In this thesis, topics in chemical hazard and risk assessment are explored through the use of multimedia mass balance models and high-throughput chemical property prediction methods. Chemical hazard metrics, as calculated by environmental fate and transport simulations, are investigated to determine the validity of two common simplifying assumptions in the underlying models; the use of octanol as a surrogate for organic matter and the use of environmental parameters that do not vary in time. A major finding is that the use of these common simplifying assumptions in multimedia mass balance models has little effect on chemical risk assessment, provided that chemicals are ranked relative to each other rather than a predetermined cutoff value. Chemical properties relevant to hazard and risk are collected, and applied in a large-scale chemical hazard assessment to derive a short list of potential Arctic contaminants. Several further data needs are identified; these are widely applicable and easily calculable metrics for chemical biodegradation, toxicity and emissions. A new method of predicting chemical properties is presented to assist in meeting these data needs. The method automatically creates predictive, quantitative relationships between the structures and properties of chemicals that are comparable to similar relationships created with expert judgement.
| Identifer | oai:union.ndltd.org:TORONTO/oai:tspace.library.utoronto.ca:1807/29670 |
| Date | 29 August 2011 |
| Creators | Brown, Trevor |
| Contributors | Wania, Frank |
| Source Sets | University of Toronto |
| Language | en_ca |
| Detected Language | English |
| Type | Thesis |
Page generated in 0.0023 seconds