Earlier today, I blogged about a Reuters report that yesterday's magnitude 5.8 Mexico earthquake did not trigger the seismic monitors at the Palo Verde nuclear generating station west of Phoenix [right]. One of the geobloggers (Ontario-geofish) called this further proof that the monitors don't work.
I brought this claim up to my personal seismological advisor (also my wife, Ann Becker, MS Stanford, PhD Berkeley) over dinner tonight, and she said to not jump to conclusions.
She's worked on the Humboldt (CA) nuclear plant and Yucca Mountain among many other projects and pointed out that Palo Verde is a 'stiff' structure that will ride out long period seismic waves while short period waves are more of what they are typically most concerned about.
If I'm quoting her correctly, P waves are shorter period and thus potentially more of an issue, but they only carry a very small part of the energy of a quake and die off quickly with distance. Given the distance of the yesterday's quake from Palo Verde, it would not surprise her if the plants seismic monitors were not triggered. They would more likely be set to respond to nearer high-frequency events where the P wave or other sources could produce enough energy in the right wavelength to be a potential problem.
Now, this is just idle discussion over dinner. I haven't talked with anyone at Palo Verde, but it's a reasonable explanation for yesterday's events. The monitors would not have been triggered because the short-wavelength energy was too small to reach the minimum threshold of concern.