André Felipe Bendix

Biological treatment of dairy wastewater depends on complex microbial communities that vary widely across reactors, influent sources and operating conditions. Whether stable treatment performance arises from conserved taxa or from redundancy in functional potential remains unclear. In this study, we conducted a meta‐analysis and reprocessing of 239 samples from nine independent studies spanning six countries and multiple reactor configurations. Taxonomic profiles showed extreme heterogeneity, with no ASV detected in at least 50% of samples and most genera exhibiting strong study‐specific patterns. In contrast, the predicted functional repertoire was highly conserved. PICRUSt2 analysis identified 196 MetaCyc pathways present in at least 90% of samples, and these pathways accounted for approximately 90% of total predicted functional abundance. Generalised linear latent variable models showed that temperature and pH created consistent but modest ecological gradients, while study‐ and reactor‐level factors dominated community structure. Despite substantial taxonomic turnover, functional signatures associated with fermentation, biosynthesis, nitrogen cycling, and degradation of complex substrates remained stable across systems. Treatment‐stage transitions also followed predictable metabolic trajectories. These results show that dairy wastewater treatment systems are taxonomically flexible but functionally robust, and that extensive redundancy supports consistent metabolic capacity across diverse operational and environmental contexts.

Wastewater‐based epidemiology (WBE) has been widely applied for SARS‐CoV‐2 surveillance. Yet, operational parameters such as influent flow rate and sewer system characteristics may reflect underlying hydraulic variability relevant to detection performance, raising questions about the representativeness of results from small and medium‐sized plants. We conducted a systematic meta‐analysis of 10 published studies assessing SARS‐CoV‐2 RNA detection across wastewater treatment plants of different sizes and flow ranges. Data were extracted into 2 × 2 tables and analyzed in R using the meta and metafor packages. Odds ratios (ORs) were pooled with Mantel–Haenszel random‐effects models and DerSimonian–Laird τ ² estimation. Subgroup analyses compared studies from developed versus emerging economies, meta‐regression tested population size as a moderator, and publication bias was evaluated with funnel plots and Egger's test. The overall pooled OR was 2.08 (95% CI: 0.99–4.37, p = 0.053; I ² = 82.3%), indicating a non‐significant trend toward higher detection at higher flow rates. Subgroup analyses showed similar pooled effects across developed and emerging economies, though heterogeneity was lower in emerging settings. Meta‐regression revealed no significant association between population size and detection (slope = 0.29, p = 0.52). Funnel plots were relatively symmetric, and Egger's test did not indicate significant publication bias ( p = 0.169). Overall, these findings suggest that influent flow rate is not significantly associated with the probability of SARS‐CoV‐2 detection in WBE studies, supporting the inclusion of small and medium‐sized treatment plants in surveillance networks. At the same time, the concentration of studies in high‐income countries highlights inequities in the global WBE literature and in site deployment, with vulnerable populations often underrepresented. Expanding WBE infrastructure in underserved regions is necessary to ensure equitable and representative surveillance capacity.