McCaffrey now consulted statistical considerations. Since there have been seismographs for only a hundred years and historical legends go back only a few centuries, it is not possible to conclude from the well-known earthquake history that some sections of subduction zones are spared by megabs. Because of the long return times, the statistical data base is insufficient for such conclusions. Since subduction zones are generally located in the ocean, tsunamis must be expected at all such plate boundaries.

Geophysicists have been trying for some time to make at least statistical forecasts of earthquake recurrence times at certain plate boundaries. The underlying calculations are relatively simple: In a magnitude 9 quake, the earth's crust shifts an area of several hundred kilometers of a plate boundary by 20 meters at a stroke. Depending on how fast the plate moves? between two and ten centimeters a year? It takes between 200 and 1, 000 years to accumulate enough tension for a mega life. If the voltage in between is reduced by smaller quakes, the intervals between two megabounds can be even longer.

Based on these considerations, McCaffrey comes to the conclusion that unusually many megabounds have occurred in recent decades. The strongest ever quake occurred in Chile in 1960 (magnitude 9.5), there were other magnitude 9 quakes in the Aleutian Islands in 1946, in Kamchatka in 1952, in Alaska in 1964 and in Sumatra in 2004. display

Robert McCaffrey (GNS Science, Lower Hutt, New Zealand) et al .: Geology Vol. 36, p. 263 Ute Kehse© science.de