Mean waiting time
between impacts (yrs)
lower curves show how far in each direction the actual curve might lie.) If we knew which craters had associated extinction events, their positions could be plotted on the impact-kill curve. As noted in the discussion of Table 4, however, the only crater about which we are certain is Chicxulub. Michael Rampino and Bruce Haggerty21 make an educated guess that three other craters can be added to the list as shown in Table 6 and Figure 24.
Is the impact-kill curve credible on its face? It predicts that the largest extinctions are associated with craters at least 140 km in diameter. This is plausible because the largest extinctions and the largest craters each occur about every 100 million years. There have been five major extinctions since the Cambrian began, and, Shoemaker's estimates tell us, about the same number of giant impacts. Note again that these two calculations are completely independent. Sepkoski's best estimate of species killed at the K-T boundary is about 70 percent, corresponding to a crater of about 150 km. (If Chicxulub is actually 300 km in diameter, as some argue on the basis of its buried topography and gravity structure, then the actual impact-kill curve would have to be closer to the lower dashed line.) None of the points other than Chicxulub is precisely located but they do fall within the upper and lower boundaries of the curve. On the other hand, as far as we know, some sizable craters have failed to produce mass extinctions. Neither the 24-km Ries Crater in Germany, nor the 45-km Montagnais Crater, located in the seafloor off Nova Scotia, is connected to an extinction. If the reasoning behind the impact-kill curve is correct, and if Ries and Montagnais have no associated extinction, then the curve must hug 0 on the Y axis until
FIGURE 24 The impact-kill curve, combining Raup's kill curve (Figure 23) and Shoemaker's estimates of the frequency of formation of craters of differing sizes (Table 5). The four sites listed in Table 6 are plotted, though only Chicxulub is confirmed. [After Raup.22]
it reaches the point corresponding to a crater of around 45 km in diameter, when it must begin to rise steeply. This could be the case, for example, if a certain critical impactor mass were required before extinctions become global and massive.
The stage, the smallest unit into which geologists subdivide the rocks of the geologic column, represents a fundamental subdivision of earth history. In the 600 million years since the Cambrian began, Sepkoski identifies 84 stratigraphic intervals, most of them stages, giving an average duration for a stage of approximately 7 million years. If impact causes all extinction, as Raup rashly considered, then craters large enough to be associated with extinction ought to have about the same waiting time as the duration of an average stage. Is that the case? To find the answer, inspect Table 5: Note that 7 million years is the mean waiting time for a crater just over 50 km in diameter. A crater of that size releases about 5 million megatons of energy, roughly 50 million times the power of the atomic bomb that
Was this article helpful?