AbstractWild bee populations are facing strong declines due to a variety of stressors, such as habitat loss and pesticide use. Research focused on wild bees has increased over the last three decades, yet species identifications are still relying on lethal sampling approaches, followed by time‐ and labor‐intensive pinning and morphological analysis by experts. The use of DNA metabarcoding for species detection offers a time‐ and cost‐effective complement to traditional approaches and can potentially facilitate nonlethal sampling. We developed a DNA metabarcoding workflow to identify cavity‐nesting Hymenoptera using environmental DNA collected from vacated nest tubes. We compared different nest sizes and sample types and assessed the effect of these parameters on Hymenoptera species detection rates. We also tested whether our method could detect multiple species in mixed‐species nests. The eDNA results were compared to data from morpho‐taxonomical species identifications. Despite limited DNA quantity, we detected Hymenoptera at the species level in 57.4% of our samples. Hymenoptera detection was also possible from single brood cells. In mixed‐species nests, we correctly detected the presence of multiple species. In nests of solitary wasps, we additionally detected other arthropods, which are probably the remains of larval provisions. Morphological and molecular data agreed in large parts. This eDNA approach offers the opportunity for the implementation of large‐scale and nonlethal monitoring of cavity‐nesting Hymenoptera that additionally provides valuable information on trophic interactions.
Read full abstract