Mycorrhizal symbioses and tree diversity in global forest communities

Abstract

Unraveling the mechanisms underlying the maintenance of species diversity is a central pursuit in ecology. It has been hypothesized that ectomycorrhizal (EcM) in contrast to arbuscular mycorrhizal fungi can reduce tree species diversity in local communities, which remains to be tested at the global scale. To address this gap, we analyzed global forest inventory data and revealed that the relationship between tree species richness and EcM tree proportion varied along environmental gradients. Specifically, the relationship is more negative at low latitudes and in moist conditions but is unimodal at high latitudes and in arid conditions. The negative association of EcM tree proportion on species diversity at low latitudes and in humid conditions is likely due to more negative plant-soil microbial interactions in these regions. These findings extend our knowledge on the mechanisms shaping global patterns in plant species diversity from a belowground view.

Fig. 1.. Hypothetical relationships between the proportion of EcM tree individuals in a community and tree species richness; and forest plot data used to test the predictions.

(A to C) Species richness can show a negative (only includes a negative first order term, hypothesis 1, or EcM dominance hypothesis), symmetrically unimodal (includes both linear and quadratic terms but the absolute coefficients of both terms have similar magnitude, hypothesis 2, or mycorrhizal mixture hypothesis), or asymmetrically unimodal (the absolute value of quadratic term being larger than that of the first-order term, hypothesis 3, or integrated hypothesis) relationship with the proportion of EcM trees. (D) The global pattern of the proportion of EcM tree species based on stem abundance in our forest plot dataset (N = 4090). The green background on the map indicates forest regions classified by (43). (E) Raw relationship between the proportion of EcM tree individuals and tree species richness using the generalized linear model without controlling for covariates (r² = 0.59).

Fig. 2.. Effects of EcM tree proportion interacting with absolute latitude and aridity on tree species richness.

(A) Standardized coefficients of generalized linear model, open and solid circles indicate nonsignificant and significant relationships (P < 0.05); (B to D) relationships between EcM tree proportion and tree species richness across gradients of absolute latitudes and aridity (aridity index = 0.5, 1.2, and 2); lines are predicted species richness (log₁₀ back-transformed) after controlling for covariates; the ranges of species richness differ among these panels due to variations in the aridity index. EcM and EcM2 indicate linear and quadratic proportions of EcM trees; AI, aridity index; soil clay, soil clay content. Lat_abs, absolute latitude.

Fig. 3.. The relative importance of predictors in explaining tree species richness in random forest models.

(A) Global model (N = 4,090, r² = 0.88); (B) boreal model (N = 579, r² = 0.55); (C) temperate model (N = 2,457, r² = 0.80); (D) tropical model (N = 1,054, r² = 0.75).

Fig. 4.. Direct and indirect effects of environmental variables on species richness in forest communities.

(A) Global model; (B) boreal model; (C) temperate model; (D) tropical model. The goodness of fit for structural equation models is indicated by the Fisher’s C value with P > 0.05 suggesting good fit of models and no missing paths. Brown and blue solid lines represent significantly positive and negative relationships (P < 0.05), respectively. Dashed gray lines represent nonsignificant relationships (P ≥ 0.05). Arrow width is proportional to strength if there is a significant relationship. SR, species richness; EcM and EcM2 indicate the linear and quadratic proportions of ectomycorrhizal trees; MAT, mean annual temperature; AI, aridity index; soil N, total soil nitrogen. The path from the linear to the quadratic proportion of ectomycorrhizal trees must be included in the model by piecewiseSEM package (72) due to their mathematical relationship (not causality). Paths linking environmental variables to EcM2 were removed because they were not significant. Including all potential paths in the models would also prevent the calculation of Fisher’s C. *P < 0.05, **P < 0.01, ***P < 0.001.