Exactly, there was a certain level of risk the Japanese authority assumed at the time in directing the engineering of the reactor.
Reiterate one more time...does one plan for a 100 year event, 1000 year event, or a million year event? Most plans I’ve ever seen require the 100 year event included as the speck of the design. This event was bigger than a 100 year event.
You get anything that exceeds the designed specs. it becomes an act of God...sh*t hit the fan....
The reality is if the tsunami had been the planned 7 meter event, none of this would be happening. The reactor would be performing beyond design specifications.
Part of the problem in these (and other) designs is the classification of “100 year events” and “500 year events” and so on relies on VERY poor modeling and knowledge of just how frequently these events happen.
Add to that the less than perfect modeling with math and we get poor requirements.
I still like how the financial boys on Wall Street had models that showed that the rate of market decline in October, 2008 couldn’t happen more than once in 4 billion years’ worth of market time (ie, it was *highly* improbable), yet we had two such declines in one week. That’s an example of poor modeling.
Consider that it was only recently that scientists admitted that mariners have been right all along and not merely telling tall sea tales: there are such things as rogue waves, waves 60’ high that come out of nowhere and can break a ship in half or roll it. Mariners have been saying this for years, and only with some very new approaches has science confirmed that “Yes, this can happen” and apparently off the coast of S. Africa, it happens often enough to make you sit up and take notice.
In Japan, the issue is one of modeling quake frequency and duration. Some stuff I’ve now been reading is about quake severity and frequency in Japan. New research indicates that severe quakes tend to ‘cluster’ in time in Japan, with long periods of ‘relative’ calm (when your big quakes are 8’s, 6’s look relatively calm) in between.
The central problem for civil engineers in Japan is that they have only 100 years (plus or minus) of really quantified data on which to base models. Coming up with a “1,000 year” design based on this paucity of data isn’t going to be easy.