The need for large-scale experiments to assess and predict the response of ecosystems to perturbation.
Ecosystem experiments are field experiments in which the experimental unit is large enough to include the relevant physical, chemical, and biotic context of the processes being studied. Whole-ecosystem experiments have yielded insights about many processes in a diversity of habitats. Successful design and interpretation of ecosystem experiments depend on connections to theoretical, long-term, and comparative studies. Ecologists have overcome many problems of inference for ecosystem experiments. However, the issue of replication is far less important than the need to compare alternative explanations for the results, which may involve reference ecosystems, premanipulation data, and additional measurements or experiments designed to compare possible explanations. Potential limitations of ecosystem experimentation include the variability and slow dynamics of ecosystems, certain aspects of academic and management culture, and institutional shortcomings. Progress in ecosystem experiments can be accelerated through dedicated sites and funding, and keystones that foster collaborations between management and science for adaptive ecosystem management.