Positive effects of trait and phylogenetic distances on tree growth and implications for biodiversity-productivity relationship

Background/Question/Methods

Numerous grassland experiments have found evidence for the complementarity effect -- an increase in productivity with higher plant species richness due to niche partitioning. However, empirical tests of complementarity in complex tropical forests are lacking. We conducted a spatially explicit analysis of 541,568 growth records of 292 species from a 50-ha tropical forest plot to test neighborhood complementarity -- whether an individual tree’s growth is enhanced by greater dissimilarity to its neighbors. We constructed linear mixed-effect models to assess how neighborhood distance measured as the distance-weighted functional trait and phylogenetic differences between focal and neighborhood trees would affect focal tree’s absolute diameter growth. Furthremore, the strength of complementarity may depend on focal tree’s shade tolerance as shade intolerant species are often more sensitive to available resources and natural enemies than shade tolerant species. To test this hypothesis, we used hierarchical Bayesian models to assess the relationship between individual species’ neighborhood distance dependent effects and shade tolerance. Finally, we used simulations to confirm whether the identified neighborhood complementarities were sufficient to produce positive biodiversity-productivity relationship at stand level.

Results/Conclusions

Both neighborhood phylogenetic distances (PD) and trait distances (TD) had significant positive effects on individual tree growth. Doubling neighborhood PD and TD increased absolute diameter growth by an average of 23.5% and 18.4%, respectively. When neighborhood distance indices were calculated without conspecifics, their effects on individual growth decreased considerably but remained positive and significant. Doubling heterospecific PD and TD increased absolute diameter growth by an average of 8.67% and 5.85%, respectively. This result indicates that both conspecifics and heterospecifcs contribute to the observed positive neighborhood complementarity effect. When TD indices were calculated with each single trait rather than all the eight traits in regular TD index, TD effects on tree growth were insignificant and much weaker. It implied that the observed positive complementarity effects should be the joint results of multiple processes (e.g. competition for different resources and natural enemy attacks) reflected by different traits. The strength of individual species’ neighborhood complementarity was negatively related to shade tolerance: more shade-intolerant species had larger coefficients of PD and TD. Finally, our simulations demonstrated that the identified neighborhood complementarities were sufficient to produce positive biodiversity-productivity relationships. Our study provides mechanistic links among neighborhood interaction, functional identity, phylogenetic relatedness, and complementarity effects.