Hi All. What do we really know about earthquakes? I’m afraid to say not a whole lot. We didn’t really know the cause of earthquakes until less then 50 years ago.

The quake history for the State of California is pretty well know for a period going back about 3000 years. That 3000-year period is for the San Francisco Bay Area. However that is only for very large quakes. The kind that rupture the surface. Trenching across a fault can reveal that history, but how correct is our interpretation of it? What do we know about the small quakes? The kind that never rupture the surface, but are still felt and do some minor damage.

We know how to design structures to withstand enormous forces. But we still don't know how to figure out when those forces will strike, what triggers the release of all that energy, and in many cases how the ground beneath our feet will perform.

We know what causes earthquakes. Quakes are caused by the sudden release of strain that builds up in the giant tectonic plates as they grind against each other while moving around the surface of the earth.

But we don't know specifically what triggers a quake. We do know it isn't the alignment of the planets, or radiation from space, or your grandmother's gout. The trigger probably varies with local conditions. What triggered one quake may not be the same for another quake even if its in the same location.

And nobody knows precisely when the next quake will occur. Instead of predicting quakes, scientists have settled for long-range forecasts. In some ways this may be better, that is if you know a major quake will occur there. It allows you to build for it, to prepare for it, to get ready for it.

So no reputable scientist is going to tell you that California's notorious San Andreas Fault is going to rip up that pricey landscape at 2 p.m. on Friday the 13th. The best they can do is say that some part of it is likely to break loose sometime in the next decade or so. But there is some pretty convincing evidence that even those long-range forecasts are built on shaky ground. Most governmental agencies around the Pacific Rim base their engineering requirements on the idea that faults tend to rupture at fairly definable intervals. It's called the "time-predictable recurrence model," but it may well be dead wrong.

Geophysicists at Stanford University have taken a hard look at that concept and concluded that nature really isn't all that tidy when it comes to scheduling earthquakes. Jessica Murray and Paul Segall examined a famous section of the San Andreas Fault that runs through the farming village of Parkfield in central California. Parkfield is the most instrumented fault zone this side of Tokyo because some decades ago scientists determined that it has ruptured on average once every 22 years since 1857. Since the last quake hit there in 1966, scientists began packing all sorts of gizmos into Parkfield in the mid-1980s. They reckoned that a quake of magnitude 6 should hit there by around 1987.

The scientists were ready for the quake. It didn’t occur in the time period they were looking at, but it did occur. So far the score is San Andreas 1; scientists 0. As David Schwartz, seismologist at USGS, Menlo Park, California has stated in the past. “Being a seismologist is a humbling experience. Were always saying we didn’t know there was a fault there, or we didn’t see the quake coming.” In this instance they knew the fault was there, but they didn’t see the quake coming. Take Care...Don in creepy town