Linking populations to landscapes: richness scenarios resulting from changes in the dynamics of an ecosystem engineer.

Predicting the effects of the loss of individual species on diversity represents one of the primary challenges facing community ecology. One pathway by which organisms of one species affect the distribution of species is ecosystem engineering. Changes in the dynamics of ecosystem engineers that lead to changes in the distribution of the patches of altered habitat are likely to lead to changes in diversity. I link data on the distribution of plant species found in the riparian zone of the Adirondacks (New York, USA) in patches modified by beaver and in unmodified forest patches to a model connecting the dynamics of ecosystem engineers to the dynamics of the patches that they create. These analyses demonstrate that changes in key parameters of the model, such as decreases in beaver colonization rates and rate of patch abandonment, lead to changes in species richness of up to 45% at the landscape scale, and that these changes are likely to occur over long time scales. This general approach of linking the population dynamics or behavior of a single species to changes in species richness at the landscape scale provides a means for both testing the importance of ecosystem engineering in different systems and developing scenarios to predict how changes in the dynamics of a single species are likely to affect species richness.