Monitoring insect pollinators and flower visitation: the effectiveness and feasibility of different survey methods

The status of pollinating insects is of international concern, but knowledge of the magnitude and extent of declines is limited by a lack of systematic monitoring. Standardized protocols are urgently needed, alongside a better understanding of how different methods and recorders (data collectors) influence estimates of pollinator abundance and diversity. We compared two common methods for sampling wild pollinating insects (solitary bees, bumblebees and hoverflies), pan traps and transects, in surveys of 1 km countryside squares (agricultural and semi-natural habitats) and flowering crop fields across Great Britain, including the influence of local floral resources (nectar sugar availability or crop flower density) on the insects sampled. Further, we compared the performance of recorders with differing expertise (non-specialist research staff, taxonomic experts and non-expert volunteers) in applying methods. Pan traps and transects produced compositionally distinct samples of pollinator communities. In the wider countryside, pan traps sampled more species of solitary bee and hoverfly. In flowering crops, transects recorded a greater number of individual bumblebees, but fewer species. Across all taxonomic groups and countryside and crop samples, transects generally had lower rates of species accumulation per individual collected than pan traps. This demonstrates that differences between methods in estimating richness are not due to sampling effort alone. However, recorders possessing greater taxonomic expertise can produce species accumulation data from transects that are almost commensurate with pan trapping. The abundance and species richness of pollinators (except solitary bees) on transects in the wider countryside was positively related to the availability of estimated nectar sugar. In crops, pollinator abundance responses to flower densities were idiosyncratic according to crop type, but overall the response was positive and negative for transects and pan traps, respectively. Given these taxonomic and context-specific differences in method performance, we assess their suitability for monitoring pollinating insect communities and pollination services. We discuss the relevance of these findings within the context of achieving standardized, large-scale monitoring of pollinating insects.