Concentrations of selected pharmaceuticals and antibiotics in south-central Pennsylvania waters, March through September 2006

Abstract

This report presents environmental and quality-control data from analyses of 15 pharmaceutical and 31 antibiotic compounds in water samples from streams and wells in south-central Pennsylvania. The analyses are part of a study by the U.S. Geological Survey (USGS) in cooperation with the Pennsylvania Department of Environmental Protection (PADEP) to define concentrations of selected emerging contaminants in streams and well water in Pennsylvania. Sampling was conducted at 11 stream sites and at 6 wells in 9 counties of south-central Pennsylvania. Five of the streams received municipal wastewater and 6 of the streams received runoff from agricultural areas dominated by animal-feeding operations. For all 11 streams, samples were collected at locations upstream and downstream of the municipal effluents or animal-feeding operations. All six wells were in agricultural settings. A total of 120 environmental samples and 21 quality-control samples were analyzed for the study. Samples were collected at each site in March/April, May, July, and September 2006 to obtain information on changes in concentration that could be related to seasonal use of compounds. For streams, 13 pharmaceuticals and 11 antibiotics were detected at least 1 time. Detections included analytical results that were estimated or above the minimum reporting limits. Seventy-eight percent of all detections were analyzed in samples collected downstream from municipal-wastewater effluents. For streams receiving wastewater effluents, the pharmaceuticals caffeine and para-xanthine (a degradation product of caffeine) had the greatest concentrations, 4.75 μg/L (micrograms per liter) and 0.853 μg/L, respectively. Other pharmaceuticals and their respective maximum concentrations were carbamazepine (0.516 μg/L) and ibuprofen (0.277 μg/L). For streams receiving wastewater effluents, the antibiotic azithromycin had the greatest concentration (1.65 μg/L), followed by sulfamethoxazole (1.34 μg/L), ofloxacin (0.329 μg/L), and trimethoprim (0.256 μg/L). For streams receiving runoff from animal-feeding operations, the only pharmaceuticals detected were acetaminophen, caffeine, cotinine, diphenhydramine, and carbamazepine. The maximum concentration for pharmaceuticals was 0.053 μg/L. Three streams receiving runoff from animal-feeding operations had detections of one or more antibiotic compound--oxytetracycline, sulfadimethoxine, sulfamethoxazole, and tylosin. The maximum concentration for antibiotics was 0.157 μg/L. The average number of compounds (pharmaceuticals and antibiotics) detected in sites downstream from animal-feeding operations was three. The average number of compounds detected downstream from municipal-wastewater effluents was 13. For wells used to supply livestock, four compounds were detected--two pharmaceuticals (cotinine and diphenhydramine) and two antibiotics (tylosin and sulfamethoxazole). There were five detections in all the well samples. The maximum concentration detected in well water was for cotinine, estimated to be 0.024 μg/L. Seasonal occurrence of pharmaceutical and antibiotic compounds in stream water varied by compound and site type. At four stream sites, the same compounds were detected in all four seasonal samples. At other sites, pharmaceutical or antibiotic compounds were detected only one time in seasonal samples. Winter samples collected in streams receiving municipalwastewater effluent had the greatest number of compounds detected (21). Research analytical methods were used to determine concentrations for pharmaceuticals and antibiotics. To assist in evaluating the quality of the analyses, detailed information is presented on laboratory methodology and results from qualitycontrol samples. Quality-control data include results for nine blanks, nine duplicate environmental sample pairs, and three laboratory-spiked environmental samples as well as the recoveries of compounds in laboratory surrogates and laboratory reagent spikes.