USGS2003
Streamflow and surface-water-quality data were collected from November 1995 through April 1998 (water years 1996-98) from a surface-water network in the Puget Sound Basin study unit of the U.S. Geological Survey National Water-Quality Assessment program. Water samples collected monthly and during storm runoff events were analyzed for nutrients, major ions, organic carbon, and suspended sediment, and at selected sites, samples were analyzed for pesticides and volatile organic compounds. Eleven sites were established in three major watersheds--two in the Skokomish River Basin, three in the Nooksack River Basin, five in the Green-Duwamish River Basin, and one site in Thornton Creek Basin, a small tributary to Lake Washington. The Skokomish River near Potlatch, Nooksack River at Brennan, and Duwamish River at Tukwila are integrators of mixed land uses with the sampling sites locally influenced by forestry practices, agriculture, and urbanization, respectively. The remaining eight sites are indicators of relatively homogeneous land use/land cover in their basins. The site on the North Fork Skokomish River is an indicator site chosen to measure reference or background conditions in the study unit. In the Nooksack River Basin, the site on Fishtrap Creek is an indicator of agriculture, and the Nooksack River at North Cedarville is an indicator site of forestry practices in the upper watershed. In the Green-Duwamish River Basin, Springbrook Creek is an urban indicator, Big Soos Creek is an indicator of a rapidly developing suburban basin; Newaukum Creek is an indicator of agriculture; and the Green River above Twin Camp Creek is an indicator of forestry practices. Thornton Creek is an indicator of high-density urban residential and commercial development. Conditions during the first 18 months of sampling were dominated by above-normal precipitation. For the Seattle-Tacoma area, water year 1997 was the wettest of the 3 years during the sample-collection period. Nearly 52 inches fell (about 14 inches above average) and monthly precipitation was often 200 percent of normal. The wet years kept streamflows generally above normal and contributed to high concentrations of pesticides, nutrients, suspended sediment, and organic carbon in samples. On the basis of chemical concentrations, dissolved oxygen concentrations, and water temperature, the relative quality of water among the 11 study sites ranged from exceptionally high in the North Fork Skokomish and the Green to fair in Springbrook and Thornton. Water in the large rivers (Skokomish, Nooksack, Green-Duwamish) and in two of the small streams in the Puget Sound Lowlands (Big Soos and Newaukum) was characterized by dilute water chemistry with dissolved solids concentrations less than 130 milligrams per liter. Water in three other small streams in the Lowlands (Fishtrap, Springbrook, and Thornton) had dissolved solids concentrations as high as 320 milligrams per liter. Nutrient and pesticide concentrations mostly were higher in the small streams than in the large rivers. Suspended-sediment concentrations, however, were highest in the large rivers, with averages ranging from 85 to 443 milligrams per liter. During storm and flood events, suspended-sediment concentrations in samples from the Nooksack were as much as 2,800 milligrams per liter, and from the Skokomish, 1,500 milligrams per liter. Out of 86 pesticides and 86 volatile organic compounds analyzed, a total of 35 pesticides and 11 volatile organic compounds were detected at concentrations above laboratory reporting levels in samples collected from the four intensively studied sites, the lower Nooksack River, Duwamish River, Fishtrap Creek, and Thornton Creek. Herbicides were detected more frequently than insecticides. The herbicide prometon was detected in 66 percent of all 124 samples collected, followed by simazine (65 percent), atrazine (64 percent), and the insecticide diazinon (50 percent). The detected volatile organic compounds belonged to one of three general classes--fuel-related, the trihalomethanes, or solvents. Water-quality standards and criteria for drinking water or for the protection of aquatic life were nearly always met, except on occasion by concentrations of dissolved oxygen, iron, and manganese in samples from Springbrook and Fishtrap, and concentrations of diazinon in Thornton and Fishtrap. Concentrations of three other insecticides (carbaryl, chlorpyrifos, and lindane) also did not meet criteria protecting aquatic life in a few samples (from 1 to 5) from Thornton and Fishtrap and from the Duwamish River. Water-quality conditions at the sampling sites resulted from natural characteristics of the watersheds and from human activities. In the Nooksack and Skokomish, natural conditions were probably important contributors to the large suspended-sediment concentrations observed during high streamflow events. However, logging and road building in the basins might also have contributed to the suspended-sediment loads. In the forested study basins, the water chemistry was more dilute, based on dissolved solids and nutrient concentrations. The North Fork Skokomish (indicator site of background conditions) and the Green (indicator site of forestry and natural conditions) were characterized by high-quality water with low concentrations of dissolved solids, nutrients, and sediment and by standards or criteria that were nearly always met. Unlike the high-quality water at the background and forestry sites, natural conditions appear to have adversely affected the quality of water in Springbrook. High concentrations of iron and manganese and low dissolved-oxygen concentrations were possibly the result of ground-water discharge at this site. Of all the sites, nutrient concentrations were highest in the agricultural streams (Fishtrap and Newaukum) and next highest in the urban streams, indicating that human activities affect the water quality of these streams. Of the three large rivers integrating a mix of land uses, the lower Nooksack, in an agricultural setting, on average had the highest total nitrogen and phosphorus concentrations, followed by the Duwamish in an urban setting, and the Skokomish in a forestry-practices setting. In both urban and agricultural settings, pesticides affected water quality at the four sites sampled for organic compounds. However, more types of pesticides were found at the agricultural sites than at the urban sites. The most pesticides were detected in Fishtrap. Many of these were pesticides applied for agricultural purposes, but several were pesticides applied in urban settings, such as those used in maintaining rights-of-way. In urban Thornton Creek, however, high application rates and retail availability of pesticides likely resulted in generally higher concentrations, the detections of certain pesticides in many samples year round, and the number of times insecticide concentrations did not meet chronic criteria for the protection of freshwater aquatic life. The two sites located in urban areas also had more detections of volatile organic compounds and at a higher detection frequency than the two sites in agricultural areas. The most compounds were detected in Thornton, and in the highest concentrations. Conversely, the least number of compounds was detected in the lower Nooksack, with only 2 percent urban land upstream. Methyl tert -butyl ether, an automotive fuel additive, was detected in a few samples from Fishtrap. Many solvents, such as trichloroethene and tetrachloroethene, were detected only in samples from the urban sites on the Duwamish River and Thornton Creek. The only volatile organic compound with a distinctly agricultural signature was 1,2-dichloropropane, detected only in Fishtrap Creek samples.