systems to health, agriculture, climate and water supply, and ecosystems are described, and wherever possible, compared. After some discussion, the major findings are summed up in nine conclusions. The reader may wish to read these conclusions first.


In ordinary language, as well as in this report, the word risk is used in two ways: to convey the possibility (probability) of loss or to denote a dangerous element or factor. This chapter examines the risks associated with the three principal groups of energy systems—fossil fuel, nuclear, and solar—particularly in the generation of electricity, as this provides a convenient base for comparison.

Risks have been grouped by origin in the various steps of each energy cycle, including extraction and processing of the energy resource; its transportation and storage; its use in the production of another fuel (liquid fuels from coal, for example), electricity, or power; the disposal of waste, and finally end-use. (In engineering literature, fuel cycle is usually synonymous with energy cycle, but in official regulatory practice,1 fuel cycle excludes mining, operation of waste disposal sites, and transportation.)

The complete evaluation of risk depends on the nature and amount of the dangerous element or factor (termed “insult” by some environmentalists) and an understanding of how it stresses or interacts with its targets, of how the targets are affected (termed “insult” in medical literature), and of how they react in turn. Such target reactions can then affect other objects or systems.2

The comparison of risks is often simplified by consistent comparisons—similar kinds of risks that arise when different energy systems are employed for the same specific purpose, such as the production of a stated amount of electricity. The matching of risks may be difficult. Consider, for example, the number of deaths associated with the production of 1 quadrillion Btu (quad) of electricity in 1 year from oil or uranium. Practically all cancer deaths due to the use of oil for 1 year would occur within the 30 years following, but those from uranium might be projected to occur over thousands of years. Would 30 deaths in 30 years be better, worse, or equal to 30 deaths in 1000 years? The decision calls for judgment, or for specific sociological information that shows how the bunching of deaths leads to more or less damage than spreading them over the years.

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement