Diffuse Pollution Conference Dublin 2003 4A SUDS 4- 1 ECOTOXIC EFFECTS OF WET WEATHER DISCHARGES IN AN URBAN STREAM Anders Baun*, Anne Munch Christensen*, Fumiyuki Nakajima** *Environment & Resources DTU, Building 115, Technical University of Denmark, DK-2800 Kgs. Lyngby, Denmark. (E-mail: anb@er.dtu.dk) ** Department of Urban Engineering, the University of Tokyo, 7-3-1 Hongo, Bunkyo, Tokyo 113-8656, Japan. ABSTRACT The urban stream Store Vejleå (Denmark), which receives discharges of urban runoff, was investigated using a combination of biological toxicity tests (biotests) and chemical analysis. The urban stormwater and road runoff did not affect the mobilization of Daphnia magna, whereas three of the samples gave low, but statistically significant, effects on the reproduction of the alga Pseudokirchneriella subcapitata. In all pre-concentrated water samples toxic effects were found with EC 10 -values ranging from pre-concentration factors of 1.7-43 times and differences in toxicity depending on time and location of sampling were identified (e.g. inlet of road runoff was more toxic than inlet of storm water from an urban area). No toxicity was detected in an unpolluted reference sample pre-concentrated 100 times. Undiluted pore water samples from sediments collected in the stream were all toxic towards the algae and dilutions from 4-14 times were needed to reduce the growth rate inhibition to 20%. To reduce the toxicity of a pore water sample from an unpolluted stream a dilution factor of only 1.6 was required. A qualitative correlation between the toxicity of the pore water and the degree of pollution as evidenced by the metal concentration was observed, but statistically significant correlations could not be established by ranking procedures of e.g. metal content or PAH-concentrations versus the observed toxicity. KEYWORDS: Biotests, pore-water, sediment, solid phase extraction (SPE), toxicity, urban storm-water INTRODUCTION In recent years the focus on environmental impacts of discharges of urban storm-water and road runoff have increased and physical, chemical, and biological problems have been addressed (Maltby et al. , 1995a; Maltby et al. , 1995b; Marsalek et al. , 1999). Discharges of urban stormwater to aquatic environments have been documented to cause accumulation of heavy metals and PAHs in sediments and findings of different organic chemicals in the water phase have be described (e.g. Maltby et al. , 1995b; Carr et al. , 2000). Many of the compounds are known to be toxic to animals and humans and have potential for carcinogenic, mutagenic and/or allergenic effects (Ledin et al. , 2003), but the ecotoxic potential of the chemicals present in urban runoff is not well described in the literature, though a number of studies dealing with toxicity assessment of urban wet-weather discharges have been published (review by Marsalek et al. , 1999). Wet-weather discharges are examples of complex environmental pollutions for which an analytical-chemical characterisation may not suffice in explaining the toxic potential (Ellis, 2000). An alternative monitoring strategy using a battery of toxicity test have been pointed out as a useful tool for screening and assessing potential receiving water impacts (Marsalek et al. , 1999), however, only few actual studies have been carried out. The present study focused on measuring toxicity both in the water and in the sediment phase in the stream Store Vejleå located 20 km west of Copenhagen (Denmark). The water in the stream originates primarily from discharges of urban stormwater and road runoff. As a result of previous investigations (Kjølholt et al. , 2001; Ledin et al ., 2003) it expected that the toxicity of stream water was low, but also that organic chemicals were present in the water phase. Therefore, two approaches for toxicity testing were applied to water samples: 1) direct testing of whole samples and 2) testing of samples fractionated and pre-concentrated by solid phase extraction (SPE) in order to quantify the toxicity of non-volatile organic chemical contaminants in the water phase. Furthermore, toxicity testing of pore water from sediments collected in Store Vejleå was included in the study. The aim of the study was to address the order of magnitude of toxic effects of chemical constituents in water samples, not to make simulations of in situ toxicity. As a consequence of this, all samples were grab samples collected on different occasions (not flow-, weather-, or time-dependent sampling) and standardized toxicity tests were applied. The biotests were used as indicators for toxicity and hazard ranking of both the sediment and the water samples and two freshwater test species, that represent the primary producers and zooplankton, were used throughout the study, i.e. the alga Pseudokirchneriella subcapitata and the crustacean Daphnia magna . MATERIALS AND METHODS Site The stream Store Vejleå, shown in Figure 1, is dry most of the year upstream the inlet from basin 4. Basin 4 is a detention pond, which receives runoff water from both a motorway and an urban area. About half of the water in the stream originates from basin 4, and around one third of the water flow in the stream can be attributed to the inlet from “Dybendalsgrøften”, which can be characterized as urban runoff. The stream Store Vejleå has several small inlets connected to the urban rainwater system (cf. Figure 1).