OOS 8-7 - Alternative stable states in soil mutualist communities and ecosystem function

Tuesday, August 9, 2016: 10:10 AM
Grand Floridian Blrm F, Ft Lauderdale Convention Center
Marie Duhamel, Biology, Stanford University, Stanford, CA and Kabir Peay, Department of Biology, Stanford University, Stanford, CA

Plants and soil microbial communities are intimately linked through their respective influence on the abiotic soil environment. Plants may influence soil chemistry through the quality and quantity of litter and rhizosphere inputs, and microbial breakdown and transformation of these organic compounds in turn can influence plant nutrient uptake and foliar chemistry. Long-term development of plant-microbe-ecosystem linkages can be hard to examine in natural systems due to confounding factors influencing plant and microbial community development. In this study we use a unique system where the past history of vegetation is well known in order to quantify how plant species establishment influences development of microbial communities and microbe mediated ecosystem functions. Twenty years after a large fire at Point Reyes National Seashore, CA, we sampled in 30 independent vegetation patches with known transitions sequences between 3 dominant, monospecific vegetation types (Baccharis pilularis, Pinus muricata, Ceanothus thyrsiflorus). We hypothesized that i) as these plants display distinct pools of root mutualists (ie. P. muricata colonized by ectomycorrhiza, B. pilularis by arbuscular mycorrhiza, C. thyrsiflorus by nitrogen fixing bacteria), changes in vegetation would be linked with a regime shift in microbial systems and ii) with strong functional impacts on nutrient cycles. To test these hypotheses, metabarcoding was used to assess bacterial and fungal identity in soil samples followed by functional characterization using RNA metatranscriptomic sequencing. We also measured changes in ecosystem properties, such as soil, litter and leaf chemistry and decomposition rates.


We found that vegetation change resulted in alternative stable states for microbial community structure, function, and ecosystem properties. Metabarcoding data showed significant differences in the fungal but not in the bacterial species richness, while both the bacterial and the fungal community compositions differed significantly between patches of the three vegetation types. The plant and microbial changes also affected C and N content from litter and soil, with mainly, an increase in N content of C. thyrsiflorus litter and soil, and an increase in C for P. muricata litter and C. thyrsiflorus soil. Decomposition rate significantly decreased for both C. thyrsiflorus and P. muricata stands compared to their initial state with the lowest value for P. muricata. Analysis of metatranscriptomes is ongoing and will be used to link these changes in ecosystem function to specific microbial functional traits. Many of the changes observed were consistent with differences between plant species in their microbial mutualists, demonstrating the importance of plant-microbe partnerships in shaping ecosystem properties.