Monitoring risk from contaminant mixtures in stormwater with water quality measurements, bioassays, and bioassessment

Abstract

Urban stormwater management ponds (SWPs) are increasingly valued not only for their role in mitigating runoff but also for the biodiversity they support in densely developed environments. However, these systems receive complex contaminant mixtures from urban runoff, including pesticides, pharmaceuticals, industrial chemicals, and metals. These pollutants can accumulate in biologically active compartments like biofilms, posing risks that are not always captured by traditional water-based monitoring. My thesis investigates the nature, accumulation, and ecological effects of contaminants in SWPs using a combination of chemical, biological, and toxicological approaches.

The objectives of my research were to: (1) characterize pesticide contamination in SWPs using water, biofilm, and passive samplers; (2) quantify pesticide accumulation in biofilms and identify influencing factors; (3) assess the toxicity of contaminated biofilms through dietary exposure; (4) survey the broader suite of urban contaminants in SWPs to develop a stormwater contaminant signature; and (5) examine relationships between environmental conditions and aquatic community composition.

In Chapter 2, I surveyed 21 SWPs in Brampton, Ontario for pesticide contamination. I compared three monitoring approaches across the ponds - time-integrated water sampling, biofilm cultured on artificial substrates, and organic-diffusive gradients in thin films (o-DGT) passive samplers - finding that o-DGTs had the highest pesticide detection rates. However, issues with reproducibility in passive sampler data highlighted the challenges of using them for quantitative risk assessment. Despite generally low concentrations in water and biofilm samples, the widespread detection of diverse pesticide classes across all three matrices emphasized the chronic, mixture-based exposures in these ponds and informed recommendations for future monitoring strategies. In Chapter 3, I further investigated the use of biofilms as a sensitive and ecologically relevant matrix for contaminant monitoring. Examining a wider set of pesticide analytes, I found that over half of the pesticides detected in biofilm samples were not detected in water, suggesting that conventional sampling approaches may overlook important alternative exposure routes. Calculated bioconcentration factors (BCFs) varied widely and were not well explained by pesticide properties or water quality variables, pointing to the complexity of contaminant uptake mechanisms. To test the potential toxicity of these contaminated biofilm samples, in Chapter 4 I conducted a series of dietary exposure assays with two invertebrate grazers. Mayfly nymphs (Neocloeon triangulifer) and juvenile freshwater snails (Planorbella pilsbryi) fed with contaminated biofilms from the SWPs showed reduced survival and growth compared to controls. Although the test results did not always correlate with measured pesticide levels, these results support the ecological relevance of biofilm-mediated exposure and suggest the presence of additional stressors not captured in targeted chemical analyses. I further expanded the chemical scope in Chapter 5 by analyzing over 700 unique urban contaminants across water, biofilm, and o-DGT samples. In total, 200 organic compounds were detected, including personal care products and traffic-related pollutants, as well as persistent elevated levels of fecal indicators and chloride. From these data, I developed the Urban Stormwater Contaminant Signature (USCS): a proposed list of common, environmentally relevant compounds to guide future monitoring and toxicity testing in urban aquatic systems. Finally, in Chapter 6 I examine how environmental variables shape aquatic community composition. Diatom and macroinvertebrate assemblages sampled from the SWPs were dominated by pollution-tolerant taxa, with diatoms responding primarily to water quality (e.g., nutrients, chloride, herbicides) and macroinvertebrates more sensitive to habitat features associated with pond naturalization. Landscape-scale metrics (e.g., impervious cover) calculated from buffer zones had limited predictive power, suggesting that local conditions and upstream drainage characteristics play a stronger role in shaping biological communities.

This research highlights the need to expand contaminant monitoring in stormwater systems beyond conventional water sampling, incorporating matrices like biofilm and tools such as passive samplers to better reflect the complexities of exposures in urban environments. The detection of numerous unmonitored or rarely assessed compounds suggests that current regulatory frameworks may underestimate the complexity and risk of urban chemical mixtures. Recognizing stormwater ponds as both infrastructure and ecosystems calls for more ecologically grounded approaches to design, management, and risk assessment; ones that support biodiversity alongside water quality improvement and flood protection.