PARENT SESSION
Oral Session # 1: Ecological Theory I: Theory; Scaling.
Presiding: M Pascual
Monday, August 4. 8:00 AM to 11:30 AM, SITCC Meeting Room 100.

Does functional biodiversity matter: A null-model analysis of functional diversity ecosystem function relationships.

Wright, Justin^*,1, Naeem, Shahid^{1, 2}, ¹ University of Washington, Seattle, WA² Columbia University, New York, NY

ABSTRACT- Recent investigations of the relationship between biodiversity and ecosystem functioning have focused on the importance of functional diversity, most often measured as functional group richness (FGR). A number of key experiments have shown that plots with higher FGR tend to exhibit higher levels of ecosystem function, leading the investigators to conclude that mechanisms associated with the complementarity of functional groups they selected are likely to be what governs significant relationships between biodiversity and ecosystem functioning. These studies have been primarily plant-based and have used a priori schemes for classifying plant species. For example, plant species are typically classified as grasses, non-leguminous forbs, or legumes. Such studies have not considered that a single a priori functional group represents only one of an extremely large number of possible ways for classifying species by function. Using data from six key studies of grassland diversity and ecosystem functioning in Cedar Creek, Minnesota and the BIODEPTH plots across Europe, we assigned species to functional groups using Monte Carlo methods. We evaluated the performance of a priori functional group classifications relative to the performance of randomly classified functional groups by comparing the former with the distribution of R² values from linear regressions between FGR and ecosystem functions. These functions included above- and below-ground biomass and soil N. Although there were statistically significant relationships between a priori FGR and ecosystem functions in 21 out of the 26 regressions, in only 6 cases did they explain more of the variance than randomly assigning species to functional groups. Furthermore, neither site nor ecosystem function predicted when a priori classification outperformed random classification. These results suggest that significant relationships between FGR and ecosystem function do not necessarily support complementarity mechanisms implied in a priori classification schemes.

Key words: ecosystem function, functional diversity, biodiversity, Monte Carlo simulation