Economics of place-based monitoring under the safe drinking water act, part III: performance evaluation of place-based monitoring strategies.

The goals of environmental legislation and associated regulations are to protect public health, natural resources, and ecosystems. In this context, monitoring programs should provide timely and relevant information so that the regulatory community can implement legislation in a cost-effective and efficient manner. The Safe Drinking Water Act (SDWA) of 1974 attempts to ensure that public water systems (PWSs) supply safe water to its consumers. As is the case with many other federal environmental statutes, SDWA monitoring has been implemented in relatively uniform fashion across the USA. In this three part series, we present over 30 years of evidence to demonstrate unique patterns in water quality contaminants over space and time, develop alternative place-based monitoring approaches that exploit such patterns, and evaluate the economic performance of such approaches to current monitoring practice. Part III: Place-based (PBA) and current SDWA monitoring approaches were implemented on test datasets (1995-2001) from 19 water systems and evaluated based on the following criteria: percent of total detections, percent detections above threshold values (e.g. 20, 50, 90% of MCL), and cost. The PBA outperformed the current SDWA monitoring requirements in terms of total detections, missed only a small proportion of detections below the MCL, and captured all detections above 50% of the MCL. Essentially the same information obtained from current compliance monitoring requirements can be gained at approximately one-eighth the cost by implementing the PBA. Temporal sampling strategies were implemented on test datasets (1995-2001) from four water systems and evaluated by the following criteria: parameter estimation, percent deviation from "true" 90th, 95th, and 99th percentiles, and number of samples versus accuracy of the estimate. Non event-based (NEB) strategies were superior in estimating percentiles 1-50, but underestimated the higher percentiles. Event-based strategies were superior in estimating 95th and 99th percentiles, and required significantly fewer samples (than NEB strategies) to estimate the "true" 95th and 99th percentiles. Incorporation of place-based information significantly improves the performance of monitoring and temporal sampling strategies in the context of surface-influenced water systems in the state of Iowa. Application of similar methods to other areas and types of water systems would likely produce similar results. Compared to current SDWA monitoring, the place based approach allows for cost-effective, enhanced characterization of local contaminants of concern.