Ecology and conservation of Canada lynx in the southern Canadian Rocky Mountains

The ecology of Canada lynx is expected to differ in key ways between the species' southwestern range and northern boreal regions, with important conservation implications. Differences are hypothesized to be due to lower and more stable densities of snowshoe hares, the primary prey of lynx, as well as ecological and human limitations to lynx habitat and population distribution. I explored these and other topics relevant to lynx conservation in the southern Canadian Rocky Mountains through a study of lynx nutritional, spatial, population, and habitat ecology. Within a 3,500 km2 field study area, I addressed the general hypothesis that lynx ecology in the southern Canadian Rocky Moutains of British Columbia and Alberta resembles that of boreal populations during the low phase of the approximate 10-year snowshoe hare population cycle. The diet, space use and demographics of four male and eight female resident lynx were consistent with this prediction during five years (1996--2001) largely coinciding with an assumed increase to high phase of a hare cycle. Lynx winter diet was diverse, with 44% of kills being species other than snowshoe hares (31% tree squirrels), but the hare:alternate prey ratio was related to prevalent habitat conditions. Hare densities closely resembled cyclic lows in the boreal forest and differed among forest seres and landscapes, while variation among years was equivocal. Annual home ranges of resident lynx were up to ten times larger than expected in boreal populations, while daily and exploratory movements were also relatively extensive. A long distance (498 km linear) dispersal of an adult male was documented, but four juvenile dispersals were much shorter (17 to 74 km) and mostly unsuccessful. Kitten recruitment to late-winter was 0.29 (n = 7) during the first three years and 0 (n = 6) during the last two years, while the largest litter documented in the study area was two. Adult survival was high (0.93) during the first three years but low (0.33) during the last two years. Survival among five juveniles monitored during the first three years was low (≤ 0.20). Most (69%) mortalities were due to starvation. Throughout a 76,000 km2 region of southeastern British Columbia, lynx distribution was uneven and explained by temperature, precipitation, forest disturbance regime, and terrain. These factors likely mediate lynx energetic requirements, prey density and distribution, and interspecific competition. Across two spatial scales, resident lynx within the field study area selected habitats on the basis of terrain and land cover features as well as forest overstorey composition and structure. An index of forest heterogeneity also contributed positively in predicting lynx habitat. Highways influenced lynx space-use and movements, although population impacts are not known. A high-volume highway caused some displacement of lynx, but some lynx selected habitats in proximity to moderate- and low-volume highways. Lynx crossed all highway types less than random expectation, but the effect of a high volume highway was notably stronger. I consider my results and those of other studies in discussing the potential influence of resource management policies and practices on lynx conservation at individual, population, and metapopulation levels.; Key words: British Columbia, Canada lynx, demographics, diet, distribution, habitat, highway, management, metapopulation, movement, population, predictive modeling, Rocky Mountains, snowshoe hare, space-use.