Sea walls and ontological risk


Did the designers of the japanese seawalls consider all the factors?

In an eerie parallel with the Blayais nuclear power plant flooding incident it appears that the designers of tsunami protection for the Japanese coastal cities and infrastructure hit by the 2011 earthquake did not consider all the combinations of environmental factors that go to set the height of a tsunami.

In the case of Blayais the plant designers overlooked the combined effects of tide, coastal storm surge and wave height in setting the height of the plant’s seawall. In the case of the Japanese tsunami it appears that the sea wall designers failed to consider the possibility of earthquake induced subsidence lowering the height of sea walls.

Simply put if the design wave height was 12 metres above sea level a subsidence of 1 metre, as experienced over large areas of the coast, would ensure that a tsunami of only 11 metres would overtop a sea wall designed to meet the 12 metre wave.

This combination of factors means that the probability distribution for effective tsunami height is skewed towards higher wave heights. That is as quake intensity increases the likelihood of an overtopping wave is increased as is the likelihood of coastal subsidence. These factors in combination in turn increase the probability (and risk) above that predicted by simple tsunami height versus frequency models.

As was the case in the Blayais flooding incident this is a classic example of ontological uncertainty, in this case the unrecognised incompleteness of the wave height model, rather than uncertainty about a specific model parameter. Both these incidents are also a salutory lesson in how vulnerable a single ‘safety barrier’ system can be to ontological risk. If you have only one barrier and therefore one chance, it had better be a good one…