ExternE: Comparing Nuclear Health and Environmental Effects

When an anti-nuclear activist says “No to nuclear power because it isn’t safe” I ask “compared to what?” Decisions about energy options always involve comparative risks and benefits. So to make informed choices the politicians need to be informed and able to evaluate relative risk/benefits. A staggering amount of research has been done to characterize the risks of nuclear, fossil fuels, bioenergy, hydro, wind and solar. From that research the conclusion I reach is that nuclear power is the safest available option to meet the energy demands of both developed and developing countries. Hydro can be similarly safe, but the hydro opportunities are largely already exploited, while we need to keep in mind that in the 1975 to 1985 period some of the biggest man-made energy-related disasters were caused by dam failures in China (The worst energy-related accident was the Banqiao/Shimantan dam failure in China in 1975 when some 30 000 people were killed) and India (Machhu II, India 2500 deaths and Hirakud, India 1000 deaths). Source: “Comparing Nuclear Accident Risks with Those from Other Energy Sources” (PDF).

In the following we will move from the severe accidents comparisons to the full life cycle long term health effects.


Diagram 2 (click for full size). Most of the health risk calculations in ExternE, presented as deaths per TWh (electricity). The diagram shows electricity production facilities in all EU states and in Norway

The graphic at left shows the Deaths/TWh in all EU states + Norway for fossil fuels, bioenergy, hydro, nuclear and wind. This is from the Swedish study “Economic Analysis of Various Options of Electricity Generation – Taking into Account Health and Environmental Effects” by Nils Starfelt and Carl-Erik Wikdahl. The authors started with the exhaustive EU ExternE-Pol studies, then expressed the health/environmental effects in the readily-understood metric of “deaths per TWh (terrawatt-hour)” of electrical generation.

If you examine the chart very carefully you should be able to detect the tiny Nuclear Power data points for Denmark and France.

Mean deaths per TWh for EU states

Diagram 3 (click for full size). Mean values of health effects, presented as deaths/TWh, for the respective forms of electricity generation throughout the EU. These calculations are based on the same data as in Diagram 2.


Some of the more straightforward conclusions that can be drawn from the results shown in Diagrams 2 and 3 are:

1. Coal, lignite and oil result in considerably greater external costs and thus health effects than do the other forms f energy. This difference becomes even greater if the greenhouse effect is also included in the results: see Diagram 7.

2. The external costs of hydro power and nuclear power are about two orders of magnitude less than those from the above-mentioned fossil fuels.

3. Among the fossil fuels, natural gas has considerably less effect on the environment than do the other forms of energy.

4. The external costs of bioenergy, as shown in the ExternE results, lie close to those for fossil fuels, but it should be noted that, in most cases, the results are based on technology for which there is a considerable potential for improvement.

The authors use the ExternE database to model Sweden as an example case. From there they can reason about the impact of closing the Swedish nuclear reactor Barsebäck:

The risk of major effects, and the need for extensive evacuations in the event of an accident at Barsebäck, have dominated the debate in Sweden and Denmark. Using the results presented in this report, it can therefore be of interest to make a comparison between the health risks resulting from a nuclear power station accident and those from normal operating emissions from Danish coal-fired power production.

Closing the Barsebäck reactors will result in a loss of production in Sweden which, during a statistically average climate year, cannot be compensated for by energy savings or by increased production of nuclear power and/or water power in Sweden and Norway. For a number of years into the future, the only possibility is a greater import of electricity produced in coal-fired power plants, mainly in Denmark.

According to data given in ExternE, the increased pollution from operation of these coal-fired power stations, which would otherwise have not been operated if Barsebäck had not been shut down, will amount to about 200 deaths per year, of which most will occur in Denmark but a few also in Sweden.

This conclusion is obvious to students of energy policy, but is never-to-be-discussed in Greenpeace circles. Friends of the Earth, Greenpeace and their ilk are directly responsible for stopping nuclear plant construction, and for the dramatic cost increases driven by activist delaying tactics, and thus the pollution and related deaths from fossil fuel generation that would have been eliminated by expanded nuclear power.

As we do not know exactly how many GW of nuclear capacity would have been constructed (instead of coal and gas), it is a counterfactual to calculate how many deaths we should credit to the accounts of the anti-nuclear activists. I am comfortable with my conclusion that their account has accumulated tens of thousands of unnecessary deaths, and delayed by nearly half a century the development of mass-manufactured modular nuclear plants.

Lastly, for reference I’ll note the authors’ comment on new capacity costs — I’ve not had time to verify their figures:

The generating cost for new capacity in Sweden has been calculated to be in the range of 2.5 to 3.5 EUcents/kWh for hydro, nuclear and gas and about twice as much for bioenergy and wind. Taxes and subsidies are not included.