A theoretical approach to using faecal indicator data to model norovirus concentration in surface water for QMRA : Glomma River, Norway

Abstract

Monitoring of faecal indicator organisms, such as Escherichia coli, in environmental and drinking waters is inadequate for the protection public health, primarily due to the poor relationship between E. coli and the occurrence of human pathogens, especially viruses, in environmental samples. Nevertheless, measurements of faecal indicator organisms within the risk based approach, can provide valuable information related to the magnitude and variability of faecal contamination, and hence provide insight into the expected level of potential pathogen contamination. In this study, a modelling approach is presented that estimates the concentration of norovirus in surface water relying on indicator monitoring data, combined with specific assumptions regarding the source of faecal contamination. The model is applied to a case study on drinking water treatment intake from the Glomma River in Norway. Norovirus concentrations were estimated in two sewage sources discharging into the river upstream of the drinking water offtake, and at the source water intake itself. The characteristics of the assumed source of faecal contamination, including the norovirus prevalence in the community, the size of the contributing population and the relative treatment efficacy for indicators and pathogens in the sewage treatment plant, influenced the magnitude and variability in the estimated norovirus concentration in surface waters. The modelling exercise presented is not intended to replace pathogen enumeration from environmental samples, but rather is proposed as a complement to better understand the sources and drivers of viruses in surface waters. The approach has the potential to inform sampling regimes by identifying when the best time would be to collect environmental samples; fill in the gaps between sparse datasets; and potentially extrapolate existing datasets in order to model rarer events such as an outbreak in the contributing population. In addition, and perhaps most universally, in the absence of pathogen data, this approach can be used as a first step to predict the source water pathogen concentration under different contamination scenarios for the purpose of quantifying microbial risks.