Abstract

In rugged terrain subject to active geomorphological processes, the species composition of forest communities changes along topographic gradients over short distances. However, the phylogenetic structure of forests on rugged terrain has rarely been examined. Understanding such structures provides insight into community assembly rules dependent on local environmental conditions. To this end, we tested the topographic trends of measurements of phylogenetic community structure [net relatedness index (NRI) and nearest taxon index] in a catchment covered by temperate forests with complex relief in Japan. We found that phylogenetic structure changed from over-dispersion to clustering with increasing slope inclination, change of slope aspect from south to north and decreasing soil depth. This result suggested that environmental filtering tended to restrict community composition at relatively stressful sites, whereas species interaction functioned more strongly at relatively stress-free sites. Relatively stressful sites were characterized by early-successional species that tended to assemble in certain phylogenetic clades, whereas highly competitive late-successional species associated with lower NRI at relatively stress-free sites. However, despite this significant topographic tendency, phylogenetic community structures were not statistically different from random assumptions in most plots. This obscuration of the phylogenetic structures at the plot level could be interpreted as species adapting to disturbances because they were abundant in this catchment. Accordingly, we suggest that a stochastic process was also important for this community at the plot level, although biotic and abiotic environmental filtering controlled the vegetation structure in the catchment, where disturbances occur frequently because of geomorphological mountain denudation processes owing to active crustal movements and abundant rainfall.

Highlights

  • Understanding the causal factors of fine-grained spatial variations in species diversity and composition within a forest is one of traditional tasks of ecology (Clark et al 1998)

  • After constructing models for all combinations of the topographical variables, i.e. 144 patterns, we identified the model with the smallest Akaike information criterion (AIC) as the best model, and detected the other top three models

  • These variables varied with a unimodal distribution. They showed a significant negative correlation with Faith’s phylogenetic diversity (PD); net relatedness index (NRI) for all species: r2 1⁄4 0.505, P, 0.001; nearest taxon index (NTI) for all species: r2 1⁄4 0.498, P, 0.001; NRI for angiosperms: r2 1⁄4 0.198, P, 0.001 and NTI for angiosperms: r2 1⁄4 0.214, P, 0.001. They did not correlate with species richness; NRI for all species: r2 1⁄4 0.004, P 1⁄4 0.947; NTI for all species: r2 1⁄4 0, P 1⁄4 0.881; NRI for angiosperms: r2 1⁄4 0.016, P 1⁄4 0.218 and NTI for all angiosperms: r2 1⁄4 0.016, P 1⁄4 0.211

Read more

Summary

Introduction

Understanding the causal factors of fine-grained spatial variations in species diversity and composition within a forest is one of traditional tasks of ecology (Clark et al 1998). Theory suggests that both deterministic and stochastic processes explain spatial variation in community patterns

Objectives
Methods
Results
Discussion
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call