Spatio-temporal variation in diet among age and sex cohorts of a model generalist bird species, the Great Tit Parus major: new insights revealed by DNA metabarcoding

Dietary variation among cohorts can have a major impact on how populations adapt to environmental variation. Although variation in diet between cohorts and across habitats has been studied in many taxa, this is not true for most birds, especially smaller generalist passerines whose feeding habits are predominantly cryptic. Here we used DNA metabarcoding with next-generation sequencing to assess spatio-temporal dietary variation among age and sex cohorts of the Great Tit Parus major, a model species in avian ecology. Most dietary species were rare but nevertheless collectively made up 30% of the diet, as expected of a generalist. Winter moth Operophtera brumata, a major focus in tit breeding phenology research, was the most prevalent dietary item, but the next ten most prevalent Lepidoterans were collectively four times more important. There was considerable variation in dietary richness and composition among seasons and years. In winter, natural plant and invertebrate species were extensively represented in the diet, despite the constant availability of supplemental food. Diet composition varied with woodland type: in conifer woodlands, birds fed on species adapted to conifer plantations, as expected, but they also fed on many species adapted to deciduous species. In winter, birds in conifers used peanut feeders more than they did in mixed woodlands, where beech was more prevalent in the diet. In winter, first-year birds consumed more invertebrate species than adults, presumably because they were less selective, and beech (Fagus) was almost twice as prevalent in the first-year diet. Our results suggest considerable spatio-temporal variation in diet and variation among cohorts, and provide insight into the diet of a key model species in avian ecology. Such variation is rarely considered, even though it is likely to have important consequences for our understanding of how populations respond to environmental change.

Methods
Dietary data was obtained from Great Tit faecal samples in 2017 and 2018. DNA was extracted from faecal samples, and DNA metabarcoding carried out to identify invertebrate and plant dietary taxa. The data underwent bioinformatic processing to obtain dietary taxa. Lab and bioinformatic work was carried out at Cardiff University in the School of Biosciences. Sampling completeness analysis was conducted. Multivariate generalised linear models and generalised linear mixed models were carried out to examine dietary composition between different age classes, sexes, habitat types, and seasons.