ation. Chronic and late effects take time to develop. We know of one study well under way that tends to meet the requirements discussed here.128 It is a prospective study of comparable populations in six small cities. One such study, however, cannot answer the massive array of questions that confront the policy maker who needs critical examination of the validity of present standards. Much more work will have to be done, and over an extended period of time. We favor conducting much of this work outside government laboratories to ensure its independence, and to provide flexibility to draw upon the country’s scientific manpower.

CATASTROPHES

Nuclear power plants, liquefied natural gas, and large dams pose a danger of catastrophic accidents. The greatest potential risk is that of catastrophic accidents in nuclear power plants.*


Gas129 The principal potential hazard of natural gas is associated with leaks in the pipeline distribution system. In the case of liquefied natural gas (LNG), its cryogenic properties create additional hazards. Contact, for example, will damage human tissue. On vaporization and exposure to oxygen and a source of ignition, the gas will burn and possibly explode. The worst LNG accident on record occurred at a storage facility in Cleveland in 1944: The accident killed 130 people and caused $10 million worth of property damage. It is highly unlikely that an accident of this type could occur today, since the development and use of materials that resist brittle failure at cryogenic temperatures have largely eliminated the cause. Nevertheless, if liquefied natural gas is released in an accident, it may form a vapor cloud that travels several miles before igniting. In 1972, 40 workers were killed by inhalation of the vapors in an emptied tank on Staten Island. Liquefied natural gas is now being shipped from foreign countries in tankships for receipt in ports with large storage facilities. The degree of hazard in shipping and storing LNG is controversial.130


Hydroelectric Power131 In the 40-yr period from 1918 to 1958, 1680 deaths occurred from the failure of five dams: a statistical average of 40 deaths per year. The total number of dam failures was 33. Assuming that the average number of dams in the United States over this period was 1000, a failure rate of about 8×104 per dam-year can be estimated, and a major disaster rate of 1.3×104 per dam-year.

*

See statement 9–8, by H.Brooks, Appendix A.



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