Trophic interaction modifications occur when a consumer-resource interaction is affected by an additional species. They have been well established as being both widespread and to have the potential to exert significant dynamic impacts on ecological communities. Despite increasing calls to incorporate these effects into our understanding of population dynamics, progress has been slow. Because of the sheer number of possible effects, prioritization of the interaction modifications that are likely to have the greatest dynamic impact would be highly valuable.
We have developed a set of new metrics to describe the strength and topology of interaction modifications within the context of the wider system. Using population dynamic models of artificial communities with randomly distributed interaction modifications we investigated the ability of these metrics to identify critical interaction modifications. Our approach tested how a reduced model, including only a subset of interaction modifications (selected by a particular metric), was able to represent the ‘true’ dynamics of the system containing all interaction modifications.
Results/Conclusions
We found a number of properties that consistently identified a subset of interaction modifications that could predict both the directional response to press perturbations and the local stability of the community better than a random subset of interaction modifications. Structural properties of key interaction modifications include low density modifier species, the modification of high biomass flux interactions, and modifications directed ‘down’ the web from top predators. Direct measures of modification strength are also valuable metrics, as quantified by the influence of the modifier density on either the interactor species growth rates or the underlying direct interaction strength. By contrast, the centrality of the interaction being modified and trophic distance between the modifier and the interactors had comparatively little significance.
This theoretical work demonstrates that analysis of interaction modifications can be tractable at the network scale. It allows us to make preliminary suggestions of where experimental emphasis should be placed in practical efforts to understand the effect of interaction modifications in ecological communities.