Quality of streams in Johnson County, Kansas, 2002--10

Abstract

Stream quality in Johnson County, northeastern Kansas, was assessed on the basis of land use, hydrology, stream-water and streambed-sediment chemistry, riparian and in-stream habitat, and periphyton and macroinvertebrate community data collected from 22 sites during 2002 through 2010. Stream conditions at the end of the study period are evaluated and compared to previous years, stream biological communities and physical and chemical conditions are characterized, streams are described relative to Kansas Department of Health and Environment impairment categories and water-quality standards, and environmental factors that most strongly correlate with biological stream quality are evaluated. The information is useful for improving water-quality management programs, documenting changing conditions with time, and evaluating compliance with water-quality standards, total maximum daily loads (TMDLs), National Pollutant Discharge Elimination System (NPDES) permit conditions, and other established guidelines and goals. Constituent concentrations in water during base flow varied across the study area and 2010 conditions were not markedly different from those measured in 2003, 2004, and 2007. Generally the highest specific conductance and concentrations of dissolved solids and major ions in water occurred at urban sites except the upstream Cedar Creek site, which is rural and has a large area of commercial and industrial land less than 1 mile upstream on both sides of the creek. The highest base-flow nutrient concentrations in water occurred downstream from wastewater treatment facilities. Water chemistry data represent base-flow conditions only, and do not show the variability in concentrations that occurs during stormwater runoff. Constituent concentrations in streambed sediment also varied across the study area and some notable changes occurred from previously collected data. High organic carbon and nutrient concentrations at the rural Big Bull Creek site in 2003 decreased to at least one-fourth of those concentrations in 2007 and 2010 likely because of the reduction in upstream wastewater discharge contributions. The highest concentrations of trace metals in 2010 occurred at urban sites on Mill and Indian Creeks. Zinc was the only metal to exceed the probable effects concentration in 2010, which occurred at a site on Indian Creek. In 2007, chromium and nickel at the upstream urban Cedar Creek site exceeded the probable effects concentrations, and in 2003, no metals exceeded the probable effects concentrations. Of 72 organic compounds analyzed in streambed sediment, 26 were detected including pesticides, polycyclic aromatic hydrocarbons (PAHs), fuel products, fragrances, preservatives, plasticizers, manufacturing byproducts, flame retardants, and disinfectants. All 6 PAH compounds analyzed were detected, and the probable effects concentrations for 4 of the 6 PAH compounds analyzed were exceeded in 2010. Only five pesticide compounds were detected in streambed sediment, including carbazole and four pyrethroid compounds. Chronic toxicity guidelines for pyrethroid compounds were exceeded at five sites. Biological conditions reflected a gradient in urban land use, with the less disturbed streams located in rural areas of Johnson County. About 19 percent of sites in 2010 (four sites) were fully supporting of aquatic life on the basis of the four metrics used by Kansas Department of Health and Environment to categorize sites. This is a notable difference compared to previous years when no sites (in 2003 and 2004) or just one site (in 2007) was fully supporting of aquatic life. Multimetric macroinvertebrate scores improved at the Big Bull Creek site where wastewater discharges were reduced in 2007. Environmental variables that consistently were highly negatively correlated with biological conditions were percent impervious surface and percent urban land use. In addition, density of stormwater outfall points adjacent to streams was significantly negatively correlated with biological conditions. Specific conductance of water and sum of PAH concentrations in streambed sediment also were significantly negatively correlated with biological conditions. Total nitrogen in water and total phosphorus in streambed sediment were correlated with most of the invertebrate variables, which is a notable difference from previous analyses using smaller datasets, in which nutrient relations were weak or not detected. The most important habitat variables were sinuosity, length and continuity of natural buffers, riffle substrate embeddedness, and substrate cover diversity, each of which was correlated with all invertebrate metrics including a 10-metric combined score. Correlation analysis indicated that if riparian and in-stream habitat conditions improve then so might invertebrate communities and stream biological quality. Sixty-two percent of the variance in macroinvertebrate community metrics was explained by the single environmental factor, percent impervious surface. Invertebrate responses to urbanization in Johnson County indicated linearity rather than identifiable thresholds. Multiple linear regression models developed for each of the four macroinvertebrate metrics used to determine aquatic-life-support status indicated that percent impervious surface, as a measure of urban land use, explained 34 to 67 percent of the variability in biological communities. Results indicate that although multiple factors are correlated with stream quality degradation, general urbanization, as indicated by impervious surface area or urban land use, consistently is determined to be the fundamental factor causing change in stream quality. Effects of urbanization on Johnson County streams are similar to effects described in national studies that assess effects of urbanization on stream health. Individually important environmental factors such as specific conductance of water, PAHs in streambed sediment, and stream buffer conditions, are affected by urbanization and, collectively, all contribute to stream impairments. Policies and management practices that may be most important in protecting the health of streams in Johnson County are those minimizing the effects of impervious surface, protecting stream corridors, and decreasing the loads of sediment, nutrients, and toxic chemicals that directly enter streams through stormwater runoff and discharges.