Biodiversity is associated with important ecosystem processes and functions. However, many species are currently threatened by human activities, making ecological restoration a major tool in conservation biology. An important component of biodiversity are the interactions between species as these provide ecological functions and services. In terrestrial systems restoration usually starts with the restoration of plant communities, as plants serve directly or indirectly as a resource for upper trophic levels. Ecological networks provide a powerful tool for describing, analysing and understanding whole communities in a restoration context, for example they can be used to identify structurally important species and to measure community robustness. In this thesis, I use plant-insect networks from natural and agricultural systems, to investigate how plant communities support biodiversity at higher trophic levels. I do this by identifying keystone resources for insect herbivores and parasitoids, and by showing that keystone roles are performed by few plant species and that these roles are context dependent. Using a field experiment, I then evaluate whether plant species network roles, i.e. central vs. peripheral, can be used to restore pollinator communities, and found that central plant species attracted a higher richness and abundance of pollinators than peripheral species. Finally, I test the robustness of pollination and herbivory networks to the loss of plant species, accounting for differences in network structure and natural history between both systems. I found that herbivory networks tend to be more robust than pollination networks to plant extinctions, but that the inverse is true when interaction rewiring is considered. Together, these three approaches extend both our current understanding of bottom-up effects in plant-insect networks and the potential to undertake restoration that targets more than one trophic level.
|Date of Award||23 Jan 2019|
- The University of Bristol
|Supervisor||Jane Memmott (Supervisor)|