While researching another matter, I came across some additional information about the seismic gap theory (or model). Inasmuch as I had posted on this subject some days ago, that post had sparked some interest, and I made at least one mistake in my original post on the subject, I figured I should post the following excerpt from "The 2004 Parkfield Earthquake, the 1985 Prediction, and Characteristic Earthquakes: Lessons for the Future," by D.D. Jackson and Y.Y. Kagan (Bulletin of the Seismological Society of America, Vol. 96, Number 4, Sept. 2006)[the mistake I made was in claiming-with the caveat that I might be wrong-that the term seismic-gap theory only applied to subduction zones. Apparently it includes pretty much all types of seismogenic faults][Capitalization for emphasis is mine - mw]:

The seismic gap hypothesis has enjoyed intuitive appeal since the early work of Reid (1910). He suggested that a large earthquake releases most of the stress on a given part of a fault and that further earthquakes could be expected when that stress has reaccumulated through tectonic motion. The acceptance of plate tectonics in the 1960s as a believable mechanism for resupplying stress added more intuitive arguments for the seismic gap hypothesis. Fedotov (1968) identified several plate boundary regions that had experienced large historical earthquakes and named several zones as likely to have earthquakes in the near future. McCann et al. (1979) adopted the gap model and produced a colored map of "earthquake potential" for nearly a hundred circum- Pacific zones. THEY ASSUMED THAT SEISMIC POTENTIAL INCREASES WITH THE ABSOLUTE TIME SINCE THE LAST LARGE EARTHQUAKE. Bakun and Lindh (1985) used the history of earthquakes at Parkfield to estimate a distribution of recurrence times and a probability for a repeat earthquake at Parkfield: THIS FORMED THE BASIS OF THE PARKFIELD PREDICTION. Nishenko (1989, 1991) for the first time refined the seismic gap model into one that could rigorously be tested. He specified the geographical boundaries, characteristic magnitudes, and recurrence times for each segment. He used a quasi-periodic recurrence model to estimate conditional earthquake probabilities for 125 plate boundary segments around the Pacific Rim. Nishenko’s regions, including Parkfield, were defined by previous earthquake ruptures zones.

The seismic gap idea has been applied to make long- term forecasts for many faults and plate boundaries around the world. Rong et al. (2003) give a summary (see also Kagan and Jackson, 1991, 1995; Nishenko and Sykes, 1993, Jackson and Kagan, 1993; but there are too many to list here). SO FAR, ONLY THE MODELS BY mCcANN ET AL. (1979) AND NISHENKO (1989, 1991) HAVE BEEN COMPREHENSIVE ENOUGH TO TEST USING LATER EARTHQUAKES, AND THEY BOTH FAILED THE TEST (Rong et al., 2003).
HOW COULD SUCH AN OBVIOUS, INTUITIVE MODEL NOT BE TRUE? Consider the chain of simplifying assumptions behind it. First, the segmentation assumption is crucial. If successive earthquakes are not confined between the same segment boundaries, there is no reason to expect similar rupture lengths, magnitudes, and so forth. Second, we must assume that the average displacement is similar for successive events, else the magnitudes will not be similar. Third, we must assume that earthquake initiation and termination depend almost entirely on processes on that fault segment; stresses from earthquakes on other segments or other faults would likely ruin the similarity of successive quakes. If the assumption of identical events is maintained, it is possible to estimate how larger off-segment earthquakes affect the timing of characteristic events (Working Group on California Earthquake Probabilities, 1990; Cornell et al., 1993), but the required assumption is questionable. Finally, to use this simplifying model, we must be able to recognize the size of the characteristic earthquake with little ambiguity. Should we be fooled by a sequence of earthquakes smaller than the characteristic magnitude and assume that they were characteristic, then we would seriously overestimate their frequency.

end of excerpt

Observant readers may also note that nothing in this discussion supports my earlier contention that the theory relates to spatial "gaps," i.e. fault segments that are bordered by segments which have recently ruptured in large earthquakes. As discussed here, the "gaps" seem to be only temporal in nature. I was working from memory, and may have conflated two or more separate but related theories, in which case I apologize for the misiniformation, or, it is also possible that Jackson and Kagan's discussion above is simply of a more limited definition of seismic gap theory.

Michael F. Williams
Arroyo Grande, CA USA