Abstract/Summary

Understanding spatial variation in the structure and stability of plant-pollinator networks, and their relationship with anthropogenic drivers, is key to maintaining pollination services and mitigating declines. Constructing sufficient networks to examine patterns over large spatial scales remains challenging. Using biological records (citizen science), we constructed potential plant-pollinator networks at 10km resolution across Great Britain, comprising all potential interactions inferred from recorded floral visitation and species co-occurrence. We calculated network metrics (species richness, connectance, pollinator and plant generality) and adapted existing methods to assess robustness to sequences of simulated plant extinctions across multiple networks. We found positive relationships between agricultural land cover and both pollinator generality and robustness to extinctions under several extinction scenarios. Increased robustness was attributable to changes in plant community composition (fewer extinction-prone species) and network structure (increased pollinator generality). Thus, traits enabling persistence in highly agricultural landscapes can confer robustness to potential future perturbations on plant-pollinator networks.