Linking Microbial Ecology to Geochemistry in Sulfate Reducing Systems

Sulfate reducing bioreactors (SRBRs) can serve as passive treatment systems for mining influenced waters (MIW). An enhanced understanding of the biogeochemistry and efficacy of SRBRs can be achieved by combining molecular biological and geochemical techniques in both field and column settings. To this end, a spatial and temporal sequence of eight pilot-scale columns were analyzed employing a multidisciplinary approach using ICP-AES, next-generation sequencing, and SEM-EDX to explore the effects of variable substrate on community structure and performance (measured by Zn removal). All pilot scale reactors contained 30% limestone by mass, 7 of the 8 had variable amounts of woodchips, sawdust, and alfalfa hay, and an 8th column where the only carbon source was walnut shells. High throughput sequencing of DNA extracted from liquid in pilot-scale columns reveals, similarly to an analogous field system in Arizona, a dominance of Proteobacteria. However, after the first pore volume, performance differences between substrate permutations emerged, where columns containing exclusively walnut shells or sawdust exhibited a more effective startup and metal removal than did columns containing exclusively woodchips or alfalfa hay. SEM-EDX analysis revealed the initial formation of gypsum (CaSO4) precipitates regardless of substrate. Zn was observed in the presence of Ca, S, and O in some column samples, suggesting there was co-precipitation of Zn and CaSO4. This is congruent with micro-XAS analysis of field data suggesting iron sulfides were co-precipitating with gypsum. A SEM-EDX analysis from a subsequent sampling event (8 months into operation) indicated that precipitation may be shifting to ZnS and ZnCO3. Biplots employing Canonical Correspondence Analysis (CCA) describe how diversity scales with performance and substrate selection, and how community shifts may result in differential performance and precipitation in response to selective pressure of bioreactor material on community composition. CCA of Shannon diversity data after one pore volume revealed that zinc removal, walnut shell content, and abundance of delta-Proteobacteria (sulfate reducing organisms) were all corresponding elements. However, after several pore volumes, the walnut shell column was no longer removing Zn as effectively, and community shifts were observed throughout the columns. Analysis of field and laboratory scale microbiological and geochemical shifts, in parallel, gives insight into key biogeochemical variables linked to the performance of passive remediation systems used for the treatment of contaminated MIW, while also providing further insight into metal immobilization at the microbe-mineral interface.