Predation risk and the functional response of elk–aspen herbivory

Hypotheses for trembling aspen (Populus tremuloides) decline in Rocky Mountain national parks include long-term high herbivory by elk (Cervus elaphus), decreased disturbance, climate change, and long-term low elk herbivory. Predation theory, as applied to herbivores, was used to describe alternative functional responses predicted by each hypothesis. Further, we postulated that elk behavioral patterns resulting from predation risk could also influence aspen herbivory. We tested predictions by sampling aspen density and browsing conditions in spatially separate watersheds in the Canadian Rockies, with three areas in each of three elk density levels (<2, 2–4, >4 elk/km2), and in each of four tree cover classes: recently disturbed (<25 years ago), and low, moderate, and high. Plot elk use was indexed with pellet group counts classed by habitat patch type, aspen condition (open or thicket), and different distances from trails or park boundaries used by predators, hunters, or park visitors. The herbivory rate on aspen saplings (>1 m height and <1 cm diameter at breast height, DBH) fitted the predictions of a Type 2 functional response (e.g. increased rate of herbivory with decreased sapling density). The Type 2 response appeared to result from low elk use in dense aspen thickets when predation risk is high. An interaction effect in the tree cover×elk density factorial analysis indicated that recent disturbances (<25 years) have incrementally increased the number of aspen saplings, but did not cause a major shift to an abundant sapling state in areas of high and moderate elk density. These results best supported the predictions of the low elk herbivory hypothesis. Anthropogenic burning combined with human and carnivore predation effects on elk (behavioral and numeric) appeared to be important in aspen’s long-term persistence, and its recent decline.