... (which depends, I guess, on how much radioactivity gets out), but so far, this is a useful, if mildly coldhearted, reminder:
Putting the Problem in Perspective
Updated March 14, 2011, 10:55 PM
Bernard L. Cohen is professor emeritus of physics at the University of Pittsburgh.
The very elaborate earthquake design requirements for U.S. reactors recognizes that a large enough earthquake would defeat any precautions.
As a fallback, the requirements are designed such that the consequences of the earthquake from reactor failures should be negligible relative to the non-nuclear consequences of the earthquake. That is certainly fulfilled in the Japanese earthquake with thousands of deaths from non-nuclear consequences and, considering the evacuations, very few if any deaths from the reactor failures. From costly property damage, the non-nuclear damage is far greater than any possible damage from reactor failures.
The Japanese earthquake was 10 times more powerful than what is considered to be the most powerful credible earthquake on the San Andreas fault.
Some other nuggets of opinion here.