John, increasing vertical load will clamp thrust faults, and weaken normal faults (as you know). I suppose I would have to pull out some old geology lab books to do effective normal stress vs. shear stress calculation for a vertical load. It's not immediately clear to me why a vertical load would clamp a strike-slip fault. I'd guess from logic (without doing the math) that a vertical load off to one side of a strike-slip fault would perturb the stress field and would either clamp or unclamp it, depending on how the gradient in load interacted with orientation of fault and existing stress field.
All: it is well understood that increasing the pore pressure along any type of fault will weaken it...that is different than what we were talking about. The pore pressure if subtracted from the normal stress (meaning, the component of stress perpendicular to fault), unclamping it. Shear stress parallel to fault drives them, normal stress (perpendicular_, friction, and cohesion hold faults together.

I heard Seeber give a talk at Lamont saying that 1/2 of intraplate earthquakes are triggered by man: quarries, reservoirs, injection. I suppose naything messing with fluids could do it: oil production, injection for enhanced recovery, and we have already discussed Geysers geothermal seismicity here. I think the distant earthquakes triggered by Landers/Denali etc show that it does not take much perterbation to set off a quake in an area where the faults are right at failures.

On Thurs this week I looked through a Seis Res, Letters article by Sylvester and Heineman..I think 1996, where there was a thrust surface ruppture for a M3+ quake near Lompoc California due to removing a load by quarrying. Probably a diatomite quarry. I think it took removing 40 m of diatomite to set off that quake. Nice scarp right across quarry. I was checking references for bedding parallel thrust slip with top towards crests of anticline...bending moment slip. Works with bending a stack of computer cards. Wait, we don't have computer cards any more

Mike: I suppose a quick pass on looking for a correlation might be simply rainfall for, say, 1 month periods (or 3 month periods). Could maybe find a few places with records scattered across 100 km or 200 km surrounding area that has had these quakes (1968, yesterday, others). Maybe first thing to do would be to look a good catalogue of quakes in the region over M3 (or 2.5) to see where they are occurring. You might want to exclude the highly active zones near New Madrid (Reelfoot etc.).

I'm not going to do anything with this except maybe email Seeber if anyone comes up with something interesting. You should pursue this if it interests you.
Chris