The first problem: Sovacool lumps common industry terms, “incidents” and “accidents,” into one venti-sized category of “accidents.” Why does he do this? I’m guessing to add to the Armageddon-like anxiety he wants his readers to feel.
Sovacool states:
Incidents are unforeseen events and technical failures that occur during normal plant operation and result in no off-site releases of radiation or severe damage to equipment. Accidents refer to either off-site releases of radiation or severe damage to plant equipment. …
Under these classifications, the number of nuclear accidents, even including the meltdowns at Fukushima Daiichi and Fukushima Daini, is low. But if one redefines an accident to include incidents that either resulted in the loss of human life or more than $50,000 in property damage, a very different picture emerges.
He continues by providing additional examples of how everything falling under his redefined “accidents” category leads to death, destruction and demise. However, a couple of key points should be made before he submits his new definition to Merriam Webster.
First, “accidents” are NOT the same as “incidents,” and should not be treated the same way. Each term is distinctly classified because they each require different responses by involved government, regulatory and other agencies in how they are addressed. For example, if you look at auto insurance—your auto insurance company will treat a minor fender bender differently than an accident involving total demolition of the vehicle and injury to the driver (and possibly others). The reason for this is NOT because the auto industry is trying to cheat you, but rather because the auto carrier has to look at the overall picture—safety implications (establish fault—with the driver or equipment?), involved parties (emergency responders, hospitals, etc.), and overall costs of repair.
In the nuclear industry, “incidents” and “accidents” are treated similarly in that they are classified based on how they should be handled and what agencies should be involved. In fact, if you look at the International Atomic Energy Agency’s International Nuclear Event Scale (INES), you will see that they have categorized the two types differently according to affects the incident or accident has on: people and the environment, radiological barriers and control, and defense-in-depth. This categorization serves an important purpose in establishing roles and responsibilities for managing the crisis and determining how to best address and fix the safety problem. For example, an incident involving a fire in one area of the facility not near the nuclear reactor will not warrant a full-scale international investigation, extensive radiation monitoring, federal responders, etc. However, if that fire sparked near the reactor and caused an “accident” at the site, those actions by the nuclear industry would probably be warranted and the global nuclear industry would be at the plant’s doorstep trying to take steps to prevent them from occurring at their own sites.
This is also not to say that the nuclear industry does not learn from “incidents” as well—or that they are any less important. The U.S. Nuclear Regulatory Commission is revered as a model to countries around the world that are creating their own nuclear regulating bodies because of its strict regulatory oversight. The regulator’s role in an “incident” would be just the same as during an “accident”—to ensure that the plant is operating safely or to shut it down immediately. No questions asked. The industry takes safety incredibly seriously and has many layers of built-in protection in reactors’ designs and reinforces best operating practices through employee reporting mechanisms and other daily checks to ensure safety is always first.
Which brings me to another gaping hole in Sovacool’s argument: he clearly ignores the evidence about the nuclear industry’s strong safety record. In paragraph four, he states that the number of nuclear accidents “is low,” but chooses to ignore this crucial fact for the sake of his flawed argument.
Let’s look at the facts. In its 2010 safety and operations report, the World Association of Nuclear Operators, an international organization that consolidates best practices from operating nuclear plants worldwide, found on the topic of safety system performance:
For the 12th straight year, key backup safety systems concurrently met their individual availability goals more than 90 percent of the time. Nuclear power plants are built with multiple safety systems and backup power supplies so these systems are available, if needed, even when maintenance is being performed on a similar system or component. The three principal backup safety systems are two main cooling systems and back-up power supplies used to respond in the event of unusual situations. Each system at every plant has an availability goal just shy of 100 percent, and 93 percent of these backup safety systems met their individual goal, assuring that multiple layers of safety were in place as designed.
On the topic of industrial safety, the report also goes on to state that:
The nuclear industry is one of the nation’s safest working environments. U.S. nuclear plants continued to post a low industrial accident rate in 2010 with 0.09 industrial accidents per 200,000 worker-hours, the lowest level in a decade and well below the 2010 goal of 0.2. Statistics from other industries through 2009, as compiled by the Bureau of Labor Statistics, show that it is safer to work at a nuclear power plant than in the manufacturing sector and even the real estate and financial sectors.
(See NEI’s press release for more information on the report.)
Coincidentally, Sovacool also mentions that the nuclear industry’s death rate—under his new definition of “accidents”—would be extraordinarily high—a claim that does not stand scrutiny if you examine the facts. A March post by Next Big Future gives the latest data in this regard, showing the nuclear industry to have the lowest average death rate per terrawatt-hour (0.04), lower than the coal, oil, natural gas, biofuels, solar, wind, and hydro industries.
Sovacool’s hasty generalizations extend beyond nuclear plants to include reprocessing facilities, and he also shares incredible, “dirty” stats on the nuclear industry:
To put a serious accident in context, according to data from my forthcoming book Contesting the Future of Nuclear Power, if 10 million people were exposed to radiation from a complete nuclear meltdown (the containment structures fail completely, exposing the inner reactor core to air), about 100,000 would die from acute radiation sickness within six weeks. About 50,000 would experience acute breathlessness, and 240,000 would develop acute hypothyroidism. About 350,000 males would be temporarily sterile, 100,000 women would stop menstruating, and 100,000 children would be born with cognitive deficiencies. There would be thousands of spontaneous abortions and more than 300,000 later cancers.
I’ll have to stay tuned to his latest book to see what kind of scientific basis there is to his figures and to investigate whether or not he includes the protective actions that would be taken by the utility or local/state/federal government in the event of an accident. But for now, I’d just like to point out that even in the case of Fukushima, there have not been any radiation-related deaths. The last of three deaths that NEI reported at the facility was not believed to be from radiation, and the other two workers died while trying to stabilize the plant during the tsunami.
Given that the overall premise of his argument is flawed and that none of his facts or stats are cited (or footnoted for that matter!), I’d caution everyone to think twice before believing his “dirt” on the nuclear industry.
Pictured: Benjamin K. Sovacool, from GoodPlanet.info.
Comments
When someone claiming authority is wrong by that many orders of magnitude, with the evidence available publicly, I conclude that they no longer care about the truth, and are embarked upon deception in some "greater cause".
Here's the link.
You missed one in your debunking, right before your eyes
Under these classifications, the number of nuclear accidents, even including the meltdowns at Fukushima Daiichi and Fukushima Daini, is low.
There was a meltdown in Fukushima Daini ??? Seriously ???? First time I hear about it. And I thought the four reactors over there had reached cold shutdown, 4 days after the earthquake. There must be some really serious, high-grade conspiring at play here, to completely bury reports of a meltdown just right next door to the largest nuclear accident ever in Japan.
Must be that shadowy, all-powerful nuclear lobby they always talk about :-)
I would attribute the low worker accident rate to the safety culture in the nuclear industry. Having worked at a few plants, I can attest to the fact that not only is safety drilled into your head from day one, but safety violations are taken very seriously and failing to report a violation is a very big deal. Even something that would seem really minor to an outsider like chewing gum or smoking in an area where it's not permitted would result in a serious reprimand and would probably be reported to the NRC.
David/Brian:
Sovacool wrote a paper a couple years ago (2008?) in Energy Policy that compiled and summarized several papers on the life cycle greenhouse gas emissions of nuclear power plants. Figures from the paper have been widely quoted in the anti-nuclear press (e.g. Gore's latest book), and in retrospect I'm surprised that NNN never picked up that story. While I disagree with some of the criteria Sovacool's paper used to exclude papers from the study, I take greater issue with emphasis on the "average" results which have been widely reported. In this case the "average" has been severely skewed by the fact that papers by Storm van Leeuwen are overrepresented and that when a range of values has been given (e.g. 10-100 gCOeq/kWh) the midpoint value was used even if it's not representative of the norm. I'd like to see NEI tackle these kind of reports in the future.