Experimental species removal alters ecological dynamics in a natural ecosystem

Theory predicts that species extinction or invasion can affect the temporal dynamics of ecological communities by altering feedback patterns and by damping or amplifying environmental variation via changes in the network of species interactions, but because of the logistical challenges of investigating temporal dynamics, evidence from natural ecosystems is lacking. In a long-term experimental manipulation of a rocky intertidal community on Tatoosh Island, Washington, U.S.A., chronic removal of the dominant species Mytilus californianus altered the dynamics of the system, causing reductions in the temporal variability of three subdominant species but no consistent change in the spectral characteristics or the order of density dependence across experimental replicates. This pattern of results suggests that Mytilus californianus impacted the temporal dynamics by amplifying environmental stochasticity, rather than by changing feedback pathways as is emphasized in most theoretical predictions and laboratory studies. Hence, further investigation of the mechanisms and implications of transmission of environmental stochasticity in natural ecosystems is merited.