A probabilistic approach to the assessment of uranium resources

Abstract

This paper presents a probabilistic approach to the estimation of uranium resources that allows for the integration of geologic observations with the experience and judgment of geologists. The paper focuses on estimating roll-front type deposits. The approach is based on a quantitative material balance model of ore formation that describes the quantity of uranium resources in terms of several key parameters constituting the quantity of uranium entering a host, and the fraction of the entering uranium that has been precipitated. The parameters cannot necessarily be measured in the field, but they can be inferred from available information and interpretation of field observations. The key to this approach is eliciting these inferences from geologists, representing the uncertainties inherent in drawing the geologic inferences as probability distributions, and combining the distributions to arrive at a probability distribution for uranium resources in a region. This paper presents the model, procedures for eliciting subjective probabilities and updating the distribution over resources within a Bayesian framework, and a demonstration of the procedure by obtaining estimates for three roll-front type deposits in Wyoming.