Watershed hydrology, not nutrient chemistry, affects arid-land stream microbial diversity

Microbiota in streams drive ecosystem functions including whole-stream metabolism and nitrogen cycling, and microbial production fuels stream food webs. The interaction between surface water and shallow subsurface groundwater sets up the oxygen and nutrient gradients that drive and maintain these microbially-mediated biogeochemical functions, and microbial nutrient processing is often limited by nutrient availability. The extent to which differences in the microbial community composition of stream surface and subsurface waters are associated with differences in stream nutrient chemistry and biogeochemical function is not well understood. This study evaluated the prediction that heterogeneity in the prokaryotic community composition in surface and subsurface water within and among stream reaches would be related to dissolved nutrient concentrations and levels of surface-subsurface hydrologic connectivity, and to reach scale ecosystem nitrogen and carbon cycling rates. Data on water chemistry, whole-stream hydrological connectivity and biogeochemical function, and surface and subsurface water prokaryotic community composition at six streams in the southwestern USA were used to evaluate these predictions. Results showed no correlation between microbial community composition and stream nutrients, and a significant but weak relationship between stream reach-scale surface-subsurface interaction and prokaryotic diversity, whereby subsurface waters tended to have higher taxonomic richness than surface waters. Instead of reach-scale nutrient and hydrological parameters, watershed size and stream water bromide concentrations were best correlated to the differences in microbial community composition among the study streams. This suggests that the longer, deeper groundwater flowpaths in larger watersheds, which promote solute accumulation, may also impact stream water microbial community composition. Microbial community composition was not related to rates of reach-scale nitrogen cycling or stream metabolism, but it was related to the compartment in which most nitrogen was immobilized (fine benthic organic matter vs. filamentous algae), which may also be related to stream water salinity. Overall, results suggest that the catchment scale history of water movement may affect both microbial diversity and the fate of nitrogen. The mechanistic links between stream microbial diversity and ecosystem function deserve more attention, and given the spatiotemporally dynamic nature of stream ecosystems, need to be considered within the template of factors driving microbial biogeography.