Early in the morning of June 30, 1908, in the Tunguska region of Siberia about 1,000 km (600 miles) north of Irkutsk, an asteroid about 60 meters (200 ft) in diameter entered the Earth’s atmosphere, resulting in an immense explosion, centered about 8 km (5 miles) above the forest below. Trees were flattened over an area about 50 km (30 miles) in diameter, several times larger than the area encircled by the Beltway around Washington, D.C. It exploded with energy in the range of a modern nuclear missile warhead, about
10 megatons, or about 500 times the energy of the Hiroshima atomic bomb. While there were few, if any, casualties from this event, if such an event were to occur in a more populated area it would be a major natural disaster, comparable to a major flood, earthquake or volcanic eruption.
For this reason there is considerable interest in assessing just how often such an event might be expected to occur. Since the last one was about a century ago, it has often be supposed that the answer is “about once a century,” but this is not necessarily so. The Near-Earth asteroid surveys in progress for the last few years (LINEAR, NEAT, LONEOS and others) are aimed mainly at discovering larger asteroids that would cause major but far less frequent damage. They also discover many smaller asteroids in the “Tunguska” size range, presenting an opportunity to assess the frequency of these smaller events. Even smaller objects pose no direct danger, as they explode higher in the atmosphere and produce little if any ground damage.
The population of near-Earth asteroids (NEAs) down to ~1 km (half a mile) in diameter is reasonably well determined. Planetary scientists now estimate that there are about 1,000 such NEAS equal to or larger than 1 km in diameter, with a resulting impact frequency of about once per half million years. However, in the size range of the “Tunguska event” NEAs (diameter ~50-75 m), estimates of population, or equivalency of impact frequency, range from once per 200 years to once per 10,000 years. The LINEAR survey has now discovered ~30 NEAs in the “Tunguska” size range; thus a better estimate is possible. However, for small NEAs, a very large simulation is needed to obtain even a few “detections” in the computer model.
Dr. Alan Harris of the Space Science Institute in Boulder, CO., recently compiled a new simulation for objects down to about 200 m in diameter, dividing them into six size bins. By comparing his relative populations with the absolute population estimates of Stuart (Science 294, 1691-1693, 2001) in his two smallest size bins, Harris extends Stuart’s curve through the size range of “Tunguska” objects. He finds a population of the order of half a million objects in this size range, corresponding to an expected impact interval of the order of once per thousand years. This estimate is uncertain by a factor of about 3, largely due to uncertainty in the actual size of the Tunguska event.
This new estimate of the impact frequency of “Tunguska-sized” events is considerably less than has often been supposed. If correct, it means that the Tunguska event having happened so recently is unusual, although not extraordinarily so, and that the risk of such events in the future is a few times less than has been assumed. This is not to say that there is no danger at all. Impacts are random events, so it is not possible to say that we are “about due” for another, or that since one happened so recently that another won’t happen soon. All we can say is that the odds are less than had been often quoted based on the assumptions that such things happen about once a century or even more often.
