[baron816]

I’m not worried about how many people nuclear power kills every year. I’m worried about how many people it kills every hundred years. A major nuclear disaster, however unlikely, has the potential to kill millions of people and leave a big portion of the Earth uninhabitable for centuries. Just because we haven’t seen one on that scale yet, doesn’t mean we never will.

[thaumaturgy]

Ok, let's do a Fermi-like estimation.

Estimates on the number of deaths related to coal generation vary from around 13,00 to 30,000 per year in the US [1] and 500,000 per year in China [2]. The current world population is 7.6 billion, of which the US and China account for approximately 1.8 billion people. Let's round the coal-related death rate way down for easy math: 10,000 per year in the US and 100,000 per year in China. Then, multiply that rate by the world's population, and you have, let's say 450,000 coal-related deaths per year worldwide. This is a really squishy number, but we only need approximations here.

You were concerned about how many people would die every hundred years from nuclear disasters, so let's see if we can work today's 450,000 per year estimated deaths backwards for the last hundred years.

The world population was somewhere in the neighborhood of 1.8 billion people in 1917. Assuming linear growth (I know, I know) and a strong r-value correlation for population vs. coal-related deaths (arguable, but again, Fermi estimate), we have to sum .00006 * population from 1.8 billion -> 7.6 billion, and we end up with approximately 28 million people.

Which is to say, if we could gather up all the deaths, worldwide, from coal, over the last hundred years, and convert it into a single disaster, it would kill the entire city of Shanghai, and New York for the apple on top.

That would have to be one hell of a nuclear disaster.

Now, there are arguments to be made that the energy we've received from coal has also powered hospitals and technology which have saved or improved people's lives. There are also arguments to be made that the side-effects of coal (hospitalization, environmental disasters) cause the death toll to absolutely pale by comparison.

And again, I've rounded these numbers down at every stage of the calculation.

[1]: http://www.catf.us/resources/publications/files/The_Toll_fro... [pdf]; it includes its own numbers, at the 13,000 estimate, and the EPA's, at 14,000 to 36,000 range.

[2]: https://www.nextbigfuture.com/2011/03/deaths-per-twh-by-ener... (blog, but includes lots of supporting links; I'm open to alternative sources that give vastly different estimates).

[nl]

There are very few people who are anti-nuclear power who are pro-coal.

Even ignoring renewables, gas solves many of the worst problems with coal.

[thaumaturgy]

Sure, but even natural gas and rooftop solar have higher rates of death per year than nuclear. (In my second link above.)

In a world in which we must choose the lesser of evils for energy, nuclear is among the least evils, yet faces the greatest overall public resistance to new installations.

[nl]

I dunno - even that source itself says that the death rates from solar would fall using better construction methods. It also notes (in the case of wind) that increased take up is associated with lower death rates ("Wind power proponent and author Paul Gipe estimated in Wind Energy Comes of Age that the mortality rate for wind power from 1980–1994 was 0.4 deaths per terawatt-hour. Paul Gipe’s estimate as of end 2000 was 0.15 deaths per TWh, a decline attributed to greater total cumulative generation.")

I'd also note that the very low rates of death from nuclear power do not appear to include construction deaths, which are the only source of death measured from wind and solar.

[trhway]

>even natural gas and rooftop solar have higher rates of death per year than nuclear.

the difference here is that no event can significantly change the death rate from gas or solar whereis one wrong cough by nuclear power plant worker and we have Chernobyl with thousands of deaths (in particular in Belarus which took the majority of the Chernobyl hit - about 5000 extra thyroid cancer cases (normally a rare cancer) in the people who were children at the time plus doubling (and in some areas tripling) of the rate of the most frequent cancer - breast cancer (typically 1 in 7 women would get it during lifetime, so doubling means additional 15% of all the women in the affected area would get it) in the areas close to Chernobyl like the Gomel and Mogilev regions plus very significant increase in other frequent cancers of internal organs like colon,etc.)

Or counting it the other way - Russia and Ukraine both have the same - 0.0034 - incidence rate of cancer per year (despite the war and economic differences it is the same people with the same behavior/habits :), while Belarus where people are basically the same as in Ukraine and Russia and drink and smoke and eat the same - has 0.0052 incidence rate, ie. 50000 new cases per year instead of 34000 if they were to have the same incidence as Russia and Ukraine. 16000 extra cancer cases per year for several decades with mortality higher than 50% ...

>nuclear is among the least evils,

hardly so, giving the numbers i referenced above (compare it to coal - the coal's 700K/year deaths means "only" 1K/year for Belarus) and potential of any nuclear power plant to repeat Chernobyl - i'm aware about technical improvements of modern reactors, yet "stupidity will find a way"...

> yet faces the greatest overall public resistance to new installations.

still wondering why ? :)

[nv-vn]

Being anti-nuclear is pretty much accepting the status quo though. Modern nuclear power is much, much more sustainable than coal, scales very well, and is very safe. If you believe in a quick solution for energy issues then you shouldn't halt all nuclear development and leave fossil fuels for the next couple decades.

[nl]

Being anti-nuclear is pretty much accepting the status quo though.

Elon Musk would disagree.

If you believe in a quick solution for energy issues then you shouldn't halt all nuclear development and leave fossil fuels for the next couple decades.

Nuclear plants take 30 years at least to get through planning and building. Gas fired plants or pump solar can be deployed within a year, Wind + Solar in even less time.

[wiz21c]

Also there was an interesting view on (against) nuclear by Naomi Klein :

https://www.youtube.com/watch?v=XLRikyCRu54

I like nuclear, but her view makes a lot of sense to me.

[toomuchtodo]

Nuclear will never be able to compete with renewables on cost or deployment speed. It’s dead tech for anything other than space probes and military uses.

[nv-vn]

How? Nuclear tech is already much more developed than other renewables and it's incredibly efficient. The cost of solar is much higher than the cost of nuclear. Hydroelectric and wind are both lower, but they also require certain geography and can't be used anywhere. Nuclear is not dead tech at all, most of France's power comes from nuclear.

[nl]

France is closing at least some its nuclear power stations (and has no plans to build new ones): https://www.reuters.com/article/us-edf-nuclear/france-to-clo...

(Also, nuclear is not a renewable by an sensible measure)

[distances]

Nuclear is very slow to build in the Western world. Finland is already in their 12th year building Olkiluoto 3

[nl]

Is that 0.04 deaths/TWh sourced anywhere?

I've seen it before, but trying to trace it down is... troubling.

There's this report[1], where they only count accidents where more than 5 people were killed. This coincidentally manages to avoid counting all accidents in nuclear plants, where in the US alone there were 8 deaths[2]. I assume other energy sources would see a jump too, but that doesn't give me great confidence. The number from that report is different anwyay.

[3] http://wps.prenhall.com/wps/media/objects/2513/2574258/pdfs/... references some authors which appear to have this 0.04 number, but I can't find the source.

[1] https://www.oecd-nea.org/ndd/reports/2010/nea6861-comparing-...

[2] https://en.wikipedia.org/wiki/Nuclear_reactor_accidents_in_t...

[3] http://wps.prenhall.com/wps/media/objects/2513/2574258/pdfs/...

[xelxebar]

"However unlikely"

This is where probabilities bite us. If the probability is small enough then the expected death toll will drop to almost zero which merits little worry, despite how horrific said event mightt be. It sounds like you're doing the natural, intuitive thing of picturing a very scary scenerio and then wanting to avoid said scenerio at all costs.

The problem is, while you're busy avoiding some almost-zero probability event, you're also nuking your ability to mitigate events with actual high expected deaths and loses.

Fossil fuels already kill on the order of millions per century [citation needed], so if you think nuclear power does the same, it means we should expect several Chernobyl level events in the next few decades. Are you sure your probabilities are well-calibrated here?

Also, some of the rough skimming I've done has me believing that cycles without possibility of medown can be built, cf. liquid salt reactors. IIRC, some of these even have really nice waste profiles with the worst stuff only having half lives in the decades and very low volume at that [citation badly needed].

Anyway, fear of nuclear power reads to me like a fear of planes over cars. Chernobyl and Fukushima were single, isolated events making them visible and viscerally scary like plane crashes. Fossil fuels, on the other hand, pick us off one by one, relentlessly, but never so loudly as to evoke that gut-level fear.

[wruza]

Every year it kills 100 times less than every hundred years, that’s how average works, no matter what exact period you’re worried about. Also, no nuclear disaster ever killed millions — it is barely possible even with nuclear weapons. Relocation and financial risks, yes.

Likelihood of such disaster is estimated to be too minor to not care about, say, “an asteroid kills the planet” movie scenario.

https://www.nbcnews.com/news/us-news/scientists-say-nuclear-...

http://www.ucsusa.org/nuclear-power/nuclear-waste/safer-stor...

[Tarrosion]

By nuclear disaster do you mean weapons use or just a power generation accident? I've never heard of a power plant accident potentially leaving a big portion of the Earth uninhabitable for centuries, but if there is something please enlighten me!

[baxtr]

Uninhabitable? Maybe for humans. Wildlife grows back stronger:

http://www.telegraph.co.uk/news/2016/04/23/wildlife-returns-...

[StanislavPetrov]

The fact that toxic radiation is less lethal than humans is an indictment of humans, not an endorsement of radiation.