Gray wolves (Canis lupus) are habitat generalists occupying a wide array of ecoregions across North America. Wolves in Southeast Alaska rely on Sitka black-tailed deer (Odocoileus hemionus Sitkensis) as primary prey, but deer abundance varies by geographical region and the use of alternative prey resources by wolves has been documented. Improved understanding of how wolves use prey species in relation to their availability can provide insights into the functional responses of prey abundance to predation, an important consideration where deer populations are low and intensive management actions are currently being considered. Additionally, wolf population densities at the regional scale (which are crudely estimated with ungulate densities) may be produced more accurately by accounting for the extent to which wolves rely on ungulates compared to alternate prey items.
One of the most common methods for profiling diet is the mechanical sorting of scat remains. Diet analysis using mechanical sorting methods can often be inefficient and can yield incomplete results. Recently, the isolation and amplification of DNA present within environmental DNA samples has presented a powerful technique for identifying species and profiling biodiversity. Using molecular methods to profile contents within a scat sample has the potential to improve efficiency, reveal rare prey species, and correct mistaken identification error. In this study, we compare diets from wolf scats profiled using mechanical sorting and metabarcoding of amplified vertebrate DNA sequences. We have two primary objectives: (1) compare findings from mechanical sorting and metabarcoding of amplified target DNA sequences to validate methods and (2) to quantify diets of coastal gray wolves in southeast Alaska.
A total of 185 scats were collected from Southeast Alaska and processed using metabarcoding and mechanical sorting. Analyses show that species identified by metabarcoding were largely comparable to species identified by mechanical sorting, particularly for scats with less than three prey species present. However, there were a number of notable differences between the metabarcoding and mechanical sorting results, and in general, metabarcoding showed a higher number of species present in a given scat. Finally, these methods revealed a large diversity of prey species including six species of fish, five species of bird, two species of marine mammal, and seven species of terrestrial mammal. These results help validate metabarcoding as a powerful method for diet analysis as well as provide an informative diet profile for wolves in Southeast Alaska, which may have implications for species management in this system.