To add to your examples, there was the very real possibility with Fukushima that Tokyo would have had to be evacuated. This was seriously considered at the highest levels of the Japanese government (i.e. the PM and cabinet):

> The 400-page report, due to be released later this week, also describes a darkening mood at the prime minister’s residence as a series of hydrogen explosions rocked the plant on March 14 and 15. It says Mr. Kan and other officials began discussing a worst-case outcome if workers at the Fukushima Daiichi plant were evacuated. This would have allowed the plant to spiral out of control, releasing even larger amounts of radioactive material into the atmosphere that would in turn force the evacuation of other nearby nuclear plants, causing further meltdowns.

> The report quotes the chief cabinet secretary at the time, Yukio Edano, as having warned that such a “demonic chain reaction” of plant meltdowns could result in the evacuation of Tokyo, 150 miles to the south.

> “We would lose Fukushima Daini, then we would lose Tokai,” Mr. Edano is quoted as saying, naming two other nuclear plants. “If that happened, it was only logical to conclude that we would also lose Tokyo itself.”

https://www.nytimes.com/2012/02/28/world/asia/japan-consider...

> Because its so simple as saying, "the wind COULD have blown the fukushima fallout southwards over Tokyo instead of relatively promptly into the Pacific." A nuclear accident COULD deliver fallout to a highly populated area that cant be evacuated in time.

So the thing that actually happened with Chernobyl then, not something significantly worse than that.

> re: desalination - fine then, add the cost of desalinating salt water to replace the increasingly scarce resource that nuclear plants evaporate, to their already uncompetitive cost.

Evaporating the water is how you desalinate it. It starts as seawater, it evaporates, you recondense it as fresh water. The equipment needed to do that costs less than the value of the water; the major expense is generating the heat which you're already doing for power generation. So I guess we could subtract the net profits from producing that valuable side product from the cost of the power generation, sure.

Do we then get to add the costs of climate change to the cost of burning coal, or are we only attempting accurate accounting for nuclear and not anything else?