THE FUTURE OF ENERGY
Conversations about carbon dioxide and energy have been unavoidable within environmental circles and are increasingly so in the media. In recent years, and in light of the realities of climate change, many respected voices have begun advocating for a nuclear path to a carbonless future. Naturally, there’s vocal opposition as well. This dispute — in addition to being highly technical and with every part of it layered with differing assumptions, conflicting methods and metrics, and curious omissions — makes this subject difficult to wade through for laypersons like myself. With an increasingly political and moral tinge the whole thing is now a great big mess.
Until this point I’ve considered myself to be on the fence, while being open about my many reservations. Though I’m enthusiastic about anything we can do to restrict our outputs of CO2 to reasonable levels, as well as being happy to hear about innovations in nuclear energy and waste disposal domains, I also harbour a west coast-borne sense that nuclear is somehow unpalatably costly, dangerous, and irresponsible. This is my bias. However, I’ve had enough casual questions and conversations on this topic now that I feel like I need to have more of an informed opinion on the matter based on something other than intuition.
Troublingly, to date I’ve found few good sources of information and fewer folks who are both knowledgeable and keen to talk. As with almost anything important and requiring conversation today, things start getting weird when you offer observations, pose questions, or share any amount of skepticism toward what's presented by noisy enthusiasts. Like my experience with nearly all contemporary discourse, if you do anything but loudly celebrate the envisioned utopia you’re outing yourself as a hostile force needing to be taken out. If you’re not strongly in favour of the long-overdue global nuclear epoch, for example, you're an enemy of the revolution, shunned as an agent of the German anti-nuclear movement or some sort of misguided hippie engaged in social media trolling on behalf of Greenpeace (on behalf of Mother Gaia.) So, it's hard to even start a conversation about this stuff.
But, as with so many other trending labels, I don’t believe myself or most folks with questions and concerns fit these profiles. In fact, on enthusiasm for most things technological or modern I tend to sit closer to the extreme end of the spectrum. (For example, I cannot wait for a time when humans are no longer allowed to drive and all vehicles are self-driving. What a dream! I’m also the only one I know getting sweaty palms watching the launch of the James Webb Space Telescope and nervously anticipating all the pending hurdles along its 1.5-million-kilometre path to Earth-Sun Lagrange point two…) And it's just this sort of sentiment and worldview, one that appreciates civilization, science, and their outputs, that regularly gets me in trouble with friends and colleagues. So, I see myself as someone who should be a pretty easy sell on nuclear energy. (Just win me over, already!)
CLEAN AND ABUNDANT ENERGY
From the few op-eds I’ve read and Twitter chats I’ve had on this topic, some common feelings and assessments emerge. Folks who are convinced by the nuclear argument tend to offer a simple case about nuclear being both high capacity and low-carbon and, thus, the best climate change-responsive investment. Makes sense. A typical analysis starts by looking at the high energy consumption in the world’s wealthiest nations and the growing global demand for power in developing regions. From there the carbon dioxide burden of various energy sources is estimated, providing an evaluation for where the world’s energy mix will need to be to land us in a more sensible place carbon-wise. Great. Rankings of power production modes typically include considerations of capacity, reliability (often termed "energy security"), and an attempt at a total accounting of the CO2 resulting from building, maintaining, and generating electricity from typical examples of the most common power sources. Great. Seems objective. And, thusly, the combination of these three factors plant nuclear in the favourable position. So, as the argument goes, you’re either for nuclear or you’re some kind of backward-looking Luddite or crazy climate denier or in cahoots with such degenerates.
THE ONLY THING TO FEAR IS IRRATIONAL FEAR
However, that’s never the end of the conversation. Discussing nuclear energy with any layperson almost inevitably raises thoughts and questions around safety. For many people the problem of nuclear waste and the tragedies at Chernobyl and Fukushima are indelible, even if most folks have no specific numbers to throw out or a crisp academic references to back up their concerns. Negative associations are effectively embedded in the culture and language at this point. Many are turned off the prospects for nuclear power with little or no further consideration. Of course, Chernobyl being synonymous with multidimensional incompetence and catastrophe, reservations are perfectly understandable, even if you feel they may be irrational here in the 21st century.
On the other side, any discussion of risk with nuclear proponents elicits instant consternation, as though you’re a flat-Earther trying to discuss topology with a geophysicist or astronaut. Just to mention Chernobyl or Fukushima tells them you have a childish radiophobia induced by some combination of environmentalist and media sensationalism or, perhaps, just your own blinding ignorance. This is because nuclear energy proponents, at least those I’ve read and spoken to, almost unanimously consider Chernobyl to have happened back in the dark ages of nuclear energy. As such, bringing this up in the modern era is seen as preposterous — like comparing a Ford model T to the Tesla model X. Fukushima, on the other hand, is said to have resulted from an unprecedented set of natural disasters; making it closer to an act of God than something foreseeable or preventable or a flaw in the technology. All of this is stated as irrefutable. Some folks go further, so far as to deny virtually any concerns with nuclear power or any related injuries or deaths of any kind. Others argue that claims are merely wildly overstated and that, at Chernobyl, say, the official Soviet fatality body-count of 31 is complete and robust.
If you haven’t engaged in these discussions, you may think the above are ridiculous outlier positions, not ones held or forwarded by serious people. Sadly, they are not. In fact, I read an article published just now by respected Canadian independent journalist, author, historian, and professor Gwynne Dyre. He's a serious person who does serious work. In discussing the tragedy of the present decommissioning of German nuclear plants, Dyre argues for the safety of nuclear power by stating that “...nobody at all died at the Fukushima Daiichi plant in 2011 (although 20,000 died as a result of the magnitude 9.0 sub-sea quake and the tsunami that devastated the city).” Bold.
A RIVER IN EGYPT
Yes, claiming that earthquakes and tsunamis were inconceivable, especially on the eastern shores of Japan, is like saying you didn’t expect to find sugar or flour at a donut shop; but nobody wants to hear that, and not from you. Similarly, putting in writing that nobody died at the Fukushima plant (in the aftermath of their level 7 nuclear catastrophe) is, well, it’s a bit like talking to someone who insists your uncle didn’t die of COVID but "from" lung failure or some undiagnosed pre-existing condition. Like, “Okay, bro. You’re right, COVID isn’t real.”
Of course, all of this is offered when anyone can easily find official inquiries into Japan’s catastrophe (including an independent national commission, international ones as well, and separate investigations into the site operator, the Tokyo Electric Power Company) all offering something very different to the above miracle hypothesis. Nearly every assessment, including the Japanese parliamentary investigation report, came back explicitly determining that the tragedy “cannot be regarded as a natural disaster”, but was instead “...a profoundly man-made disaster — that could and should have been foreseen and prevented”; not caused by an earthquake or tsunami but, most fundamentally, by Japan’s culture of “reflexive obedience” and involving gross negligence and corruption on the part of business and government. Most damning, it has come to light that several reports from the 2000s warned of threats posed to this very plant by earthquake, tsunami, and seawater flooding and that none of these warnings of catastrophe were taken seriously or acted upon. So, I don’t know why one would deny or dispute any or all of this. Needless to say, no formal refutations from fans of nuclear power ever arrive.
Similarly with Chernobyl, though it’s true that we have moved on from the technology employed in the Soviet Union in the middle of last century, many of the old problems associated with nuclear power remain. Nuclear has a whole host of unique safety problems associated not with accidents but with normal operation (such as mill tailings, spent fuel, reactor decommissioning, waste disposal, and more.) This is abundantly clear within publicly available documents from the energy industry and nuclear societies themselves. (But don’t take my word for it, just go look.) Critically, these health and safety concerns have nothing to do with CO2 or immediate deaths nor are they present with other energy production technologies. You might have noticed that ten thousand generations don't need to sequester, monitor, and defend the fuel or waste streams from solar, wind, tidal, or geothermal energy production. (What am I missing?) This is a real problem. It was a problem in my grandparent’s time and it is a problem today. And, of course, none of this gets into the fact that most of the nuclear reactors on the planet are old models running on outdated technology.
FRAMING AND FACTS
All this is to say that, when questioned about safety, proponents are quick to note the current state of nuclear technology is nothing like the past and that any fears are irrational and needlessly cloud what should be a crystal-clear evidence-based picture. However, to me, framing the discussion around deaths in this way and as people having irrational fear of exploding reactors is what seems most irrational. In fact, this all comes across as a straw man argument that fans of nuclear really need to stop employing. Most obviously, even in such a rare worst-case scenario as a reactor meltdown, I'm not worried about myself or anyone else being taken out in a sudden gamma radiation dump, hydrogen explosion, or other associated major traumatic events. Being snuffed out would be ideal. Wouldn’t it? Instead, I’m concerned about everyone I know living with some terrible illness or just the horrific and compounding psychological, social, economic, and ecological aftermath of an unmitigated disaster. The death bit doesn't even really come into the equation for me. So, even framing this around deaths feels more like diverting the discussion than honest dialogue or debate. Further still, as we know, evidenced by Fukushima and Chernobyl, the greatest impact on lives from nuclear accidents to date (certainly in the nearest term and that we can all agree about) comes from forced dislocations and the abandonment of the affected region. Nuclear has this well-documented legacy. Not acknowledging these realities or simply insisting they’re at best externalities feels rather cold and detached.
Still, if you can press them to steer away from insisting you’re paralyzed by irrational fear or blindly hateful of technology, you may be able to get nuclear energy supporters to enlighten you with their wisdom (or just some simple references.) When they do so they all tend to offer the same sources of data and analysis on nuclear safety. Generally, these studies compare various energy platforms, with each type being linked to associated deaths (whether that’s your primary concern or not.) But, of course, to do so they can’t just count up deaths. Fossil fuels have been around far longer and are far more widely used still today; so, instead, deaths from accidents and air pollution are chosen and then translated into a "deaths per terawatt-hour" figure for each power source to give some rational measure for comparison. The results clearly frame nuclear as extraordinarily safe. (Yes, nuclear is technically less safe than wind or solar by most of these assessments but only by a near-meaningless fraction when compared to, say, fossil fuel sources such as coal or gas.) On the surface at least, particularly if this framing makes sense to you and is all that matters, it’s all very compelling.
RELIABLE SOURCES
That said, specific sources are worth exploring here. The most frequently shared assessment I’ve come across is what’s effectively a meta-analysis done by the good folks at Our World in Data. Aside from this, the two most popular studies I’ve found shared online are from Markandya and Wilkinson (2007) and Sovacool, et al. (2016). (These also happen to pop up in the analysis from Our World in Data, which could be why they are so commonly referenced elsewhere. I’m not sure.) In addition to these, what also comes up frequently in reports, op-eds, and social media furies alike is a favoured modern nuclear reactor and waste disposal project: Finland’s brand new Okiluoto 3 reactor and its Onkalo spent fuel repository. These newest and best-case models are presented as working examples for the 21st century and a tangible way to help right the planet’s waywardly carbonful energy grid. I’ve investigated all of these leading sources and examples and offer my evaluation in some detail below.
OUR WORLD
To begin, I’m a huge fan of Our World in Data, its team of researchers, analysts, and data communicators, including its progenitor Dr. Max Roser. Huge fan. With that said, you rightly anticipate that what follows is disagreement. Where things get interesting, and I get bogged down, is in their analysis of safety. They pose the question, “How do fossil fuels, nuclear energy and renewables stack up in terms of safety?” And they answer this by stating that nuclear has saved millions of lives and will only save millions more. The argument is largely that climate change endangers us all and the health and safety profile for low-carbon nuclear is far better than for any fossil fuel mode. They remind us that over the last several decades nuclear replaced or stood-in for coal and gas all over the world. This prevented significant volumes of carbon dioxide and air pollution that would have otherwise harmed us (and many other beings) and contributed more warming than we’re already in for. Very obviously then, newer and better nuclear plants will only do the same.
Their text and visuals offered on this topic all reference Benjamin Sovacool and company’s 2016 study in the Journal of Cleaner Production, titled Balancing safety with sustainability: assessing the risk of accidents for modern low-carbon energy systems. The other key document is the Markandya and Wilkinson piece from 2007, titled Electricity generation and health, published in The Lancet. The graphs produced from these data are compelling and show the starkest possible contrast between fossil fuel sources and alternatives. Their summary of these sources (the quickest, most easily digestible bit most folks read and share) notes that:
[A]n estimated 4,000 deaths from the 1986 Chernobyl disaster in Ukraine (based on estimates from the WHO); 574 deaths from Fukushima (one worker death, and 573 indirect deaths from the stress of evacuation); and estimated occupational deaths (largely from mining and milling)...
Without even diving in, there are some curious issues here on the surface. Obviously, any reference from 2007 cannot deal with the realities of Fukushima, which happened in 2011. Also problematic is that the Sovacool et al. paper doesn’t account for anything prior to 1990. So it omits one of the two worst nuclear accidents in history and other fatal accidents (such as Mayak and Sellafield in ‘57, Idaho Falls in ‘61, Bohunice in '76 and ‘77, and Surry in ‘86 — which, excluding Chernobyl, total an estimated 450 additional and likely deaths.) Whatever legitimate justifications there may be for these omissions, all this makes for a curious starting place in assessing risks and deaths. But, as you will see, even without these blind spots, it’s all still much messier and more shocking.
CHERNOBYL
The team at Our World in Data is happy to point out that this issue is complex. Still, muddying their straightforward assertions (that “31 people died as a direct result of the Chernobyl accident” and another 4,000 died indirectly) are a whole plethora of conflicting reports. In fact, to go looking is to discover that this is one of the most contested issues out there. Still, in their analysis, significant reports are mentioned but disregarded. Why?
The best examinations of Chernobyl I’ve found are a pair from 2006 (a year after the WHO report reference above.) One is an independent study by UK radiation scientists Fairlie and Sumner, evaluating the 2005 WHO health report and International Atomic Energy Agency's environmental assessment of Chernobyl. Like the above WHO estimate noted by Our World in Data, this report accounts for those who have and may eventually die from radiation exposure. Here, Fairlie and Sumner’s findings place the number of Chernobyl-caused excess cancer deaths between 30,000 and 60,000 — up to fifteen times higher than the WHO report from a year earlier. The Our World in Data team puts these estimates at the top of their list, suggesting these are among the highest figures for legitimate estimates of total deaths and not merely cancer-related ones. They are not. (More on this in a moment.)
The other report, from a separate team of radiation epidemiology experts led by Dr. Elizabeth Cardis, was published in the International Journal of Cancer. It’s also cited by the Our World in Data team. The associated graph produced by Our World is titled “Estimated number of deaths from the Chernobyl nuclear disaster.” Except, this is not what’s being assessed in this or other cited studies. And then they summarize the report as offering “16,000 deaths across Europe.” The Cardis et al. paper looks at current and likely future cancer deaths — that's "cancer deaths" not "all Chernobyl-related deaths." Why is this important? Well, for Fukushima they tell us that the overwhelming majority of deaths, in the hundreds at least, resulted just from the stress of the event felt by the population. Deaths directly associated with the trauma of a local nuclear catastrophe and forced evacuations as well as the abandonment, homelessness, joblessness, substance abuse, and suicides that followed Chernobyl are significant and not at all accounted for in the Cardis paper.
You can find news articles, books, reports, poems, and songs from survivors and the family members of those who died directly as a result of the Chernobyl disaster, all insisting their sufferings and deaths have always been disregarded. And we know Chernobyl saw roughly 300,000 people forcibly resettled from the most contaminated area, some 28,000 square kilometres. Further, the National Research Center for Radiation Medicine in Kiev, Ukraine estimates that 3,000,000 there, 800,000 in Belarus, and maybe another million in the rest of the former USSR suffered directly from the nuclear disaster. So, presuming not even one death out of this base of hundreds of thousands or millions, to me, reads as silliness — especially when we have the example of significant stress-induced deaths around Fukushima (among a population hardened by regular earthquakes and tsunamis...) Further still, though distasteful in some circles, it seems just as ridiculous to totally disregard the tragedy of what some have estimated to be 100,000-200,000 panic-induced abortions resulting from daily media reports of plumes of radioactive material alighting across Europe. Whatever you consider the status of fetuses to be, and even if the figure turns out to be just a fraction of the above, I hope we can agree that this was both a very real impact from this nuclear disaster and something needing to be accounted for in some way in evaluations of health and safety (by someone smarter than me.)
So, what does the Cardis paper say about cancer cases and fatalities? The paper starts by noting:
The full extent of the health impact of Chernobyl on the population is difficult to gauge. Ten years ago, Cardis and collaborators estimated that about 9,000 deaths from cancers and leukemia might be expected over the course of a lifetime in the most exposed populations in Belarus, the Russian Federation and Ukraine. The objective of the present study was to evaluate the human cancer burden in Europe as a whole from radioactive fallout from the Chernobyl accident. The exposure of Chernobyl cleanup workers [a contested number ranging between 300,000-830,000] is not considered here.
From that starting place, Cardis and team go on to estimate, very roughly, 15,700 deaths. They point out that this figure is terribly unreliable, with significant gaps in information and large uncertainty intervals. (For example, thyroid cancer case estimates range between 3,400 and 72,000. Meaning, we don't know if those gifted with thyroid cancer alone could populate a small high school or fill to the rafters four professional hockey arenas.) Here I read the authors as confidently admitting we do not know how many died from cancer as a result of Chernobyl and that good estimates are not possible at this time nor are they likely to emerge in the future. But the Our World team offers none of this, cites 16,000, which they strongly disagree with, and move on — all while confusing readers into thinking cancer was or will be the overwhelming cause of mortality from this event.
In light of all this, it’s interesting to note that 40,000 (similar to Fairlie and Sumner's radiation fatality guess) was the projected total death rate offered back in ’86 by the lead investigator for the Soviet Union’s official Chernobyl disaster commission, Valery Legasov. Curiously, this official figure and convenient convergence didn't make it into the Our World in Data assessment. What of that? And what other numbers are out there?
Well, trying to learn anything at all about this event on your own, you quickly discover that Russia, Ukraine, and Belarus (where the vast majority of Chernobyl’s fallout alighted) have remained secretive about related fatalities. Some published reports suggest that, in these most heavily impacted regions, deaths could be in the hundreds of thousands. The outliers doing work on this talk in the millions. These numbers are heavily contested, of course, and it may turn out that they are rightly ignored. I don't know. What we do know for sure is that a report by the National Academy of Medical Sciences and the National Research Center for Radiation Medicine of Ukraine, titled Thirty Years of Chornobyl Catastrophe: Radiological and health effects, and arriving in 2016 (a decade after both of the studies cited above by Our World in Data) states that nearly two million Ukrainians were victim of the Chernobyl disaster. From this same report we learn that, all these decades later, the government of Ukraine still pays compensation to some 36,500 people whose spouses died due to their involvement at Chernobyl. Other scholars and authors have published similar figures but also warn that scientists tell us the real number could be as high as 150,000 in Ukraine alone.
So, though perhaps only to me, I think it’s safe to say that "36,000 in Ukraine" feels nothing at all like Our World in Data's "4,000 total." Although I’ve read Our World in Data’s assessment and explanation for their numbers, I don’t follow how they arrive at a figure a whole order of magnitude below the above studies and this more recent and widely known and reported assessment out of Ukraine. Obviously, all of this educated guesswork is messy business, but what is clean and clear is that deaths reasonably attributed to this one accident alone are effectively unknown and could be very significant, indeed. If we’re only talking about death, not illness or terrible suffering, and the spectrum is something like 4,000 to 40,000 (but perhaps 400,000 or maybe even somewhere in the millions) I think it’s fair to admit our view of this catastrophe is closer to opaque than otherwise. And then, resultantly, I think it's fair to say that looking at all this and applying the lowest assessment anywhere, as done by Our World in Data, seems somewhat disingenuous.
That said, even if the number of cancer deaths resulting from Chernobyl is very high, at the population level we can be sure that the increased rate of cancers across Europe would remain close to zero, probably up less than 1% or even just a fraction of that. There have and will be hundreds of millions of non-Chernobyl cancer deaths across Europe over the lifetime of those who lived through it. But, for me, this only highlights how silly it is to be talking about cancer deaths while ignoring other deaths, to say nothing of the rest of the health and safety picture.
FUKUSHIMA
If that’s Chernobyl, well then what of these figures offered for Fukushima — only the second max-level nuclear catastrophe in my lifetime? Our World in Data offers “one worker death, and 573 indirect deaths from the stress of evacuation.” A quick search finds that Kazuhiko Kokubo, age 24, and Yoshiki Terashima, age 21, were killed while conducting emergency investigations and repairs at the site on the day of the disaster. I also find that, in the harsh conditions immediately after the disaster, there were 33 cases of heat stroke among workers on-site, two of which resulted in deaths from heart failure. So, just to start, and without really looking, the number of immediate deaths of workers at the nuclear power plant appears to have quadrupled. What, is your assessment that we should be counting abstract and theoretical carbon dioxide-related deaths from natural gas-fired power plants but not the direct deaths of nuclear power plant employees operating on the job during and immediately after a multi-reactor nuclear meltdown? Again, we’re off to a rough start.
But these are not the only strange omissions. The Chernobyl fatality figures include likely future excess deaths from radiation exposure. Of course, I think we can all agree that in the 21st century, and especially in light of the '86 disaster, there should be no emergency responders or employees of any power plant anywhere being exposed to unsafe doses of radiation or anything else. But that's not what happened at Fukushima.
To begin with, the international community suggests a safe human-made radiation exposure level for civilians is 1 millisievert (mSv) per year. (For comparison, a typical spine x-ray or CT scan of your head doses you with around 2mSv.) Workers at nuclear energy plants and other folks working with radioactive material have an acceptable dose limit that’s higher, set at 20mSv per year, averaged over five years. So, in these settings, typically, if a dosimeter measures a level of 20mSv an alarm sounds and the site is evacuated. As reported in local and international papers, during the emergency following the events of March 11th, three workers were exposed to radiation levels of up to 180mSv, two of whom were rushed to hospital. The same reporting notes that a total of 17 workers were exposed to radiation doses exceeding 100mSv at that time. And in just the first few months of the ongoing nuclear disaster, more than 300 workers were reported to have received significant radiation doses, with six of those exceeding lifetime limits of exposure to anthropogenic radiation — above 250mSv. Later we learned that Japan’s nuclear regulator temporarily changed the acceptable radiation exposure limit from 100 to 250mSv in order to keep people on the ground and prevent a far worse disaster. A 2013 report of the United Nations Scientific Committee on the Effects of Atomic Radiation offers further accounting. The report spells out that nearly 25,000 workers had been involved in mitigation at the failed power station and that over the first 19 months roughly 8,750 workers received doses of greater than 10mSv and 170 of them exceeded dosages of 100mSv.
How does all of this come across to you? It reads to me like perfect chaos reigned for more than a year after the "accident." Sure, none of these are fatalities but I want to ask how any of this can be the current state of affairs and acceptable at present? (And this is Japan we're talking about, where safety, diligence, adherence to rules are respected.) Moreover, why determine in advance that none of these exposures, nor any unreported worker or civilian exposure, could possibly result in a single premature fatality? After all, that's not where expert opinion sits on the matter.
In June of 2011, those notorious haters of anything nuclear *wink* at the American Nuclear Society published slides from their special session on the accident at Fukushima. There, Dr. Peter Caracappa shared that deaths from accumulated radiation exposures “can’t be ruled out” and that “conservative risk estimates” put the number in the realm of hundreds of cases. I’m happy to listen to the expertise of the chief radiation safety officer at Columbia University; or, at least, I’m not sure why he would be blindly disregarded and without comment. But the Our World in Data team omits other legitimate and readily available assessments of harm, too. For example, a 2012 Stanford University study by Ten Hoeve and Jacobson, from the Department of Civil and Environmental Engineering, tells us radioactivity released from the Fukushima failure is estimated to result in between 24 and 1,800 excess cancers (confidently around 180; but, again, note the huge uncertainty intervals we're dealing in.) From that number, the report suggests we can expect 130 deaths (perhaps as low as 15 but potentially as high as 1,100.) This fatality estimate is in the vicinity of Dr. Caracappa’s earlier evaluation for the ANS. Neither best-guess appears outlandish to me.
Still, in the case of Chernobyl, the Our World in Data team took the upper bound estimates for future deaths (maybe in the hundreds of thousands) and reduced this number (not to what teams of experts converged on: a doubtful figure in the tens of thousands, but) down to just a few thousand. I found that odd. But, note that here they haven't even done that. Not even one likely excess death from cancer is added to their already strange death count. What of that? Further, how can we be confident the number is unlikely to be zero?
Well, an article in New Scientist from March 2011, titled Caesium fallout from Fukushima rivals Chernobyl, documents how levels of caesium-137 were high enough to cause concern among Japanese and US Department of Energy officials surveying the area. Gerhard Proehl, the International Atomic Energy Agency’s head of assessment and management of environmental releases of radiation, is quoted as saying the measurements at Fukushima and outlying areas were “really high.” The piece notes how radioactive caesium poses a health risk, has a half-life of 30 years, and how those most highly contaminated areas at Chernobyl were ones with 1,490 kilobecquerels (kBq) of caesium per square metre. The author also explains how food growing in Soviet soil with measurements of 550kBq/m^2 were destroyed rather than being decontaminated. According to the article, regular measurements from 45km away from Fukushima came back with this same reading that had food plants destroyed near Chernobyl: 550kBq. However it turns out that reading was actually low. Caesium recordings as high as 1,752kBq (!) were taken in the town of Kawamata (roughly 60km northwest) as well as 1,816kBq (!!) in Nihonmatsu City (almost 70km west-northwest). The highest reading came back from a location about 35km from the power plant. There they hit 6,400kBq (!!!). I don’t read this reporting as obviously inaccurate or sensationalist. Other publications, such as one from Scientific American, also report very unexpected findings of radioactive particulate making its way into air filters in Tokyo, some 270 km south of Fukushima. If nothing else, this highlights that CO2 production and immediate deaths cannot be the only measures of concern when thinking about health, safety, or sustainability around energy generation.
Further, as with Chernobyl, many folks were forcibly removed from the area around Fukushima due to the above radiation readings. The reports I’ve read suggest approximately 164,000 people within a 20 kilometre radius of the power plant were dislocated. And we also know that five years later 100,000 still hadn’t returned. A decade on and it looks like 35,000 or so have still not gone back. So how insignificant is all of this that it forms an externality in one’s impact assessment? (And one might wonder how many people have to abandon their homes and livelihoods when a wind farm fails; or how many hectares of agricultural land or public forest has to be forfeited, and for how long, when a solar array goes down?)
Many nuclear proponents insist the events of Fukushima are in the past; but even the most cursory look reveals that nothing of concern has concluded. For instance, leakage of radioactive materials into the groundwater and ocean has continued since the day of the disaster. From the beginning, because they had no plan, no tools, and a total lack of storage or treatment capacity, radioactive wastewater was being deliberately (and secretly) released into the sea by the electric company. ("The solution to pollution is dilution.") As hundreds of tonnes of water were being poured onto the damaged reactors to cool them from meltdown, all while hundreds of additional tonnes of water were leaking into the basements on-site buildings, the the power company and emergency crews were pumping out nearly a thousand tonnes of toxic water each day to being with. In the first months alone, we know that hundreds of thousands of tonnes of contaminated water (perhaps 300,000 tonnes) was sent directly into the ocean.
This problem was so unanticipated (despite being warned of in safety reports for at least a decade) and so far over everyone’s heads (despite nuclear power plants commonly being build at the water's edge) that in 2013 World Nuclear News, an information outlet supported by the World Nuclear Association, shared that an information workshop and request for proposals was being organized, with invitations for anyone anywhere in the world to submit suggestions for how to deal with the catastrophe to a website of the newly-created International Research Institute for Nuclear Decommissioning. So, far from being under control, as the Japanese government and power company insisted, nobody on planet Earth had any idea what was going on or what to do about it. No one. Worse, as late as 2014 Japanese media reported that the Tokyo Electric Power Company was improperly measuring radioactive strontium in wastewater and discharging concentrations more than 160,000 times (yes you read that right one-hundred-sixty-thousand times) above permissible limits.
Eventually the company began treating and storing their wastewater. But, of course, that became its own problem. With no plan in place and no real strategy, this improvisation was only intended to buy some time. In the decade since the disaster began, more than a thousand holding tanks (most bigger than your house) have accumulated on site to hold toxic wastewater — now amounting to some 1.25 million tonnes worth. In April of 2021, as capacity approached its limits, the Japanese government authorized plans to (officially) begin the release of a million tonnes of tritium-laden wastewater into the ocean. And that's a plan that most of the world has voiced some objection to.
There has been and will continue to be no immediate or long-term health impacts resulting from of all the above nonsense? That would be a miracle. Regardless, I think this water aspect gives us all a good sense for where the cutting edge of nuclear energy disaster planning, safety, mitigation, decontamination, and waste disposal is at: "I dunno, how ‘bout we just dump it into the sea!" And then one wonders if this reality tilts nuclear power further into or away from the "perfectly safe" category for the folks at Our World in Data and other proponents? And some of us might wonder how solar or geothermal stack up by comparison?
But all of that is just contaminated water. Clean-up of the Fukushima power plant site is far more outlandish. Any disaster recovery will be a significant undertaking but Fukushima promises to be far bigger than first imagined (and once again beyond the understanding of nuclear energy insiders and experts.) Initial estimates said clean-up would take 30-40 years. Quickly that evaluation was revised to around a half-century, with full decontamination and reclamation expected to arrive somewhere around the 2060s; though, some at that same time claimed it could take that long just to develop the technology to begin the task.
Then, in July of 2020, almost a decade after the disaster and following four years of labour, a study group with the Atomic Energy Society of Japan offered a revised date. And their evaluation is truly bleak. Chaired by Hiroshi Miyano, former professor of nuclear engineering at Hosei University, the study proposed two scenarios for decommissioning. Their shortest timeline for returning the site to a usable vacant lot (involving the demolition of all facilities and clearing out more than 7 million tonnes of contaminated waste) is a century, so 2120 (+/- a few decades.) If only partial removal happens, requiring continual monitoring and management, and facility tear-down is delayed until radiation levels drop to safe levels, the timeline stretches out beyond 300 years. But it gets worse.
In 2017, workers at the disaster site were unexpectedly measuring regular radiation levels many times that of the previous highs immediately after the original failure: 530 sieverts per hour rather than the previous peak of 73. For scale, that's enough radiation to render even a robot inoperable in roughly two hours of exposure or 10,000 times the annual allowable dose for a radiation worker. As such, a decade on and the electric company admits it still doesn't even know the extent of the disaster. To make this clearer still, in September of 2020 Japan’s Nuclear Regulation Authority conducted a study of radiation at the Fukushima site using a new, specialized robotic snake (because all their previous devices died from radiation.) On this outing they recorded radiation levels still and again many times higher than expected at the reactor site. Yes. That's right: a full decade after the catastrophe they still didn't even know what they were dealing with (and, obviously, no prior assessment could have taken into account such radiation levels.) More troubling still, you can find reports with annotated camera images from the site, showing nuclear experts speculating about nature of materials captured there. Yes, the world's nuclear scientists appear to have uncovered materials unknown to science at the Fukushima site. So what does all of this mean for estimates of time and cost? What are we at now? 600 years? A thousand?
However the numbers work out, no part of the above frames nuclear as low-risk or safe for me. Even today, it seems clear experts in this field (and anything adjacent) are closer to having no idea how long decommissioning and decontamination will take on a failed nuclear reactor site. And that's the easy and least consequential bit. It doesn’t sound to me like anyone has a solid grasp of the real human, ecological, or merely the dollar cost of all of this, either. On that note, in 2013, an estimate for the cost to taxpayers alone for decontamination, decommissioning, nuclear waste storage, and compensation at Fukushima landed around ¥11 trillion ($120 billion). Just three years later a reassessment doubled that number closer to ¥21 trillion ($235 billion) along with the power company agreeing to contribute another estimated ¥16 trillion (or $175 billion). But that wasn't the end of the estimates. In 2019 a revised figure from the Japan Center for Economic Research suggested, considering that ¥12 trillion has already been spent, the clean-up costs would be closer to somewhere between ¥35 trillion and ¥80 trillion ($734 billion) over just 40 years of decommissioning.
And you'll notice these numbers arrive before the centuries-long timelines I've spelled out above. I've not been able to find updated cost estimates based on more recent information. Could it double again? If nothing else, I think we should be able to agree that the near-bottomless money-hole that this disaster has become — with unknown future costs and hazards and a closing date tentatively set in the 24th or maybe 29th century — the real impact will be non-zero. For nuclear power proponents this should be very concerning as such a financial burden must cut into any budget for power plants and will inevitably be added to risk assessments for any future project. (Unless, as with direct and indirect fatalities, we just choose to disregard everything we know going forward.) More to the issue here, in this light, the Our World in Data evaluation of Fukushima looks to me to be at best extraordinarily weak.
TO THE FUTURE, AND BEYOND
If the disasters of ‘86 and ‘11 are a worst-case (and we’re omitting most human deaths and suffering, as well as all other significant nuclear power plant accidents prior and since) what is the best-case? Celebrated by many is the first nuclear power plant to arrive in Europe in more than a decade: Finland’s fifth reactor and the world’s premiere third generation pressurized water reactor: the 1,600MW Olkiluoto 3. Proponents also boast of Finland’s cutting edge spent fuel disposal facility. To talk to proponents, all of this is meant to usher in a new wave of reactors across Europe and maybe the world, offering modern, safe, reliable, affordable energy all while being a climate change-responsive low-carbon source.
Well, what do we know about this reactor? Olkiluoto 3 was ordered in 2003, construction of the reactor began in 2005, and completion was scheduled for 2009. We also know that, to date (January of 2022), the plant still has to be connected to the grid. The reactor is built and has been turned on but production of electricity is (re)scheduled to begin sometime in mid 2022, well over a decade late. Luckily for Finns, the taxpayer cost for the reactor was fixed at €3 billion ($4.3 billion), meaning that all cost overruns are shouldered by the contractor.
And, on that note, reports suggest the contractor has lost at least €5.5 billion ($7.9 billion) on the project, making the reactor almost three times the original price tag. However, this may be nothing near the real figure. If you go looking, back in 2008 the World Nuclear Association confirmed that “...each year of additional delay in the construction of a nuclear power plant adds another estimated $1 billion [USD or €885 million] to the cost.” So, according to the industry itself, it seems we should expect the final bill to be closer to €15 billion, five times the initial cost. Most significantly, as with many infrastructure projects, the lowest of preliminary estimates for nuclear facilities are often seen as prohibitively high, making other sources much more palatable. Whomever is footing the bill, multiplying the cost by three or five is not irrelevant and doesn't make nuclear an obvious bet or even plausible for many.
In addition to the delays and cost overruns, there was also every manner of bungling associated with Olkiluoto in the intervening years. For example, Finnish regulators found the contractor had no experience building nuclear power plants, was unprepared for such a project, and the company and key personnel lacked the basic management experience to pull off the project. (Does this give anyone confidence? Does it make the project look nothing like Fukushima or Chernobyl?)
My favourite report on Olkiluoto was one showing parts of the project were completed by cheap, non-unionized foreign labour. It came to light early on in the project that a transnational labour subcontractor working on the project was operated by the Bulgarian mafia. That's real, not a bad TV plotline. Reports emerged that some workers at Olkiluoto were not being paid social security and at times went unpaid entirely, while also being forced to fork over their income for mafia protection. The Finnish Minister of Labour, Lauri Ihalainen, was quoted by local news explaining that, “These problems have continued for many years. Promises to improve conditions have come and gone. There is something very wrong here.” But all of this is just further insight into the state of the nuclear power industry; and the Finnish gong show that's being offered as a model for Europe and the world.
Learning all of this caused me to seek out what Finland — this tiny, remote, sparsely populated nation (similar in population and density to Queensland, Australia or British Columbia, Canada) — gets for its five nuclear reactors and at a cost of untold billions? You may be surprised by the answer.
Finland's nuclear reactors are celebrated as being among the world's most productive, with an average capacity factor around 93% (meaning that if it's rated at 1,600 MW its actual energy output is pretty close to that.) The nuclear plant at Olkiluoto has two existing reactors built in the '70s. Each is rated at 890 MW and together are said to produce 22% of Finland's energy. Reactor three, the new one, should it produce power for the grid in coming months, will alone cover about 20% of the nation's energy needs (or electricity for roughly 1.1 million people.) The other two reactors composing Finland's nuclear grid are found at the Loviisa Nuclear Power Plant. Both are '70s era, Soviet-designed reactors with capacities of 507 MW each, making up a combined 14% of the grid; however, these older plants are set for decommissioning in the next few years. All of this means that, temporarily, this five reactor nuclear power system should supply enough power for less than 60% of the tiny population of Finland. To me that's hugely unimpressive and nothing like what's suggested by nuclear energy enthusiasts.
In this light, if those oldest reactors were replaced by two or three modern ones and the Finns then topped up their system to be 100% nuclear (some people's dream) they would need an additional two or three reactors on top of this just to meet the needs of today. That would be as many as six more reactors, at a cost of another €18-30 billion and take maybe another fifteen to thirty years (if each isn't delayed a decade.) Maybe that makes sense for wealthy Finland (or states or provinces in Australia or Canada), where land is cheap and reactors can be situated distant from population centres and vital resources. But is this what the world needs to get us to a more carbon-sensible place? Isn't Finland (like Canada and Australia) nearly irrelevant to the bigger picture? And is any of this context at play where low-carbon energy sources are needed most: where populations are dense and booming and future need is greatest? The number of reactors on Earth at present sits around 440. Just to replace our worst coal and gas plants the world would have to accept hundreds or even thousands of new nuclear reactors and in less well-equipped situations than the former USSR or Japan.
For example, the fastest growing populations on the planet are in nations with relatively small area and dense populations around equatorial Africa such as South Sudan, Uganda, Benin, Cameroon, and Nigeria. Not one of these countries have a single nuclear reactor, nor do any of their bustling neighbours, and the region today booms with half a billion people. Uganda alone (with among the lowest electrification on Earth and 30% less land than Finland but 41 million more people) would need dozens of nuclear reactors to replace their small number of fossil fuel plants and provide just some power to only some of the population. That, to me, is crazy-talk. If single new reactors, like Finland’s Olkiluoto 3, provide power for maybe a million people, are advocates hoping to add 500 reactors (at $4B-$10B a pop) to this region alone and maybe 1,000 in Asia and another 500 or so in the Americas? If not, why is that? Doing so is perfectly safe, after all. Further, if they aren’t calling for this, what are we even talking about?
Because you can't mention it without having proponents freak out, none of this even touches on the reality that when we give nations nuclear reactor technology for benign purposes, as Canada did with India at Rajasthan in the 1960s, it often results in those countries developing nuclear weapons. You can't bring those cases up or pose them as real threats to security or life or you're engaged in irrational scaremongering. So, you may not want Benin or South Sudan to have nukes at this time but it's just best not to bring it up. Trust me.
STORAGE
Well, maybe proponents are concerned about the safety and storage of spent fuel and mining waste. All these reactors, or just some of them, would require dozens or hundreds more high-level waste sites. Unlike much of our burdensome waste, high-level nuclear waste requires treatment, management, isolation, and defence. The biggest problem with such operations is the extremely long periods of time some material remains potentially hazardous to life. Short-lived products, like strontium-90 and caesium-137, have a half-life of decades. Long-lived ones, such as iodine-129 and technetium-99, persist in a problematic state far longer. How long? 15,700,000 years and 210,000 years respectively. And what is the cutting-edge containment plan associated with Olkiluoto, at Onkalo, that everyone is excited about?
The Finnish site is similar to the Waste Isolation Pilot Plant (WIPP) opened in New Mexico in 1999. Both are examples of a "deep geological repository" (modelled after two similar sites built in Germany the early and late 20th century.) If "deep geological repository" sounds to you like "big hole in the ground" (and calls to mind Fukushima’s innovative strategy of "just dump it in the ocean") you would be correct. Yes, it sounds a little primitive, but maybe simple is safe. Right? Right?!
Well, in 2012 the Royal Institute of Technology in Sweden published a report, from researchers Szakálos and Seetharaman, arguing that the copper capsules to be used in Finland will not serve the intended purpose. Though their analysis has been disputed, I'm unable to find a retraction or reassessment from the authors or the Royal Institute in the decade since. That and any other evaluation would be very interesting to look at. Given the total chaos that seems to pervade the entire nuclear industry, I would not be surprised to learn that we haven't even developed a theoretical method for containing nuclear waste.
Also concerning is that in 2014 the $25 billion dollar New Mexico repository had a little $2.5 billion dollar accident that saw the site closed for three years. There, a waste explosion occurred and subsequent fire raged for hours deep underground. The fire and explosion resulted in the release of smoke laden with toxic and radioactive materials, including plutonium and americium, exposing dozens of employees, including 13 above ground workers (seven of whom were treated on site, six others were rushed to hospital.) This accident was not contained to the site, however, and actually scattered trace amounts of waste materials up to a kilometer from the site.
Of the event, a report from the Bulletin of the Atomic Scientists arriving a month later explained that "Wastes containing plutonium blew through the WIPP ventilation system, travelling 2,150 feet to the surface, contaminating at least 17 workers, and spreading small amounts of radioactive material into the environment." Following the accident, the New Mexico Environment Secretary, Ryan Flynn, told reporters "Events like this simply should never occur. From the state's perspective, one event is far too many."
An official report arriving the following year revealed the situation on the ground was completely insane. As NPR explained, the immediate cause of the fire was due to nuclear waste disposal technicians randomly switching from using inorganic clay cat litter for soaking up liquid nuclear waste spills to an organic wheat-based cat litter. Following the stupidity-induced accident, experiments showed:
Various combinations of nitrate salt, Swheat Scoop® [cat litter], nitric acid, and oxalate self-heat at temperatures below 100°C. Computer modeling of thermal runaway was consistent with the observed 70-day birth-to-breach of Drum 68660 ... releasing radioactive uranium, plutonium and americium throughout the underground facility.
A 2019 study from Stanford University's Center for International Security and Cooperation cited three failings at the WIPP stie that resulted in this "accident" [sic]: mismanagement at the Los Alamos National Laboratory; structural problems within the National Nuclear Security Administration, the Energy Department, and the Office of Environmental Management; and a breakdown in “nuclear culture.”
Of course, this one-too-many event arrived just 15 years into the intended 10,000 year lifetime of this "stable and safe" storage site for radiological materials. In this way, nuclear truly seems like it is a lightning rod for "unlikely" or "impossible" events. Will Onkalo be different?
All I know for sure is that the 2005 International Atomic Energy Agency report, Cost estimate of Olkiluoto disposal facility for spent nuclear fuel, spelled out the price tag for the Finn's big hole like this:
The total investment costs of the disposal facility are estimated to be 503 M euro (Million Euros), the total operating costs are 1,923 M euro and the decommissioning and the closure costs are 116 M euro totaling 2,542 M euro. The investment costs of the above ground facility are 142 M euro, the operating costs are 1,678 M euro. The repository investment costs are 360 M euro and the operating costs are 245 M euro. The decommissioning costs are 7 M euro and the closure costs are 109 M euro.
Given that every cost estimate for building, maintenance, and decommissioning of nuclear facilities in every report I've ever come across turns out to be so far below the final figure, it's probably safe to say these numbers are laughable. (See Fukushima above.) When you add this to the fact that that these figures are decades old, it's probably safe to say Onkalo will cost many billions ($5B? $25B?) more than anticipated.
CONCLUSION
Nuclear is perfectly safe and clean. Obviously. Oh, and it's inexpensive as well.
+ + +
ADDITIONAL RESOURCES
“The true toll of the Chernobyl disaster” - BBC - July 2019
United Nations Scientific Committee on the Effects of Atomic Radiation - Report to the UN General Assembly - 2013
“Thirty years of Chornobyl catastrophe: Radiological and health effects” - National Report of Ukraine - 2016
"Into Eternity" - Michael Madsen - 2010
Comments