4 ms·
Worth noting from a system safety perspective that it's not just about how to prevent the accident, but also how to limit the consequences once it does happen.
by kqr 4y ago
Worth noting from a system safety perspective that it's not just about how to prevent the accident, but also how to limit the consequences once it does happen.
From what I remember of the Fukushima accident, one of the quick remedies they could have started with immediately is pump sea water in as a coolant. The management didn't want to do that, because sea water corrodes equipment and makes it more expensive to restore afterwards.
This is a natural thing for management to say -- they're sitting there in Tokyo not fully experiencing the severity of the problem.
The problem is that Japanese culture puts an emphasis on obeying your superiors, so it took a long time until people disregarded those orders and started cooling with sea water anyway. That sort of obedience culture is not really compatible with systems designed to suffer rare but potentially catastrophic failures.
- LargoLasskhyfv 4y ago> they're sitting there in Tokyo... I wonder if that had been the case if the wind had blown in another direction at the time, towards Tokyo, instead of mostly the open sea. I also think it's difficult to come to conclusions in seeing this in an isolated way. Everything was in chaos and much destroyed because of the tsunami, surely that delayed communications, and assessment of the 'real' situation? Sometimes when I read about the history of all that, and all the talking about, I get the feeling like the commentariat views this like a clean room laboratory exercise. Like in a model, this happened, what to do now? Chaotic reality doesn't work like that.
- kqr 4y agoThe reason I brought that specific thing up is that it's not an isolated thing. It happens over and over in complex situations when the system is top-heavy in its steering. In any complex situation, you need to have people on the ground with the trust and authority and safety to do what they think is best, precisely because of the reasons you mention with the chaos and delayed communications, and so on.
- panick21_ 4y agoWe would have great scientific data by now how the deadly nuclear cloud isn't actually that deadly. But it would have caused panic and that wouldn't have been good.
- LargoLasskhyfv 4y agoI'm not so sure about that. https://en.wikipedia.org/wiki/USS_Ronald_Reagan#2011_deployment https://en.wikipedia.org/wiki/USS_Ronald_Reagan#2011_deploym... https://en.wikipedia.org/wiki/USS_Essex_(LHD-2)#2011 https://en.wikipedia.org/wiki/USS_Essex_(LHD-2)#2011 Some searching for 'Fukushima US-NAVY cancer' will bring up endless stuff from all sorts of sites. The thing is, regardless of the sites reputations, there is a signal that those ships took a large dose, because some of their sailors claim their sudden leukemia was caused by that. I don't know. Can one call that psychosomatic? Did they sniff too much JP-7 instead? IMO that signal is too strong to be disregarded.
- walnutclosefarm 4y ago> From what I remember of the Fukushima accident, one of the quick remedies they could have started with immediately is pump sea water in as a coolant. The management didn't want to do that, because sea water corrodes equipment and makes it more expensive to restore afterwards. Well, not really. Some of the early damage might have been prevented by this (at the cost of destroying the reactors), but Fukushima really had no way to reliably cool their reactors without auxiliary power. Once that was gone, there was going to be serious trouble, and likely destruction of some of the reactors.