5 ms·

I attended a talk by a nuclear expert at the University of Cape Town right after the Fukushima incident. He said that the risk of a Chernobyl given current hig

I attended a talk by a nuclear expert at the University of Cape Town right after the Fukushima incident. He said that the risk of a Chernobyl given current high safety standards was 1 every 20 years. At the end of his talk he remained bullish on nuclear, suggesting that doubling the number of nuclear plants would solve the coal problem. In the Q&A that followed, I asked him to confirm whether the 1 in 20 years probability of a Chernobyl scale disaster was correct and he said yes. I then asked him to confirm that his suggestion was to double the number of nuclear plants from around 250 to 500 and he said yes. I then suggested that this would increase the risk of a Chernobyl to 1 every 10 years. He was stunned. He had never put two and two together. A young nuclear engineer in the room quickly suggested that building more plants would increase safety...

davidivadavid 10y ago
That doesn't sound like a very well thought out argument. There's such a thing as a learning curve.
vidarh 10y ago
Given that the historical data we have give no basis for suggesting the risk of a Chernobyl-level event is as high as 1 every 20 years, I wonder why you didn't question that (yes, I realise we're talking averages, and you could suddenly get more, but fact remains that while our past experience is still short, we have nothing to suggest the risks would be that high, not least the knowledge that the number of plants of a flawed design like Chernobyl's is low, and that we know Chernobyl didn't fail under normal conditions). But, indeed, one every 10 years would still be far less destructive and lethal than continuing our current use of coal. Btw. we're well on our way to that 500 number - by 2014 there were 435 plants - so I guess we'll see soon enough. There is, however, every reason to assume that your flippant assumption that doubling the number of plants would double the risk is wrong, for the simple reason that the safety profiles of the older plants like Chernobyl and Fukushima is drastically different, and the then-current set of plants included a lot of old plants of old designs. Even for Chernobyl itself it took massive human error - the plant was intentionally and explicitly operating under non-standard conditions far outside of normal operating parameters as part of a test. Meanwhile many modern designs are getting close to walk-away safe.

davidivadavid 10y ago

That doesn't sound like a very well thought out argument. There's such a thing as a learning curve.

vidarh 10y ago

Given that the historical data we have give no basis for suggesting the risk of a Chernobyl-level event is as high as 1 every 20 years, I wonder why you didn't question that (yes, I realise we're talking averages, and you could suddenly get more, but fact remains that while our past experience is still short, we have nothing to suggest the risks would be that high, not least the knowledge that the number of plants of a flawed design like Chernobyl's is low, and that we know Chernobyl didn't fail under normal conditions). But, indeed, one every 10 years would still be far less destructive and lethal than continuing our current use of coal. Btw. we're well on our way to that 500 number - by 2014 there were 435 plants - so I guess we'll see soon enough. There is, however, every reason to assume that your flippant assumption that doubling the number of plants would double the risk is wrong, for the simple reason that the safety profiles of the older plants like Chernobyl and Fukushima is drastically different, and the then-current set of plants included a lot of old plants of old designs. Even for Chernobyl itself it took massive human error - the plant was intentionally and explicitly operating under non-standard conditions far outside of normal operating parameters as part of a test. Meanwhile many modern designs are getting close to walk-away safe.