Hacker News new | ask | show | jobs
by troupo 943 days ago
This was literally the definition of a black swan event: "an event that comes as a surprise, has a major effect, and is often inappropriately rationalized after the fact with the benefit of hindsight"
1 comments

No, Japanese authorities and TEPCo knew tgat earthquakes and tsunamis could happen, and usually together. They planned for it, they simply planned insufficiently, so quite the opposite of a black swan.
The definition of a catastrophic one-in-a-lifetime event: you can't plan for it.

Fukushima had been in operation since 1971. You truly really believe that there were no earthquakes in the meantime, and the only reason for failure in 2011 is that they didn't plan for the largest earthquake and one of the worst tsunamis in Japanese history?

Just found another highlight from the executive summary:

>> Prior to the earthquake, the Japan Trench was categorized as a subduction zone with a frequent occurrence of magnitude 8 class earthquakes; an earthquake of magnitude 9.0 off the coast of Fukushima Prefecture was not considered to be credible by Japanese scientists. However, similar or higher magnitudes had been registered in different areas in similar tectonic environments in the past few decades.

Now, tell me again, how the Fukushima earthquake and tsunami were totally impossible to predict events?

And it gets better:

>> In addition, a number of trial calculations were performed by the operator before the accident, using wave source models or methodologies that went beyond the consensus based methodology. Thus, a trial calculation using the source model proposed by the Japanese Headquarters for Earthquake Research Promotion in 2002, which used the latest information and took a different approach in its scenarios, envisaged a substantially larger tsunami than that provided for in the original design and in estimates made in previous reassessments. At the time of the accident, further evaluations were being conducted, but in the meantime, no additional compensatory measures were implemented. The estimated values were similar to the flood levels recorded in March 2011.

So, in 2002 they predicted pretry accurately the strength of potentail tsunamis. And then they did nothing for next nine years. The USSR knew about the control rod issue of RBMKs, and as TEPCO in Japan, did nothing until Chernobyl. See the inherent risk of nuclear power here? Because both, Chernobyl and Fukushima were pretty much predictable (as the risk was prpoerly identified before both accidents) and preventable if people in charge, and the organizations, would have put counter measures in place.

Edit: And now you can google the material difference between common and special causes.

>> The common cause failures of multiple safety systems resulted in plant conditions that were not envisaged in the design.

Well, according the IAEA, the planning against, and risk evaluation of, earthquakes and tsunamis for the Fukushima Daiichi NPP were insufficient:

>> The seismic hazard and tsunami waves considered in the original design were evaluated mainly on the basis of historical seismic records and evidence of recent tsunamis in Japan. This original evaluation did not sufficiently consider tectonic-geological criteria and no re-evaluation using such criteria was conducted.

Source:

https://www-pub.iaea.org/MTCD/Publications/PDF/Pub1710-Repor..., page 48.

Stop speculating and assuming, stop thinking from first principle and do the following instead:

- read text books on basic engineering, electricity generation and grid operation

- read some basics regarding nuclear reactors (or, in all seriousness, watch Chernobyl which provides some really good basic explanations)

- read the publicly available incident reports from the IAEA on thebaccidents of your choosing

All the questions you might have about any nuclear accident are answered in those reports. Those reports are prepared and investigated by experts in their field, going painstackingly over documents, event logs, design documents, meeting minutes (like a real and thorough investigation you know). Try understand all of that, and then draw whatever conclusion you want. Because by doing that, we have a basis for discussion, one rooted in reality in facts, as opposed to the headline-talking points fairytale stuff we have now, where people without any basic knowledge how things work draw assumptions from some headlines...

Once HN was a place where people applied critical thinking to topics, and where curious to learn from people knowing more (of which you find a ton around here). Since COVID, more and more people seem to think some general smartness allows them to understand even the most complex issues better than anyone else based on headlines, opinion pieces and social media talking points. I hope we can return to how things were before, it is getting really frustrating at times.

Edit: Some general engineering advice: Those once in a life time event are actually planned for, the USSR had rules for that in the 70s as well. The approach od identifying those risks was formalized as Failure Modes and Effects Analysis, FMEA. They are done for the design, production and operation, for processes, in order to identify those risks, the likelihood, the criticality and the ability to detect them. Following that, mitigation actions are defined and put into place. And, very important, they revisited regularly.

This wasn't done neither for Chernobyl nor Fukushima (just because the formal process didn't exist back then doesn't mean engineers didn't do those exercises, and the USSR had regulations in olace basically asking for the same thing. Regulations that were ignored by everyone for Cherbobyl unit No. 4). The IAEA reports for Chernobyl and Fukushima point thaz insufficient planning out explicitely, as do the USSR reports on Chernobyl. And if you think those fuck ups only happened in the distant past, the Boeing 737 Max can basically be traced back to the same root cause. Because these failures are not as much about people as they are about organizations. Hence processes and rules to follow in safety critical industries, and other large scale organizations. The cowboy style of operating in a start-up doesn't translate well to those places.

This is what I don't like about anti-nuclear FUDers: overconfident post-hoc rationalisation completely ignoring everything (context, history etc.).

Not a single one of these FUDers said anything about Fukushima in the 40 years of its operation in an active seismic zone (translation: multiple earthquakes) in a country that had at the time over 50 operating nuclear reactors.

But the 2011 happened. And look at them crawling out of the woods with armed with overconfidence, thinly veiled ad-hominems and TV shows.

Ad-homones and TV shows? I litterally cited the IAEA report on Fukushima... But I agree, cunducting regular audits, similar to after incident investigations, is something the IAEA should do. Similar to what the FAA and EASA do.

But let me guess, you don't even know where to find said IAEA report, do you?