Microbial diversity of biofilms at stream confluences

A traditional focus of ecology has been the study of biodiversity and community composition patterns. Only recently have ecologists begun to study biodiversity patterns in dendritic landscapes, such as stream networks. Confluences, where two or more streams mix, are generally assumed to be important in shaping the biodiversity of a fluvial network by way of accumulating species form multiple catchments. Ecological theory predicts that biodiversity increases downstream; however, it remains unclear whether the dendritic geometry of fluvial networks constrains the distributional patterns of microbial diversity in a similar manner to plants and animals. Over the last years, I had the opportunity to create a unique dataset comprising gene sequences of microbial biofilms the dominant form of microbial live in streams and rivers from more than 100 streams in a fluvial network (River Ybbs, Austria). Contrasting ecological theory and patterns found for animals, I found that biodiversity of microbial biofilms was lower downstream of confluences throughout the fluvial network, ultimately causing biodiversity to decrease downstream. However, the mechanisms underlying these patterns remain elusive. Understanding these mechanisms is of central importance given the pivotal role of biofilms for biogeochemical cycles and ecosystem processes in streams. The overall objective of this proposal is to gain mechanistic understanding of the impacts of confluences, as conspicuous nodes in fluvial networks, on biofilm diversity. Conceptually, I will adopt a metacommunity approach to study hyporheic biofilms in experimental bioreactors mimicking confluences. Using a controlled experimental setup, I will be able to isolate the effects of the metacommunity on the biodiversity of biofilm communities, by excluding potential changes of environmental conditions from upstream to downstream of confluences. I will combine high- throughput sequencing, quantitative PCR and metagenomics to comprehensively describe biofilm community structure, taxonomic and functional diversity. The obtained data will then be used to develop a mathematical community model. This model should enable me to disentangle stochastic demographic processes, as encapsulated in the neutral theory of community assembly, and deterministic processes, such as niche separation and biotic interactions, as potential drivers of biofilm biodiversity patterns. The project will be conducted at the University of Glasgow under supervision of Dr. Christopher Quince and Prof. Dr. William Sloan, who both figure among the premier scientists for theoretical microbial ecology, bioinformatics and community modeling. Upon my return to Austria, I will apply the acquired expertise in community modeling to the extensive data set of the fluvial network of the River Ybbs and, thus, set the experimental results into a broader ecological context. This research will equally contribute to microbial ecology, stream ecology and to general ecology.

Microbes are crucial for the functioning of all ecosystems. In streams and rivers, microbial biofilms are major engines of carbon and nutrient cycling and are responsible for the self- purification ability of these ecosystems. The diversity and structure of these microbial biofilms has important consequences for the functioning of streams and rivers. A recent study in a stream network in Austria found that biofilm diversity decreased at confluences, thereby contradicting patterns observed for fish, for instance. To assess the role of confluences for stream biofilm diversity under controlled conditions, microbial biofilms were grown from stream water in an experimental setup mimicking stream confluences. In contrast to the patterns observed in the real stream network, biofilm diversity increased at the experimental confluences; however, this increase was weaker than expected from random mixtures of the tributary communities. Analysis of the phylogenetic structure indicated that competition between microbial species could be one mechanism reducing the diversity of biofilms at stream confluences; however, additional environmental factors, like changes in hydrological conditions may be necessary to explain the decreasing diversity at natural stream confluences. Community composition in the experimental confluences also deviated clearly from a random mixture of the tributary communities. Generally, confluence communities tended to be dominated by the tributary communities stemming from more nutrient rich streams, indicating strong effects even of small changes in nutrient concentrations. Furthermore, functional diversity as assessed by metagenomic analysis of functional genes was correlated to species diversity, underlining the importance of microbial diversity for ecosystem functioning. Collectively, these findings suggest that mixing of microbial communities at stream confluences is an important factor for biofilm community assembly, but its effect on diversity is counteracted by competitive interactions and potentially by environmental conditions. Such mechanistic understanding of microbial diversity patterns will help us to comprehend and manage stream ecosystems and the ecosystem services they provide.