Out of all major energy sources, nuclear is the safest

I guess you could argue that GPS has killed as many children now, by way of guided missiles and missiles fired by drones - but there's still a gulf between unleashing terror on a divine scale as a part of a publicity stunt and geopolitical maneuvering, to making warfare incrementally more efficient.

https://lh3.googleusercontent.com/-Ndd82xj3j1o-3KcyvBkLTB1mR...

In 1975, it was the site, or perhaps more accurately, region, of the worst power plant disaster in all history: the Banqiao dam failure. News of this only fully emerged after over two decades.[1] You can spot the reservoir itself at the far left of the images, at mid-height.

In the disaster, a confluence of events lead to the deaths of approximately 171,000 people, with 11 million displaced. There's considerable uncertainty in those numbers.

The causes were multiple: siting, improper engineering, unheeded warnings, a (literal) perfect storm (tropical typhoon striking a cold front and lingering over the region for a full day, dropping over 1 meter of rain), improper emergency plans, failed communications, situational confusion, nightfall, and a hopelessly inadequate response and recovery. Of the deaths, "only" -- a term used advisedly -- 25,000 or so were due to direct flooding. The remaining 150,000 or so succumbed to starvation or disease in the weeks following the events.

And yet: the book as been closed. The cities in the floodplain are rebuilt. The dam itself has been rebuilt. Over 7 million people live in Zhumadian City, 95 millions in Henan Province total.

There is no disaster exclusion zone.

There is no disaster exclusion zone which will persist for the next three centuries.

There is no molten reactor core.

There is no coreium.

There is no radioactive waste which will persist for 10,000 to 1 million years.

The book is closed.

Proponents of nuclear power assume that we can assess risks with tails not of the decade or so of Banqiao, but of 100, 1,000, 1 million years. Utterly outside the scope of any human institutions, or of the human species itself.

Our models of risks and of costs fail us.

(They've failed us as well in the case of fossil fuels, and, quite possibly, for hydro power -- I'm not giving this example as endorsements of either, but to give the story of risk and closure, or its lack. Those are other stories, for other posts.)

The problems with nuclear power are massive, long-tailed, systemic and potentially existential. The same cannot be said of a wind farm or solar array. There is no significant 10,000 year threat from wind power, or solar power. We're not risking 30 - 60 km exclusion zones, on an unplanned basis, of which we've created at least four in the half-decade of significant nuclear energy applications: Hanford, Washington, Three Mile Island, Pennsyvania, Chernobyl, Ukraine, and Fukushima, Japan. And this is with a global plant of some 450 operating nuclear power plants as of 2017[2]

(This compares with over 7,600 power plants in the United States alone.[3])

None of these sites has been fully remediated. In the specific case of Hanford, the current management plan is budgeted at $2 billion, and there is no final management plan in place. This eighty years after the facility first opened.

If the total experience has been, say, 500 reactors, over 50 years, or 25,000 reactor-years of experience, and we've experienced at least four major disasters, then our failure rate is 0.016%.

The global share of nuclear power generation in 2012 was about 10%.[4] Which means that without allowing for increased electrical consumption within existing or extending to developing nations, the plant count would have to increase tenfold.

Holding the reactor-year failure rate constant would mean 80 core meltdowns per century.

Reducing that to the present rate of four meltdowns/century would require reducing the failure rate to 0.0008%. That's five nines, if anyone's counting.

Five nines on a process involving weather, politics, business, social upheaval, terrorism, sabotage, individual psychology, group psychology, climate, communications, response, preparedness....

And ... the involvement of the Japanese Mafia, the Yakuzi, in the management of TEPCO, who operated the Fukushima nuclear power plant.[5]

All of which played a tremendous role in how badly the Banqiao disaster itself played out -- everything which happened at Banqiao by dynamics could just as well have happened in a nuclear plant.

But it wasn't a nuke, it was a dam. And after a few hours, the waters receded, and after a few weeks, the land dried, and after a few months, recovery could start, and after a couple of decades ... even in what was still a poor country ... the recovery was complete.

Banqiao was a disaster, no doubt.

But what it wasn't was a nuclear disaster.

________________________________

Notes:

0. https://en.m.wikipedia.org/wiki/Zhumadian

1. http://www.sjsu.edu/faculty/watkins/aug1975.htm

2. https://www.nei.org/Knowledge-Center/Nuclear-Statistics/Worl...

3. https://www.eia.gov/tools/faqs/faq.php?id=65&t=3

5. http://www.iea.org/publications/freepublications/publication...

4. https://www.theatlantic.com/international/archive/2011/12/ya...

It has significant federal R&D and liability-related subsidies, as well as other federal and, in several states, state-level subsidies. Nuclear is by no means deprived of subsidies.

If it really is more along the lines of nuclear is safe, but once every five years a worker at the plant might slip up and be exposed to dangerous levels by accident, then I guess I think the comparison makes more sense.