Analysis of sampling strategies for pulse loads of SARS-CoV-2: implications for wastewater-based epidemiology

A faecal transport model was applied to a 11.3 km(2) wastewater servicing area in Toronto, Ontario, Canada to explore the role different wastewater sampling campaigns have on estimating the prevalence of SARS-CoV-2 in a population of 60,000. A stochastic wastewater and water quality model was used to evaluate the effectiveness of 11 common sampling strategies during periods of high and low COVID-19 infection among the population, tested using virtual sampling during dry-weather flow. The virtual sampling campaigns were based on the most common automatic sampler programming capabilities and widely used Wastewater-based epidemiology (WBE) sampling campaigns most commonly reported in the literature. Including changes to the weighting method (time, volume, or flow-weighted sampling), sample count, collection period, or sample time. Results suggest that grab samples should be avoided and that sampling campaigns with the greatest sample counts and durations are the most robust at capturing COVID-19 infection among the population. Most surprisingly, changes to the weighting method were negligible indicating that a greater number of samples, and larger sample volumes are preferred. This work suggests that investment in flow monitoring equipment for flow- or volume-weighted sampling will not improve WBE results, and that standard time-based sampling is sufficient.