Comment by nwallin | Hacker News Reader

Comment by nwallin | Hacker News Reader

Parent Full thread

> in Fukushima Dai-ichi's case, for some reason (possibly financial?), it was decided that the tsunami barrier was sufficient (it wasn't) for the job,

Plate tectonics was not understood or scientifically described until 65-67. It was not a widely accepted theory until later, at least the 70s. By that point, Fukushima had already been permitted and built.

Without plate tectonics to create the plate shift earthquake which resulted in the tsunami which caused the disaster, the only mechanism to create tsunamis known was underground rockfalls from steep slopes. Afaik the builders of the plant studied the bathymetry around the area, and saw limited risk of tsunamis.

The primitive science of the time ruled out the possibility of a tsunami overcoming a tsunami wall that was X feet tall. So they built a tsunami wall that was X+k feet tall, and called it a day. When later scientific theory showed that larger tsunamis were not just possible, but inevitable, the plant was not retrofitted to deal with the new reality. Oops.

> When later scientific theory showed that larger tsunamis were not just possible, but inevitable, the plant was not retrofitted to deal with the new reality.

You are saying the process of risk assessment and mitigation was not regularly actualized with new scientific knowledge. How is that possible ? Is there communication issue between different fields ?

For our information, does someone know : As soon as the engineers / executives were informed of this new risk, what was their decision at the time ? (Maybe this was asked during a court hearing)

You typically don't have a design engineer actively monitoring safety features that are permitted by the regulator, built and maintained to standard. In a dynamic environment where the situation changes regularly there is a little more defense for catching that type of thing. In a static environment like fixed plant it could well be they just weren't ready for that sort of change.

That style of hazard, where existing controls passed through a rigorous system but become insufficient with time, are really insidious. It is certainly easy to understand why they would have missed it; even though it isn't an acceptable outcome. Especially given that communication in the 70s/80s involved a lot more talking to people and reading the right journals.

Indeed!

And so goes hindsight.

And we (the human race) have learned.

> And we (the human race) have learned.

yet, this is specific to this one incident, one specific threat.

The issue is more likely to be systemic. Once a production system is up and running, it is not consistantly being improved and iterated upon, but instead left to stagnate technologically (and in this case, in safety in light of new research etc).

But the argument is that such level of continuous improvement/iteration is costly, and the owners would not recouperate the costs! And so hence the economics goes, and so does the human condition continues to revolve around profit and loss.

Thus regulate mandatory re-certified every 10 years. Also that such process must involve meeting or exceeding //current// engineering standards and safety, with the potential for limited exceptions if approved by need and actively in progress plans for achieving the specifications.

Can't a new president quickly overturn such regulations?

Also, wasn't fukushima tsunami the 'biggest' in recent centuries ?

In Japan? Maybe. Globally? Not even close. As recently as 2004, there was a tsunami in the Indian Ocean that caused over 200,000 casualties (compared to the 15,000-20,000 casualties of the 2011 Japanese tsunami).

Death toll really isn't a good number to use for gauging the risk to nuclear reactors. Indonesia was more vulnerable to tsunami damage and deaths than Japan for basically the same reasons why Japan has nuclear power plants and Indonesia doesn't.

Sorry if it seemed insensitive, I didn't mean the most harmful, but the most powerful in terms of physics, as an engineering reference point if you will.

The tsunami which hit Fukushima was 43-49 feet tall. For seismic originated tsunamis, this is near the upper limit.

Not all tsunamis are caused by earthquakes though. On July 9th, 1958, there was a rockfall in Lituya Bay, Alaska. A cliff face broke off, and fell 3000 feet into the ocean. This resulting in a tsunami which stripped vegetation at up to 1720 feet in the surrounding areas.

https://geology.com/records/biggest-tsunami.shtml

Reply on news.ycombinator.com