Jeffrey C. Lewis
Defense Presentation: 3:30-4:50pm on Monday, 12 May 2014, at the Wildlife Seminar (147 Architecture Hall)
Committee members: Drs. Steve West (chair), Keith Aubry, Kurt Jenkins, Toby Bradshaw, and Aaron Wirsing
Title: The post-release movements, survival, and landscape-scale resource selection of fishers (Pekania pennanti) reintroduced to Olympic National Park.
Abstract: The fisher (Pekania pennanti) is a mid-sized carnivore in the weasel family (Mustelidae) that occurred throughout the temperate and boreal forest ecosystems of North America. Because of the extremely high value of the fisher’s pelt in the 1800s and early 1900s, fisher populations were overexploited and extirpated throughout much of the southern portion of the historical fisher range (southern Canada and northern U.S). Translocations have been successful at restoring fishers throughout much of the vacant portions of the historical range, however little information from these translocations is available in the published literature. As part of the fisher recovery process in Washington state, we translocated 90 fishers (50 F, 40 M) from central British Columbia to the Olympic Peninsula. We radio-collared each fisher to allow investigations of post-release movements, survival and resource selection of a large sample of translocated fishers. Fishers moved extensively after their release and the extent of these movements was best explained by sex, and the month when movements occurred. Similarly sex was the most influential factor in explaining the distances between release sites and established home ranges of 48 fishers (27 F, 21 M) that established home ranges. Mean distance (±SE) from a release site to an established home range was 44.5 ± 6.4 km for males and 30.1 ± 3.6 km for females; however the mean number of days from release until home range establishment was similar for the sexes. Twenty-six of 27 females established home ranges from December to October of their first year and the distribution of establishment dates did not differ from uniform distribution (χ² = 7.00, df = 10, P = 0.725). Conversely, 19 of 21 males established home ranges between April and July of their first year and this distribution of establishment dates differed from a uniform distribution (χ² = 46.571, df = 10, P < 0.001). Twenty-one of the 27 females (78%) and 9 of 21 males (43%) established home ranges within the Olympic Fisher Recovery Area (i.e., Olympic National Park and Olympic National Forest). Mean home range sizes for males (128.3 km2) and females (63.5 km2) were among the largest reported for the species.
The survival of the founders was most strongly influenced by the year in which they were released (cohort), season of the year, sex, and age; duration of captivity, weight, release-date and number of canines did not influence survival. The importance of influential variables was best explained by the lower survival rates of fishers released in year 2 (cohort), the lower survival rates of fishers during the breeding season (season), the higher survival rates of males (sex) and the higher survival rates of juveniles (age). Point estimates for survival rates were highest for juvenile males (0.65-0.94; range of annual survival for the 3 cohorts), followed by adult males (0.69-0.91), juvenile females (0.46-0.89) and adult females (0.37-0.86); this finding is the opposite of what is commonly found in established fisher populations, where adult females often have the highest survival rates and juveniles have the lowest. Cause of mortality was determined for 24 of 35 recovered fishers; predation was the leading cause of mortality (40%) followed by vehicle strikes (20%), drowning (6%) and trapping-related (3%). Predation and vehicle strikes appeared to be more common as causes of mortality in our study than in most other studies of fisher mortality.
The selection of established home ranges differed between the sexes as well as the degree of selectivity. Most (79%) females established home ranges within the unmanaged forest landscapes of the Olympic Fisher Recovery Area. A resource selection model that included the amount of mid-seral forest and mean elevation was clearly the best among the 17 models that we evaluated, indicating substantial selection by females. Because most females established home ranges within unmanaged forest landscapes in Olympic National Park, they selected unmanaged mid-seral forests within landscapes dominated by unmanaged mid and late-seral forests. Most (56%) males established home ranges on managed forest landscapes and exhibited a selection for home ranges with smaller overstory trees and greater amounts of early seral forest than was available within the study area; however all 17 models of resources selection received some level support by the data, indicating weaker selection by males for the variables we included in our models. The amount of natural open space was the factor that best distinguished male and female use of home range core areas, with females using cores area with substantially less natural open areas than males.
This work represents the first investigations of survival and resource selection and the second investigation of post-release movements of a translocated fisher population, and it provides insights into the factors that influence translocation success. Our findings should be particularly useful for managers that are assessing the feasibility or developing an implementation plan for a fisher translocation.