You're probably looking at best-case vs. worst-case here. The expense of getting sufficient wind power in areas without wind or solar power in low-light areas can get pretty crazy. And a large portion of the expense of nuclear power is regulatory hurdles we impose on it.
> Its failure modes have so much more potential to be catastrophic than the alternatives.
The success mode of the current most popular alternative (fossil fuels) has proven much more catastrophic. As for other things like wind and solar power, I think a lot of people are unconvinced that it's fair to call them "alternatives" yet. Nuclear has proven itself as a good general-purpose solution and they have not, and nuclear seems to be a lot better than what we are doing now.
Hydro power is responsible for the most fatal electricity generation incident ever: Banqiao Dam. It catastrophically failed and killed an estimated 171,000 people.
Fukushima was a flea bite by comparison.
> According to the Hydrology Department of Henan Province, in the province, approximately 26,000 people died[14] from flooding and another 145,000 died during subsequent epidemics and famine.
Video with computer simulation: https://www.youtube.com/watch?v=ctPLXim9WG8
It's also not particularly environmentally friendly to dam rivers everywhere.
Some of the wave-action stuff is interesting, but you've obviously got to be near a large body of water for that to work.
Technically, no. Artificial hydro (more commonly known as pumped storage) means you can essentially store energy in gravity, anywhere. i.e. build a container at an elevated position, pump water into it using intermittent energy (e.g. wind at night, excess solar etc), drop water down through a turbine to generate electricity again to provide a low baseload.
The reality is that 1) this is hugely expensive and so generally pumped storage is restricted to exploit existing natural resources with lots of elevated mountains between which one can store ridiculous volumes of water rather than build your own giant water container.
So generally hydro is limited to existing water resources (e.g. large rivers), or storage resources (e.g. mountains) which are quite limited.
Costa Rica uses its large amount of rainfall and dams to use generate hydro power, and this article discusses a drought they had in the last year that meant they had to fire up the old fossil fuel plants: http://www.theguardian.com/commentisfree/2015/mar/30/truth-b...
So I'd still consider having some base load capacity available fairly important.
[citation needed]
Your metrics for "success" are interesting:
> It's more expensive than the alternatives, including wind and solar now.
According to [0] only onshore wind beats nuclear in terms of cost. However, that does not include the scaling issues wind has if you want to produce as much total MWe as a nuclear reactor (ex: there is only so much high quality wind onshore land).
> Its failure modes have so much more potential to be catastrophic.
This is the same logic for many anti-terrorism bills, but I will ignore that for now. This is also why they are regulated to such lengthy extremes, such as having backup power both onsite and offsite. It is also required to withstand winds that occur only once in 10,000 years.
> The industry has a terrible record of cheating on nuclear waste disposal. Much of it is dumped at sea in secret because the cost of legitimate disposal is so expensive it would make the industry commercially unviable.
Sources? Much of it I know of is stored on site in giant pools because the waste belongs to the DoE, and the DoE still cannot find a compromise on how it should be handled.
> Human error is common and impossible to eliminate. The nuclear industry is somewhere the human factor has been historically disastrous and it doesn't seem likely that will change.
Replace "nuclear" with anything else and it is just as true. For example, structural engineering has had millenia lengths of time over nuclear to rack up disasters.
Ultimately, nuclear accidents amplify human error to a magnitude that many people find uncomfortable. Which is why the industry is regulated to a magnitude more than any other. It is not like they hand out reactor operator licenses to just anybody, either. 1 out of my undergraduate peers of ~7 got licensed to operate NCSU's PULSTAR research reactor, which has much simpler systems than a commercial power-producing reactor. It is also easy to forget that the nuclear industry in the USA has been near-paralyzed at the 1950's and 1960's level of technology, which is only about 20 years into the commercial reactor field. Unfortunately, there really is not any other industry I know of that, at 20 years into its growth, was paralyzed and allowed to operate for an additional 60-70 years. In that context, I think the US has done well given the little wiggle room regulation has allowed growth in.
[0] https://en.wikipedia.org/wiki/Cost_of_electricity_by_source#...
Energy Source Death Rate (deaths per TWh)
Coal – world average 161
Coal – China 278
Coal – USA 15
Oil 36
Natural Gas 4
Biofuel/Biomass 12
Peat 12
Solar (rooftop) 0.44
Wind 0.15
Hydro 0.10
Hydro - world including Banqiao) 1.4
Nuclear 0.04
Note that rooftop construction is one of the most dangerous jobs in the US, with high mortality rates from falls.[1]Coal mining is extremely hazardous, it would top this chart even just looking at production. Folding in air pollution deaths from burning coal and you add thousands of deaths per day.[2]
Each year coal and oil kill more people than the entire 60 year history of nuclear, including every nuclear disaster.
Even if we focus on worst case scenarios, nuclear is safer. The impact of Chernobyl is wildly overestimated by the public.[3]
For hydro, let's just note that a dam failure in Banqiao killed 171,000. The highest death estimates for Chernobyl (and this is including expected early deaths from radiation exposure that still haven't appeared a few decades after the incident) and you're looking at maybe 4,000.
You could add a Chernobyl every year, even take the most aggressive estimates for early deaths caused by that accident, then ignore all air pollution deaths, and coal and oil would still be more dangerous. They're just that insanely deadly an industry. Nuclear's worst case is safer than coal and oil's average.
On cost I think you're mistaken too, nuclear remaining an incredibly cheap source of power per watt, but power economics are complicated and that'd take another discussion. All I can say is that this was the standard wisdom of everyone I knew who worked in the energy industry, despite occasional scare stories in magazines about how the "true" cost of nuclear is disguised by subsidies - the math doesn't quite add up.[4]
[0] Energy deathprints: http://nextbigfuture.com/2008/03/deaths-per-twh-for-all-ener...
[1] http://www.bls.gov/iif/oshwc/cfar0020.pdf
[2] Studies on air pollution deaths (spoiler: among the leading contributing factors to mortality worldwide): http://www.thelancet.com/journals/lancet/article/PIIS0140-67... ; http://www.nature.com/nature/journal/v525/n7569/full/nature1...
[3] An overview of the actual mortality impact of Chernobyl found it responsible for a few dozen deaths in emergency workers (fewer than 50, and some of those dying 20 years after the accident). It might be possible to eventually tie a few thousand early deaths to the incident, but we're still waiting, decades after it happened. In general, radiation hazards were widely exaggerated. http://www.who.int/mediacentre/news/releases/2005/pr38/en/
[4] Admittedly biased source, but it runs a few publically available numbers, even nuclear projects with cost overruns look better than some renewable projects: http://nuclearradiophobia.blogspot.com/p/cost-of-nuclear-pow... I mean, ideally you get a mix of both sources, because they serve different purposes on the grid, but your bursty diurnal power isn't meant to serve as your cheap main source of power.
The WHO gives the number as "up to 4000 could eventually die" [2].
And some researchers give the number as "985,000, mostly from cancer" [3].
Since it's so hard to tally long term harm from such a hard to pin down source such stats are very hard to accurately quantify.
[1] https://en.wikipedia.org/wiki/Deaths_due_to_the_Chernobyl_di...
[2] http://www.who.int/mediacentre/news/releases/2005/pr38/en/
[3] http://www.globalresearch.ca/new-book-concludes-chernobyl-de...
The 985,000 figure comes from an interesting source. It originated in a book that peer reviewers savaged, for, among other things, making up numbers without providing any justification whatsoever and ignoring the scientific method.
M. I. Balonov decried the book, claiming "it has very little scientific merit while being highly misleading to the lay reader." He called it science fiction.
https://en.wikipedia.org/wiki/Chernobyl:_Consequences_of_the...
While you're right that there are hard problems in long term forecasts, but there are entire fields of study, dosimetry and epidemiology, that have had smart people toiling away on those problems for their entire career, and this is maybe the most widely studied industrial disaster in history. We have made enormous strides in how to measure environmental radiation, doses received, and the correlation between those and increases in mortality. Individuals are hard, but population estimates are actually easier to calibrate over time. So if you write a book containing a sensationalist number to get some headlines, researchers will be able to tell you're pulling their leg.
As rtpg says though, this is a bit of a moot point, because, sure, let's call it 985,000, then nuclear remains still the safest form of power by an enormous margin.
* It's more expensive than the alternatives, including wind and solar now.
* Its failure modes have so much more potential to be catastrophic.
* The industry has a terrible record of cheating on nuclear waste disposal. Much of it is dumped at sea in secret because the cost of legitimate disposal is so expensive it would make the industry commercially unviable.
* Human error is common and impossible to eliminate. The nuclear industry is somewhere the human factor has been historically disastrous and it doesn't seem likely that will change.
There's really not a lot to like about the nuclear power industry. Can anyone explain why we shouldn't just leave this method of power generation behind as a tragic mistake?