However, without direct observations of foraging, species-specific foraging behaviour is difficult to quantify. As a result we know relatively little especially about species-specific patterns in systems with several browsers that are perhaps competing for resources.
A group of Swedish researchers have now developed a browser specific species diagnostic kit that utilizes environmental DNA (eDNA). They took advantage of the fact that during browsing, a small amount of saliva containing some cells is deposited at the bite site, providing a source for DNA that can be used for species identification. They were able to distinguish between four sympatric ungulates (moose, Alces alces; roe deer, Capreolus capreolus; fallow deer, Cervus dama; and red deer, Cervus elaphus) that occur in deciduous forests in central Sweden. I would call that DNA Barcoding if they hadn't decided to use Cytochrome b primers for their study but that's more a problem of standardization. The group was able to show that false positives were extremely uncommon and that about half of the samples could be amplified even if the browsing happened up to 12 weeks before. Some samples as old as 24 weeks could also be amplified.