Still underway, this is one of the few, tremendously important field studies of wolverines, located in southeastern British Columbia. Results are still preliminary, but indicate a low density population with a low reproductive rate, even relative to other wolverine populations. Key findings include the importance of protecting wolverines from trapping, the importance of remote, undisturbed, roadless areas for security, especially natal den sites, and the conservation of a healthy ungulate prey base.
The author was kind enough to email PCA a copy of this report, which greatly facilitated the posting of these excerpts.

Abstract (in its entirety)
"We are currently completing a multi-year project investigating demography and habitat use of a harvested population of wolverine in a 7000 km2 area within the Kootenay Region of British Columbia. The study area contains 10 registered traplines, two National Parks, a major transportation corridor, two hydroelectric reservoirs, active logging areas and supports a substantial helicopter skiing industry. Wolverine are considered "vulnerable" by the BC Wildlife Branch and consequently vital rate data are important for evaluating management/conservation options. A total of 39 (14F, 25M) adult and subadult wolverine have been captured over 3700 trap-nights using log box traps baited with available road-killed wildlife. Population estimates for the core 4000 km2 study area based on 4 years of live trapping data are 25.6 (95% CI: 15.6-55.3) and 24.0 (95% CI: 14.7-44.3) for 1996 and 1997 respectively. Six of 11 mortalities detected during the study to date have been human caused. Annual survival rate was estimated to be 0.77 (95% CI: 0.66 - 0.88) for all age and sex categories combined. Reproductive data are being gathered through radio-telemetry and follow-up ground investigations of breeding females after den abandonment. To date, 3 2-kit litters have been produced in 14 adult female reproductive seasons. Four of these juveniles have been captured and implanted with radio transmitters. Natal densites have been in the ESSF biogeoclimatic zone; associated with woody debris and/or large boulder talus in undeveloped drainages. Home ranges of males (x = 1005 km2) were significantly larger than females (x = 310 km2). At the study area scale, the distribution of wolverine use is highly clumped in four distinct utilization peaks. The two largest utilization peaks occur within Glacier and Mt. Revelstoke National Parks; disproportionate to their land base within the study area. The focus for the remainder of the project is on the monitoring of existing females and kits, identification and characterizing natal dens in spring 1999. A population census using motion sensitive cameras will be conducted March - April 1999. Final report and management recommendations will be completed in 2000/01.

Other Notable Excerpts
(if you're pressed for time, look for excerpts marked with an "!", including the management recommendations at the end)

Introductory Discussion (in its entirety)
"Wolverine have disappeared from almost half of their former range (Paquet and Hackman 1995). In Canada, the eastern wolverine is listed by COSEWIC as endangered and the western wolverine as vulnerable, yet wolverine are the least known of Canada's large carnivores and conservation research needs are extensive (Hummel 1990). An understanding of vital rates is fundamental to the evaluation of conservation options.
In southeastern BC and elsewhere in western Canada, persistence of wolverine and other wide-ranging carnivores depends on suitable habitat inside and outside protected areas. Pressures of human use in both protected and unprotected landscapes are important carnivore conservation issues, since potential mortality/fragmentation sources such as highways, railways as well as disturbance/displacement agents such as human recreation occur with increasing intensity within protected areas; and additional industrial (logging, hydroelectric generation, mining) and commercial (trapping, helicopter skiing) land use occur in surrounding lands (Paquet and Hackman 1995; Hummel 1990).
Since wolverine are so wide ranging (home ranges 48-2000 km2; Hornocker and Hash 1981; Magoun 1985; Whitman et al. 1986; Banci 1987; Hatler 1989), they present a problem of scale for managers and trappers alike. Wolverine do not exist in manageable numbers within individual traplines and therefore may be vulnerable to overharvest. In eastern Canada, increased human activity and development has resulted in declining wolverine populations and subsequently, listing of the species as "endangered" by COSEWIC (van Zyll de Jong 1975; Dauphine 1989). Western populations of wolverine may be heading in a similar direction if measures are not taken now to protect wolverine and wolverine habitat. The juxtaposition of National Parks, transportation corridors as well as resource extraction activities and commercial trapping within our study area provide an opportunity to examine how these land uses influence wolverine vital rates. Specific objectives are:
(1) Estimate vital (birth, death) rates for wolverine within the North Columbia Mountains, SE B.C.
(2) Identify landscape/habitat characteristics and human use activities which correlate with use by wolverine
(3) Develop a population model to estimate current and potential rates of increase of wolverine
(4) Produce biologically-based management recommendations for the conservation of wolverine in the NCM
This report presents the initial results from the first 45 months of field activity up to January 31, 1999. Data analysis is not yet complete, but preliminary results are presented." (2-4)

Results section

Mortality
"A total of 11 radio-collared animals have died during the study, 4 were commercially trapped (M207, M217, M246, F206), 1 was killed on the Trans-Canada Highway in GNP (M212), 1 was killed on the CP Rail line in GNP (F223), 2 died of natural causes (F204, F229) and 2 were killed by other predators (M230, M208). One male (M224) died as a result of injuries received from either a fall or being kicked by a moose. At least three additional wolverine which had lost their radio-collars have also been commercially trapped (M201, F219, M232) and 1 other ear-tagged animal was recovered by backcountry skiers in Mt. Revelstoke Park (M209). Using only the radio-telemetry located mortalities, annual survivorship for all collared wolverine up until January 1999 was 0.77 (95% CI: 0.66 - 0.88). Estimate for females only was similar (0.74, 95% CI: 0.55 - 0.93); kit survival was slightly lower (0.61, 95% CI: 0.21 - 1.0) but is based on a limited sample size" (9).

Reproduction
"Three litters of 2 kits each have been produced over 14 adult female reproductive seasons (Table 1). Limited data suggest reproductive rates varied greatly between years (Table 1). Females used dens during 5 seasons but either did not produce young that survived to weaning age or young were not detected. During the remaining 6 seasons females did not establish dens." (9)

Characteristics of Natal Dens
! "Four suspected and 3 confirmed natal den sites have been found between 1995 and 1998 (Table 2). All dens have been found within roadless, tributary valleys in the ESSFvc biogeoclimatic subzone under woody debris or a combination of woody debris and large boulders. Females occupied dens as early as late February and used them until mid May in some cases. Den sites were not re-used in subsequent years. Four of the seven den sites were located in National Parks." (11)

Home Range Size and Habitat Use
! "Home ranges (Table 3) of males (1005 km2) were significantly greater than females (311 km2; t1,18 = 2.85, p < 0.01). Subadult males had larger average home ranges than adult males (1611 km2 vs. 601 km2; t1,8 =2.18, p = 0.058), whereas subadult female home ranges did not differ significantly from adult females (274 km2 vs. 335 km2; t1,8 = 0.66, p = 0.27 ). Male home ranges appear to overlap those of 1 or more females and those of other males, while ranges of females are exclusive except for accompanying young of the year and in some cases non-breeding sub-adults. Home range boundaries were defined by geographic (eg. watercourses) and manmade features (eg. reservoir, highway). Three animals crossed the Trans-Canada highway at least 4 times; one was struck and killed." (12)
"Use of available habitat differed between males and females (G = 68.7, df =2, p<.005) and by season. In winter, males used ICH habitats proportionately greater than expected (Figure 3; G = 126.6, df = 2, p <0.001) whereas females used ESSF habitats more (Figure 3; G = 61.0, df = 2, p < 0.001). In summer, males spent more time in ESSF habitats while females used higher elevation AT habitats (Figure 3). The effects of human activities and land use have not been analysed." (13)

Distribution of Animal Use Within Study Area
"Between January 1995 and October 1998, 1256 radio telemetry locations were obtained from 43 individual animals radio collared/implanted... Four clear "peaks" in use (25% probability contour) are evident; the largest located in Glacier National Park, followed by Mt Revelstoke National Park, Bigmouth/Windy Creek (north) and Downie/Goldstream (centre). Overlay techniques clearly demonstrate that high use areas ("peaks") are found in protected areas in greater proportion than expected based on relative trapping effort and park areal extent. Protected areas comprise approximately 20% of the study area; include approximately 11% of the high trapping effort area (25% contour), but contain over 68% of the high use area (Figure 2 & 4)." (14-15)

Discussion section

"... Magoun (1985), suggested that annual survivorship need be 0.906 or greater for a hypothetical wolverine population in NW Alaska to be stationary or stable. Our estimate of 0.77, falls well below this threshold which may indicate a decline. Closer examination of the assumptions used in the Alaskan analysis will be required to determine if this is the case. In Idaho, Copeland (1996) documented a reproductive rate of 0.67 kits/female/year; Magoun (1985) in NW Alaska reported 0.69 kits/female/year. Our present estimate of 0.43 kits/female/year is lower but is based on a low sample size. Differences arise through strong year effects, presumably related to food availability. In 1997 we had 2 litters from 5 adult females wheras in 1998 only 1 litter was produced from 6 females. These rates are well below the averages reported using corpora lutea, placental scars and fetus counts from carcasses (Rausch and Pearson 1972; Liskop et al. 1981; Banci and Harestad 1988)." (16)

"Reproductive females established dens in areas with little or no human disturbance, below treeline under avalanche debris or large boulders. Although at the stand or patch scale, all dens located to date have been found in non-forested habitats similar to those reported by Magoun and Copeland (1998), our data differ when viewed at the landscape scale. Results clearly suggest that the upper elevation forested zone (ESSFvc) [Englemann Spruce-Subalpine Fir], not the alpine/parkland zone are most used for denning. Additional factors such as human activity, distribution of prey/carrion and presence of other predators likely affect the suitability of an area as denning habitat but have not yet been investigated." (17)

"... Since wolverine operate at the landscape scale, with home ranges averaging 300km2 for females and 1000km2 for males in our study area, habitat selection is unlikely to be strongly tied to stand or patch level attributes. Rather, habitat patch size, juxtaposition and prey density/distribution may be more important factors. Final habitat use analysis will evaluate these factors in a GIS environment." (18)

"Density estimates reported from published studies utilizing radio-telemetry and mean home range size tend to yield higher and more variable density estimates (1/65 km2, Hornocker and Hash (1981); 1/48 km2, Magoun (1985); 1/177 km2, Banci 1987; 1/198 km2, Copeland 1996) than those based on snow-tracking (1/207 km2, Quick 1953; 1/193 km2, Becker 1991). Whether these differences are related to the estimation techniques or actual differences in wolverine density is not certain. Our estimate (1/167 km2) using four years of live-trapping information falls within the range of those reported above." (18)

! "The uneven distribution of wolverine use across the study area (Figure 4) suggests that habitat quality is not equal across the area. The two largest "peaks" in use are centred on Glacier and Mt. Revelstoke National Parks. The Upper Bigmouth/Windy Creek "peak" area appears to support a high proportion of the wolverine population in the northern portion of the study area and may function as a refugia (Hatler 1989), similar to the national park areas... At present most of this area is a defacto wilderness with access only in Bigmouth Creek." (19)

Management Implications
! "Recommendations are preliminary at present pending completion of field data collection.
1) Human caused mortality of wolverine from trapping and transportation corridors is the largest factor influencing survivorship. Trapping restrictions may be warranted if rates are found to be unsustainable. Tracking harvest sex and age through compulsory inspection would assist management decisions. Carrion along the road and rail right-of-ways needs to be disposed of rapidly to avoid collateral kill of carnivores such as wolverine.
2) National Parks and unroaded wilderness areas appear to act as refugia at present. Pressures from commercial backcountry use, snowmobiling and logging may erode the capacity of these areas to support wolverine, particularly reproductive females.
3) Maintenance of an abundant, diverse ungulate community is a necessary precursor to persistence of wolverine. Habitat and harvest management strategies which maintain moose, mountain goat and caribou populations will benefit wolverine." (19-20)

---

| Home | Join Us | Clearinghouse Main Menu |

| Wolverine Clearinghouse | Wolverine Information |

Copyright © 2000 Predator Conservation Alliance. All Rights Reserved.

---

P.O Box 6733
Bozeman, MT 59771
Phone: (406) 587-3389 | Fax (406) 587-3170 | Email