Metagenomic insights into the changes of runoff water quality in a deep tunnel drainage system.

Abstract

Deep tunnel retrofitting of conventional urban drainage systems represents a pivotal strategy for mitigating stormwater pollution and combating flooding. While microbial-driven biogeochemical cycles in stormwater are constrained by taxonomic diversity and environmental variability, the interplay between hydrogeochemical dynamics and microbial functional genes during storage remains poorly characterized. In this study, an in situ stormwater self-purification system was constructed to investigate seasonal water quality evolution, microbial community dynamics, and functional gene regulation in Shenzhen, China. Compared with continuous rainfall events, initial postdrought stormwater events resulted in significantly elevated pollutant loads. Dissolved organic matter analysis revealed that endogenous contaminants accounted for 76 % of the total contaminants, characterized by high microbial bioavailability and low humification after 14 days of storage. The storage of samples favors the enrichment of functional microorganisms such as Plancomycetota, Verrucomicrobiota and Proteobacteria. A quantitative assessment of 62 functional genes linked to carbon (C)/nitrogen (N)/sulfur (S) cycling identified temperature, oxidation‒reduction potential ammonia nitrogen, chemical oxygen demand and total nitrogen as critical drivers of microbial community succession and gene abundance. N cycle genes presented heightened sensitivity to environmental fluctuations, with increased stability and metabolic activity observed in wet season samples. Comparative analysis demonstrated that deep tunnel samples presented more stable functional gene profiles and enriched microbial consortia relative to their surface counterparts. These findings elucidate the mechanistic relationships between hydrogeochemical variables and microbial functional resilience in stormwater storage systems. This work advances the process-level understanding of biochemical cycles mediated by C, N and S transformations, offering actionable insights for optimizing urban drainage infrastructure and microbial-mediated pollution control strategies.