Optimization of sewage sampling for wastewater-based epidemiology through stochastic modeling

The proliferation of the SARS-CoV-2 global pandemic has brought to attention the need for epidemiological tools that can detect diseases in specific geographical areas through non-contact means. Such methods may protect those potentially infected by facilitating early quarantine policies to prevent the spread of the disease. Sampling of municipal wastewater has been studied as a plausible solution to detect pathogen spread, even from asymptomatic patients. However, many challenges exist in wastewater-based epidemiology such as identifying a representative sample for a population, determining the appropriate sample size, and establishing the right time and place for samples. In this work, a new approach to address these questions is assessed using stochastic modeling to represent wastewater sampling given a particular community of interest. Using estimates for various process parameters, inferences on the population infected are generated with Monte Carlo simulation output. A case study at the University of Oklahoma is examined to calibrate and evaluate the model output. Finally, extensions are provided for more efficient wastewater sampling campaigns in the future. This research provides greater insight into the effects of viral load, the percentage of the population infected, and sampling time on mean SARS-CoV-2 concentration through simulation. In doing so, an earlier warning of infection for a given population may be obtained and aid in reducing the spread of viruses.