The Shocking Reasons Why We Should Go Nuclear

I once wrote about how we aren’t really doomed, per se. Things could get bad, even really bad, owing to climate change and irrational energy and environmental policies, even economic crashes, even war, even worldwide NBC war. But it won’t be an apocalypse. It won’t wipe out the human race. It won’t even likely send us to Road Warrior for a century or two; although even if it did, that would just be a brief dip in the upward march of progress of human civilization in its four thousand year history—since we’d just recover eventually and start marching even further on; like happened in the Middle Ages: an advanced civilization collapsed, but eventually was restored, and then even its scientific and technological advances were far exceeded.

That’s still true. But one thing I said in the course of making that point is this:

[The] innovation of solar-thermal power is actually a better alternative to nuclear (and won’t be blocked by superstitious fear), and such plants are now starting to appear in the U.S. and will inevitably replace all coal and natural gas plants within fifty to a hundred years.

Well, turns out, that’s not quite true. Sure, we may expand and benefit from solar-thermal. But it actually isn’t better than nuclear. And actually, any sane energy policy would devote considerable resources into expanding our nuclear power infrastructure instead. The people’s fear of it just has to go. This is, after all, the same ignorant populace that is so scientifically illiterate they think “irradiated food” will be radioactive or poisonous; when in fact it would be neither, and actually save the lives of thousands of people every year by nearly ending food-born illnesses and extending the shelf-life of perishables. By the same reasoning, we need to replace the bulk of our power generation with nuclear plants. And we need that badly. Because all the other options, “(like solar, wind, seismic, and thermal) and alt fuels (like algal biofuels, hydrogens, and other artificial fuels),” even smarter uses of old-school petrofuels (from coal to oil to gas), are actually super shitty. Yes, even solar and wind. Super, super shitty.

Let me explain. Because the explanation might shock you. For you might not have thought of it. Yet it’s almost face-slappingly obvious.

The Problem in a Nutshell

This all came about when discussing energy policy with a savvy Canadian engineer one day as I was staying at his place. He was a lovely host. And also knew his shit. He had the research to back up his claims. And I followed up even to double check. And it checks.

Of course, no surprise Canadians know more about this stuff than Americans. Americans are idiots. Almost half of us still want to elect Donald Trump to be our President. We’re not, let’s be honest, smart.

But back to the point. We’ll start with the example of solar power. That’s awesome, right? Zero pollution, doesn’t hack the heads off of birds, just sits there and generates totally free energy from the sun. Wellll… About that zero pollution thing. Uh, actually, making solar panels pollutes the shit out of the planet; recycling them is not an option at this stage (since we need vastly more of them than there are to recycle), nor is that pollution free, either; and maintaining them consumes energy and resources, and that also generates pollution down the line. It’s also expensive. The amount of energy that goes into both making and maintaining solar panels is so enormous, that the net gain we get back from those panels before they expire is paltry. Shockingly paltry.

The same plays out for wind power or any other energy source: when you step back and count the resources and energy consumed, and pollution generated, making and maintaining every other source of alternative power, from wind to solar, even solar thermal, the same shocking result comes out of the math: they all suck. Well, actually, hydro doesn’t suck so much, but alas, its capacity potential is woefully insufficient to solve our energy needs (there just aren’t that many rivers to damn; even if damning rivers wasn’t environmentally destructive, a serious downside to hydro people often forget). Ditto geothermal, where though capacity potential may be theoretically adequate, the costs of making and sustaining such plants on the required scale is on the order of ridiculous—which is not to slight its tremendous advantage where it’s most easily tapped—just like hydro, use it when you can; but it’s not going to supply the world.

People often forget, building and replacing solar panels or wind turbines requires mining (which entails toxic waste, pollution, energy costs, and environmental damage), manufacturing (which entails toxic waste, pollution, energy costs, and environmental damage), transportation and emplacement (which entails toxic waste, pollution, energy costs, and environmental damage), maintenance (parts and labor and cleaning, which collectively entail toxic waste, pollution, energy costs, and environmental damage), and disposal or recycling at end-of-life (which entails toxic waste, pollution, energy costs, and environmental damage).

It also entails hidden costs in power storage. Solar panels generate only during the day (and wind likewise only blows some of the time), but you need some of that power at night (likewise when the wind isn’t blowing), so you have to store it; but batteries or any other storage system costs again in mining, manufacturing, transportation and emplacement, and disposal or recycling; and they also lose a lot of what they hold (no battery system is 100% efficient; especially when you count the energy cost to make them, and their variable efficiency at temperature), so they waste energy, too (and hence, they also are consuming energy).

(The delivery system, e.g. the U.S. power grid, also wastes energy, as all those miles of wires lose electricity as waste heat etc., but as that’s the same loss for all generation types, it doesn’t count against any one power source over another, so we can ignore it for the purpose of comparing the sources of power generation.)

We forget to add all that up. And when we do, solar panels start to look like a really shitty idea. They have their uses (e.g. it can’t hurt to roof your house with some, since it’s not like you’d realistically line your roof with natural gas combustion generators instead), but they really truly suck for large scale energy solutions. On such scales they are bad for the environment. And they waste almost as much as they produce. Wind turbines are only marginally better. So, what then?

Oh Right, We Have to Subtract the Energy Put in from Energy Produced

This is all analyzed in such research reports as “Energy Intensities, EROIs (Energy Returned on Invested), and Energy Payback Times of Electricity Generating Power Plants” by Dr. Weißbach et al. There, for example, they find that on just energy return alone, “nuclear, hydro, coal, and natural gas power systems (in this order) are one order of magnitude more effective than photovoltaics and wind power.” They find that we get 28 times more energy out of a natural gas system than we put in (counting all energy costs to create and sustain it). Solar panels, even in the best cases, get us only 4 times more energy than we put in; and most systems don’t even get twice as much out as we put in! Solar panels, in other words, royally suck as an energy source. They waste far too much energy just to exist. They have their own unique advantages (e.g. they are quiet, they don’t locally pollute, etc.). But they cannot, on scale, solve the world’s energy needs. That would be the dumbest policy ever. There is a reason gas and oil still kick solar’s ass: they are just vastly more efficient in practice.

Solar thermal gets us a return of between 8 and 21 times the energy we put in (or 9 to 19, depending on which data you look at). Which is way better than solar panels, but still not as good as natural gas. Solar thermal also consumes considerable quantities of water—though much of what it consumes is reusable, and it’s comparable to nuclear in this regard if we measure it against the well-known and still widely used light water reactor model; new nuclear reactors consume less water (I’ll get to that).

So apart from the water consumption, I was right that solar thermal was the shit. Sort of. Compared to solar panels anyway. But you can see why it doesn’t quite compete with gas, which on average doubles the output. Wind power gets us 4 to 16 times the energy out that we put in, so not exactly good competition either. But notably, better than solar panels; although wind doesn’t blow enough everywhere. But contrast hydroelectric: 35 to 50 times the energy out as in! Even when this requires building and maintaining giant damns (it doesn’t always), hydro kicks even gas’s ass. But alas, as noted, there just isn’t that much hydro energy to capture. For comparison, coal is about an even 30. So it’s already comparable, and far more plentiful, even still after all the extraction we’ve done so far.

So, what about nuclear? Guess what. It gets us 75 times more energy than we put in. And yes, that even includes all the energy costs of mining (on the front end) and waste storage (on the back end). And with known advances in the pipe, that figure will exceed a factor of 100 in a few decades. This kicks the ass of every energy source we have. And it’s almost endlessly scalable. Nuclear power thus returns over three times more energy than even the most efficiently emplaced solar thermal. Five times wind. Twice hydro. And forty times more than any realistic solar panel solution. The following table (from the Weißbach paper, p. 29) illustrates the significance of this:

Table of Energy Returned on Invested, showing EROI unbuffered and EROI buffered, buffered meaning using storage tech to account for when power is generated vs. when it is needed. The economical threshold is at around seven. The bars for unbuffered show solar panels at 3.9, biomass (corn) at 3.5, wind at 16, solar thermal at 19, gas at 28, coal at 30, hydro at 49, nuclear at 75, and buffered it's the same except for solar panels, which drop to 1.6, wind, which drops to 3.9, and hydro, which drops to 35.Note that “economical threshold” bar. What does that mean? It means that if that energy source is relied on in the aggregate for our nation’s electrical power needs, and yet its energy return is below that amount (which is a factor of about 7 times, energy produced to energy put in) we will actually be wasting so much money generating the energy we use to run our economy that our economy will tank. Notice this nixes solar panels. They are far too wasteful to even be useful as an energy policy solution. If we tried it, we’d crash our economy.

Likewise biofuels. And also wind; though unbuffered wind power can beat the mark, most wind power has to be buffered—because the wind doesn’t blow all the time, so you have to store the energy you need, for when the wind isn’t blowing, and that adds economic costs that plummet wind power’s utility. Solar thermal is just barely useful. Gas and coal are great (hence why we are using them). Hydro is better, but again, not a practical solution, owing to limited emplacement. But nuclear rocks over everything, twice as good as even hydro, more so coal and gas, and far better than solar thermal, and vastly better than wind (even if we could get wind to always work at its max, and we can’t).

Pollution & Environment

And that’s just by energy cost. If we factor in pollution and other environmental costs—e.g. coal and gas both increase global warming; and to an extent so do the others, due to the carbon emissions they entail at the stages of manufacturing, maintenance, transport, disposal, etc.—alternative energy sources don’t look so great as you may have imagined.

This is particularly important when considering nuclear. Everyone freaks out at the prospect of storing all that radioactive waste. But like people who freak out about terrorist attacks but not about driving (despite vastly higher odds of dying from the latter), this is an irrational focus on the sensational at the expense of reality. In reality, the pollution caused by the manufacturing and deployment of solar power is worse than nuclear waste. It’s greater in quantity (per terawatt hour); it’s greater in lethality (per terawatt hour); and it’s greater in personal odds of exposure (just like you are very unlikely to be present at a terrorist attack, you are very unlikely to be affected by nuclear waste; whereas you are far more likely to be affected by all the air and water pollution caused by solar panel manufacturing).

And that’s not all. See, for example, this analysis of Lifetime Deaths per TWH from Energy Sources. Here is one table produced there (the data is crude but the proportions realistic):

Table showing deaths per terawatt hour of energy produced. Coal worldwide is 161 deaths. Coal in China is deadlier due to lack of pollution and safety controls, at 278 deaths. Coal in the US is far safer for converse reasons, at just 15 deaths. Oil is 36. Natural gas is 4. Biofuel is 12 (agriculture is actually a dangerous activity). Peat the same. Solar panels is 0.44. Wind is 0.15. Hydroelectric is 0.10 when safely conducted; 1.4 when not. Nuclear is 0.04.

You’ll notice coal-sourced energy, even in a safety and environmentally concerned state like the U.S., will kill 375 times more people than nuclear would, if we replaced all our coal capacity with nuclear. So if you had to choose, the nuclear waste is really an awesome deal. It’s far less deadly. Likewise natural gas would kill 100 times more people than nuclear power. Solar panels kill 10 times more people than nuclear (and that’s just deaths from workplace accidents etc.; that number isn’t counting pollution deaths from manufacturing, transport, etc.). Wind power kills 4 times more people than nuclear (ditto). Even hydroelectric kills over 2 times more people than nuclear. Nothing in fact is safer than nuclear power.

Even if we count shitty nuclear, e.g. Chernobyl—which we shouldn’t, because as an applied energy policy in a major developed nation today, we would not build shitty plants like that, so we would not face Chernobyl-level risk factors (more on that point below)—nuclear still is safer than everything else. That’s right. Everyone freaks out about “the next Chernobyl,” but in fact, Chernobyl didn’t really do all that much harm. And it was the worst in history.

After Fukushima, whose effects were substantially less significant than Chernobyl, and Three Mile Island, which caused no statistically measurable deaths from its far smaller leak of radioactive material, the worst ever incident in any Western democracy in the last half century was a tornado strike at Brown’s Ferry, Alabama, which caused zero deaths, and no radiation leak, and led to significant improvements in nuclear plant safety design—as did the incident at Three Mile Island, which, BTW, occurred half a century ago.

Meanwhile the still over-cited incidents at Windscale, England, and Fast-Chalk River, Canada, occurred seventy to eighty years ago. Ancient history, by technology standards. And yet just try comparing that with the death rate in and because of the coal industry eighty years ago, and the comparison is still night and day. Similarly, there have been a few worker deaths in the nuclear industry over the last thirty years; but the number of workplace deaths from implementing the solar industry have been much greater (likewise every other alternative).

There actually hasn’t been a seriously deadly nuclear accident, ever, in human history (not counting deliberately, which has happened only once: when the U.S. nuked Japan in 1945). That is, compared to, for example, even just net mining deaths in the coal industry, total deaths from nuclear meltdowns and waste disposal has always been trivial; vastly more so when we include deaths from the pollution using coal generates. And environmentally, from Exxon Valdez to Deepwater Horizon, oil has done far more damage; and with the vast scale of aging oil caps and rigs, the damage inevitably to come will exceed even that. Moreover, nuclear deaths have always been the product of correctable bad design, and thus not actually a scalable risk. To the contrary, the risk plummets vastly when you take safe design and disposal seriously; this is as true of nuclear as of coal, as one can see comparing the death rate caused by American coal-use with Chinese.

And so the same analysis found, that even adding in the total lifetime deaths creditable to Chernobyl, nuclear power is still safer than even hydro; twice as safe as wind; six times safer than solar; and over 200 times safer than even the safest and least pollutive of coal industries. And that’s just as things are; and even not counting a lot of deaths in industries like solar caused by associated pollution in manufacturing and transport etc. Again, solar power, not that clean. Neither is wind power. Of course, what matters is the differential, i.e. the amount of pollution entailed along the entire production train for competing sources, nuclear vs. wind for instance. But that differential does not come out so well for any source over nuclear, because so many more solar panels and wind turbines have to be made to add up to a single nuclear plant, and that’s even after factoring in the peculiar potency of nuclear waste—simply because the amount of nuclear waste generated is also so vastly small.

Then There Is As Things Could Be

So even as things are, nuclear outperforms every other source of electrical power, in efficiency, pollution, and public risk. That’s right. It pollutes far less, it kills far fewer people, and it generates far more energy in proportion to input. But we actually can do better at nuclear than that. Most nuclear plants today are old. Because of irrational public fear, new plants get nixed at the proposal stage and old plants are left to be kept running with shoestrings and glue. And yet that’s doubly irrational. Because the old plants are the most dangerous. New plants implement safety and security and efficiency advances that vastly reduce the dangers and risks of nuclear power; and which also greatly reduce the amount and potency of nuclear waste as well.

Thus, the very thing that causes people to fear nuclear power so much as to stop its development, is precisely what its development would have solved. People who protest nuclear power are therefore increasing public risk. If they would instead promote a new plant replacing an old one, they would be greatly reducing public risk. Indeed, they would be doing so even if they support the building of several new plants in place of an old one, the improvements in design are that significant. There are literally no alternatives to replace old nuclear plants with that won’t kill far more people.

This reality played out recently in Japan, where the Fukushima plant disaster was entirely due to its aged design (built in 1971). Newer plants were immune, and thus survived the earthquake and tsunami essentially undamaged, posing no risk to the public. Even Fukushima’s sister plant, still quite aged (though still newer, built a decade and a half later), only required staff management to handle a minor coolant emergency. Others built around the same time were fine and could have resumed operation immediately, but public fear kept them off line until fairly recently (after reports showed they were not in danger from earthquake or tsunami; and they were upgraded anyway to be even less so). And these plants were built thirty years ago. Newer plants are even more superior. Similarly, Chernobyl was so badly designed it would never have been erected in the U.S., in any decade. It therefore cannot be used as representative of the dangers of nuclear power. Nor can Fukushima. As no new plant built will be at all as badly build as that antiquated plant was.

What we have today is a much safer plant design, much harder to damage with natural disasters or terrorist attacks, and producing much less waste material with a much lower radioactivity or lifespan. In fact, the newest designs actually consume nuclear waste and convert it into energy! And further investment will only lead to more improvements in these respects. The readiest version is the molten salt reactor, which has been around a long time, but has recently been upgraded into the most advanced nuclear reactor in tested use. It has numerous safety advantages (e.g. it is meltdown proof; it consumes no water; etc.); and by actually consuming nuclear waste as a fuel it reduces the environmental impact of even old reactors, and produces less environmental impact itself. Moreover, these models allow numerous smaller plants to operate efficiently and safely, reducing the danger of concentrating a large region’s power supply in a single plant vulnerable to a disabling attack. Other new designs, especially with more R&D, have the same or similar merits (integral fast reactors, including lead-cooled and gas-cooled; and currently in development, wave reactors).

The new tech has convinced even stalwart anti-nuke environmentalists to change their mind about the viability and indeed urgency of going nuclear, as the only plausible alternative renewable and clean energy source there is for the world’s power needs. See, for example, How Humans Save Nature, a TEDx talk by Michael Shellenberger. All of it is good, but pertinent here is his reversal on nuclear power, which he explains starting at timestamp 13:52.

On balance, being against nuclear power is just like being an anti-vaxxer: it is based on a false assessment of risk. Both exaggerate immensely the actual dangers of these technologies. And both fail to assess risk differentially. Even if a vaccine can cause death, the probability it will is vastly less than the probability of dying from the disease it inoculates against. The net risk is therefore a no brainer in Game Theory: only a fool wouldn’t vaccinate. You are trading an enormous risk for a trivial one. In other words, you might not be eliminating risk, but you are vastly reducing your risk by making use of the technology rather than opposing it.

(Oh, and BTW, the logic is identical for whether you vote this month; if you don’t go out and cast a vote for Clinton, unless you are already in a totally Trump-safe state, you may as well go join the anti-vaxxers; even just on the single factor of appointing judges to the Supreme Court, upon which all human and civil rights depend—never mind the dangers to our economy, security, and effectiveness and efficiency of government. This is true both individually—your own interests will be far more damaged by Trump winning than Clinton—and in terms of herd immunity: if you let Trump win, you are fucking over the rest of us, just like sending your unvaccinated kids to a public school. And let’s be honest: that kind of makes you an asshole.)

Similarly, opposing nuclear only furthers dependence on mass-scale use of other technologies that are and have been causing far more deaths than the technology you are opposing would. This is true hundreds of times over for fossil fuels, still our primary energy source, and the only practical and efficient energy source available (because wind and solar don’t even approach their return on investment; and hydro and geo can’t be scaled up enough to replace fossil fuels). But even these alternative energy sources cause more deaths than nuclear. So if protecting human life is your goal, you should be pro-nuclear. The more so if you want to save the environment, end global warming, and yes, even reduce poverty (because cheaper access to energy has a direct correlation with reduction in poverty).

Conclusion

Ironically, almost all energy is ultimately a form of solar power. Photovoltaic and solar-thermal power are obviously so, directly converting sunlight into electricity. But even wind power only taps the motion of air that is ultimately the result of the heating of that air by the sun. Hydroelectric taps the falling of water that is continually raised to higher ground by being evaporated by the heat of the sun (and then falling as rain or snow). Fossil fuels (and all their organic substitutes), including oil, coal, and natural gas, are only “fuels” at all because of biological processes that stored solar energy in plants (and thence the animals that ate them, or ate the animals that ate the animals that ate them), which ultimately came from the photosynthesis of solar rays. In other words, even oil is just a liquid battery for solar energy.

That’s right. No sun, no fossil fuels. Nor wind. Nor hydroelectric. There would have been some fossil fuel without the sun; there are a scattered few geothermal ecosystems on earth, but you need space-age technology just to visit them, and the petrochemical sedimentation they’ve deposited throughout the earth’s history probably wouldn’t have lasted beyond the 18th century—even if we could have gotten to it. Hence, realistically, nearly all our sources of electrical and mechanical energy (both actual and potential) are just different kinds of solar power. Geothermal and nuclear are the only significant exceptions, not deriving from our sun’s energy. Although they also got their energy from stars, probably stars previous to our sun, which generated the requisite elements in their own nuclear furnaces, and supplied the heat and rock (whose stress and pressure produces even more heat) that has become concentrated in the core of our planet. But geothermal is not commonly accessible. And nuclear is what everyone is afraid of.

And though all the other sources drawing energy from the sun are, in fact, far less efficient than direct solar power (when measured solely by how much electricity they generate per hectare of sunlight), they are ironically still vastly cheaper and thus more efficient for us to access. Yes, photovoltaic energy (solar panels), though less efficient than the heated-salt solar-thermal steam turbine (generators that convert reflected solar heat into kinetic and then electrical energy), converts more sunlight to electricity. It’s just that it does so at such great expense to us that it falls far below all other sources in utility. A solar panel can’t compete with an entire planet generating millions of years and millions of hectares of fossil fuel. Nor can it compete with an entire region’s climatic redistribution of rainwater to power hydroelectric plants. That process may be inefficient. But it operates automatically (costing us nothing), and concentrates the resultant stored energy from such a vast area, that it costs us little to tap it. Wind power similarly beats solar by tapping an automatic collection of solar energy over a vast area (all the sunlight hitting all the air of an entire region, thereby pushing it on as wind).

This is why solar power is simply not a viable model for world energy policy. Some limited use of solar is practical. But the scale of power output we need is too vast. If you want to get away from the deadly and destructive use of fossil fuels, hydro and geothermal are not scalable enough (they work cost-effectively only in a few places, so if we invested more in them, we’d quickly reach practical capacity; worth doing, but it’s not the end solution), wind and solar thermal are too weak to compete with fossil alternatives (they have a place, but it’s limited), and solar panels are not even economically viable at all (it’s at best a localized solution). Nuclear is the only power source that generates huge returns, far exceeding even fossil fuels, and more cleanly than any alternative that exists, with far less threat to human life and the environment. There isn’t any reason not to invest in it as our primary energy solution. Particularly with the new and advanced technologies available.

You probably didn’t know solar and wind generated so much less energy after subtracting the energy we put into them. You might not have thought how limited hydro and geo are, because we are already close to capacity with them. And you probably didn’t think about the deaths, dangers, and pollution that a solar and wind power industry entail. They are still better than fossil fuels in terms of environment and safety, but come not even close in terms of net energy return. Nuclear is considerably safer than solar and wind, less toxic to the environment, and far exceeds fossil fuels in energy generated even after subtracting the energy we have to use to build, supply, and maintain the plants. And this will become even more the case as we continue to tap out easily accessed fossil fuel deposits and have to burn more and more energy accessing them in harder to reach places. Whereas the opposite is happening with nuclear: new plants burn nuclear waste, so nuclear fuel will actually become far more accessible.

It’s time to give up our irrational fear of nuclear power. It’s time to invest in the new advanced reactor technologies and begin eliminating our dependence on fossil fuels for our electricity nationwide, and worldwide. For our health. For our economy. For our environment.

Liked it? Support Richard Carrier on Patreon!

14 comments

  1. Richard Johnson November 1, 2016, 3:49 pm

    Everyone needs to know about Thorium reactors and how they are safer, produce less radioactive waste, and both the fuel and byproducts are not usable for bombs. The US halted research on them in 1973, mostly because they wanted the plutonium from uranium reactors for use in making bombs. Look up the Wikipedia article about it.

    Reply
    1. Thorium is promising if the (massive) problem of finding a material that will hold the hot mess in place can be solved. India has been researching it for decades and China now have a super-ambitious project, but the problem is very hard and even in the best case it will be decades until it becomes an alternative. On paper, it is very promising though.

  2. Your use of a single study to draw such a far-reaching conclusion is not serious. Other studies have come to very different conclusions. It should be noted that this paper is written by researchers of nuclear energy, so we can hardly expect them to have an unbiased attitude to different energy sources.

    The most obviously controversial point in this study is that wind and solar have been (massively) penalised by assuming that they will be the only form of power generation, which would require some sort of (very expensive) system for storing energy to be used when it is dark or the wind is not blowing. In my opinion, this is so completely unrealistic as to render the methodogical approach unserious. We don’t choose one energy source to produce all energy, we choose the source that is most suitable for producing the next marginal unit that we need. For instance, if we have a very good place to build a hydropower plant we may build that first, and if we have a windy location we may next build some wind mills, etcetera.

    I only briefly checked a draft of the paper, but the results for wind power are unchanged. Even removing the ridiculous penalty for storage, all other studies I can find have significantly better EROI than this paper.

    I also strongly doubt the numbers reached for nuclear. The authors assume a mean lifetime of 60 years for nuclear power plants before decomissioning, even though existing or planned plants almost never have such long life spans even on paper.

    In short, they are comparing theoretical future-tech nuclear power plants with tried and tested, measured results from wind power generated with existing and to a large extent outdated technology. They also assume that when a wind turbine is ‘decomissioned’ the entire investment is lost. For purposes of long-term EROI we should probably assume that the greatest part of the *energy* investment, the superstructure of the wind mill, will be reused, while the blades and turbine may be replaced.

    Reply
    1. Other studies have come to very different conclusions.

      Those are actually surveyed in the paper I cite. They show why those results aren’t adequate, and what they did to correct for what they excluded.

      It should be noted that this paper is written by researchers of nuclear energy, so we can hardly expect them to have an unbiased attitude to different energy sources.

      The data is still the data. You don’t get to pick what’s true based on which bias you prefer. You have to settle on what the facts are. So if you can’t challenge the facts, you don’t have an argument.

      The most obviously controversial point in this study is that wind and solar have been (massively) penalised by assuming that they will be the only form of power generation, which would require some sort of (very expensive) system for storing energy to be used when it is dark or the wind is not blowing.

      That is accounted for in the paper. They give results both for buffered and unbuffered power gen. So there has been no unfair penalization. The buffering still exists as a problem anyway. So you have to count that in any comparison of options. It doesn’t go away because you use less of it. It’s proportional. It remains at all scales.

      In my opinion, this is so completely unrealistic as to render the methodogical approach unserious.

      That a single solar panel only generates energy half the day is unrealistic? That wind doesn’t blow 24/7 is unrealistic?

      If you want to see an example of bias generating results contrary to reality, you just gave us an exhibit.

      We don’t choose one energy source to produce all energy, we choose the source that is most suitable for producing the next marginal unit that we need.

      That doesn’t matter to these results. The ROI remains the same no matter what percentage of generation comes from a given source. The fact is, the amount of energy we need is vastly greater than any practical wind-solar solution can address. It is beyond economic viability in its sheer scale. It’s fine to have some. But it won’t get us anywhere near where we need to be, without costing vastly more (or even tanking our economy) and polluting the environment more and killing far more people. The reality exists at all scales. Nuclear is vastly (vastly) safer, cheaper, and cleaner per TWH, regardless of how many TWH you decide to generate with solar or wind. For example, if you decide to generate only 1TWH with wind-solar, you could replace that 1TWH with a small nuclear plant and you would still kill fewer people, spend less money, and pollute the environment less. The effect remains at all scales. So you are always choosing how many people to kill, how much pollution to generate, and how much money to lose.

      For instance, if we have a very good place to build a hydropower plant we may build that first, and if we have a windy location we may next build some wind mills, etcetera.

      I already addressed this in the article. This is simply not scalable. We will reach capacity rapidly on sources like hydro. And as I said, sure, tap where we can. But our energy needs are simply too vastly greater than these solutions can accommodate. I don’t think you understand this. Or the scalability of the costs, deaths, and pollution. Which remains. Per above. You can’t avoid the consequences of choosing a TWH generation source that kills, pollutes, and costs. Your choices are simply how much killing, pollution, and costs. And the difference remains. At all scales. Per above.

      I only briefly checked a draft of the paper, but the results for wind power are unchanged. Even removing the ridiculous penalty for storage, all other studies I can find have significantly better EROI than this paper.

      Then you didn’t read the paper. They give values for both buffered and unbuffered. I even mention this in my article. It’s even shown on the chart I have in my article. This is starting to look like you are so biased and careless you can’t even “see” evidence and literally just “hallucinated away” what was actually in my article and that paper in agreement with the results you want to be true, rather than what actually is true. You need to stop and ask yourself why you just did that. This indicates your judgment in this issue is seriously unreliable and needs a major recalibration.

      I also strongly doubt the numbers reached for nuclear. The authors assume a mean lifetime of 60 years for nuclear power plants before decomissioning, even though existing or planned plants almost never have such long life spans even on paper.

      Check their sources. They are right. Current plant designs have sixty year lifespans, owing to new technologies. This ignoring of technological progress and assuming plants built today are just like plants built thirty years ago is simply more evidence of irrational bias against nuclear. Update your knowledge. This is not 1970 anymore. This is not even the same century!

      In short, they are comparing theoretical future-tech nuclear power plants with tried and tested, measured results from wind power generated with existing and to a large extent outdated technology.

      Nope. Their reference plant is an actual plant in service. Not hypothetical. The ones being tested now and actually operational are also not hypothetical but nearly ready to go online as soon as there is investment. And I treated them separately. They are even vastly better than the reference plant used in the study.

      They also assume that when a wind turbine is ‘decomissioned’ the entire investment is lost. For purposes of long-term EROI we should probably assume that the greatest part of the *energy* investment, the superstructure of the wind mill, will be reused, while the blades and turbine may be replaced.

      There is no evidence that’s true. The superstructure ages same as any tower or bridge; and is the least expensive component of the machine. The lifespan is about the same and comes to a total time replacement average. The replacement of blades and turbines and associated parts over the lifetime of the machine is already in the measured annual maintenance costs, for example. Some parts may last longer, others less. The average lifetime of the whole, remains as reported. The data is the data. You can’t invent data. Just because you “want” wind turbines to last longer, doesn’t miraculously cause them to.

  3. Your other source, about fatalities/twh is also known for making the rounds in nuke-fundie circles. It is not serious.

    Actually a serious calculation would still look *very* good for nuclear, but the ways in which the nuke-fundies ignore science in order to get an ultra-super-awesome number rather than just a really impressive number speaks volumes. The reason they get almost no deaths from nuclear is that they are only counting ‘proven’ deaths, from acute radioactive poisoning and such. This is not how we treat calculations for other power sources such as coal. Typically, we have no way to prove that someone who dies from lung problems did so because of particles emitted by coal. It could have other reasons. Instead, we get the numbers through statistics, we know that a certain level of particle emission leads to a certain number of deaths.

    But this is really the same for radioactivity. If we do the calculations seriously, we would expect a few thousands (maybe 2000-5000) deaths from the Fukushima disaster due to exposure to low doses of radioactivity. This may sound like a lot, but we should note that there were about 16000 recorded immediate deaths from the earthquake that caused that accident, and it is the only other accident than Chernobyl that has caused mass health effects. It should be noted that claims by anti-nuke-fundies that the Fukushima radiation has caused health risks on other continents are completely unfounded by science. Also, most of those who will statistically die from the exposure will live for many years in good health, although that can also be true for eg coal emissions. If you discuss this with a nuke-fundie they will often completely deny mainstream science which says that the risk is of exposure is fairly linear to dosage with no lower cutoff level. This conclusion is strongly supported by statistics for medium dosages, and for low dosages where statistical studies become impractical, but theoretical findings which indicate that radiation cancer is, for at least some types of radiation, caused by a single energy quantum.

    Reply
    1. The reason they get almost no deaths from nuclear is that they are only counting ‘proven’ deaths, from acute radioactive poisoning and such.

      Again, you don’t get to invent data. And the study included statistically expected costs to life expectancy. You aren’t being honest about how these results were generated. Your bias seems so extreme you don’t even check to see what numbers they are using or even look to see if there is any evidence-based reason to increase them.

  4. And, sorry for posting so much, but now for the reasons why nuclear sucks, the reasons omitted by your no doubt nice but still clearly irrationally nuke-loving friend.

    First let me say that nuclear power has a very important place, for the near and medium future and, depending on technology advancement, maybe in the long run as well. I’m very sad to see how Germany dismantled their nuclear plants and used more coal instead. There is no sensible and realistic option to get France off nuclear power either.

    But there are a number of very serious problems with nuclear that show why it is not the awesome universal energy solution nuke-fundies want people to believe.

    The first is simply cost: it is not a cost-effective solution compared to alternatives such as wind power. Nuclear power plants are pretty much always (I know of no exception) built as government investments. As a business investment they just don’t cut it and never have. Nuke-fundies have always projected that this will change now because of the latest technological breakthrough, but these predictions have so far been wrong every time. Meanwhile, technologies such as wind that were also introduced with government subsidies have seen technological breakthroughs that make them competitive on their own merits.

    The second major reason, which is related but hard to quantify, is that nuclear power plants are behemoths. A single modern plant generates so much energy that it can power a small nation. This gives rise to massive problems of corruption both during the building phase and the production phase. What, for example, if an operator is so unlucky as to experience a shutdown of a reactor during a period of peak energy usage, causing a massive spike in electricity price? Suddenly the plant gets massively profitable, what a shame really!

    This puts nuclear in contrast with most other forms of generation, with the possible exception of hydroelectric. Wind and solar benefit from being an excellent fit for market based innovation and competition, both for production of the units and for operation and maintenance.

    Reply
    1. The first is simply cost: it is not a cost-effective solution compared to alternatives such as wind power.

      100% false. And proven false with evidence. You have no evidence to the contrary. The one who asserts facts contrary to reality with no evidence after just being presented with extensive peer reviewed evidence? That’s the fundamentalist in the room. Please give that some thought.

      As a business investment they just don’t cut it and never have.

      Also false. You can’t just assert things without evidence. A wind farm the size sufficient to replace a single nuclear plant is going to cost vastly more and generate vastly less. That’s what ROI is. So if wind farms are practical business investments, so are nuclear plants. And as one would thus expect, there are numerous successful, profitable private nuclear plant companies in just the U.S. alone. There are also successful, revenue-neutral state-owned nuclear power companies, all over the world. And the same state incentives given to wind can be given to nuclear. And so on. This is therefore no argument.

      The second major reason, which is related but hard to quantify, is that nuclear power plants are behemoths.

      The new plants ready to go online will not be, or needn’t be. As I explicitly said in the article above that you evidently didn’t actually read.

      This gives rise to massive problems of corruption both during the building phase and the production phase.

      If that were true, it would show in the data. In other words, the ROI calculated, already inherently includes any such effect, whether it exists or not. So this is not an argument either.

      What, for example, if an operator is so unlucky as to experience a shutdown of a reactor during a period of peak energy usage, causing a massive spike in electricity price? Suddenly the plant gets massively profitable, what a shame really!

      Gee. I wonder if maybe we have fifty years of worldwide operational evidence with which to answer that question?

      Do you think?

      Maybe?

      Like, maybe you could check the evidence to see what the answer to your question is?

      You know, evidence? Of which we have tons and tons, for numerous nations and decades, and hundreds and hundreds of plants?

      This puts nuclear in contrast with most other forms of generation, with the possible exception of hydroelectric. Wind and solar benefit from being an excellent fit for market based innovation and competition, both for production of the units and for operation and maintenance.

      The evidence does not back this up. The ROI remains what it is.

      You don’t get to make up data. There is no evidence of any tech in the pipe that will substantially alter the ROI for wind or solar. There is actual substantial evidence of tech in the pipe that will substantially improve nuclear ROI, safety, security, and cleanness.

      I base my beliefs on evidence. You might want to decide some day to do that as well.

  5. Forgive me if I’m not going to spend hours of research for writing a blog comment pointing out some errors in an area where I actually have the required expertise (I have an MSc in mechanical engineering) and I can see that you don’t have the required understanding of the subjects to draw the conclusions you do. Obviously I can document every single claim I have with peer-reviewed research, if you ask. But you did not even bother to ask.

    There is no sensible system on this board for quoting, so I’ll limit myself to two points:

    First, the supposed costs for storage are *not* proportional as you claim. Only a person completely ignorant of how energy systems work would claim something so completely ridiculous.

    This is how it works: let’s say we have a simple system with only two sources of power: hydroelectric and wind. In this system, we will use the hydroelectric plant as what is known as *regulating power*. This means that we will activate more turbines when the wind is not blowing so much. When it is very windy, we may deactivate the hydropower completely. This is typically not a loss, because the limiting factor of hydropower capacity is simply the available amount of water. How much wind power we can have given available regulating power is a very complicated subject, but there is no doubt that wind power can give a significant contribution without any use of buffering whatsoever. Going beyond that, the added cost is *still* not proportional, because it is possible to export electricity from windy areas to less windy areas, which for a larger area cancel each other out for the most part. This results in transfer losses, which is why we will see a sub-linear reduction in overall efficiency in this area. Only if we build a wind-only solution (say, on a small island), buffering becomes the problem you believe exists everywhere.
    Second, no, we do not replace towers and bridges every 20 years! You really have no clue what you are talking about here. The reason a period such as 20 years is typically used for the projected lifespan is that such projects are determined by *costs*, not by EREI. Even though the superstructure is the most *energy intensive* part, it is cheap, owing in no small part to the low energy prices we have today. Because very long running investments are complicated to manage, a relatively short life span is used since it makes little difference economically. If we care about EREI, we should take this into consideration.

    Third, again, they are comparing nuclear future-tech nuclear plants with outdated wind plants. There are tried and tested wind power parks that have been measured, over a period of decades, to have an EREI of 50x+. You have nothing such for nuclear, where you are using projections (very optimistic ones) based on shorter runs of model plants, or even theoretical predictions of dream projects that would surely be super efficient if it was not for the global conspiracy that somehow prevents these awesome new plants from actually being built (if you had followed this debate for a bit longer, you would know that these claims of new generations being vastly more economical are not new, they are as old as nuclear power itself. Newer generations have indeed shown great improvements, but they have not become more economically feasible).

    Fourth, about economics: you make the mistake of thinking that because a private company is *operating* a power plant, they paid for the construction of that plant. I just checked the first plants in your links, constructed back in the 60’s with massive government subsidies. I’m not saying that government investment in nuclear is necessarily a bad idea, in fact I would say it is frequently a great idea, but you can’t just ignore this. This is the reason why almost no nukes are built in the US anymore and most plants are from the Cold War era: there is simply much less government funding available nowadays, probably largely due to the reduced need for production of plutonium for nuclear weapons.

    Reply
  6. I have now read the Weißbach et al paper more closely. There are, in fact, additional flaws in this paper that make the results completely useless. When assessing research papers, it is a good idea to also check if there have been other papers published that comment on them. Such is indeed the case for this article, where a peer-reviewed comment claims, in summary, that “Serious methodological errors in a paper by Weißbach et al. invalidate their results.”

    “Comments on “Energy intensities, EROIs (energy returned on invested), and energy payback times of electricity generating power plants”—Making clear of quite some confusion” by Marco Raugei, with a number of other experts joining him on followups.

    Apart from the ‘buffering’ already discussed, it is clear that Weißbach are also using completely unorthodox methods, seemingly of their own invention, to ‘correct’ the numbers they are using as their sources. These ‘corrections’ appear to stem from a fundamental lack of understanding of what the numbers represent, how they are arrived at, and even their purpose.

    There is also a remarkable degree of arrogance on part of Weißbach, perhaps best illustrated in the following quote:

    “In opposite to that, the IEA advises to consider the backup outside the system borders without any scientific justification [6].”

    Here, the authors summarily dismiss the advise and expertise of the International Energy Agency, believing they have found some fundamental flaw in how everyone else does these calculations. And this, even though none of the authors are experts in energy systems! This type of behaviour is what we would expect from hobbyist climate revisionists.

    Reply
  7. Steven C Watson November 3, 2016, 9:13 pm

    Ὀψὲ λόγῶν ἀλέουσι μύλοι, ἀλέουσι δὲ λεπτά. With apologies to Sextus Empiricus.

    We all do well to change our minds when the facts change. Dr Carrier does it often. I ask, after John Maynard Keynes, what do the rest of you do?
    Very informative and I enjoyed it immensely, Tah very muchly.

    Reply
  8. Steven C Watson November 3, 2016, 10:30 pm

    Johan Rönnblom, did you miss where Richard writes:

    “This is all analyzed in such research reports as…”? I take that to mean at least several such reports have been examined. On previous form I am probably well under-estimating. You can only use the word thorough about Dr Carrier in a sense of English understatement. Very little of what he has argued here is unfamiliar to me. The data and sources may be different but the arguments and conclusions are the same as I have been seeing and hearing since at least the eighties.

    I’ll check your citations, as I always do where I can, but this is another of those arguments that has been rehearsed that many times to the same conclusion – a win for nuclear – that you really should get over it already.

    Reply
    1. To begin with, the whole framing of the discussion is ridiculous. The ‘my tech is better than your tech’ argumentation style is for fanboys (I have never seen a woman engaging in such fights).

      Whether the fight is between PC or Mac, vinyl vs cd, car brands, or nuclear vs wind vs solar vs whatever: it’s silly, and anyone taking it seriously is a clueless fanboy.

      The reason there are different technologies in use out there is not that everyone else is stupid or that there is some grand conspiracy, but that different tech has different advantages and disadvantages. The only correct answer is ‘it depends’.

      It’s very unlikely that Carrier has done any serious survey of other relevant studies, because serious studies give very different results. Now, this is not to say that nuclear does not do reasonably well in some circumstances. But no, apart from very confused or biased studies, or outright fabricated propaganda, you will not see the claim that solar or wind power does not carry its own costs in terms of energy usage, carbon footprint, or any other environmental metric you might imagine. Certainly, for some segments of energy production, there are other technologies performing better.

      To finish, let’s just consider one more factor of real world energy production: the daily electricity usage pattern. In the evening, a lot of industries are shut down, and in the night the use of household appliances is also much reduced. The result is that the demand is highest during the day, and lowest in the middle of the night. Now, typical energy systems do not have any dedicated ‘buffering’, but rather the production is adjusted so that every hour, minute and second, the production is matched with the supply. This is a problem not only for wind power, but actually for most forms of power, and more so for nuclear. Because it takes a lot of time to stop and start a nuclear power plant, and for technical and economical reasons you want to run it 24/7 with as few interruptions as possible. So nuclear is only ever suitable for the ‘base’ rate of lowest electricity consumption during the daily and even yearly cycles. To some extent this can be evened out by exporting during the night to neighbouring regions, but only if these regions are not themselves running on nuclear power.

      Conversely, while solar power is not necessarily the most efficient technology per kWh generated, it has the property of generating the most electricity on sunny days, and if peak electricity usage is caused by air conditioning systems, as is the case in many places, this makes it a very good fit to help even out these peaks. In economic terms, they will generate power at those hours where the electricity price is highest, and in ecological terms, they will reduce the need for the type of dirty energy production typically employed during peaks, such as cheap and relatively inefficient gas turbines.

Add a Comment

I only publish comments by my patrons and anyone who or whose work I discuss in the article commented on. Comments must also follow good etiquette. Those who engage in dishonest, abusive, or harassing behavior may even be banned as commenters and patrons.

 

*