Biodiversity and Ecosystem Functioning: Experimental Systems

Experimental studies of the relationship between biodiversity and ecosystem functioning have taken two main approaches: the experimental removal of species from real communities and the construction of model communities to form gradients of diversity. The fundamental difficulty in designing biodiversity experiments is the large number of possible species combinations that make simple, unambiguous fully-factorial designs impossible with all but the smallest species pools. Nevertheless, experimental and analytical methods have developed to include the replication of both diversity and composition, multi-site experiments and the simulation of non-random extinction scenarios. Methods that calculate deviations from expected null values and that employ covariance approaches have also been developed and these can distinguish between some of the underlying mechanism. Experiments that omit key functional groups can demonstrate that biodiversity effects occur in their absence but approaches that can identify how much different species contribute to the overall relationship between biodiversity and ecosystem processes remain a challenge for future experimental design.