Association of ectomycorrhizal trees with high carbon-to-nitrogen ratio soils across temperate forests is driven by smaller nitrogen not larger carbon stocks

1. The distribution of mycorrhizal associations across biomes parallels a distinct gradient of soil carbon (C) and nitrogen (N) stocks, raising the question of how mycorrhizal traits relate to ecosystem properties. Arbuscular mycorrhizal (AM) and ectomycorrhizal (EM) hosts and fungi employ contrasting strategies for N acquisition, which may manifest in differences in soil C and N pools and/or soil C: N. However, cross-biome comparisons are confounded with climatic and edaphic gradients as well as phylogenetic and functional trait distributions of component plant species. Here, we test emerging hypotheses that soil C, N and C: N are related to the dominance of EM trees within a temperate forest region where AM and EM trees largely coexist but vary in local abundance. 2. To determine the importance of mycorrhizal type on soil C and N, we analysed data from c. 1,000 forest inventory plots in the eastern United States. For each plot, we quantified the dominance of trees with different mycorrhizal associations and accounted for potentially confounding variables including phylogeny (angiosperm or gymnosperm), leaf N, soil clay content and climate. We used hierarchical Bayesian models to determine how these variables explained the patterns of soil C and N in the forest floor and mineral soil layers. 3. Increasing EM dominance was associated with higher C: N across all soil layers. This relationship remained even after accounting for tree phylogeny, leaf N content, soil clay content, temperature and precipitation, which were all important for explaining soil C: N. However, this mycorrhizal pattern of soil C: N was not related to increases in soil C content; rather, increasing EM dominance was associated with reductions in soil N. 4. Synthesis. Our findings are consistent with the proposition that mycorrhizal associations are related to terrestrial ecosystem properties. The mycorrhizal effect on soil C: N may result from differences in how arbuscular mycorrhizal and ectomycorrhizal plants interact with their fungal symbionts, decomposers and organic matter, to sustain differential cycling of C and N. Alternatively, these patterns could arise from differential success of the two mycorrhizal types in contrasting soil conditions; both processes may occur simultaneously, leading to a self-reinforcing positive feedback.