Abiotic rather than biotic filtering shapes the arbuscular mycorrhizal fungal communities of European seminatural grasslands.

Although it is well known that arbuscular mycorrhizal fungi (AMF) play a key role in the functioning of natural ecosystems, the underlying drivers determining the composition of AMF communities remain unclear. In this study, we established 138 sampling plots at 46 grassland sites, consisting of 26 acidic grasslands and 20 calcareous grasslands spread across eight European countries, to assess the relative importance of abiotic and biotic filtering in driving AMF community composition and structure in both the grassland soils and in the roots of 13 grassland plant species. Soil AMF communities differed significantly between acidic and calcareous grasslands. In root AMF communities, most variance was attributable to soil variables while very little variation was explained by host plant identity. Root AMF communities in host plant species occurring in only one grassland type closely resembled the soil AMF communities of that grassland type and the root AMF communities of other host plant species occurring in the same grassland type. The observed AMF-host plants networks were not modular but nested. Our results indicate that abiotic conditions, rather than biotic filtering through host plant specificity, are the most important drivers in shaping AMF communities in European seminatural grasslands.

Trait assembly in grasslands depends on habitat history and spatial scale.

During the past century, grasslands in Europe have undergone marked changes in land-use, leading to a decline in plant diversity both at local and regional scales, thus possibly also affecting the mechanisms of species sorting into local communities. We studied plant species assembly in grasslands with differing habitat history and hypothesised that trait divergence prevails in historical grasslands due to niche differentiation and trait convergence prevails in more dynamic grasslands due to competitive exclusion and dispersal limitation. We tested these hypotheses in 35 grassland complexes in Estonia, containing neighbouring grassland habitats with different land-use histories: continuously managed open historical grassland, currently overgrown former grassland and young developing grassland. We assessed species assembly patterns in each grassland type for finer scale-a 2 × 2 m plot scale from a local community pool and for broader scale-a local community from the habitat species pool for that grassland stage and observed changes in trait means at finer scale. We found that grasslands with long management history are assembled differently from former grasslands or young developing grasslands. In historical grasslands, divergence or random patterns prevailed at finer scale species assembly while in former or developing grasslands, mostly convergence patterns prevailed. With increasing scale convergence patterns become more prevalent in all grassland types. We conclude that land-use history is an important factor to consider when assessing grassland functional trait assembly, particularly at small scales. Understanding the mechanisms behind species assembly and their relationship with land-use history is vital for habitat conservation and restoration.