In natural forests, and in many tree biodiversity experiments, functional and phylogenetic diversity are confounded with each other and with species richness. As a result, it is difficult to disentangle the consequences of these dimensions of biodiversity for the services that forests provide, including production of fiber, contribution to inorganic nutrient cycling through decomposition, and regulation of invertebrate herbivore populations. Though recent tree diversity experiments have begun to address these biodiversity-ecosystem service relationships, few are designed to disentangle the role of functional from phylogenetic diversity or distinguish their roles from those of species richness or functional trait identity. In short, many diversity experiments are not designed to address these biodiversity-ecosystem functioning relationships across phylogenetic scales. We present an analysis of the links between these dimensions of diversity and forest functionality over the first several years of the Forests and Biodiversity (FAB) and Biodiversity in Willows and Poplars (BiWaP) experiments at Cedar Creek Ecosystem Science Reserve in Eastern Minnesota, USA. These experiments are designed to disentangle the consequences of functional diversity from phylogenetic diversity (FAB) and taxonomic from intraspecific diversity (BiWaP) for ecosystem function. We expected that biodiversity across phylogenetic and functional axes would enhance productivity, nutrient cycling, and generalist herbivory, while suppressing specialist herbivory.
We found that species richness, along with functional diversity and functional trait identity, of tree communities increased stem biomass growth (FAB), although dominant species suppressed growth of neighbors (BiWaP). Though functional traits also predicted decomposition of leaf litter in experimental litterbags, decomposition was driven by trait identity and was not affected by diversity per se. Finally, herbivory responded in complex ways to diversity, with generalist leaf removal decreasing (FAB) or showing no change (BiWaP) with phylogenetic diversity and specialist damage to leaves decreasing (both) with host dilution. These results indicate that the consequences of diversity are not straightforward and that, though phylogenetic diversity may serve as a useful proxy of biodiversity, functional trait information is still critical for predicting ecosystem service provision. We conclude that biodiversity-ecosystem functioning relationships vary with phylogenetic scale. Future tree diversity experiments should be designed such that multiple dimensions of biodiversity, from intraspecific genetic diversity through phylogenetic diversity across evolutionary lineages, can be manipulated independently.