Decay rates of leaf litters from arbuscular mycorrhizal trees are more sensitive to soil effects than litters from ectomycorrhizal trees
AbstractWhile it is well established that leaf litter decomposition is controlled by climate and substrate quality at broad spatial scales, conceptual frameworks that consider how local‐scale factors affect litter decay in heterogeneous landscapes are generally lacking. A critical challenge in disentangling the relative impacts of and interactions among local‐scale factors is that these factors frequently covary due to feedbacks between plant and soil communities. For example, forest plots dominated by trees that associate with ectomycorrhizal (ECM) fungi often differ from those dominated by trees that associate with arbuscular mycorrhizal (AM) fungi in terms of their litter quality, microbial community structure and inorganic nutrient availability. Here, we evaluate the extent to which such factors alter leaf litter decomposition rates. To characterize variations in decomposition rates, we compared decay rates of high‐quality litter (maple; AM) and low‐quality litter (oak; ECM) across forest plots representing a gradient in litter matrix quality and nitrogen (N) availability driven by the relative proportions of AM and ECM trees in each plot. In experiment two, we added litter from two AM and three ECM tree species to forest plots with either a high‐quality litter matrix and high N availability (i.e. AM‐dominated plots) or a low‐quality litter matrix and low N availability (i.e. ECM‐dominated plots). In both experiments, we found that AM litter decomposed more rapidly than ECM litter, and this effect was enhanced in AM‐dominated plots. Then, to separate the contributions of litter matrix effects from N availability effects, we added N fertilizer to a subset of plots from experiment two. Nitrogen addition increased decay rates of high‐quality litter across all sites, but had no effect on low‐quality litter, suggesting that low N availability, not litter matrix quality, constrains decomposition of high‐quality litters. Hence, N availability appears to alter litter decomposition patterns independently of litter matrix properties. Synthesis. Our results indicate that shifts in the relative abundance of ECM‐ and AM‐associated trees in a plot or stand have the potential to affect litter decay rates through both changes in litter quality as well as through alterations of the local‐scale soil environment.
Volume, Issue, Page Number103, 6, 1454-1463