Estimating joint distributions of contaminants in United States community water system sources

The 1996 amendments to the U.S. Safe Drinking Water Act mandate revision of current maximum contaminant levels (MCLs) for various harmful substances in community drinking water supplies. A revised MCL for any contaminant must reflect a judicious compromise between the potential costs and benefits of lowering exposure. This evaluation requires detailed information about the occurrence of the contaminant and the costs and efficiencies of the available treatment technologies. Although community water systems must comply concurrently with the MCLs for over 80 regulated substances, regulations generally are set one contaminant at a time. The failure to consider the joint behaviors of multiple contaminants during the regulatory process can lead to mischaracterization of the actual costs and benefits.; In order to estimate more effectively the true costs and benefits of simultaneous compliance with standards for several contaminants, the U.S. Environmental Protection Agency is attempting to expand existing regulatory evaluation methods to account for multiple contaminants. Such technology requires not only the joint consideration of treatment options, but also the joint occurrence distributions of the contaminants. The purpose of this thesis is to develop and implement methods to assimilate contaminant co-occurrence information from a number of water quality databases.; This thesis begins by reviewing existing water quality modeling methodology, highlighting a number of trends that are not ideal from a statistical perspective. Through a detailed model comparison study for arsenic, it next establishes a reasonable structure for modeling a single contaminant. It then provides the requisite statistical and implementational theory for an extension of this model to the simultaneous consideration of multiple contaminants. It examines the performance of this extension relative to independent models, and presents modifications to the model that can be used to account for database heterogeneity. It then uses the models to make inferences about national contaminant levels, showing how these inferences can be sensitive to marginal versus joint modeling. Finally, it derives statistical theory allowing finished water concentrations to inform the model via the synthesis with models for treatment mechanisms. The methods presented in this thesis make significant progress in redressing all of the enumerated shortcomings of existing analyses.