sons. For example, USGS collected water quality data through the stream benchmark program in largely pristine small watersheds, in the National Stream Quality Accounting Network (NASQAN) program at the mouths of major river systems, and in its many cooperative study projects with states and local governments where sampling designs and constituents measured were largely problem-driven and particular to the place. EPA, states, and local governments collected water-quality data for monitoring and compliance purposes. Sampling at a given site was often started and stopped, depending on the project duration and funding, and hence few sites had sufficiently long records of consistent analysis to enable valid trend analysis at a national scale. Inconsistencies in data collection included differences in how a sample was taken from a stream, how the sample was handled after collection, and what analytical methods were used to measure chemical and biological constituents. A compelling original argument for NAWQA was USGS’s ability to sustain a consistent, geographically diverse, and quality-assured data collection over decades, and follow through on a scientifically valid study design.
Since the program’s infancy, NAWQA has standardized sampling regimes and network design to enable cross-site comparisons to meet local and regional stakeholder needs, but at the same time to enable a national water-quality assessment. NAWQA brought order to a wide range of practices and motivations in water-quality sampling and analysis. NAWQA uses USGS approved methods that have been developed and tested by USGS researchers and approved for use at a national scale. These methods are periodically published in the USGS National Field Manual for the Collection of Water-Quality Data (USGS, variously dated). NAWQA now provides a nationally consistent data collection and analysis of water-quality samples (Gilliom et al., 1995). In setting this example and working with other groups on consistent practices, NAWQA has also helped to improve the water-quality monitoring efforts of other entities. This is a significant and enduring accomplishment.
NAWQA products are used to assess status and trends in water quality, to evaluate the effectiveness of regulatory programs, to inform policy analysis, and to support ecological risk assessment. For each of these applications, it is essential that data from a limited sampling of environmental attributes be put in geographical and climatic context with the uncertainty of inferences reported. NAWQA has developed and applied robust extrapolation and inference-based techniques that are statistical, geospatial, and/or process-based. These various models support inferences from recent