Riverine sediment environments have extremely high microbial diversity yet the role of microorganisms in freshwater ecosystem resilience is poorly understood. The dynamic nature of lotic environments may offer continuous replenishment of microbial services and thereby create high resilience. However, this hypothesis cannot be addressed without basic knowledge of microbial biogeographic patterns and the factors controlling them. Spatial patterns are expected to depend on both environmental drivers and microbial traits. Potential drivers include water flow, sediment load, local geochemistry, land-use practices, and pollutants. Microbial traits, such as stress tolerance and resource specialization, may also be important determinants of geographic patterns, but traits can have widely varying phylogenetic associations making it difficult to predict distributions across complex environments. Our long-term research seeks to understand the impact of land-use practices on watershed microbial ecosystem processes at a range of scales. For this study, we focused on testing whether microbial phylogenetic diversity is correlated with sediment geochemistry in a heavily industrialized river. The Kanawha River watershed, encompassing more than 10,000 square miles of Appalachian Mountains in West Virginia, is our model system. The river runs through the Charleston metropolitan area and has yearly peak flow of about 60,000 cubic feet/sec. The river has been severely impacted by a variety of disturbances including acid mine drainage, logging, municipal inputs, and an eighty-year chemical industry that inspired the moniker “Chemical Valley.” We collected replicate sediment samples along a 60 km span of the river using a 8 cm diameter corer. Cores were separated into two layers (0-5 cm and 6-10 cm). DNA extraction was done with the PowerSoil DNA Kit (Qiagen). Bacterial and archaeal diversity was assessed with Illumina sequencing of 16S rRNA genes following Earth Microbiome Project protocols. More than 3 million paired-end reads were obtained. Coordinated measurements of 20 chemical variables were done with Inductively Coupled Plasma Spectroscopy and Dionex Ion Chromatography.
Results/Conclusions
Ordination analysis (PCA) showed that sediment geochemistry displayed spatial patterns. Canonical Correspondence Analysis of bacterial OTUs (species) of the dominant phyla showed that Proteobacteria, Chloroflexi and Bacteroidetes had distinct distribution patterns related to geochemistry. Sodium, chloride, bromide and electrical conductivity were prominent variables affecting OTU distributions. In conclusion, we found that the high-flow and high-sediment load of the river did not homogenize microbial diversity in the upper sediment layer, and that phylogenetic niche coherence was associated with local geochemistry.