Talk about a missed forecast. Some 20 years ago, it was announced that accurate earthquake prediction was not only possible but right around the bend. Yet in the time since this proclamation was made, scientists have learned instead that the way the Earth moves is much more complex than they ever anticipated. Various quakes in the past year or so from Northridge to the Shikotan quake in the Kuril Islands to Kobe have indeed borne out the complexity of earthquake prediction and hazard assessment.

Consequently, the community has started taking another look at longstanding models and theories about the movement of the Earth. Seismic gap theory, for one, has been widely debated in the Earth science community in recent years. The reigning wisdom holds that long, quiet fault segments—or seismic gaps—are the most likely sites for large earthquakes to occur in future decades along a given boundary zone. On the other hand, faults that have recently experienced a large quake, by the gap model, should rest quietly for many decades or centuries to come. Yet several teams of scientists maintain that there are serious shortcomings to this model.