Time-variable factors in earthquake hazard

The development of a model for time-variable hazard estimation involves both subjective and objective elements in eight identifiable stages; the final stage—adoption of the model for practical use—is reached only after the conclusion of a successful performance test. At the present state of knowledge such a test consists in measuring the performance of the time-variable model against that of the present static model (i.e., seismic zoning). Precursory phenomena provide a basis for the quantification of the time-variable hazard provided the data are sufficient to allow certain intermediate quantities—the probability distributions for location, magnitude and time of occurrence, the valid alarm rate, and the failure rate—to be evaluated. Similar quantification should also be attainable on the basis of various types of regularity which have been suggested in the occurrence of large earthquakes, but the details of these models have yet to be fully ascertained.