Print PageThe concept of resilience has a long history in ecology both as a way of understanding the response of ecosystems to perturbations and also as a way of predicting the characteristics of ecosystems. This theory has been carefully developed for the linear dynamics of small perturbations, but in natural systems, ecologists have increasingly recognized that large perturbations and nonlinear responses are the rule, not the exception. Quantifying resilience to large perturbations clarifies the impact of nonlinearity and density dependence on ecosystem function. Starting with two species models and then scaling to large food webs, we map all perturbations that return to a specified neighborhood of a steady state within a given amount of time. For linear dynamics, the boundaries of this resilience map can be calculated analytically. For nonlinear dynamics, boundaries are identified with a Bayesian sampling scheme.
Results/Conclusions
We find that resilience maps for large perturbations to nonlinear food webs form highly skewed regions in the multidimensional phase space of population abundances. This result demonstrates that nonlinear food webs are highly sensitive to the direction of a perturbation. Recovery times along two distinct directions may differ by several orders of magnitude. As a result, food webs that evolve a high degree of resilience to a recurring set of disturbances may be highly vulnerable to other, less frequent disturbances. In particular, food web adaptation to prevailing conditions may diminish resilience in the face of global change.
