Jump to Main Content
Understanding human infectious Cryptosporidium risk in drinking water supply catchments
- Swaffer, Brooke, Abbott, Hayley, King, Brendon, van der Linden, Leon, Monis, Paul
- Water research 2018 v.138 pp. 282-292
- Cryptosporidium, drinking water, genotype, grazing, humans, land use, lifestyle, livestock, microbiological risk assessment, models, oocysts, pathogenicity, pathogens, polymerase chain reaction, risk, runoff, surface water, water quality, water reservoirs, water supply, water treatment, watersheds
- Treating drinking water appropriately depends, in part, on the robustness of source water quality risk assessments, however quantifying the proportion of infectious, human pathogenic Cryptosporidium oocysts remains a significant challenge. We analysed 962 source water samples across nine locations to profile the occurrence, rate and timing of infectious, human pathogenic Cryptosporidium in surface waters entering drinking water reservoirs during rainfall-runoff conditions. At the catchment level, average infectivity over the four-year study period reached 18%; however, most locations averaged <5%. The maximum recorded infectivity fraction within a single rainfall runoff event was 65.4%, and was dominated by C. parvum. Twenty-two Cryptosporidium species and genotypes were identified using PCR-based molecular techniques; the most common being C. parvum, detected in 23% of water samples. Associations between landuse and livestock stocking characteristics with Cryptosporidium were determined using a linear mixed-effects model. The concentration of pathogens in water were significantly influenced by flow and dominance of land-use by commercial grazing properties (as opposed to lifestyle properties) in the catchment (p < 0.01). Inclusion of measured infectivity and human pathogenicity data into a quantitative microbial risk assessment (QMRA) could reduce the source water treatment requirements by up to 2.67 log removal values, depending on the catchment, and demonstrated the potential benefit of collating such data for QMRAs.