Expert judgment in assessing radwaste risks: What Nevadans should know about Yucca Mountain; [Final report]

For phenomena characterized by accurate and largely complete data, quantitative risk assessment (QRA) provides extraordinarily valuable and objective information. However, with phenomena for which the data, models, or probabilities are incomplete or uncertain, QRA may be less useful and more questionable, because its conclusions are typically empirically and theoretically underdetermined. In the face of empirical or theoretical underdetermination, scientists often are forced to make a number of methodological value judgments and inferences about how to estimate and evaluate the associated risks. The purpose of this project is to evaluate instances of methodological value judgments and invalid or imprecise inferences that have occurred in the QRA done for the proposed Yucca Mountain high-level radioactive waste facility. We shall show (1) that questionable methodological value judgments and inferences have occurred in some Yucca Mountain QRA`S; (2) that questionable judgments and inferences, similar to those in the Yucca Mountain studies, have occurred in previous QRA`s done for other radiation-related facilities and have likely caused earlier QRA`s to err in specific ways; and (3) that, because the value judgments and problems associated with some Yucca Mountain QRA`s include repetitions of similar difficulties in earlier studies, therefore the QRA conclusions of some Yucca Mountain analyses are, at best, uncertain