Pesticide fate during drinking water treatment determined through passive sampling combined with suspect screening and multivariate statistical analysis

Abstract

Emerging contaminants such as polar pesticides pose a potential risk to human health due to their presence in drinking water. However, their occurrence and fate in drinking water treatment plants is poorly understood. In this study we use passive sampling coupled to suspect screening and multivariate analysis to describe pesticide fate throughout the treatment stream of an operational drinking water treatment plant. ChemcatcherÒ passive sampling devices were deployed at sites (n = 6) positioned at all stages of the treatment stream during consecutive deployments (n = 20) over a twelve-month period. Sample extracts (n = 120) were analysed using high-resolution liquid chromatography-quadrupole-time-of-flight mass spectrometry and compounds identified against a commercially available database. A total of 58 pesticides and transformation products from different classes were detected. Statistical analysis of the qualitative screening data was performed to identify clusters of pesticides with similar fate during ozonation, granular activated carbon (GAC) filtration, and chlorination. The performance of each treatment process was investigated. Adsorption to GAC media was found to account for the greatest proportion of pesticide attenuation (average removal of 70% based on detection frequency), however, operational performance varied for certain pesticides during periods of episodic and sustained pollution. GAC breakthrough occurred for 21 compounds detected in the GAC filtrate. Eleven pesticides were found to occur in potable water following treatment. We developed a management plan containing controls, triggers, and responses, for five pesticides and a metabolite (atrazine, atrazine desethyl, DEET, dichlorobenzamide, metazachlor, and propyzamide) prioritised based on their current and future risk to treated water quality.