Wastewater Speaks: Evaluating SARS-CoV-2 Surveillance, Sampling Methods, and Seasonal Infection Trends on a University Campus

Abstract

Wastewater surveillance has emerged as a cost-effective and equitable approach for tracking the spread of SARS-CoV-2. In this study, we monitored the prevalence of SARS-CoV-2 on a university campus over three years (2021-2023) using wastewater-based epidemiology (WBE). Wastewater samples were collected from 11 manholes on campus, each draining wastewater from a corresponding dormitory building, and viral RNA concentrations were measured using reverse transcription-quantitative PCR (RT-qPCR). Weekly clinical case data were also obtained from the university health center. A strong positive and significant correlation was observed between Grab and Composite sampling methods, supporting their robustness as equally effective approaches for sample collection. Specifically, a strong correlation was observed between Aggie Village 4 Grab and Aggie Village 4 Composite samples (R² = 0.84, p = 0.00) and between Barbee Grab and Barbee Composite samples (R² = 0.80, p = 0.00). Additionally, higher viral RNA copies of SARS-CoV-2 (N1 gene) were detected during the Spring semester compared to the Fall and Summer semesters. Notably, elevations in raw N1 concentrations were observed shortly after the return of college students to campus, suggesting that these increases were predominantly associated with students returning at the beginning of the Fall and Spring semesters (January and August). To account for variations in fecal loading, SARS-CoV-2 RNA concentrations were normalized using Pepper Mild Mottle Virus (PMMoV), a widely used viral fecal biomarker. However, normalization using PMMoV did not improve correlations between SARS-CoV-2 RNA levels and clinical case data. Despite these findings, our study did not establish WBE as a consistently reliable complement to clinical testing in a university campus setting, contrary to many retrospective studies. One key limitation was that numerous off-campus students did not contribute to the campus wastewater system corresponding to the monitored dormitories. However, some off-campus students were still subjected to clinical testing at the university health center under mandated protocols. Moreover, the university health center discontinued reporting cases per dormitory after 2021, making direct comparisons more challenging. Nevertheless, this study highlights the continued value of WBE as a surveillance tool for monitoring infectious diseases and provides critical insights into its application in campus environments.

Keywords: PMMoV normalization; SARS-CoV-2 RNA detection; grab and composite wastewater sampling; university public health surveillance; wastewater-based epidemiology.

Figure A1

Amount of N1 gene copy numbers in wastewater from all buildings between 2021 and 2023.

Figure A2

Predictions of the incidence rates based on observed amount of N1 gene copy numbers in wastewater from all buildings in 2021.

Figure 1

Campus map of North Carolina A&T State University with marked dormitories for sample collection (retrieved and modified from Campus website).

Figure 2

(a) Clinical cases on NCA&T campus between Spring 2021 and Spring 2023; green bars represent Spring semesters and yellow bars represent Fall semesters. (b) Scatter plot showing the correlation (Rs² = 0.04, p = 0.29) between the total number of clinical cases in NC state and NCA&T campus between Spring 2021 and Spring 2023.

Figure 3

(a) SARS-CoV-2 RNA copies between Spring 2021 and Spring2023. The yellow area represents the Spring semesters, and the green area represents the Fall semesters. (b) Histogram of Spearman’s rho for all pairwise correlations between dormitories. (c) Heatmap of p-value on negative log 10 scale of the correlation between each pair of dormitories as assessed using the spearman’s correlation test. (d) Heatmap representation of the pairwise spearman’s rho corresponding to the p-values in (c) and the distribution in (b). The high correlation values indicate similar contributions of RNA concentrations among the dormitories, with no significant differences observed. This suggests a consistent distribution of SARS-CoV-2 RNA levels across sampling sites over the analyzed period.

Figure 3

(a) SARS-CoV-2 RNA copies between Spring 2021 and Spring2023. The yellow area represents the Spring semesters, and the green area represents the Fall semesters. (b) Histogram of Spearman’s rho for all pairwise correlations between dormitories. (c) Heatmap of p-value on negative log 10 scale of the correlation between each pair of dormitories as assessed using the spearman’s correlation test. (d) Heatmap representation of the pairwise spearman’s rho corresponding to the p-values in (c) and the distribution in (b). The high correlation values indicate similar contributions of RNA concentrations among the dormitories, with no significant differences observed. This suggests a consistent distribution of SARS-CoV-2 RNA levels across sampling sites over the analyzed period.

Figure 4

(a) Predictions of the incidence rates based on observed amount of N1 gene copy numbers in WW in selected buildings for 2021 (Rs² = 0.03, p-value = 0.006). (b) Scatter plot between NCA&T campus-wide clinical cases with mean of SARS-CoV-2 RNA concentrations in Spring 2022–2023 on a log scale (Rs² = 0.007, p-value = 0.65).

Figure 5

(a) Scatter plot between PMMoV and SARS-CoV-2 RNA concentrations for Spring 2022–2023 (Rs² = 0.002, p-value = 0.29). (b) Scatter plot showing correlation between clinical cases and normalized RNA concentrations by PMMoV in 2022–2023; mean.logRatio represents mean of “log10(SARS RNA copies)/log10(PMMoV RNA copies).

Figure 6

(a) Spearman’s correlation analysis between SARS-CoV-2 RNA concentrations of two sampling methods (Grab and Composite) for Aggie V4 hall and normalized by PMMoV. (b) Spearman’s correlation analysis between SARS-CoV-2 RNA concentrations of two sampling methods (Grab and Composite) for Barbee Hall and normalized by PMMoV.