Introduction

Background

Modern pharmacologic research since the middle of the 19th century has resulted in the development or discovery of numerous drugs for the treatment of disease and the relief of ailments. The numbers and types of these drugs, or pharmaceutical chemicals, and the amounts used, have increased greatly in the last several decades, to the point where thousands of drugs are used today. Many are in common or daily use in the world’s developed countries (Kaufman and others, 2002). Increased use can lead to increased concentrations in waste streams and the potential for release of ever larger loads of these compounds to the environment. Prior to the last 10 years, few studies had been performed to determine the sources, transport, and fate of such compounds. An array of published studies now are available showing that pharmaceutical chemicals and their metabolites are present in streams at a wide range of concentrations, particularly downstream of human population centers (Halling-Sørensen and others, 1998; Daughton and Ternes, 1999). Researchers have investigated the occurrence of pharmaceuticals in surface waters (Buser and others, 1998; Kolpin and others, 2002; Calamari and others, 2003; Löffler and others, 2005) and groundwater (Seiler and others, 1999; Hinkle and others, 2005; Barnes and others, 2008), the presence of such compounds in and their removal from wastewater (Ternes, 1998; Ternes and others, 2004; Jones and others, 2005), and the ecological effects of these compounds (Wilson and others, 2003; Gagné and others, 2006; Kim and others, 2007).

Despite these research efforts, few data exist to quantify the occurrence, concentration, and likely ecological effects of pharmaceutical chemicals in surface waters. A national reconnaissance was carried out in 1999–2000 by the U.S. Geological Survey (USGS) to assess the presence and concentrations of a suite of pharmaceuticals and other compounds in 139 streams across the United States (Kolpin and others, 2002). That study provides a good framework and baseline for developing a better understanding of the occurrence of pharmaceuticals and their metabolites in the Nation’s streams, but was not designed to determine the occurrence and effects of these compounds on a local scale; for example, only three water samples from Oregon were analyzed as part of that study. Given a growing and aging population with increased reliance on pharmaceuticals for the treatment of medical conditions, increased use of pharmaceuticals in agriculture, and greater public interest in the quality of drinking water and the health of aquatic species, the need to learn more about the occurrence and concentrations of pharmaceutical chemicals in Oregon’s streams is becoming increasingly important.

Some important source pathways of pharmaceutical chemicals, such as wastewater treatment facilities (WWTFs, which receive waste from hospitals, households, and municipal industries), have been identified in previous studies and need to be considered and assessed locally. Ternes (1998) was one of the first to document the presence of pharmaceuticals in raw and treated wastewater, using samples from selected WWTFs in Germany. Most modern WWTFs were not designed specifically to remove pharmaceutical chemicals from the waste stream. Many such chemicals are removed fairly efficiently through standard wastewater treatment, but others are not. Even if a large fraction of a pharmaceutical is removed through treatment, the remaining load still may constitute a large source to the receiving water body. Similar conclusions were reached in other studies in which higher concentrations of pharmaceuticals were measured downstream of WWTFs or population centers (Buerge and others, 2003; Kolpin and others, 2004; Glassmeyer and others, 2005; Gagné and others, 2006; Han and others, 2006; Karthikeyan and Meyer, 2006; Wilkison and others, 2006; Guo and Krasner, 2009). Research by Bedner and MacCrehan (2006) suggested that the chlorination process used in some WWTFs might transform acetaminophen, a common analgesic and one of the most widely used pharmaceutical chemicals, into a more harmful form. It is not just the parent pharmaceuticals, therefore, but their metabolites and degradates that must be studied to better understand the full effect of pharmaceuticals in the environment.

The potential for ecological effects caused by the presence of pharmaceuticals and their metabolites in aquatic systems is poorly understood at this time (2009). Because these compounds inherently affect physiological processes, it is likely that a measurable and potentially harmful ecological effect could occur at some concentration, but more research is necessary to define the ecological risks. Gagné and others (2006), for example, detected a wide range of pharmaceuticals in municipal treatment facility effluent and determined that such compounds have the potential to produce a toxic response in rainbow trout. Kim and others (2007) examined the toxicity of four widely used pharmaceuticals and six sulfonamide antibiotics and determined that several might have acutely toxic effects at concentrations greater than 1 mg/L and that some potential ecological risk is present at lower concentrations. Han and others (2006) assessed the ecotoxicological effect of a handful of pharmaceuticals on a common plankton species, determined that toxic effects can occur at sufficiently high concentrations, and found that although typical pharmaceutical concentrations downstream of certain WWTFs did not result in a significant risk, the potential for significant risk did exist. The ecological effects of pharmaceuticals in the environment is a subject with many unexplored topics such as synergistic effects, the risks of chronic and multiple-year exposure to trace concentrations, and the risks resulting from pharmaceutical metabolites and degradates.

In addition to WWTF sources, pharmaceuticals and other biological and chemical contaminants can enter surface waters through accidental or illicit dumping, poorly managed or failing on-site or septic systems, storm-sewer/sanitary-sewer cross connections, and unmanaged pet and animal wastes, among other sources. Proactive management and protection of aquatic resources would greatly benefit from the development of one or more definitive methods of identifying and tracking these separate sources. A wide variety of techniques and tracers are being developed in response to this need. Genetic techniques, for example, are becoming increasingly useful in identifying the sources of bacteria detected in streams (Stoeckel and others, 2004). Chemicals such as caffeine historically have proven useful as markers of human-related contaminant sources (Buerge and others, 2003), but that use may be diminishing because of the widespread consumption and careless disposal of caffeinated beverages. Other studies have found that, in addition to caffeine, certain anionic surfactants and fluorescent whitening agents (the “optical brighteners” in some laundry detergents) are good indicators associated with fecal coliform contamination (Sankararamakrishnan and Guo, 2005). Standley and others (2000) found that certain fragrance compounds, in conjunction with caffeine, can be used as tracers of human-related contaminant sources. The identification of a suite of chemical markers that is unique to human-related sources of stream pollution would prove invaluable to investigations of contaminant fate and transport as well as the management of water resources.

Pharmaceutical chemicals represent a compound class with great potential for use as tracers of specific sources of anthropogenic pollution. For use as a tracer of human-related contamination, a candidate compound should have few or no natural sources, no normal means of entering the stream of interest, a well-defined usage pattern, a sufficiently long lifetime to allow detection in the environment, and a reliable and accurate means of detecting and quantifying its presence in water, sediment, or tissue samples. Many pharmaceutical chemicals fit this general profile, although it is unclear which candidate pharmaceuticals might make the best tracers for sources of human-related stream contamination.

Study Area

The Tualatin River basin in northwest Oregon includes the western edge of Portland (Oregon’s largest city) and Portland’s western suburbs and outlying communities. In 2002, the basin was home to a rapidly growing human population of about 500,000 (U.S. Census Bureau, 2006). The great majority of those people live within a designated urban growth boundary which contains the cities of Portland, Beaverton, Tigard, Tualatin, Hillsboro, and Forest Grove, to name just a few (fig. 1). The cities are located primarily on the valley floor and are gathered mainly toward the middle and eastern edge of the basin. From its headwaters in the forested Coast Range mountains to the west, the Tualatin River meanders east through agricultural areas on the valley bottom before skirting the southern edge of the urban area and joining the Willamette River south (upstream) of Portland.

In a basin that is characterized by cool, wet winters and warm, dry summers, the Tualatin River typically has its highest streamflow of several thousand cubic feet per second in the winter during large eastward-moving Pacific storms, and its lowest flow of less than 200 ft³/s during late summer. Summer streamflow is reduced by irrigation withdrawals, but augmented by upstream reservoir releases and treated effluent from two large WWTFs. The Rock Creek and Durham WWTFs, operated by Clean Water Services, process an annual average of 60 Mgal/d of wastewater for more than 480,000 customers. During the low-flow summer period, the WWTFs add about 70 ft³/s (45 Mgal/d), or as much as 35 percent, to the flow of the Tualatin River (Bonn, 2008a). The Rock Creek and Durham WWTFs are advanced tertiary treatment facilities that use activated sludge treatment and chemical precipitation to remove nutrients and organic matter, followed by chlorination, filtration, and dechlorination for disinfection. The Rock Creek WWTF discharges to the Tualatin River just south of Hillsboro, and the Durham WWTF discharges to the Tualatin River near the mouth of Fanno Creek near Durham (fig. 1).

Tualatin River tributaries are readily grouped by characteristics that mirror the predominant land use within their drainages. Fanno, Rock, and Beaverton Creeks are the major streams draining urban areas of the Tualatin River basin. Fanno Creek has a well-established and completely (100 percent) urbanized drainage. The presence of a large human population, with a relatively dense urban development structure, sanitary and storm-sewer network, and extensive impervious areas, results in a diverse array of potential sources of stream pollution that are different from those in agricultural or forested drainages. In contrast to Fanno Creek, the Gales Creek drainage is predominantly forested (70 percent forest, 27 percent agricultural) and the Dairy Creek drainage has a large percentage of agricultural land use (50 percent) with most of the rest being forested (41 percent; data from 2001 National Land Cover Database, see Homer and others, 2007).