Bacterial community composition of New Zealand groundwater

Abstract

To address groundwater ecosystem health as part of freshwater management, it is crucial to understand microbial diversity in these ecosystems, as microbial communities mediate biogeochemical processes. However, to date, only limited information is available on groundwater ecosystems. The objectives of this study are to: undertake the first survey of New Zealand groundwater bacterial taxonomic diversity; link this diversity to metabolic potential to inform freshwater management, and investigate the use of hydrochemistry and aquifer characteristicsto predict microbial taxonomy and metabolic function. ¶ Bacterial assemblages from 35 groundwater monitoring sites encompassing varying hydrogeochemical conditions across the country were analysed using 454 pyrosequencing of the V5 V7 region of bacterial 16S rRNA gene. Individual site microbial diversity was estimated using diversity indices whereas diversity relative to all sites was investigated using hierarchical cluster analysis. The influence of hydrochemistry on diversity was tested using canonical correspondence analysis. Metabolic modelling was performed using METAGENAssist. ¶ New Zealand groundwaters contain a rich bacterial taxonomic diversity with limited overlap between sites. Proteobacteria was the most abundant phylum, and Variovorax the most common genus. Pseudomonas, Burkholderia, Acidovorax, Janthinobacterium, Polaromonas and Caulobacter were the other common related genera. Samples with similar microbial communities formed “bioclusters”, grouped by combination of more abundant Operational Taxonomic Units (OTU) with differing rare OTUs. Groundwater chemistry was the main factor influencing bioclustering, explaining 59% of the variation in the relative abundance of sub sampled OTUs, with NO3-N, pH, DO, NH4-N, Fe, Br and SO4 being the strongest controlling parameters. In contrast, bioclusters were not correlated to aquifer lithology, confinement, well depth, geographic region or land use in the site’s recharge zone. Predicted oxygen requirements, metabolic potential and bacterial energy sources were associated to each biocluster. This information may be added to on-going monitoring of the health of a groundwater ecosystem, to support freshwater management.