THESIS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY Sources and Fluxes of Organic Contaminants in Urban Runoff

Urban runoff quality is recognized as one of the most significant pressures on aquatic ecosystems worldwide. Research into pollutants in urban runoff has traditionally focused on nutrients, suspended solids and metals and consequently knowledge of anthropogenic organic pollutants is limited. The aim of this research was to investigate the occurrence and identify the sources of certain selected organic contaminants in urban runoff, and to evaluate tools for predicting the fluxes of these pollutants in urban catchment areas. Alkylphenols and phthalates were selected for further study since they are used in large quantities and emissions of these compounds are likely to end up in urban runoff. The occurrence of alkylphenols and phthalates was investigated in urban snow, stormwater and sediment. In general, 4-nonylphenol showed high concentrations and detection frequencies compared to most other alkylphenols. Among the phthalates, diisononyl phthalate (DINP) was detected at the highest concentrations in all matrices, followed by diisodecyl phthalate (DIDP) and di(2-ethylhexyl) phthalate (DEHP). Nonylphenol, octylphenol and DEHP were repeatedly detected in stormwater and snow at concentrations exceeding the European water quality standards. This suggests that measures to reduce the discharge of anthropogenic substances to urban areas are necessary to achieve good water status. Substance flow analysis (SFA) was used to map the sources and quantify the loads of phthalates and nonylphenols in urban catchment areas. The calculated loads of the contaminants were in agreement with measured loads in a studied catchment area and SFA was thus considered efficient for identifying the most important sources of phthalates and nonylphenols. The emission factors used in the calculation of the pollutant loads were also used in a process-based stormwater quality model for predicting nonylphenol and phthalate concentrations in runoff. The model revealed low predictive power; the simulated concentrations were generally one magnitude higher than the measured concentrations. In future studies, it is recommended to link the outcomes from the SFA to a fate model. This integrated model would provide a holistic overview of the sources and sinks of pollutants in urban catchment areas and could be used to evaluate both source control and end-of-pipe mitigation practices.