europe or coastal Asia -- the fatalities could easily rise into the tens of millions.
Neither can we take refuge in the fact that 70% of the Earth is covered by oceans. Impact-induced tsunamis could wipe out coastal cities over a wide area. So, to have practical value, surveys should not be limited to the (observationally easy but numerically rare) 1-km NEOs, but should instead catalogue objects at least down to the fewhundred-metre size range5. What is needed is a more ambitious survey to completely identify the population of small, potentially threatening NEOs.
The strategy for such a survey has been explored by Alan Harris of the Jet Propulsion Laboratory6. He argues that the whole sky must be surveyed on a monthly basis with a sensitivity about 100 times greater than current NASA-sponsored surveys. How can this be done? A large (6-8-metre) telescope is required, with a wide field of view tiled with CCD (charge-coupled device) optical detectors and connected to a massive computer array capable of meeting the huge data-processing demands. The technology exists and tentative designs are beginning to appear7-9. Such a telescope, which would have many applications in other branches of astronomy, is projected to cost about $100 million (about half the price of a Jumbo jet). What is missing is any sign that such a facility will be funded by governments and their agencies. Perhaps astronomers can attract the interest of private donors in the search for threatening NEOs. If not, it seems we will have to face the asteroidal impact hazard with our eyes wide shut.
David Jewitt is at the Institute for Astronomy, University of Hawaii, 2680 Woodlawn Drive, Honolulu, Hawaii 96822, USA. e-mail: email@example.com
Nature 403, 145 - 148 (2000) (c) Macmillan Publishers Ltd.
1. Rabinowitz, D., Helin, E., Lawrence, K. & Prado, S. Nature 403, 165-166 (2000).
2. Alvarez, L. W., Alvarez, W., Asaro, F. & Michel, H. Science 208, 1095-1108 (1980).
3. Morrison, D. The Spaceguard Survey: Report of the NASA International Near-Earth Object Detection Workshop (Jet Propulsion Laboratory, Pasadena, 1992).
4. Chapman, C. R. & Morrison, D. Nature 367, 33-40 (1994).
5. Binzel, R. P. et al. From the Pragmatic to the Fundamental: The Scientific Case for Near-Earth Object Surveys (1999).
6. Harris, A. Planet. Space Sci. 46, 283-290 (1998)