Species interactions, environmental conditions, and stochastic processes affect community composition; however, the strength and relative importance of these forces and how they interact to influence community dynamics remain poorly understood. We have previously demonstrated the potential for phytoplankton assemblages to effect change in lake bacterial communities at surface conditions. Temperature and light availability vary with depth as well as season in darkly stained humic lakes, and have the potential to affect bacterial communities both directly and indirectly via altered interactions with phytoplankton. Our objective was to determine the relative contribution of phytoplankton assemblages, temperature, light and the interactions among phytoplankton, temperature, and light in structuring bacterial communities. Bacterial communities (<1μm size) from two humic lakes were combined with phytoplankton assemblages (20-100μm) from each lake or a no-phytoplankton control. Each combination of bacteria and phytoplankton was incubated in situ for 5 days under all combinations of light (surface, ~25% surface irradiance) and temperature (5 levels from 10ºC to 25ºC) (n=3). Bacterial communities from before and after incubation were characterized using automated ribosomal intergenic spacer analysis (ARISA) and DNA sequencing of 16S small subunit rRNA genes.
Results/Conclusions
Permutational multivariate analysis of variance indicated that the largest influences on bacterial community composition following incubation were phytoplankton source (R2=0.233, p<0.001), bacterial source (R2=0.155, p<0.001), temperature (R2=0.108, p<0.001), and the interaction between phytoplankton source and temperature (R2=0.081, p<0.001). As incubation temperature increased, divergence of mesocosm bacterial communities from their initial community composition increased and dissimilarity between communities incubated with phytoplankton and no-phytoplankton control communities increased. Bacterial community response to light was small relative to other factors, but significant (R2=0.012, p<0.001), as were the two-way interactions between light and phytoplankton (R2=0.015, p<0.001), bacteria (R2=0.006, p=0.016), and temperature (R2=0.013, p=0.032). The effect of light was highly context-dependent. These results demonstrate the importance of interactions among ecological drivers for structuring bacterial communities and highlight the necessity of studying multiple potential drivers simultaneously to account for context-dependence. A generalizable understanding of the effects of species interactions and environmental conditions on community composition will require elucidating mechanisms underlying observed differential community responses.