The Effects of Motorway Runoff on Freshwater Ecosystems: 3. Toxicant Confirmation A. B. A. Boxall*, L. Maltby Department of Animal and Plant Sciences, The University of Sheffield, Sheffield, S10 2TN, United Kingdom Received: 26 July 1996/Revised: 28 November 1996 Abstract. Previous studies have demonstrated that small streams receiving road runoff have reduced water and sediment quality. These changes in quality are associated with alterations in the structure and functioning of stream communities. Laboratory studies have indicated that the community changes are due to sediment-associated contaminants, and toxicant identification evaluations have shown that the major toxicants are contained probably in a fraction of sediment extract that contains polycy- clic aromatic hydrocarbons (PAHs). The aim of the present study was to determine whether PAHs were indeed the major toxicants in sediment extracts. Toxicity tests were performed with PAH mixtures, the toxic fraction of an extract of runoff- contaminated sediment, and a whole sediment extract. These indicated that three PAHs accounted for the toxicity of a sediment extract: pyrene, fluoranthene, and phenanthrene. The possibility of spatial or temporal variation in major toxicants was also investigated and tests on a number of sediment extracts obtained from a number of sites at different times demonstrated that the three PAHs accounted for 30.8 to 120% of an extract’s toxicity. When the PAHs were considered individually, pyrene was shown to account for most of the toxicity (44.9%), followed by fluoranthene (16%) and phenan- threne (3.5%). Road runoff contains a variety of contaminants including oil, tar products, dioxins, oxygenated compounds, halogenated phe- nols, metals, and deicing salts (e.g., Maltby et al. 1995a). Small streams receiving road runoff have both elevated levels of contaminants in sediment and water and an altered structure and functioning of benthic communities, suggesting a possible causal relationship between contaminants in road runoff and biological effects (Maltby et al. 1995a). Such a causal relation- ship has been further supported by laboratory studies confirm- ing that sediment contaminated with road runoff is toxic to benthic macroinvertebrates (Maltby et al. 1995b). Runoff from roads can contain hundreds of compounds and consequently sediments in streams receiving road runoff can contain a complex mixture of contaminants. The toxicity of complex mixtures is usually due to only a small proportion of the contaminants present. If these toxic contaminants can be identified, measures can be taken to reduce or eliminate the toxicity. One approach to identifying toxicants in complex mixtures is to perform a toxicant identification evaluation (TIE). Toxicant Identification Evaluation is a three stage process that combines toxicity testing with an analysis of the physical and chemical characteristics of mixtures to identify potentially causative toxicants (Mount 1989a, 1989b; Mount and Anderson-Carnahan 1988; USEPA 1991a, 1991b). In the first phase of the TIE process, the physical and chemical characteristics and thus the nature of the major toxicants are determined by combining toxicity testing with physical and chemical manipulations. These manipulations either remove the toxicants or render them nonbioavailable. In the second phase of the TIE process, the major toxicants are isolated using fractionation techniques. The toxicities of the fractions are then determined and chemical analyses are performed on the toxic fraction(s) in order to identify probable individual toxicants. The first two phases of the TIE process have been performed on the sediment obtained from a small stream that receives road runoff from the M1 motorway in the UK. The studies indicated that the toxicity of runoff-contaminated sediment was due to organic compounds. Fractionation studies have demonstrated that the major toxicants were in a fraction containing three to five ring PAHs (Maltby et al. 1995b; Boxall and Maltby 1995). The major PAHs in this fraction were phenanthrene, anthra- cene, fluoranthene, pyrene, benzo(a)anthracene, chrysene, ben- zo(a)pyrene, and benzo(b&k)fluoranthenes. Three to five ring PAHs in sediment extracts have been shown to be bioavailable to Gammarus pulex (Maltby et al. 1995b). While the toxicity of PAHs to G. pulex is unknown, individual PAHs have been shown to be acutely toxic to a range of other aquatic organisms, including molluscs, crustaceans, and fish (e.g., Donkin et al. 1989, 1991; Landrum et al. 1991; Suedel et al. 1993). While benzo(a)pyrene and benzo(b&k)fluoranthenes were major components of the toxic fraction (benzo(a)pyrene account- ing for 5.3% and benzo(b&k)fluoranthene 11.4% of the total PAHs in the fraction), a further fraction containing higher concentrations of these two compounds was not toxic to G. pulex, indicating that they were not major toxicants. The * Present address: Water Research Centre, Henley Road, Medmen- ham, Marlow, Bucks, SL7 2HD, UK Correspondence to: A. B. A. Boxall Arch. Environ. Contam. Toxicol. 33, 9–16 (1997) ARCHIVES OF E nvironment al Cont aminat ion and T oxicology r 1997 Springer-Verlag NewYork Inc.