The role of aquifer microorganisms in controlling arsenic contamination of Bangladesh aquifers via oxidation of organic carbon coupled with reduction of sedimentary iron oxyhydroxides and concomitant arsenic dissolution is generally accepted. What remains to be ascertained is the in situ biogeochemical mechanisms of cycling different carbon sources and directly relating indigenous microbiota to arsenic release. Using biomarker fingerprint approaches, this dissertation expanded the presently growing research in the biogeochemical carbon cycling controlling arsenic contamination in Bangladesh aquifers. Comprehensive profiles of microbial cell membrane components (PLFA and sterols) at three different aquifers tested the regional distribution of aquifer microbial community abundance, structure, and organic input potential across Araihazar. The highly variable bulk viable microbial biomass observed across these three sites confer both regional-scale and localized heterogeneous distributions of in-aquifer microbial communities which control carbon cycling in the aquifer. The lack of correlation between PLFA biomarkers and dissolved arsenic challenges the assumption that greater extent of microbial community metabolism results in an increase in arsenic in groundwater. Natural abundance radiocarbon isotope Δ14C analysis of cell membrane PLFA and available carbon pools (SOC, DOC, DIC) confirmed that young organic carbon substrates are being cycled at two of the three sites investigated here. This corroborates previous reports at nearby sites (Site B and F) thereby contributing to a well-constrained carbon source which actively support microbial metabolism over a regional scale. Sterol biomarker distributions were characterized to determine potential sources of organic input into the aquifer. In particular, the importance of raw human and/or animal sewage waste as a source of labile carbon was assessed by measuring the faecal biomarker Coprostanol and comparing its abundance to other sources of biogenic sterols using sewage input proxies (Sewage Contamination Index, Coprostanol/Cholesterol ratio). This was motivated by previous findings which correlated sewage contamination with dissolved arsenic at depth at nearby sites. While sewage contamination was low in the shallow aquifers at these sites, it is more likely that plant organic matter supported the elevated microbial abundance at shallow depths. On the other hand, evidence presented in this project suggests that sewage contamination intrudes into deeper aquifers (e.g. buried Pleistocene) and contributes to the vulnerability of previous pristine aquifers to future arsenic contamination. / Thesis / Master of Science (MSc)

Investigation of life’s limitations on Earth provides the necessary information to constrain where life outside of Earth may be proliferating or previously existed. This Master’s thesis applied phospholipid fatty acid (PLFA) analysis in combination with organic carbon and 16S rRNA gene data to assess and characterize microbial communities through both microcosms and in situ samples of bentonite clay, which is an intended barrier component for the long-term storage of high-grade nuclear waste. Microcosm experiments were set up to test the impact of water activity in as-received, uncompacted bentonite clays using a high (0.99) and low (0.93) water activity over a one month period. Under aerobic incubation water activities of 0.93 and 0.99 had no resolvable effect between water activity levels on the growth of cells of indigenous communities of microbes in as-received uncompacted bentonite. Growth was detected under both water activities by a significant increase in total PLFA abundance. The increase in PLFA over the period of the study suggested an approximate increase in cells from 4x10^6 to 2x10^7 E.coli equivalent cells/g. The distribution of the PLFA and genetics data suggests the community is composed predominantly of gram-positive aerobic heterotrophs with lesser amounts of anaerobic bacteria and eukaryotes potentially in the form of fungi. Similar cell abundances and community structures were identified in the Tsukinuno Mine bentonite DGR analogue site which is a ~12 to 16 Ma deposit approximately 200 m below the surface. Total PLFA recovered from the core subsamples ranged from 32 pmol PLFA/g to 431 pmol PLFA/g, which corresponds to a range from 7.5x10^5 to 1.2x10^7 E.coli equivalent cells/g, across all cores. The community was composed of both aerobic and anaerobic bacteria consisting of gram-positive and gram-negative bacteria, as well as possible sulfate-reducing bacteria and eukaryotes. / Thesis / Master of Science (MSc)