Restoration of plant-pollinator interaction networks via species translocation

The recent decline in pollinator biodiversity, notably in the case of wild bee populations, puts both wild and agricultural ecosystems at risk of ecological community collapse. This has triggered calls for further study of these mutualistic communities in order to more effectively inform restoration of disturbed plant-pollinator communities. Here, we use a dynamic network model to test a variety of translocation strategies for restoring a community after it experiences the loss of some of its species. We consider the reintroduction of extirpated species, both immediately after the original loss and after the community has reequilibrated, as well as the introduction of other native species that were originally absent from the community. We find that reintroducing multiple highly interacting generalist species best restores species richness for lightly disturbed communities. However, for communities that experience significant losses in biodiversity, introducing generalist species that are not originally present in the community may most effectively restore species richness, although in these cases the resultant community often shares few species with the original community. We also demonstrate that the translocation of a single species has a minimal impact on both species richness and the frequency of community collapse. These results have important implications for restoration practices in the face of varying degrees of community perturbations, the refinement of which is crucial for community management.