Lecture 14: Is Nuclear Power Safe? Calculating the Number of Deaths Caused by Nuclear Accidents...
MIT OpenCourseWare · 1:15:46 · Yesterday
Nuclear power is a statistically safe energy source with a mortality rate comparable to hydroelectric or natural gas power, even when accounting for the potential of large-scale accidents. While rare "beyond design basis" events can cause hundreds of thousands of latent cancer deaths over a century, these fatalities, when averaged across total electricity production, do not make nuclear power an outlier in terms of public health risk.
-
Modeling approach — Calculations rely on the "Linear No Threshold" model, which assumes that every incremental unit of radiation exposure increases cancer risk, regardless of how small the dose .
-
Primary isotopes — Cesium-137 acts as the main long-term hazard due to its multi-century presence in the environment, whereas Iodine-131 creates an acute radiation dose immediately following an incident .
-
Accident severity — A "typical" major reactor accident is modeled using a weighted average of Chernobyl-like and Fukushima-like events, resulting in an estimated release of approximately 18 petabecquerels of radiation .
-
Fatalities — Applying these exposure models results in an estimated 230,000 total cancer deaths over a century following a major reactor event .
-
Risk assessment gaps — Industry standard safety tools, known as Probabilistic Risk Assessment, significantly underestimate actual historical accident frequencies, often by a factor of approximately 50 .
-
Comparative safety — When normalized for total energy generation, nuclear power results in roughly 1.4 to 2.3 deaths per terawatt-hour, placing it alongside other common energy forms rather than as a uniquely dangerous technology .
-
How does the Linear No Threshold model influence the calculation of total cancer fatalities?