Pharmaceutical compounds in Merrimack River water used for public supply, Lowell, Massachusetts, 2008-09

This report presents results of a study conducted by the U.S. Geological Survey (USGS), in cooperation with the Massachusetts Department of Environmental Protection, to determine the occurrence of 14 commonly used human-health pharmaceutical compounds and fecal-indicator bacteria in Merrimack River water used as a drinking-water source by 135,000 residents in eastern Massachusetts. The study was designed to complement the USGS National Water-Quality Assessment Program's Source Water-Quality Assessment, which identifies patterns of occurrence of 280 primarily unregulated organic wastewater contaminants in source water used by community water systems and determines whether these patterns also occur in treated drinking water prior to distribution. The study involved periodic collection and analysis of raw Merrimack River water and treated drinking water over the course of 1 year. Water samples were collected periodically without regard to flow regime or antecedent weather conditions at the Lowell Regional Water Utility's Merrimack River intake upstream from Lowell, Mass. The same parcel of water was then sampled as finished water following treatment. Despite the presence of many potential sources of contamination in the drinking-water source area, only 2 of the 14 pharmaceutical analytes were detected at reportable concentrations in the source-water samples, and these occurred in only one set of periodic samples. Acetaminophen, a nonprescription analgesic, and caffeine were detected in the September source-water samples at concentrations of 0.084 and 0.068 micrograms per liter, respectively. Three other compounds-carbamazepine, an antiepileptic; cotinine, a metabolite of nicotine; and diphenhydramine, a nonprescription antihistamine-were detected in source-water samples, but at concentrations too low to be reliably quantified. None of the 14 pharmaceuticals was found in the finished water at a reportable concentration, defined as two times the long-term detection limit used by the analytical laboratory. In addition to the pharmaceutical analyses, measurements of fecal-indicator bacteria (Escherichia coli) concentrations and several physical characteristics were made on all source-water samples. Values for these constituents were consistently within State standards. It is possible that the monthly sampling schedule missed hydrologic events that would have transported greater concentrations of sewage contaminants to the sampling site, or that the large flow volume of the river at the study site effectively diluted the contaminant signal, but it is also likely that recent efforts to separate stormwater- and wastewater-discharge systems in the reaches upstream from the Lowell Regional Water Utility have greatly reduced the potential for sewage contamination at the intake.