Jacqui Frair

Survival and movement behaviour of resident and translocated wapiti (Cervus elaphus): Implications of their management in west-central Alberta, Canada

Picture1Much of our knowledge of wapiti ecology has been acquired from predator-free systems. Restoration of large carnivores and rapid landscape changes from industrial activities are changing the environmental conditions for this important ungulate species in western forests. The goals of this research were two-fold: 1) increase our understanding of the consequences of competing mortality risks and continued landscape changes on wapiti, and 2) put tools into the hands of managers trying to balance competing demands on the forest ecosystem. Because my research relied heavily on wapiti location data collected by global positioning system (GPS) collars, I first conducted a simulation study to test the effects of collar bias on habitat selection patterns and the efficacy of two bias-correction techniques – sample weighting and simulation via multiple imputation – both of which effectively removed the effects of bias from inferences of habitat selection. Between Jan 2000-Dec 2004,1 conducted a translocation study, releasing wapiti from 4 different source populations into the central Rocky Mountain foothills of Alberta, Canada and compared movements and survival to that of resident wapiti in the area. In so doing, I observed spatial and temporal trade-offs associated with the risk of mortality from aboriginal hunters and wolves in the area that have implications for managing the resident wapiti population as well as for improving translocation programs. Roads played a defining role in the survival patterns of wapiti in this system. Using a technique called first-passage time, I detected different scales of movement by resident wapiti that reflected patterns of landscape heterogeneity caused by timber harvest in the central foothills. Based on this information, I divided wapiti trajectories into behavioural “states”, which fluctuated diumally and responded to local variation in vegetation and terrain conditions, predation risk by wolves, and the proximity of roads and trails. Ultimately, I devised a state-based movement model for wapiti, using a correlated random walk to evaluate the combined effects of habitat amount, alternative road development schemes, and increasing road density on habitat effectiveness for wapiti in the central Rocky Mountain foothills of Alberta.