The development of sanitary sewer networks has been one of the cornerstones of modern civilisation. The economic investment in these networks is significant, both in their initial construction and their ongoing maintenance. Microbially induced concrete corrosion (MICC) within sewer systems leads to pipe failure, which has important financial, logistical and environmental implications. To date, most of the research into MICC has used culture-dependent methods, and despite the limitation of these approaches they have provided insight into this process and the organisms involved.
The development of culture-independent methods, in particular high-throughput sequencing, presents an opportunity to more clearly elucidate the diversity of microbes involved in MICC. In this thesis, I substantially improve the understanding of the microbial communities responsible for MICC through a detailed study of sites from around Australia. Ultimately, it is my hope that this improved knowledge will help in the development of effective mitigation techniques. While the use of advanced molecular ecology techniques will improve our knowledge of sewer MICC communities, it is critical to also understand the physicochemical or operational conditions in the sewer, which may influence these communities and the rate of corrosion.
This PhD has revealed several key discoveries, chiefly, that the community is not uniform in its composition and is instead comprised of several communities. In the sewers examined in this study, these communities can be clustered based upon their position within a pipe and are remarkably consistent both spatially and temporally. Here, the use of pyrosequencing, and the systematic sampling regime, has allowed at least an order of magnitude deeper examination of individual communities than any previous work. Consequently, the results presented here have a level of confidence which has not been previously possible. This examination of corrosion associated microbial communities strongly implicates four bacterial lineages, Acidithiobacillus, Mycobacterium, Acidiphilium and Xanthomonadaceae, as key members of a consortium of microbes responsible for MICC. The repeated collection of samples, as well as physicochemical and operational data, from the same site over a period of two and a half years and at multiple points within the system has given an unprecedented comprehension of the spatial and temporal community fluxes.
The exploration of community development on samples of fresh and pre-corroded concrete has allowed revision of the commonly accepted model of sewer MICC outlined by Islander et al. (1991). Due to the detailed profiles of the in-situ corrosion communities generated in this project, comparison of the developing and mature communities is also possible. Here, a comprehensive examination of the Sydney sewer communities and environment was undertaken, along with a detailed investigation of the Gold Coast, Melbourne and Perth sewer environments.
Despite difficulties generating a complete picture of the microbial communities in the Melbourne and Perth sites, this research has enhanced our understanding as to how a generalized sewer MICC model could be applied in a broad range of sewer environments. The diverse range of sewer environments examined in this PhD has revealed that, irrespective of variation in examined environmental conditions, the lineages of Acidiphilium, Acidithiobacillus, Mycobacterium and Xanthomonadaceae appear to be ubiquitous in MICC communities. Perhaps even more importantly, the roles of two of the four key community lineages, Mycobacterium and Xanthomonadaceae, are unknown. This suggests that the commonly accepted process of simple autotrophic sulfur-oxidation within an incidental heterotrophic community may be misleading, and instead sewer MICC is possibly the result of a more complex network of microbial processes. The findings presented here have significantly improved on our knowledge of sewer MICC communities and provided a solid foundation that will inform further research within this environment.