Utilising affordable smartphones and open-source time-lapse photography for pollinator image collection and annotation

Monitoring plant-pollinator interactions is crucial for understanding the factors influencing these relationships across space and time. Traditional methods in pollination ecology are resource-intensive, while time-lapse photography offers potential for non-destructive and automated complementary techniques. However, accurate identification of pollinators at finer taxonomic levels (i.e., genus or species) requires high enough image quality. This study assessed the feasibility of using a smartphone setup to capture time-lapse images of arthropods visiting flowers and evaluated whether these images offered sufficient resolution for arthropod identification by taxonomists. Smartphones were positioned above target flowers from various plant species in urban green areas around Leipzig and Halle, Germany. We present proportions of arthropod identifications (instances) at different taxonomic levels (order, family, genus, species) based on visible features in the images as interpreted by taxonomists. We document whether limitations stem from the automated setup (e.g., fixed positioning preventing capture of distinguishing features despite high image resolution) or from low image quality. Recommendations are provided to address these challenges. Our results indicate that 89.81% of all Hymenoptera instances were identified to family level, 84.56% of pollinator family instances to genus level, and only 25.35% to species level. We were less able to identify Dipterans to finer taxonomic levels, with nearly 50% of instances not identifiable to family level, and only 26.18% and 15.19% identified to genus and species levels. This was due to their small size and the more challenging features needed for identification (e.g., in the wing veins). Advancing smartphone technology, along with their accessibility, affordability, and user-friendliness, offers a promising option for coarse-level pollinator monitoring.