Well I’ve had ummm call them discussions with Strat about these things. He is very, very educated and up on the subject.
My thing is that the whole mess is really based more on some type of chaos theory than any type of linear prediction model.
And because of that I tend to not pay attention to any models that say “Well, historically we have one 5.0 quake here every 300 years”
Long term, that model kinda works, I mean you can go down and look at the San Andreas fault and literally see where a road that went across the fault year after year gets displaced a couple centimeters.
But precisely because I believe in a chaos model, then things like subtle changes in tidal influences, the electrodynamic properties of the strata (remember piezoelectricity?), all these what might be thought of as non-geologic effects may come into play.
So I try to keep an open mind. Thats all.
The severity-frequency data is used by insurance companies but it doesn’t mean you will have a 5.0 quake every 300 years. That’s a common misunderstanding, just like flood frequency-severity data.
As for chaos theory, most definitely applicable in an engineering analysis. We know where many of the cracks are in the structure, relative movement of some of the pieces, and even an idea of how much energy may be stored, but we don’t know how much stress the structure can take and how much of a force bump it will take before it fails.
I just find it fascinating!