Abstract
Injection of waters of varying quality into ground waters is increasingly being considered in water resource management, though potential public health risks associated with pathogens are uncertain. The objective of this study was to develop a microbial dose-response relationship to reflect uncertainty as well as variability in health responses. The results were applied to a ground water model to assess public health risks of ground water injection projects. The major contribution of this study was the analytical derivation of a predictive Bayesian dose-response function, based on the Pareto distribution and a bi-variate prior distribution for use with microorganisms. The use of predictive Bayesian methods has been previously proposed only for chemical risk assessment, though their capability for integrating subjective and numeric information make them attractive for microbial dose-response assessment. The resulting predictive Bayesian CDF was applied to the contaminant concentration contours predicted by computer model to be associated with a salinity barrier ground water injection program, to yield the risk at a given distance from the injection well. The dose-response function developed in this dissertation has general applicability to all injection programs using waters of impaired quality.