UC Davis

Treatment systems for wastewater and agricultural runoff serve to protect environmental water quality. However, with the proliferation of new compounds in consumer products and growing acknowledgment of the environmental implications of transformation products of parent compounds, conventional targeted mass spectrometry approaches are no longer sufficient in assessing the efficacy of these systems. Therefore, broad-scope methods such as suspect screening and nontarget analysis must be employed to gain a better understanding of the complex processes occurring within these systems.This dissertation considers both a municipal wastewater catchment and a scaled-down version of a woodchip bioreactor, and finds that both are comprised of competing physical, chemical, and biological processes. In the first chapter, suspect screening is used to find many different classes of compounds within a set of samples collected at seven different time points and eight different locations within a sewer catchment and at the wastewater treatment plant. In the second chapter, the same set of samples is considered in terms of the bulk nontarget features. However, due to the duration of the sampling campaign, wastewater samples had been run on the LC-QTOF-MS in multiple batches, which then required implementation of an empirical Bayes method to correct for batch effects. The success of this method is explored. Finally, the third chapter uses time-series data from microcosm-scale versions of woodchip bioreactors for agricultural run-off to identify examples of key processes that occur within the system.