Deterministically selected rare taxa drive changes in community composition in drinking water biological activated carbon filters

Background Biofiltration offers a sustainable, low-energy solution for drinking water treatment but suffers from inconsistent performance due to complex microbial dynamics. Current studies lack insight into early biofilter microbial community assembly. Here we perform a high-resolution spatial and temporal investigation of biomass accumulation and community development within biological activated carbon (BAC) filters over the first six months of operation. Results We found that initial biomass accumulation is not linear, instead characterised by periods of growth and decay. Mass balance identified an estimated + 6.54x10 ⁸ new cells daily during the growth phase (days 34 to 62), falling to a loss of 1.69x10 ⁹ by the decay phase (days 83–162). There was no significant increase in richness until the decay phase (ANOVA p-values > 0.05 between days 34, 62 and 83). Significant stratification (ANOVA p-values < 0.05) was observed with bed depth with 79% (± SD2.7%) of biomass found in the top 15cm of the filter bed, the bottom section (90cm) had 36.5-fold lower biomass. An abundant community of 20 primary colonisers made up to between 20 ± SD8% and 80 ± SD5% of the total community and persisted over time. This community increased in absolute number during the growth phase (140% increase) however remained stable after this. Conversely rarer taxa were found to continue to increase into the decay phase (131% between days 62–162). Core community analysis and neutral modelling of the seeding influent water and the biofilter found that the abundant taxa stochastically assembled early from the water, while the rarer taxa driving changes in diversity, were selected by deterministic factors within the filter bed with 38% advantaged by the filter environment, compared to only 20% of the persistent abundant community. Conclusion This study demonstrates that biofilter microbial communities undergo dynamic changes, with abundant early colonizers persisting steadily while rare taxa drive fluctuations in biomass through phases of growth and decay. Understanding these microbial dynamics and ecological interactions can inform engineering strategies to optimize biofilter performance, enhancing water treatment efficiency by targeting key microbial groups throughout filter maturation.