Suspended solids and metals in highway runoff : implications for treatment systems

Sammanfattning: It is known that traffic is a source of pollutants and that pollutant loads increase with elevated traffic densities. Studies executed in Stockholm, Sweden advocate that highway runoff from roads with an annual average daily traffic (AADT) exceeding 30,000 vehicles need treatment before discharge to the receiving water. It is common knowledge that sedimentation is the most expedient method for stormwater treatment. However, sedimentation units are area demanding and in highly urbanised watersheds the land use is often restricted. Studies have implied the occurrence of first flush, i.e. an initially higher pollutant load in the beginning of the runoff event, in highway watersheds. With an emphasized first flush it would be possible to treat only a part of the total runoff volume reducing the area needed for a sedimentation basin. In general two methods are used to design stormwater treatment ponds. One method is based on the reduced catchment area and pond surface and the other is based on an average runoff volume and a permanent pond volume. The methods are relaying on data from routine monitoring of various treatment systems and suggest removal efficiencies for pollutants. Applying general removal efficiencies for design it can be intricate to estimate an outlet concentration when the specific removal efficiency may be dependent on the initial concentration of the pollutant. Consequently, knowledge of the removal efficiencies dependence on initial concentration would be helpful to optimise stormwater treatment systems. This research has studied runoff from highly trafficated watersheds. The aim has been to evaluate the mass transport, stormwater quality and sedimentation behaviour and their implications for stormwater treatment. The study sites, Eugenia and Fredhäll, are located along the six-lane highway E4 through Stockholm that has an AADT load of 120,000 vehicles and a speed limit of 70 km/h. In lack of a unified definition of first flush the mass transport was studied using the EU directive 1991/271/EEC discharge demand for TSS of 60 mg/l. It was found that for the majority of the runoff events during winter the event mean concentration exceeded 60 mg/l suggesting that the complete runoff volume should be captured during winter. The dissolved concentration of metals showed significant variations between winter and summer, as did the concentration in the particulate matter (mg/kg). It was possible to correlate total metal concentration to total suspended solids with good correlation (r2 >0.90) for the majority of studied metals in winter and summer. The findings would imply that a successful treatment of the studied metal pollutants could be carried out by sedimentation. However, depending on discharge criteria, the elevated levels of dissolved matter, especially during winter, have to be considered with regards to the selection of the appropriate water treatment process. The sedimentation process could be described by a logarithmical function and initial turbidity. Good correlation (r2 >0.90) was indicated between turbidity and TSS. The sedimentation process of the studied highway runoff varied significantly (p<0.05) when elevated levels of NaCl could be found in the runoff. A significant difference (p<0.05) was shown for turbidity and TSS between summer and winter, which was assumed to be related to the use of studded tires. This study implies that the entire runoff volume must be treated and that the use of first flush as a design criterion is less applicable for the winter period. The study implies good correlation between total metal concentration and TSS. In addition the indicated correlation between turbidity and TSS would point to the possibility to use turbidity as a surrogate measurement for TSS and the studied metals. Moreover, the possibility to describe the sedimentation process by the initial concentration of turbidity would infer the utilisation of turbidity as a tool for process control for stormwater treatment systems. In addition, the novel results for the dependence on the sedimentation process could be incorporated in existing models for design of stormwater treatment systems in similar watersheds.