MEPS 311:263-271 (2006)  -  doi:10.3354/meps311263

ABSTRACT: The extent to which changes in biodiversity are causally linked to key ecosystem processes is a primary focus of contemporary ecological research. Highly controlled manipulative experiments have revealed significant and positive effects of increased diversity on ecosystem functioning, but uncertainties in experimental design have made it difficult to determine whether such effects are related to the number of species or to effects associated with species identity and density. Using infaunal marine invertebrates, we established 2 parallel laboratory experiments to examine the hypothesis that changes in the composition of benthic macrofauna alter the biogeochemistry of coastal intertidal mudflats. Our study identified clear effects of increased infaunal species diversity on nutrient generation. However, significant species identity and density effects underpin the observed response, reflecting species-specific traits associated with bioturbation. Post-hoc examination of our conclusions using power analysis revealed that, given our experimental design, the probability of finding a correct significant effect, the minimum detectable difference necessary to detect a significant effect, and the minimum number of replicates necessary in order to achieve an acceptable power, all differed between species. Our study has important implications for the design of biodiversity–ecosystem function experiments because the disparity between the contributions that individual species make to ecosystem function demands the use of different levels of replication for each species within an experiment.

KEY WORDS: Biodiversity · Ecosystem function · Species richness · Experimental design · Nutrients · Bioturbation

Full text in pdf format

Export citation Share:    Facebook - - linkedIn