|Title||A Review of Geological Evidence for Recurrence Times of Large Earthquakes|
|Publication Type||Book Chapter|
|Year of Publication||1981|
|Authors||Sieh KE, Simpson DW, Richards PG|
|Book Title||Earthquake Prediction|
|Publisher||American Geophysical Union|
|City||Washington, D. C.|
The geological record of the past several thousand years contains valuable information for evaluating the earthquake potential of the earth's major fault systems. Geologists have begun to characterize past and, presumably, future behavior of active faults and recurrence intervals for large earthquakes by studying 1) uplifted marine terraces, 2) fault-scarp morphology, 3) physiographic features offset along faults, and 4) faulted or otherwise deformed young sediments.
Along the convergent plate margins of Alaska and Japan, for example, studies of uplifted marine terraces have aided in evaluating the likelihood of imminent rupture of faults in two seismic gaps. In Nevada, Utah, and eastern California, detailed studies of scarp morphology along normal faults of the Basin and Range Province are beginning to reveal the recurrence intervals, sizes, and patterns of prehistoric earthquakes. Studies of offset stream channels along the San Andreas fault have shown that right-lateral events of as much as 10 m have occurred repeatedly in the past with an average frequency of about two hundred years. Elsewhere along the San Andreas and on other faults in California and Japan, studies of faulted and deformed young sediments have enabled dating of specific prehistoric earthquakes or, at least, a determination of the minimum number of events that occurred during the deposition of the strata.
In the western U.S. and Japan and perhaps in other seismically active regions as well, there is good reason to believe that within the decade we will know the average recurrence intervals, regularity, and sizes of past large seismic events at several localities. Hopefully, this will enable forecasts of some future large earthquakes with uncertainties measured in decades rather than centuries and will provide a sound basis for hazard mitigation and for directing short-term predictive efforts to those fault segments in imminent danger of rupture.