Since their inception, the national networks have provided water-quality data for numerous watersheds with diverse climatic, physiographic, and cultural characteristics, which has benefited efforts to investigate the causes of spatial and temporal variations in water quality (Smith and others, 1987a; Smith and Alexander, 1983). The watersheds range from the relatively small, minimally disturbed basins in the HBN to the larger, more culturally-influenced drainage basins of NASQAN. During the operation of the networks, considerable changes have occurred in terrestrial and atmospheric sources of water pollutants in the United States brought about, in part, by regulatory actions. Examples include changes in fertilizer use, atmospheric sources of sulfur and nitrogen, and wastewater discharges from municipal sewage treatment plants. The national networks have provided some of the best available information for investigating the influences of these pollutant sources on national and regional water quality (e.g., Smith and others, 1987a; Smith and Alexander, 1986; Kramer and others, 1986; Lins, 1986; Smith and others, 1993). In addition, national network data have been used to quantify trends in stream water quality (e.g., Lettenmaier and others, 1991; Smith and others, 1993), estimate the rates of chemical flux from watersheds (e.g., Alexander and others, 1996a; Peters, 1984), and investigate relations between water quality and streamflow (e.g., Smith and others, 1982) and between water quality and various physical characteristics of the watersheds (Biesecker and Leifeste, 1975; Peters, 1984). Numerous state and local investigations of stream water quality have also used NASQAN and HBN data to examine stream water-quality conditions and trends (e.g., Schertz, 1990; Schertz and others, 1994; Hay and Campbell, 1990; Hainly and Ritter, 1986).
The national networks have provided stream measurements for a relatively consistent and comprehensive set of water properties. Approximately 85 physical, chemical, and biological properties have been analyzed during more than 60,000 visits to 679 stream locations. Relatively consistent sampling and analytical methods have been used at national network sites during their operation. The water-quality data reflect sampling over a wide range of streamflow conditions. As of 1995, expenditures for the two national networks have totaled approximately $95 million, including $80 million for NASQAN and $15 million for the HBN.
Documentation and quality-assurance information on network operations and historical records of sample-collection agencies, laboratories, analytical methods, and estimates of laboratory measurement error are available to assist in the proper use of the stream water-quality data. This information is especially useful for documenting changes in network operations and sampling and analytical methods that may affect the use and interpretation of the ambient water-quality data. Much of this information is contained in USGS publications and previously unpublished USGS internal technical memoranda. Estimates of laboratory measurement error (i.e., variability and bias) have been previously published as part of the Blind Sample Program, an external USGS laboratory quality- assurance program (Maloney and others, 1994). These data have been used to systematically assess the accuracy of USGS laboratory methods and national network stream water-quality data (see for example, Maloney and others, 1994; Alexander and others, 1993).
We have assembled in an easy-to-use format on two CD-ROMs (Compact-Disk Read-Only Memory) historical water-quality and streamflow data and supporting documentation and quality- assurance information for NASQAN and HBN (these data are available on CD-ROM from the U.S. Geological Survey Digital Data Series; see Alexander and others, 1996b). The water-quality data include all measurements made at the national network stations for selected constituents during the periods of 1962-95 for the HBN and 1973-95 for NASQAN. These data on CD-ROM are collectively referred to as WQN, water-quality networks. This report provides the necessary background for understanding the characteristics of the WQN data and the retrieval system for accessing the data.
The first section of this report gives a brief overview of the purpose and content of the CD-ROMs. A list of the key features of the CD-ROM data sets is provided.
The second section of the report describes the objectives, history, and characteristics of WQN data. This section includes information on the field and laboratory methods used in the collection, processing, and analysis of the data and the procedures used to quality assure the water-quality data. We also describe the specific data selected for publication on the CD-ROMs.
The third section of the report describes the history and characteristics of a USGS quality-assurance program, the Blind Sample Program, responsible for independently monitoring the quality of laboratory analytical methods and environmental measurements since 1981. We describe the characteristics of the measurement bias and variability data produced by this program and published on the CD-ROMs.
A fourth section of the report discusses the various ways of using the Blind Sample Program data to evaluate the accuracy of national network stream water-quality data. We discuss characteristics of the quality-assurance data as they relate to those of the stream water-quality data, and describe various techniques for using the quality-assurance data to improve interpretations of the water-quality data.
In a final section, we briefly describe the organization of the WQN data sets and the characteristics of the DOS software supplied on one of the CD-ROM discs for retrieving and outputing the data.