Description
The effect of fluctuations in the emergence and abundance of arthropod species on diets of insectivorous birds is largely unknown. DNA barcoding of fecal samples is a promising technique to track diet composition across habitats and time, but how accurately can we estimate the biomass contribution of prey species in fecal samples using amplification-based barcoding? As a validation study, we COI-barcoded 63 fecal samples of Pied Flycatcher (Ficedula hypoleuca) chicks, whose diet was determined by camera observations of prey items and sizes delivered to the brood. We optimized DNA extraction, PCR conditions and sequencing depth until 90% of the samples yielded on average 14,351 reads of which 96.3% were assigned to Animalia. We modified published COI primers (LCO1490 and HCO1777) to reduce the reads “lost” to bird DNA (from 6% to 0.001 %) and to improve amplification of spiders. Among the 27 most common orders within Arachnida, Insecta, Chilopoda, Diplopoda, and Malacostraca (representing 98% of all Animalia reads), the average proportion of reads in the fecal samples was strongly correlated with the relative biomass scored in the camera sessions (R2 = 0.82). Both datasets identified Lepidoptera, Diptera, and Coleoptera as the most important prey taxa. But does the match between barcodes and camera observations hold at lower taxonomic levels? We will present detailed analyses at the family and genus level. A teaser: as expected, Hymenoptera was underrepresented on the cameras, as we missed parasitic wasps, but both methods identified the same Lepidoptera species as most commonly eaten.| Period | 28-Jun-2019 |
|---|---|
| Event title | 137th American Ornithological Society Annual Meeting: Birds on the Edge - Dynamic Boundaries |
| Event type | Conference |
| Conference number | 137 |
| Location | Anchorage, United States, AlaskaShow on map |
| Degree of Recognition | International |