Microbiological Water Quality and Health Risk Assessment of Urban Rainwater Tanks

Abstract

The utilisation of roof collected rainwater to supplement conventional drinking water supplies is currently advocated as a means to conserve existing urban drinking water supplies in Australia. Whilst the use of rainwater for garden irrigation and toilet flushing is universally endorsed by health authorities throughout Australia this is not the case for use for household hot water systems and for delivery to indoor household taps (eg bathroom and laundry). This is because there is still some uncertainty relating to the potential health risk associated with personal washing and the incidental household consumption of rainwater. The aim of this project was to perform a descriptive study to ascertain the prevalence and concentration of a variety of bacterial contaminants in roof collected rainwater supplies in 6 localities throughout Australia. Samples were analysed for E.coli, total coliforms, enterococci, Clostridium perfringens, Salmonella spp, Campylobacter spp, Legionella spp and Aeromonas spp. In addition, the study also aimed to investigate the association between roof catchment and rainwater tank characteristics, maintenance protocols and use of devices (e.g. leaf litter diverters, first flush devices) and microbial water quality. A total of 36 tanks were surveyed (total number of samples analysed was 67). Results showed that rainwater tank supplies appear to provide water of relatively poor microbiological quality when compared with conventional Australian urban water supplies. Furthermore, the detection of Campylobacter spp and Salmonella spp bacteria in some rainwater tank waters confirmed the plausibility of gastrointestinal infection arising from these bacterial enteric pathogens when rainwater tank water is consumed as drinking water, or incidentally from domestic hot water use in instances where hot water is not heated sufficiently to achieve enteric pathogen inactivation. Results from this study supplement existing data about the prevalence of bacterial enteric pathogens in Australian roof collected rainwater tanks. However further research is required for input to Quantitative Microbial Risk Assessment models to estimate the health risk associated with drinking rainwater or the incidental consumption of hot water, as might occur in the domestic situation. The relatively small number of rainwater tanks surveyed and the variability of the tanks with respect to materials, roof catchment characteristics and cleaning regimes in this descriptive study hindered the detection of relationships between rainwater tank characteristics and operating protocols and microbial water quality. Experimental studies using indicator micro-organisms fit for purpose potentially provide the best and most economical means to investigate strategies that minimise microbial contamination of roof collected rainwater. This is because such studies allow levels of contamination and rainfall to be manipulated to reflect worst-case scenarios and rainwater tank variables can be controlled.