result in sufficient environmental modification to cause catastrophic mortality. One computer simulation portrays the ejection of a dust layer into the high atmosphere that would cut out sunlight for months at a time, ending photosynthesis and sinking all latitudes into a deep cold. After the dust settled, high levels of atmospheric carbon dioxide would warm the Earth by producing a greenhouse effect. Another model suggests that the fast-traveling ejecta from the impact site could have been at temperatures high enough to cause atmospheric oxygen and nitrogen to combine, forming clouds that would precipitate into nitric acid rain. A third hypothesis, which is gaining increased acceptance, is that an impact of the proper age, formed the 180-km-diameter Chicxulub crater in the Yucatan of Mexico. Chicxulub strata is composed of thick sulfate-rich evaporite and carbonate deposits; an impact into such deposits could eject huge amounts of sulfate aerosols into the stratosphere, resulting in large-scale global cooling and several years of acid-rich rains as the aerosols settled out of the atmosphere—both the cooling and acid rains could have devastated the food chain. Several issues complicate the impact scenario—one strong argument cites evidence of animal groups that suffered heavy extinction before the primary iridium anomaly settled into place.

Regardless of what the eventual consensus on its cause may be, the iridium anomaly has stimulated intense geological research. Not only have the contemporary rocks been studied closely—the anomaly has been documented at about 100 sites throughout the world—but the deep ocean record has been scrutinized, revealing a perturbation of the earth system that lasted half a million years. The most significant result of this intense geological research may be that serious consideration of sudden global catastrophes is now acceptable.

All meteorite impacts are now receiving great attention. The roughly 100 known impact craters on Earth are being looked at anew, their ages are being reassessed, and attempts are being made to see whether their incidence could be periodic. An impact site in Sweden, 50 km in diameter, has been drilled to a depth of more than 6 km, allowing observations about the effects of large-body impact at depth. The discovery of iridium anomalies in some of the oldest rocks strongly suggests a flurry of meteorite impacts as late as 3.6-million-years ago. And the most drastic of catastrophes, collision between the Earth and a Mars-sized body before 4-billion-years ago, is now considered a possible explanation for the Moon's formation.

Observations of asteroids show that the current meteorite flux is likely to be higher than had been considered, and in 1989 one carefully observed asteroid passed closer to us than any other in the past 50 years. There is a calculable, if remote, chance that the Earth will be struck by an object more than a few kilometers in diameter in the foreseeable future. Such an event would cause massive destruction, and a case can be made for assessing the possibility of diverting an incoming object to avert such a potential collision.

No other global iridium anomalies have been recognized, although a locally defined anomaly dated at 34-million-years ago corresponds to a substantial extinction. No iridium or shocked quartz horizons have been found for the greatest extinction ever, which was 240-million-years ago; however, the absence of evidence, especially in older rocks, should not lead to a positive conclusion. There is certainly evidence that other ancient mass extinctions were complex events, extending over intervals of several million years.

The mainstream of the earth sciences has shifted away from the extreme uniformitarian position, which attempts to explain all phenomena in terms of directly observed processes. Now researchers, confident in the soundness of their inductive methodology, can consider the possibility of exceptional events. There is, of course, an appropriate reluctance to invoke exceptional events as causal mechanisms, and intense skepticism deservedly awaits all such suggestions. The study of catastrophic effects on earth systems requires further work. Promising directions for general research include their continuing consequences, such as how perturbations propagate through the earth system and how secular change is altered permanently by catastrophe.

MODELING THE EARTH SYSTEM

An Incomplete Record

Solid-earth scientists, like all other scientists, have long used conceptual models as an aid in understanding. For example, it was recognized in ancient times that the occurrence of seashells in rocks on the continents required a model in which the disposition of sea and land changed with time. By the nineteenth century this idea had developed further with the recognition that there had been widespread episodes of continental flooding that were correlatable over great distances. In the mid-twentieth century the repeated flooding and emergence of North America over the past 550-million-years were found to provide a strong framework for under-



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement