Effects of between-site variation in soil microbial communities and plant-soil feedbacks on the productivity and composition of plant communities.
A critical challenge in the science and practice of restoration ecology is to understand the drivers of variation in restoration outcomes. Soil microbial communities may have a role in explaining this variation due to both site-to-site variation in the composition of soil microbial communities and due to variation that can arise due to plant-soil feedbacks. We tested the relative importance of between-site variation in soil microbial community composition and plant-soil feedbacks in shaping plant community composition and ecosystem function. We used a standard two-phase plant-soil feedback design. Soil inoculum was collected from four tallgrass prairie sites. Then, soils were conditioned separately with nine plant species, and conditioned soils were used to inoculate prairie community mesocosms. In a separate experiment using soil from an additional site we tested conditioned soil samples for the abundance of arbuscular mycorrhizal fungi (AMF) and rhizobia. Site of soil origin and plant-soil feedbacks both had effects on the composition and productivity of our plant communities, and the magnitudes of these effects were similar. We also found changes in the abundance of AMF and rhizobia due to plant-soil feedbacks and that AMF abundance were associated with differences in plant community composition. These results indicate that the composition of soil communities due to site-to-site variation and plant-soil feedbacks are both important determinants of plant community composition and productivity. Our results also suggest that AMF and rhizobia are key microbial functional groups underlying plant-soil feedback effects. Synthesis and applications. Site-to-site variation in soil communities can explain some variation in restoration of plant communities. Since plant-soil feedback effects of restored plant species do not overcome this variation, knowledge of soil microbial communities present at a site prior to initiation of restoration efforts may improve predictability of restoration outcomes, and reintroduction of some components of the soil community may be necessary to achieve restoration goals. Additionally, by understanding variation due to plant-soil feedbacks, restoration practitioners can choose plant species for reintroduction that will create favourable soil conditions, including promoting microbial mutualists. Plant-soil feedbacks should also make it possible to increase heterogeneity in soil microbial communities, leading to increases in beta diversity in plant communities.