The Oregon Bee Atlas (OBA) is a statewide citizen science initiative that aims to document the diversity and floral associations of Oregon’s native bee fauna. Through a standardized protocol, volunteers collect bee specimens and record their floral host data using the iNaturalist platform. This extensive dataset has enabled the development of the Melittoflora, an interactive analytical tool that visualizes the relative frequency of bee-flower associations and provides insights into the floral preferences of individual bee species across space and time.
The Melittoflora has the potential to serve as a predictive model for pollinator community composition based on known floral resources, which could be invaluable for habitat restoration planning, conservation prioritization, and pollinator-supportive land management. However, the predictive accuracy of this tool has not yet been empirically validated. To address this gap, we conducted a “ground-truthing” study of the Melittoflora system within a restored grassland prairie in Oregon. Using the same specimen collection and documentation protocols established by the Oregon Bee Atlas, we sampled bee communities and compared observed floral associations with those predicted by the Melittoflora database. Here, we present our findings on the concordance between predicted and observed bee-flower interactions, discuss factors influencing model accuracy (e.g., floral phenology, habitat context, sampling biases), and evaluate the tool’s applicability for forecasting pollinator presence in restored or managed landscapes. Our results contribute to the refinement of Melittoflora as a practical decision-support tool for restoration ecologists and land managers working to enhance native pollinator habitats.
.png)
