Goal, scope and background: The potential negative impact of urban storm water on aquatic freshwater ecosystems has been demonstrated in various studies with different types of biological methods. There are a number of factors that influence the amount and bioavailability of contaminants in storm water even if it is derived from an area with a fairly homogenous land use such as a roadscape where a variation in toxicity during rain events might be expected. There are only a few previous investigations on the toxicity of highway runoff and they have not explored these issues extensively. The main objective of this study is therefore to characterize the potential toxicity of highway runoff during several rain events before it enters a detention pond in Västerås, Sweden, using laboratory bioassays with test organisms representing various functional groups in an aquatic ecosystem. The results are to be used for developing a monitoring program, including biological methods.
Materials and methods: The storm water was sampled before the entrance to a detention pond, which receives run-off from a highway with approximately 20,000 vehicles a day. The drainage area, including the roadscape and vegetated areas, is 4.3 ha in size. Samples for toxicity tests were taken with an automatic sampler or manually during storm events. In total, the potential toxicity of 65 samples representing 15 different storm events was determined. The toxicity was assessed with 4 different test organisms; Vibrio fischeri using the Microtox comparison test, Daphnia magna using Daphtoxkit-F agna, Thamnocephalus platyurus using the ThamnotoxkitF and Lemna minor, duckweed using SS 028313.
Results and discussion: Of the 65 samples, 58 samples were tested with DaphniatoxkitF agna, 57 samples with the Microtox comparison test, 48 samples with ThamnotoxkitF and 20 samples with Lemna minor, duckweed. None of the storm water samples were toxic. No toxicity was detected with the Lemna minor test, but in 5 of the 23 samples tested in comparison to the control a growth stimulation of 22-46% was observed. This is in accordance with the chemical analysis of the storm water, which indicated rather large concentrations of tot-N and tot-P. In addition to the growth stimulation, morphological changes were observed in all the 5 samples from the winter event that was sampled. The lack of toxicity observed in our study might be due to a lower traffic intensity (20,000 vehicles/day) at the site and the trapping of pollutants in the vegetated areas of the roadscape, resulting in much smaller loads of pollutants in the storm water than in some previous studies.
Conclusions: Ecotoxicological evaluations of storm water including run off from rain events from urban roadscape studies clearly reveal that toxicity may or may not be detected depending upon site, storm condition and the test organism chosen. However, storm water might not be as polluted as previously reported nor may the first flush be such a widespread phenomenon as we originally expected. In this study, there was also a good correlation between pollutant load measured and the lack of toxicity. The test organisms chosen in this study are commonly used in effluent control programs in Sweden and other countries, which makes it possible to compare the results with those from other effluents. In this study, only acute toxicity tests were used and further studies using chronic toxicity tests, assays for genotoxic compounds or in situ bioassays might reveal biological effects at this site. Furthermore, most of the samples were taken in spring, summer or fall and it is possible that winter conditions might alter the constituents in the storm water and, thus, the toxicity of the samples.
Recommendations and perspectives: Considering the complex nature of run off from urban roadscapes, it will be virtually impossible to evaluate properly the potential hazard of particular storm water and the efficiency of a particular treatment strategy from only physical and chemical characterizations of the effluent. Therefore, despite the lack of toxicity detected in this study, it is recommended that toxicity tests or other biological methods should be included in evaluations of the effects of runoff from roadscapes.