In dueling studies released yesterday, scientists are debating whether a projection of 130 future cancer deaths from the meltdowns at the reactors in Fukushima last year indicates that nuclear power is good or bad overall to the health of people.
The argument centers on a paper by meteorologist John Ten Hoeve and environmental engineer Mark Jacobson, who are both at Stanford University in California, published online yesterday in Energy & Environmental Science. (A copy of the paper is available.) The pair started with estimates of the radioactive nuclides released when meltdowns occurred in multiple reactors at the Fukushima Daiichi Nuclear Power Plant due to the loss of power to cool nuclear fuel after the earthquake and tsunami of 11 March 2011.
They plugged that nuclides data into a three-dimensional global atmospheric model and produced a global map of the radioactive fallout, comparing their model's predictions to actual measurements of airborne radionuclide concentrations and contamination deposited on the ground as an accuracy check. They then computed the estimated doses of radioactive iodine-131, cesium-137, and cesium-134 individuals around the world received, using a model developed by the U.S. Environmental Protection Agency, and finally estimated the likely incidences of radiation-induced cancers, following the theory that any exposure above background radiation proportionately increases the risk of cancer—the linear no-threshold (LNT) model that is widely used, though still controversial. They conclude that the Fukushima accident is likely to cause between 15 and 1100 additional cancer deaths, with their best estimate being 130 deaths. The number of likely additional cancer cases ranges from 24 to 1800, with 180 being their best estimate. (The wide range in the estimates stems from uncertainties in the models.) In both cases, Japan would bear the brunt of the cancers. Europe and North America would see about one additional cancer death each.
Going a step further, the pair applied the same modeling to a hypothetical accident at the Diablo Canyon Power Plant in California. They concluded that an accident there with the same emissions as in Fukushima would cause 25% more deaths, despite the region's much lower population density, because of differing meteorological conditions.
In an "opinion" on the paper appearing online in the same journal at the same time, Nobel Prize-winning physicist Burton Richter, former director of what is now called the SLAC National Accelerator Laboratory in Menlo Park, California, sees the cancer estimates as evidence of the relative safety of nuclear power. "What struck me ﬁrst on reading the Ten Hoeve-Jacobson paper was how small the consequences of the radiation release from the Fukushima reactor accident are projected to be compared to the devastation wrought by the giant earthquake and tsunami," he wrote, contrasting the 20,000 deaths attributed to the earthquake and tsunami to the relatively small number of deaths expected to result from the radiation. "It made me wonder what the [health] consequences might have been had Japan never used any nuclear power."
Following up on that musing, Richter then presents a calculation of the years of life that would have been lost due to the health effects of pollutants emitted by fossil fuel-burning power plants if Japan had relied on coal or gas instead of nuclear power. And by that measure, Richter says, nuclear power beats gas by a nose and coal by a mile. For each terawatt hour (TW-h) of electrical output, the use of coal causes the loss of 138 years of life, he estimated; the comparable number for gas is 42 years; and for nuclear power, 30 years, including losses attributed to the Fukushima accident by the Ten Hoeve-Jacobson paper. "The obvious conclusion is that nuclear power is better for your health than other choices, a conclusion that may come as a surprise to many," Richter wrote.
Richter added that the Fukushima analysis by Ten Hoeve and Jacobson "is a ﬁrst rate job and uses source of radioactivity measurements that have not been used before to get a very good picture of the geographic distribution of radiation." He also endorsed the team's use of the LNT model "to give an upper bound to the biological effects."
Yet University of Tokyo physicist Ryugo Hayano, who has been measuring the actual radionuclide exposure of Japanese citizens, took issue with the team's approach. "In my view, the [Ten Hoeve and Jacobson] paper applied the LNT model inappropriately," he tells ScienceInsider. Hayano pointed to International Commission on Radiological Protection recommendations that the LNT model be used to set limits for occupational exposure, but that because of the uncertainties involved, it should not be used "to calculate the hypothetical number of cases of cancer or heritable disease that might be associated with very small radiation doses received by large numbers of people over very long periods of time."
Ten Hoeve and Jacobson also had a chance to reply to Richter's thoughts, again online yesterday in Energy & Environmental Science. First, they noted that any cancer mortalities would come on top of the approximately 600 people, mostly elderly and infirm, who died due to the hardships of evacuating from around the Fukushima plant. Second, they then take Richter to task for ignoring that wind, solar, geothermal, hydroelectric, tidal, and other alternative energy sources could have replaced nuclear power instead of simply coal and gas. Third, they also note that neither they nor Richter factored in the environmental effects of mining and drilling for fossil fuels and disposing of nuclear waste. But they note that these issues were beyond the scope of their paper.
Stay tuned for round 2.