Data from: Negative frequency-dependent foraging behaviour in a generalist herbivore (Alces alces) and its stabilizing influence on food-web dynamics

• Main Authors: Hoy, Sarah R., Vucetich, John A., Liu, Rongsong, DeAngelis, Don L., Peterson, Rolf O., Vucetich, Leah M., Henderson, John J.
• Format: Article in Journal/Newspaper
• Language: unknown
• Published: 2019
• Subjects: consumer-resource dynamics; context-dependent decision-making; diet choice; frequency-dependent foraging strategies; selective foraging behaviour; Manly’s selection index; predation risk; snow depth; Alces alces
• Online Access: http://hdl.handle.net/10255/dryad.216161; https://doi.org/10.5061/dryad.6v2k320

1.Resource selection is widely appreciated to be context‐dependent and shaped by both biological and abiotic factors. However, few studies have empirically assessed the extent to which selective foraging behaviour is dynamic and varies in response to environmental conditions for free‐ranging animal populations. 2.Here, we assessed the extent that forage selection fluctuated in response to different environmental conditions for a free‐ranging herbivore, moose (Alces alces) in Isle Royale National Park, over a 10‐year period. More precisely, we assessed how moose selection for coniferous versus deciduous forage in winter varied between geographic regions and in relation to: (i) the relative frequency of forage types in the environment (e.g., frequency‐dependent foraging behaviour), (ii) moose abundance, (iii) predation rate (by gray wolves), and (iv) snow depth. These factors are potentially important for their influence on the energetics of foraging. We also built a series of food‐chain models to assess the influence of dynamic foraging strategies on the stability of food‐webs. 3.Our analysis indicates that moose exhibited negative frequency‐dependence, by selectively exploiting rare resources. Frequency‐dependent foraging was further mediated by density‐dependent processes, which are likely to be predation, moose abundance, or some combination of both. In particular, frequency‐dependence was weaker in years when predation risk was high (i.e., when the ratio of moose to wolves was relatively low). Selection for conifers was also slightly weaker during deep snow years. 4.The food‐chain analysis indicates that the type of frequency‐dependent foraging strategy exhibited by herbivores had important consequences for the stability of ecological communities. In particular, the dynamic foraging strategy that we observed in the empirical analysis (i.e. negative frequency‐dependence being mediated by density‐dependent processes) was associated with more stable food‐web dynamics compared to fixed foraging strategies. 5.The results of this study indicated that forage selection is a complex ecological process, varying in response to both biological (predation and moose density) and abiotic factors (snow depth) and over relatively small spatial scales (between regions). This study also provides a useful framework for assessing the influence of other aspects of foraging behavior on the stability of food‐web dynamics.