ABSTRACT
The question of how to define and measure different types of uncertainties is particularly critical in the analysis of high-consequence phenomena (e.g., global climate change or failures of nuclear reactors) because of public sensitivities to the magnitude of the potential outcomes. In such cases, policy making is generally guided by a mix of public concerns and scientific research results, with, one might hope, a link between the two. Identification and quantification of uncertainties then helps clarify what is known and what is not.1 *4 In other instances, however, full uncertainty analysis is not needed, for example, because the best risk management strategy is fairly obvious.
