The concept of resilience has become an important focus in the study of ecosystems that have multiple stable states. If the resilience is high, such systems can recover from relatively large perturbations, whereas a tiny perturbation can tip a low-resilient system to an alternative state. Intuitively it is clear that the spatial scale of the perturbations relative to that of ecological processes is important. However, our fundamental understanding of the spatial aspects of resilience is still limited. We analyze how the effect of local disturbances plays out depending on the scale of feedbacks, using three simple spatially explicit models. 

Results/Conclusions

We show that depending on the spatial scale of perturbations and  ecological processes three distinct responses can be observed: the perturbed patch may be repaired, it may stay unchanged or a domino effect may cause a gradual expansion of the damage till the whole system shifts to the alternative state (we call this a “domino collapse”). Large scale ecological coupling through exchange of organisms or matter enhances the probability of repair of local damage, but at the same time increases the risk of a large scale system collapse. This implies a trade-off between large-scale and small-scale resilience seen also in other complex systems ranging from ecological networks to financial systems.  Our results further indicate that many small-scale perturbations may have less effect than one medium-scale perturbation with the same total disturbed area.