Water and sediment samples were collected by the U.S. Geological Survey in 12 watersheds in Johnson County, northeastern Kansas, to determine the effects of nonpoint and selected point contaminant sources on stream-water quality and their relation to varying land use. The streams studied were located in urban areas of the county (Brush, Dykes Branch, Indian, Tomahawk, and Turkey Creeks), developing areas of the county (Blue River and Mill Creek), and in more rural areas of the county (Big Bull, Captain, Cedar, Kill, and Little Bull Creeks). Two base-flow synoptic surveys (73 total samples) were conducted in 11 watersheds, a minimum of three stormflow samples were collected in each of six watersheds, and 15 streambed-sediment sites were sampled in nine watersheds from October 2002 through June 2004.
Discharge from seven wastewater treatment facilities (WWTFs) were sampled during base-flow synoptic surveys. Discharge from these facilities comprised greater than 50 percent of streamflow at the farthest downstream sampling site in six of the seven watersheds during base-flow conditions. Nutrients, organic wastewater-indicator compounds, and prescription and nonprescription pharmaceutical compounds generally were found in the largest concentrations during base-flow conditions at sites at, or immediately downstream from, point-source discharges from WWTFs. Downstream from WWTF discharges streamflow conditions were generally stable, whereas nutrient and wastewater-indicator compound concentrations decreased in samples from sites farther downstream. During base-flow conditions, sites upstream from WWTF discharges had significantly larger fecal coliform and Escherichia coli densities than downstream sites. Stormflow samples had the largest suspended-sediment concentrations and indicator bacteria densities. Other than in samples from sites in proximity to WWTF discharges, stormflow samples generally had the largest nutrient concentrations in Johnson County streams.
Discharge from WWTFs with trickling-filter secondary treatment processes had the largest concentrations of many potential contaminants during base-flow conditions. Samples from two of three trickling-filter WWTFs exceeded Kansas Department of Health and Environment pH- and temperature-dependent chronic aquatic-life criteria for ammonia when early-life stages of fish are present. Discharge from trickling-filter facilities generally had the most detections and largest concentrations of many organic wastewater-indicator compounds in Johnson County stream-water samples. Caffeine (stimulant), nonylphenol-diethoxylate (detergent surfactant), and tris(2-butoxyethyl) phosphate (floor polish, flame retardant, and plasticizer) were found at concentrations larger than maximum concentrations in comparable studies.
Land use and seasonality affected the occurrence and magnitude of many potential water-quality contaminants originating from nonpoint sources. Base-flow samples from urban sites located upstream from WWTF discharges had larger indicator bacteria densities and wastewater-indicator compound concentrations than did base-flow samples from sites in nonurban areas. Dissolved-solids concentrations were the largest in winter stormflow samples from urban sites and likely were due to runoff from road-salt application. One sample from an urban watershed had a chloride concentration of 1,000 milligrams per liter, which exceeded the Kansas Department of Health and Environment's acute aquatic-life use criterion (860 milligrams per liter) likely due to effects from road-salt application. Pesticide concentrations were the largest in spring stormflow samples collected in nonurban watersheds. Although most wastewater-indicator compounds were found at the largest concentrations in samples from WWTF discharges, the compounds 9-10, anthraquinone (bird repellent), caffeine (stimulant), carbazole (component of coal tar, petroleum products), nonylphenol-diethoxylate (detergent surfactant),