There is growing recognition that soil bacteria play important roles in ecological communities as plant mutualists, pathogens, and facilitators of chemical cycling. Like all organisms, the presence and abundance of bacteria partly depends on their resource use, physiological abiotic tolerances, and biotic interactions and dependencies on other microbes, plants, or animals. Over time, both living and nonliving factors potentially shape bacterial communities in detectable ways, though the strength and direction of many of these effects remain unknown or not well-supported. Bacterial assemblages are often considered as a whole, partly due to their size and complexity, but diverse and important patterns of response are likely to be revealed by analysis of these communities at multiple resolutions. We hypothesize that abundant and clear relationships between bacterial taxa and environmental characteristics exist, even if their significance is not apparent in whole-community analyses. To this end, we utilized 16S metagenomic techniques coupled with quantitative measurements of biological and chemical variables to gain insight into bacterial response to key environmental features. We used constrained ordinations of bacterial community response to environmental variables, random effect linear models of bacterial abundance and diversity, and indicator species analysis to assess whole-community and taxon-specific responses to environmental variables.
Results/Conclusions
In contrast to previous studies, vegetation type (native or invaded), salinity, and pH were not found to be significantly associated with changes in whole-community diversity. Soil phosphorus, potassium, and water depth were significantly and positively related to diversity. Constrained ordinations identified soil potassium, sodium, and nitrogen as primary drivers of community structure, though these results were statistically weak. We identified significantly more taxa associated with salinity than with vegetation type. As bacteria are considered important plant symbionts, it is surprising that a chemical feature is more broadly relevant than a major biotic feature. Despite a lack of evidence for the role of pH as a driver of community diversity, acidity was significantly related to abundance of several taxa, demonstrating that while pH may not drive community diversity at our sites, it nonetheless plays a role in shaping bacterial abundance. We found few strong relationships between environmental parameters and bacterial communities when these communities were considered as a whole. In contrast, many significant relationships do exist between the same tested parameters and specific taxa. This result highlights the distinctive nature of bacterial life and demonstrates that taxon-specific analysis is an important complement to whole-community characterization.