Herbivore-induced dissolution of the dimensional structure of biodiversity’s association with ecosystem functioning: Impacts of deer exclusion on biodiversity and ecosystem functioning in a temperate forest

Background/Question/Methods

The influence of biodiversity on ecosystem properties has been examined primarily by varying single dimensions of diversity (e.g., taxonomic, functional, genetic) and then monitoring ecosystem responses.  In general, these uni-dimensional studies reveal that biodiversity is likely to show a positive, saturating association with ecosystem functioning.  It is unlikely, however, that changes in one biodiversity dimension will not result in changes in others which implies that one may not readily extrapolate findings from uni-dimensional studies to situations in which multiple dimensions of biodiversity have simultaneously changed.  To explore this issue, we examined whether changes in the dimensional structure of biodiversity impacts ecosystem properties in ways not readily discerned from uni-dimensional studies.  Specifically, we examined the impact of deer browsing on understory vegetation in a temperate deciduous woodland at Black Rock Forest in the Hudson Highlands of New York.  White-tail deer (Odocoileus virginianus) were excluded from replicate plots for five years which allowed understory vegetation to recover and exhibit changes in biodiversity.  We examined three biodiversity dimensions:  taxonomic (TD), functional (FD), and phylogenetic (PD) diversity.  We then used structural equation modelling (SEM) to determine the simultaneous impacts of herbivore-induced changes in these three dimensions on understory canopy cover, our chosen ecosystem property. 

Results/Conclusions

SEM revealed multiple, simultaneous, positive associations between TD, FD, and PD of the understory vegetation in plots where deer were excluded for five years.  These dimensions additionally showed significant differences in associations with canopy cover that varied in magnitude and sign (TD = 0.48, p<0.05; FD = -0.66, p<0.01, PD = 0.59, p<0.01).  These significant associations between multiple dimensions of biodiversity and canopy cover, however, were not found in plots exposed to herbivory for five years.  While the dimensions remained positively associated with one another, there were no significant influences on understory canopy (TD = -0.21, p>0.5; FD = -0.27, p>0.4 p<0.01, PD = -1.10, p>0.7).  We conclude that herbivore-induced changes in multiple dimensions of understory biodiversity exhibited a dissolution of biodiversity’s influence on canopy cover.  These findings suggest that herbivory, or other trophic interactions, may render biodiversity’s impact on ecosystem functioning less predictable than uni-dimensional studies currently suggest.