Hi Roger,

I agree for large quakes. Where is pretty much where there is strain built up on a fault or in a seismic gap; magnitude is based pretty much on the length of the fault and what strength earthquakes have occurred on the fault in the past in the area in question, size of potential foreshocks, etc. I am probably leaving out many other factors a seismologist would take into consideration, but to my mind, when definitely seems to be the missing element. Hence my interest in animal behavior precursors, earthquake weather, people's sensitivities, etc. (By the way, the coyotes were silent tonight. Haven't heard a yip, let alone any howling.

It's interesting, Don in Hollister (sadly gone now) claimed his method could predict where and when, but not how big. Of course, he was looking at a specific set of data. Too bad all the data couldn't be combined such that one might have all three criteria for an impending quake. I would have liked to have seen that comem about, but everyone is always so private about their prediction methods. If everyone would pool their data, some really valid predictions might ensue.

Roger, did you ever do an analysis of Don's prediction rate? If so, what were your results?

I'm really going to miss Don's presence here. What a shame he is gone.

Thanks,

Cathryn