Abstract
Boreal forests are threatened worldwide due to land use change and climate change. In these habitats, the moose (Alces alces) functions as an ecosystem engineer by influencing plant community structure, and consequently is of significance when managing forests for resource use and conservation goals. In this study, we used vegetation survey data and moose diet data, derived from metbarcoding of fecal samples, to investigate links between the moose diet and plant community composition at fifteen sites in Trøndelag, Norway. We extracted taxonomic data from 257 moose fecal samples and compared with plant abundance data carried out in concurrent vegetation surveys from an existing exclosure experiment. We identified sixteen plant taxa that were both prevalent in the moose diet and abundant in the survey sites. Both simple and mixed linear models were used to examine relationships between the frequency of a species in the moose diet and its abundance. The results indicate that there is no strong evidence to support a dietary link with direct browsing impacts on plant abundance overall, though certain species are impacted more heavily by moose presence than others. The slow-growing woody species, Pinus sylvestris, exhibited significant negative response to moose browsing, while species with more competitive traits such as Trientalis europaea, Melampyrum sylvaticum, Avenella flexuosa, and Luzula pilosa all exhibited positive response to moose presence. This indicates that the impacts of moose browsing are more influenced indirectly, such as through trampling, increased light availability, and reduced competition. Management schemes in boreal forests should account for these direct and indirect impacts, developing holistic approaches toward ecosystem function and multi-use management. Potential relationships between biomass in the diet of large herbivores and plant community structure should be further investigated as quantitative approaches to DNA metabarcoding improve.