Abstract
Though it is widely argued that antipredator responses carry nutritional costs, or risk effects, these costs are rarely measured in wild populations. To quantify risk effects in elk, a species that strongly responds to the presence of wolves, we noninvasively monitored diet selection and nutrient balance in wintering elk in the Upper Gallatin, Montana, USA, over three winters while quantifying the local presence of wolves at a fine spatiotemporal scale. Standard nutritional indices based on the botanical and chemical composition of 786 fecal samples, 606 snow urine samples, and 224 forage samples showed that elk were generally malnourished throughout winter. Increased selection for dietary nitrogen within forage types (e.g., grasses) led to ∼8% higher fecal nitrogen in the presence of wolves. However, urinary allantoin : creatinine and potassium : creatinine ratios decreased in the presence of wolves, suggesting large declines in energy intake, equal to 27% of maintenance requirements. Urinary nitrogen : creatinine ratios confirmed that deficiencies in nitrogen and/or energy were exacerbated in the presence of wolves, leading to increased endogenous protein catabolism. Overall, the nutritional effects of wolf presence may be of sufficient magnitude to reduce survival and reproduction in wintering elk. Nutritionally mediated risk effects may be important for understanding predator–prey dynamics in wild populations, but such effects could be masked as bottom‐up forces if antipredator responses are not considered.