Can earthquakes and volcanic eruptions be "due" or "overdue"? As in "They typically happen every X years, we haven't had one for Y years, so we're more likely to have one now"?

I heard Kevin McCloud saying something along that line in "Escape to the Wild", about a volcano in Chile, but also lots of other times. Example claim

Though often overshadowed by the San Andreas, it was the Hayward Fault running through the East Bay Hills that shook.

Scientists across the Bay agree on one thing: the Hayward fault is overdue for a big one.

They say any day now.

The fault typically erupts with a major shaker every 140 years and it's been 147.

  • 2
    Surely these statements are to be understood as statistical in nature. It's not like any of these features keep time. I mean, Old Faithful, noted because it is regular, and that is on a much smaller scale. – dmckee Aug 29 '15 at 0:56
  • 2
    In case of big earthquakes that happen periodically on the long term, big stresses need to build up and takes time. In case of the recent earthquake in Nepal, it's caused by India moving north by a few centimeters per year but the North is stuck and doesn't move smoothly, it moves abruptly every time the stresses that have accumulated are sufficient to cause a massive earthquake. Unfortunately, in the recent earthquake not enough stress as relieved, stresses have been transferred to other parts of the region that haven's moved in 600 years. Seismologists fear a magnitude 9 quake there . – Count Iblis Aug 29 '15 at 2:55
  • 2
    Stresses build up over a number of years. If the last relief via earthquake was some time ago, stress can be assumed to be high and in need of relief. – SQB Aug 30 '15 at 12:43
  • 2
    That leaves the interesting question "is it feasible to intentionally set off a hopefully minor quake to bleed off stress in a controlled fashion and avoid a big, uncontrolled quake a few years down the road?" – Shadur Aug 31 '15 at 14:48
  • There are 2 notions: pure statistics, and physical thresholds. As a statistics, any singular event can be in advance or late compare to the expected mean. Depending on the probability law involved, it gives indication or not on the future. As a physical phenomena where a continuous quantity accumulates up to some breaking threshold, then it sure increases the odds. Still, any similar looking events all around can modify the local odds (rearangment of tensions around the various faults), to lower or increase them. – Fabrice NEYRET Oct 11 '15 at 19:40

You must log in to answer this question.

Browse other questions tagged .