Spatially heterogeneous plant-soil feedbacks promote plant diversity and hamper monocultures

Plant-soil feedback (PSF) studies have become a central component of our
understanding of many terrestrial plant community processes. Recent work has
shown that spatially interacting PSFs can strongly influence the plant performance. However, emperical data on the role of the spatial scale (grain) of PSF heterogeneity is lacking. In a greenhouse experiment we placed soils with different PSF in patches in fine- (6x6cm patches), coarse-grained (12x12cm) and spatially homogeneous treatments. Monocultures and mixtures of six common grassland species were planted on the soil treatments. We tested how the grain of spatial heterogeneity affected plant performance both with intra- and interspecific competitors. Spatial PSF heterogeneity had pronounced effects on plant performance. For monocultures performance was reduced in the heterogeneous PSF treatments (~10% reduction in biomass). Although the strenght of the effect of PSF heterogeneity varied across test species and soil conditioning, most were negative. When grown in plant mixtures, however, spatial heterogeneity seemed to offer refuge from antagonists for at least some species, as their direct PSF changed from strongly negative to neutral or even positive in the fine scale heterogeneity treatment. We conclude that spatial PSF heterogeneity strongly affects plant performance, particularly in interspecific competition. We think that both the likelihood of encountering patches with strong direct and/or indirect PSF as well as the potential rate of transmission of antagonistic effects (e.g. pathogen infection) importantly determine plant performance in competitive mixtures. While theory seems to hold up in some instances, however, in many cases the responses remained poorly predictable. Consequently, there is a need for studying plant-soil interactions in space and to do so by mimicing field conditions in more realistic ways.