Noninvasive genetic survey of wolverines and their response to the Banff-Yoho transportation corridor

Annual Report

Year 1 - 2010-2011

Western Transportation Institute – Montana State University

Parks Canada Agency

Miistakis Institute for the Rockies

Suggested citation: Clevenger, AP, B Dorsey and N Heim. 2011. Noninvasive genetic survey of wolverines and their response to the Banff-Yoho transportation corridor. Annual report: Year 1, 2010-11. Cover photo: Wolverine in larch near Consolation Lakes, Banff National Park, Alberta (16 July 2011). Credit: Keith Webb.

Content

Executive summary Background Purpose and objectives Study design development, research coordination and logistics Study area Methods Results Future work Outreach and communications Partnership coordination and funding Acknowledgements Literature cited

Executive summary Background Wolverines (Gulo gulo) are rare, occur at low density, and move widely within very large home ranges. They are highly sensitive to human disturbance including transportation infrastructure. They are latently becoming recognized throughout Yellowstone to Yukon (Y2Y) region as genuine indicators of healthy, connected ecosystems. The Trans-Canada Highway (TCH) in the Canadian Rocky Mountains has long been recognized as a lethal barrier to wildlife and an acute fracture zone for population connectivity in the ecoregion. More than 25 years mitigating the TCH in Banff National Park has essentially restored habitat connectivity across mitigated sections of this major transportation corridor. However, the combined effects of an additional 30 kilometers of twinned TCH with more lanes of traffic in Banff National Park threaten to fragment and isolate populations of wide-ranging, fragmentation-sensitive species, such as wolverines and lynx. The scale and magnitude of the TCH ìPhase 3Bî Twinning Project and measures designed to mitigate environmental impacts are unprecedented for Parks Canada. Numerous wildlife-crossing structures along Phase 3B are designed to mitigate the impacts by enhancing connectivity and link habitats of key wildlife species over time. Purpose and objectives The TCH Phase 3B twinning is the first highway mitigation project within the range of wolverines in North America. Our objective is to collect baseline information on wolverine occurrence and assess the potential effects of a major east west transportation corridor on wolverine movements and gene flow in the Canadian Rocky Mountains. Using two noninvasive methods (genetic sampling and remote cameras) our main objectives are to estimate population size, model habitat occupancy, and assess fine-scale genetic structure across the TCH and other potential barriers. From 2010-2014, our research will address for the first time the crossing structure design requirements of wolverines in western North America in addition to providing the first information on wolverine occupancy and habitat relationships in Banff, Yoho and Kootenay National Parks. Study area The survey area encompasses approximately 6000 km2 and is located in parts of Banff, Yoho and Kootenay National Parks and the Columbia Valley near Golden, British Columbia (Figure 1). The study area was overlaid with a 12 x 12 km grid; approximately 50 grid cells covered the study area. Noninvasive hair traps were used to ìsampleî wolverines in the population. One sampling location or hair trap was located in each grid cell. However, to increase probability of detecting wolverines and cross-highway movements within the TCH corridor we placed an additional sampling site in select grid cells overlaying the highway. Methods We used two methods of detecting wolverine occurrence in the survey area: 1) Hair traps consisted of a whole skinned beaver carcass nailed to a tree and secured with baling wire. Barbed wire was wrapped from the carcass down to ground (snow) level. Wolverines

climbed the tree several times before removing the carcass, and in doing so left hair on the barbs. 2) Remote infrared-operated cameras were placed at each hair trap site to photo-document wolverine visitation and behaviour. Hair traps were checked three times at monthly intervals after they were set up. Hair traps were removed after the third and final sampling session (mid-March to mid-April). Results A total of 48 hair trap sites were established in the study area. Each hair trap site was visited four times (set-up plus three sampling sessions). The visitation rate to the hair trap sites increased during the three sampling sessions. Session 1: 38% (18 of 47 sites) were visited by wolverines. Session 2: 71% of the sites (34 of 48 sites) were visited.Session 3: 79% (37 of 47 sites) of the sites were visited. Overall, 85% of the sites were visited during at least one session. Of 142 sampling opportunities during the three sessions, wolverines visited the sampling sites 89 times (63%). Seven sites were not visited by wolverines during the survey. Two wolverines were photographed at the same site at the same time on four occasions. Many non-target species photographed at the hair traps. More than 900 hair samples were collected during the three sampling sessions. Future work Summer-Fall 2011: Analysis of hair samples: Expected completion November 2011. Report writing: Preparation of summary report and short manuscript. Winter 2011-12: Addressing the data gaps - A limited amount of survey work will take place, focusing on resampling areas that were under-sampled or were not sampled in 2010-11. Experiment with alternative hair capture techniques. Feasibility of remote cameras to document locations of reproductive females ñ Test feasibility of using cameras as method to document occurrence of reproductive females in study area. Winter 2012-13: Repeat winter 2010-11 survey. Outreach and communications Presentations: 5 public and professional presentations on the wolverine research were delivered. Wolverine Watch: More than 50 people registered to assist with the noninvasive surveys. Media: The wolverine research garnered attention from 7 different media during the course of the year (6 newspaper articles, 1 radio interview). Highway Wilding Webisodes: Filmmaker Leanne Allison created several short videos about road ecology, the mitigations for wildlife on the TCH, and the wolverine research.

Background

Wolverines (Gulo gulo) are rare, occur at low density, and move widely within very large home ranges (Banci and Harestad 1990, Lofroth and Krebs 2007). They inhabit alpine and subalpine zones throughout the Rocky Mountain cordillera. Their populations have experienced considerable range reduction over the last 50 years (COSEWIC 2003, Laliberte and Ripple 2004). Loss of habitat and barriers to movement along with a warming climate are recognized as threats that further diminish and fragment the critical landscapes wolverine utilize and need to disperse through (Krebs et al. 2004, Schwartz et al. 2009, Copeland et al. 2010). Presently little is known about the status of wolverines in the national parks of the Canadian Rocky Mountains (Suitor 2005). Federally, the wolverine is listed as a species of Special Concern, while in Alberta it is listed as Data Deficient (Alberta Fish and Wildlife 2008). Thus, monitoring populations of wide-ranging species, such as wolverine, has been identified as a critical management objective in Banff and Yoho National Parks (Parks Canada 1997, 2007). Expanding highways and increasing vehicle traffic have been identified as one of the most severe human-caused impacts in the entire Yellowstone to Yukon region (Weaver et al. 1996). The Trans-Canada Highway (TCH) in the Canadian Rocky Mountains has long been recognized as a lethal barrier to wildlife and an acute fracture zone for population connectivity in the ecoregion (Banff Bow Valley Study 1996). More than 25 years mitigating the TCH in Banff National Park has essentially restored habitat connectivity

across mitigated sections of this major transportation corridor (Ford et al. 2009, Sawaya et al. in prep). However, the combined effects of an additional 30 kilometers of twinned TCH with more lanes of traffic in Banff National Park threaten to fragment and isolate populations of wide-ranging, fragmentation-sensitive species (Golder Associates 2004). The scale and magnitude of the TCH ìPhase 3Bî Twinning Project and measures are unprecedented for Parks Canada. Numerous wildlife-crossing structures along Phase 3B are designed to mitigate the impacts by enhancing connectivity and link habitats of key wildlife species over time. Currently there is little information on the distribution and life requisites of wolverines in Banff, Yoho and Kootenay National Parks (Suitor 2005). Recent research to the north in central Alberta suggests that the abovementioned national parks may be a source population for unprotected areas in British Columbia (BC) and Alberta (Fisher et al. 2009). With regards to highways, a localized study in the Kicking Horse Pass (Yoho NP) found that wolverines tended to avoid crossing the TCH (Austin 1998). Other studies have anecdotally shown that wolverines are averse to crossing two-lane highways (Copeland 1996, Krebs and Lewis 2000). It is unclear what impact the TCH has had on wolverine dispersal and population genetics until now or how they will respond to a fenced highway with engineered wildlife crossing structures. The Phase 3B section of TCH is roughly 30 km from Castle Jct to the BC border (Kicking Horse Pass) and is entirely within Banff National Park. Presently Phase 3B is largely completed with new

fencing and crossing structures, save for a 15 km section west of Lake Louise scheduled for completion in 2013. The TCH west of Phase 3B in Yoho National Park is still unfenced and two lanes, however, highway twinning is expected within the next 5-10 years. Parks Canada will need information that identifies the most suitable locations for crossing structures for this section of TCH. The current mitigated-unmitigated highway design presents an opportunity to obtain pre-mitigation baseline data for the 50 km stretch of TCH in Yoho National Park prior to twinning. Because of their wide-ranging movements and low-density populations, historically gather information on wolverine distribution and abundance has proved challenging. Currently DNA-based techniques enable the identification of species, individuals, their genders, and genetic relatedness from hair samples collected through non-invasive genetic sampling methods (Foran et al. 1997, Gompper et al. 2006). These techniques enable biologists to monitor population trends, estimate population size, model habitat occupancy, and assess population viability (Schwartz et al. 2007, Long et al. 2008). Further, noninvasive genetic data can be used to understand dispersal-related gene flow by modeling the effects of different landscape features (Cushman et al. 2006, Garroway et al. 2011).

Purpose and objectives Our objective is to collect baseline information on wolverine occurrence and assess the potential effects of a major east-west transportation corridor on wolverine movements and gene flow in the Canadian o noninvasive methods (genetic sampling and remote cameras) our main objectives are to estimate

population size, model occupancy and assess fine-scale genetic structure across the TCH and other potential barriers. (2) With data obtained from the regional-scale sampling, we will model gene flow and identify potential dispersal corridors across the unmitigated TCH in Yoho National Park. (3) Last, we plan to investigate causes of the observed pattern of genetic structure in the population by linking behaviour and movement of individual wolverines to population level patterns of gene flow by snowtracking within the TCH corridor. From 2010-2014, our research will address for the first time the crossing structure design requirements of wolverines in western North America in addition to providing the first information on wolverine occupancy and habitat relationships in Banff, Yoho and Kootenay National Parks. Staff Three full-time employees worked on the wolverine research project in winter 2010-11: Project leader Tony Clevenger and research assistants Ben Dorsey and Nikki Heim. Research assistants Barb Bertch, Jen Reimer and Alex Taylor worked intermittently and shared one full-time position. Jim Zettel, Cathy Gill and Andrea Kortello volunteered their time regularly to the project.

Study design development, research coordination and

logistics The research and monitoring plan addresses how roads (TCH, Hwy 93) and other man-made and natural barriers

influence wolverine movement and gene flow. Preparatory work consisted of formalizing the methodology for the project, purchasing equipment and supplies to conduct the fieldwork, and securing internal support from Parks Canada field unit staff (public safety, trail maintenance, resource conservation and volunteer programs). Researchers were consulted on wolverine survey study design, methods and analysis: Dr M. Schwartz (Conservation Genetics Team Leader and Research Ecologist at the U.S. Forest Service Rocky Mountain Research Station, Missoula, MT), Jason Fisher (Mammal Ecology Research Scientist, Alberta Innovates, Victoria, BC), Garth Mowat (Carnivore biologist, Ministry of Environment, Nelson, BC), and Kyran Kunkel (World Wildlife Fund, Missoula, MT). Areas where potential hair sampling sites would be placed were visited and GPS’ed within the survey area. Equipment and supplies required for the winter surveys were purchased. Beaver carcasses were obtained on a regular basis by staff from Animal Damage Control and provincial Conservation Officers in Canmore and Ponoka County. Protocols for data collection, backcountry travel etc. were developed in consultation with Lake Louise, Yoho, Kootenay (LLYK) and Banff Field Unit’s wildlife and public safety staff. Avalanche safety training: Field staff took part on a weeklong Avalanche Safety Training Course (Level 1 & 2) in early December taught by ACMG ski guide Felix Camire (El Gato Adventures, Canmore, AB). A daylong meeting with Banff and LLYK Field Unit staff took place in mid-October at the Banff Warden Office. The project objectives and methods were

presented to Parks Canada staff. We requested input from Resource Conservation staff with regard to proposed hair trap locations; with public safety staff regarding avalanche safety in accessing the proposed sampling sites; regular communications with public safety during the winter survey; communications with Banff dispatch; training requirements for volunteers assisting with the survey and internal communications.

Study area

The survey area was delineated by creating a 30 km buffer around the TCH from Castle Junction (Banff NP) to the west boundary of Yoho NP. The area encompasses approximately 6000 km2 and was overlaid with a 12 x 12 km grid. The grid network allowed for a stratified, systematic sampling approach. The 12 x 12 km grid cell size is slightly larger than the standard 10 x 10 km grid used for most surveys to monitor wolverine populations (Koen et al. 2008). The slightly larger cell size was chosen to increase the sampling area. Since a substantial part of the study area consists of rock and ice we felt the increased cell size would not likely reduce our sampling success. Noninvasive hair traps (see below) were used to sample wolverines in the population. One sampling location or hair trap was identified in each grid cell. However, to increase probability of detection and movements within the TCH corridor an additional sampling site was placed in select grid cells that overlaid the highway. The location of potential sampling sites was determined by situating them along areas either known or presumed to be travel corridors for wolverines; typically valley bottoms leading to subalpine areas

and high alpine passes. Field data from wolverine observations and snowtracking (Bertch 2003) were used to guide preliminary site selection. More than one potential sampling site per cell was identified if there was uncertainty regarding the suitability of the selected site. Criteria for site selection consisted of forest stand density, terrain and forest age. We strived to place sampling sites in mature forest with low stand density (good visibility), and in relatively level

terrain. Initially we identified potential sampling sites using topographic maps in a GIS. Once we had located at least one site for each cell, we then reviewed each site using aerial orthophotos in GIS to make local-scale refinements to the sampling site locations. Distancing criteria were used to locate hair-sampling sites. Sampling sites in adjacent grid cells were > 6 km apart. We tried to minimize human visitation or disturbance to the sampling sites by

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?!?

!?

!?

!?!?

!?

Numa-Vermillion

Windy

Wapta-Southside

Flints

Taylor Ck

Spilly

Stoney

Tokumm

Cuthead

Hoodoos

Redearth-Massive

Simpson R

Brewster

Redearth

Paradise

Bow Lake

Bow River

Otterhead

Nicholson

BC-West Gate

PipestoneHidden Jct

SherbrookeBath Creek

Forty-Mile

Hawk Ck

Castle camp

Baker Lower

Mosquito Ck

Healy Ck

Porcupine Ck

Castle N

Emerald Lake

Moraine Lake

Amiskwi Lower

McArthur Pass

Beaverfoot-Ice

Hospital Ck

Johnson-Mystic

Ottertail Lower

Parson

Assiniboine slides

Cyclone

Fish Lake

Baker Upper

Tak Falls

Amiskwi Upper

Ottertail Upper

??

Wolverine Baited Hair Trap Sites - Winter 2010-11

Golden

Lake Louise

Banff

!? Baited hair trap sites

TCH Castle to Yoho West Gate

National Parks

30 km TCH Buffer- 7138 sq. km.

12 km grid 1:450,000 E

Map of study area including 30 km buffer around the Trans-Canada Highway, 12 x 12 km grid, and 48 hair trap sites sampled during winter 2010-11 in Banff, Yoho, Kootenay National Parks and province of British Columbia. placing them >500 m from heavily used trails or backcountry sites, and >200 m from any other trails. We placed hair sampling sites >2 km from highways in order to avoid risk of attracting wolverines to sites on the opposite side of

a highway. Parks Canada Public Safety required that all hair-sampling sites be outside avalanche terrain exposure scale (ATES) Class 2 (challenging) or 3 (complex) areas. Access to any sampling sites that required travel through or near

Class 2 or 3 areas had to be approved by the Mountain Park Visitor Safety manager. Final site determination of all samping sites was made through a day-long meeting with Banff and LLYK Field Unit public safety and resource conservation staff in mid-October (see above).

Methods We used two methods of detecting wolverine occurrence in the survey area: 1) Hair traps consisted of a whole skinned beaver carcass nailed to a tree and secured with baling wire (Fisher et al. 2009). Barbed wire was wrapped from the carcass down to ground (snow) level. Wolverines climb the tree several times before removing the carcass, and in doing so leave hair on the barbs. For some hair trap sites we stored beaver carcasses in lockable 45 gallon drum barrels. This was done for the more remote sites and helped minimize the amount of weight to carry traveling to the site.

Wolverine hair traps consisted of a hung beaver carcass and barbed wire wrapped below to the base of tree.

2) Remote infrared-operated camera (Reconyx Silent Image and PC85) were placed at each hair trap site to photo-document wolverine visitation and behaviour. Cameras were equipped with Lithium AA batteries and a compact flash card for data storage. We used a hand-held PDA to make sure the camera was aligned with the hair trap. To attract wolverines to the hair trap site a commercial trapping lure (Craven’s Gusto or O’Gorman’s LDC) was liberally smeared on a cloth and hung high in the air to disperse the lure in the area. Nearly all hair trap sites were accessed on skis. However, monthly Alberta Snow Survey flights provided helicopter access to two remote hair trap sites (Fish Lakes, Upper Baker) and snowmobiles were used on a few occasions in conjunction with cross-country ski tracksetting work (Moraine, Pipestone) and Public Safety snow survey work (Tak Falls).

Lockable 45 gallon drums were used to cache beaver carcasses near the most remote hair sampling sites.

Hair traps were checked three times at monthly intervals after they were set up. During each monthly check we rebaited the traps with a fresh beaver carcass and hung a fresh lure rag at the site. If the previous beaver carcass was not fully consumed, we nailed a fresh beaver carcass next to the old one. Hair traps were removed after the third and final sampling session (mid-March to mid-April).

Remote cameras were installed at each hair sampling site.

Results A total of 48 hair trap sites were established in the study area. Hair trap sites were set up throughout the study area during the first month of the survey starting 14 December 2010 and finishing up in mid-January 2011. One site (Amiskwi Upper) was not set up during the initial set up period, but one month later and therefore sampled only twice. Similarly, the Windy only sampled twice as it was taken down after the second sampling session. Sampling sessions 1 and 3 were comprised of 47 hair trap

sites, while sampling session 2 consisted of 48 hair trap sites. The total distance (round trip) to the hair trap sites in Banff, Yoho and Kootenay Parks was 499.2 kms. The roundtrip distance to the five sites located outside the national parks was 45 kms. Each hair trap site was visited four times (set-up plus three sampling sessions), thus the total distance traveled by research staff was over 2000 kms during the four-month period.

Fieldwork was conducted in teams of at least two people and nearly all sampling sites were accessed by skis. Of the 48 sites, 40 (83%) required travel in Class 1 avalanche terrain, six sites (13%) in Class 2, and two sites (4%) were accessed by travel in Class 3 terrain. Nearly all sites were located in Class 1 avalanche terrain; only one (Macarthur Pass) was in Class 2 terrain. The visitation rate to the hair trap sites increased during the three sampling sessions, similar to what was reported by Fisher et al. (2009) in the Willmore Wilderness Area. Session one had a percent visitation rate of 38% (18 of 47 sites), 71% of the sites (34 of 48 sites) were visited in session two, while 79% (37 of 47 sites) of the sites were visited by

wolverines during the third and final session.

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?!?

!?

!?

!?!?

!?

!(

!(

!(

!(

!(

!(

!(!(!(

!(

!(

!(!(

!(

!(

!(

!(

!(

Numa-Vermillion

Windy

Wapta-Southside

Flints

Taylor Ck

Spilly

Stoney

Tokumm

Cuthead

Hoodoos

Redearth-Massive

Simpson R

Brewster

Redearth

Paradise

Bow Lake

Bow River

Otterhead

Nicholson

BC-West Gate

PipestoneHidden Jct

SherbrookeBath Ck

Forty-Mile

Hawk Ck

Castle camp

Baker Lower

Mosquito Ck

Healy Ck

Porcupine Ck

Castle N

Emerald Lake

Moraine Lake

Amiskwi Lower

McArthur Pass

Beaverfoot-Ice

Hospital Ck

Johnson-Mystic

Ottertail Lower

Parson

Assiniboine slides

Cyclone

Fish Lake

Baker Upper

Tak Falls

Amiskwi Upper

Ottertail Upper

??

Wolverine Baited Hair Trap Sites - Winter 2010-11 - Session 1

Golden

Lake Louise

Banff

!( Session 1-wolverine visit (n=18)

!? Baited hair trap sites

TCH Castle to Yoho West Gate

National Parks

30 km TCH Buffer- 7138 sq. km.

12 km grid 1:450,000 E

Results of session 1 and hair trap sites visited by wolverines.

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?!?

!?

!?

!?!?

!?

!(

!(

!(

!(

!(

!(

!(

!(

!(

!(

!(

!(

!(

!(

!(

!(

!(!(

!(

!(

!(

!(

!(

!(

!(

!(

!(

!(

!(

!(

!(

!(

Numa-Vermillion

Windy

Wapta-Southside

Flints

Taylor Ck

Spilly

Stoney

Tokumm

Cuthead

Hoodoos

Redearth-Massive

Simpson R

Brewster

Redearth

Paradise

Bow Lake

Bow River

Otterhead

Nicholson

BC-West Gate

PipestoneHidden Jct

SherbrookeBath Creek

Forty-Mile

Hawk Ck

Castle camp

Baker Lower

Mosquito Ck

Healy Ck

Porcupine Ck

Castle N

Emerald Lake

Moraine Lake

Amiskwi Lower

McArthur Pass

Beaverfoot-Ice

Hospital Ck

Johnson-Mystic

Ottertail Lower

Parson

Assiniboine slides

Cyclone

Fish Lake

Baker Upper

Tak Falls

Amiskwi Upper

Ottertail Upper

??

Wolverine Baited Hair Trap Sites - Winter 2010-11 - Session 2

Golden

Lake Louise

Banff

!( Session 2 -wolverine visit (n=33)

!? Baited hair trap sites

TCH Castle to Yoho West Gate

National Parks

30 km TCH Buffer- 7138 sq. km.

12 km grid 1:450,000 E

Results of session 2 and hair trap sites visited by wolverines.

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?

!?!?

!?

!?

!?!?

!?

!(

!(

!(

!(

!(

!(

!(

!(

!(

!(

!(

!(

!(

!(

!(

!(

!(

!(!(

!(

!(

!(

!(

!(

!(

!(!(

!(

!(

!(

!(

!(

!(

!(

!(

!(

Numa-Vermillion

Windy

Wapta-Southside

Flints

Taylor Ck

Spilly

Stoney

Tokumm

Cuthead

Hoodoos

Redearth-Massive

Simpson R

Brewster

Redearth

Paradise

Bow Lake

Bow River

Otterhead

Nicholson

BC-West Gate

PipestoneHidden Jct

SherbrookeBath Creek

Forty-Mile

Hawk Ck

Castle camp

Baker Lower

Mosquito Ck

Healy Ck

Porcupine Ck

Castle N

Emerald Lake

Moraine Lake

Amiskwi Lower

McArthur Pass

Beaverfoot-Ice

Hospital Creek

Johnson-Mystic

Ottertail Lower

Parson

Assiniboine slides

Cyclone

Fish Lake

Baker Upper

Tak Falls

Amiskwi Upper

Ottertail Upper

??

Wolverine Baited Hair Trap Sites - Winter 2010-11 - Session 3

Golden

Lake Louise

Banff

!( Session 3-wolverine visit (n=37)

!? Baited hair trap sites

TCH Castle to Yoho West Gate

National Parks

30 km TCH Buffer- 7138 sq. km.

12 km grid 1:450,000 E

Results of session 3 and hair trap sites visited by wolverines.

Overall, 85% of the sites were visited during at least one session. Of 142 sampling opportunities during the three sessions, wolverines visited the sampling sites 89 times (63%). Seven sites were not visited by wolverines during the survey; four in the national parks (Baker Lower, Hoodoos, Porcupine Cr, Amiskwi Upper [sampled only twice]) and three outside the national park (Nicholson, BC West Gate, Spilly). Within the national park boundary, 90% of the sites were visited by wolverines. Of the 41 sites visited by wolverines, 7 (17%) were visited only one time, 20 (49%) were visited twice, and 14 (34%) sites were visited all three sampling sessions.

Jeremy Guth and Peter Poole inspecting the Upper Amiskwi hair trap In reviewing the photographs obtained from the hair trap sites we were able to summarize some of the incidental data collected during the survey. Two wolverines were photographed at the same site at the same time on four occasions (5% of wolverine visits). Several non-target species were photographed at the hair traps during the survey:

• American marten: n=130 of 142 sampling opportunities; 92%.

• Red fox: n=8; 6% • Coyote: n=10; 7% • Wolf: n=11; 8% • Lynx: n=15; 11% • Cougar: n=1; 1% • Moose: n=2; 1% • White-tailed deer: n=3; 2% • Snowshoe hare: n=52; 37% • Striped skunk: n=2; 1% • Golden eagle: n=1/1%

Two wolverines were photographed at the same site at the same time on several occasions.

Many non-target species were photographed at the hair traps during winter 2010-11, including lynx, wolves, cougars. More than 900 hair samples were collected during the three sampling sessions. Not all samples were from wolverines since we collected all hair samples found at the hair traps. Although not quantified in this report, hair samples were obtained at nearly all sites with

confirmed visits by wolverines. The amount and quality of hair samples collected at each site will ultimately determine whether there is extractable DNA from the samples and the ability to genotype individuals in the study area.

Future work Summer-Fall 2011: Analysis of hair samples - Hair samples are currently being analysed at the US Forest Service Conservation Genetics lab in Missoula, Montana. Analysis is expected to be completed by December 2011.

More than 900 hair samples were collected during three sampling sessions during winter 2010-11. Report writing - The results from the winter 2010-11 survey will be summarized in a stand-alone technical

report and a short manuscript will be prepared describing occupancy and behaviour of wolverines at noninvasive sampling sites in the Canadian Rocky Mountains. Winter 2011-12: Addressing the data gaps – A limited amount of noninvasive genetic survey work will be conducted this winter. During this period, attempts will be made to try and sample areas that were under-sampled or were not sampled at all (Moose Creek, Glenogle, Upper Ice) during the previous winter. As part of coordinated non-invasive genetic sampling of wolverines in neighbouring Kananaskis Country we plan to sample a small number of grid cells on the eastern boundary to fill sampling gaps and create a much larger and seamless study grid. We hope to coordinate some of the above sampling with the monthly Alberta Snow Survey flights. Experiment with alternative hair capture techniques – We would like to investigate improvements in hair capturing techniques. The barbed wire technique did not consistently capture hair from wolverines. Other barbed wire configurations (loose wrap vs. tight wrap), varying tree circumference for barbed wire, and capture methods such as gun brushes, may provide better hair capture success. We plan to test alternative techniques by setting up pilot or test hair traps in areas with relatively easy access and with high probability of attracting wolverine visits, e.g., sites visited by wolverines during three sessions in winter 2010-11. Feasibility of remote cameras to document locations of reproductive females. - A key component of viable wolverine populations is the survival of

reproductive females (Persson 2003). Magoun et al (2011) recently developed an integrated system for photographing a wolverine’s ventral pattern while concurrently collecting hair for microsatellite DNA genotyping. In their pilot study they were able to identify three lactating females using the cameras during March–May. Piloting this technique during winter 2011-12, will allow us to evaluate the feasibility of using cameras as a method for documenting the proportion of reproductive females in the study area. The pilot work would take place at the last sampling visit of the sampling season (mid-March to mid-April), i.e., the months of year when breeding females would be lactating if they had kits. Instead of removing the entire hair-sampling set-up, at each site we would install a run pole, support structure, bait and position a camera opposite the run pole (see Magoun et al. 2011). Camera batteries would be replaced with new batteries. The site would be checked and taken down after 4-6 weeks. This first year of fieldwork we found there was a high rate of repeat visits to successful hair trap sites. Thus we believe that if lactating females are visiting hair traps in our study area, there is a high likelihood that they would continue to visit the sites once the system is in place at the end of winter. Magoun et al. (2011) encouraged others to make modifications of the system to improve the design and to suit different habitat types (e.g., tundra vs. forest) and different wolverine populations where behavior of wolverines might differ (protected vs non-protected areas). We view this effort as a valuable opportunity to document a minimum estimate of the number and locations of breeding females in the study area by using a

relatively simple method and capitalizing on established sampling sites. Winter 2012-13: Repeat winter 2010-11 survey – The 2010-11 survey will be repeated throughout the same study area in order to increase sampling of missed individuals during winter 2010-11, and new individuals recruited into the population during the previous two years. Given the results from the analysis of the 2010-11 samples, we will be able to identify and prioritize areas where more sampling effort is needed. These will be areas where wolverines are known to regularly occur but DNA extraction from previously collected hair samples was not possible, or areas within the study area that weren’t sampled. Documentation of reproductively active females will be conducted depending on the results of the pilot study in winter 2011-12.

Outreach and communications

Presentations

Public and professional Lake O’Hara Trails Club. 4 August and 1 September 2010. “Wolverines: Shrinking spaces for a wilderness wanderer”. Le Relais Shelter, Lake O’Hara, Yoho National Park. (T Clevenger) Alpine Club of Canada, Rocky Mountain Chapter. 5 September 2010. “ “. Monthly meeting of chapter. Canmore Seniors Centre, Canmore, AB. (T Clevenger) Wolverine Watch (WW) Orientation. 4 March 2011. Presentation to WW volunteers. Banff Seniors Centre, Banff, AB. (T Clevenger)

Bow Valley Naturalists (BVN). 25 April 2011. Short presentation to BVN summarizing results of first years survey. Banff Senior’s Centre, Banff, AB. (T Clevenger) Cross-boundary Workshop to Identify and Prioritize Key Information Needs for Wolverine. 7 June 2011. “Noninvasive survey of wolverines and their response to Banff-Yoho transportation corridor”. Presentation to workshop. Provincial Bldg., Blairmore, AB. (T Clevenger)

Wolverine Watch This citizen science-based program to assist wolverine research began with seed funding from the Alpine Club of Canada (ACC) to T Clevenger, B Dorsey and T Hewitt. The aim of Wolverine Watch (WW) is to add information on wolverine occurrence in the Canadian Rockies, increase awareness regarding their tenuous status, and recruit people interested in assisting us with the noninvasive genetic survey work described above. During the course of the winter 2010-11, more than 50 people registered to assist with the noninvasive surveys. Most volunteers were from Banff and Canmore, however, some from as far away as Calgary, Alberta and Golden and Invermere, B.C. Volunteers had a range of skills and backcountry winter travel experience. Parks Canada initially agreed to have the WW volunteers sign on as part of the Banff National Park Volunteer Program, but in early December the agency realized that this would violate the Parks Canada Labour Code. Thus, the project was unable to use volunteers until we had our own liability insurance.

WolverineWatch website Delays using the volunteers to assist on the project lasted until March 2011 when the project obtained insurance from the State of Montana. Soon after, we contacted the volunteers registered in December to determine their availability and interest in participating on the project. Roughly 30 or more volunteers expressed interest in assisting the project. An orientation session was held in Banff for those volunteers and soon thereafter, they began assisting with the hair traps. A dozen or more volunteers eventually assisted with the fieldwork during March and April. The interest in participating in WW was exceptionally high and the citizen science-based project will continue to collect information on wolverine observations and use volunteers in the future.

Volunteer Recruitment Volunteers were actively recruited for assisting with the research through a variety of means, including public presentations (ACC, Le Relais) and posters distributed to all the backcountry lodges and ACC huts in the study area. A recruitment postcard and poster also were created, and distributed through numerous outlets including: U of C Outdoor Program Centre, Mountain Equipment Co-op, Out There, Kananaskis Country Visitor Centres – Barrier Lake

and Peter Lougheed, Banff Mountain Film Festival, Gear-Up Sports, Switching Gears, Banff Mountain Magic and Vertical Addiction. Volunteer recruitment notices were distributed via email to numerous outdoor clubs (including nordic, backcountry, snowshoeing and ice climbing enthusiasts), naturalist clubs and other interested contacts.

Media

Rocky Mountain Outlook (Banff-Bow Valley) – 2 September 2010. “Presentation features tough, elusive wolverine”. Article about upcoming presentation at the ACC monthly meeting in Canmore. Banff Crag and Canyon – 14 September 2010. “The rare and elusive wolverine”. An article on the wolverine research and Wolverine Watch appeared in Banff’s Crag and Canyon and the Canmore Leader after the talk at the monthly ACC meeting. Rocky Mountain Outlook (Banff-Bow Valley) – 16 December 2010. “Wolverine Watchers wanted”. Article about upcoming wolverine research and Wolverine Watch citizen science program. Rocky Mountain Outlook (Banff-Bow Valley) – 23 December 2010. “Parks Canada not involved with new Wolverine Watch program”. Letter to editor from T Clevenger and D Duke correcting 16 December 2010 article (above) that Wolverine Watch is not a Parks Canada program or supported by Parks Canada. CBC Radio – 23 December 2010. T Clevenger was interviewed by Jim Brown on CBC’s Eye-opener radio program about the wolverine research and

Wolverine Watch. Calgary Herald and the Vancouver Sun – 26 April 2010. “Biologists gather genetic data on elusive wolverines”. An article on the wolverine research appeared in the Calgary Herald and the Vancouver Sun on April 26 and 27, 2011. The Crown of the Continent E-magazine featured a story on the project in their spring 2011 edition. Rocky Mountain Outlook (Banff-Bow Valley) – 25 May 2011. “Wolverine study yields valuable fur and facts” (Lynn Martell).

Highway Wilding Webisodes We have partnered with Leanne Allison, an award-winning filmmaker, to create several short videos about road ecology, the mitigations for wildlife on HWY 1, and the wolverine research. These “webisodes” will be hosted on our Highway Wilding website (highwaywilding.org), in addition to on the Highway Wilding channel on YouTube.com.

Webisode on You Tube Please visit: http://www.youtube.com/user/highwaywilding to view and share five short videos about the research project. The videos discuss the need for this research, give you a real sense as to what the field work

entails and showcases some of the animals that visited our sites this winter. Also, please continue to check out our website at www.wolverinewatch.org.

Partnership coordination and

funding The larger partnership enables the investment of each partner to be leveraged. As well, it facilitates the creation of other projects deemed important and relevant by the Steering Committee to be created. The role of maintaining the existing partnership, as well as being vigilant for worthwhile opportunities to expand the partnership, is shared by the Western Transportation Institute (WTI) and the Miistakis Institute (MIR).

Fundraising The primary sources of funding for the Project include Parks Canada Agency and the Western Transportation Institute, who have each committed significant multi-year support. Beyond that, a number of philanthropic foundations have made year-to-year contributions (as required by their structure), and a number have been, and will be, courted on an on-going basis. The committed funding is largely directed at the research and monitoring activity, leaving the communications and outreach activities as the main area in need of foundation support. In 2010/2011 resources were secured from the following: Woodcock Foundation Mountain Equipment Co-op Wilburforce Foundation McLean Foundation

Alberta Sport, Recreation, Parks and Wildlife Foundation TD Friends of the Environment Foundation Lake O’Hara Lodge Private Donation

Acknowledgements Many people contributed to making the first year a successful one. Jason Fisher (Alberta Innovates) provided guidance on study design and with Matt Wheatley (Alberta SRD) both graciously loaned us 30+ remote cameras. Bill Abercrombie (Animal Damage Control) generously provided time and trapper support to keep our beaver carcass supply on track. Don Gorrie assisted with the placement of barrels and beaver carcasses in key locations. Reg Hawryluk was always ready to help out with checking hair traps or give advice on access or logistics. Jim Zettel, Cathy Gill, Andrea Kortello and Reg Bunyan were instrumental in consistently helping us carry out the survey throughout the winter. Wayne Shibley was more than generous with time, local advice and snowmobile support to help get us up Hospital Creek. Cal Sime helped in many ways and provided guidance on tackling some of the more difficult sites to access west of Yoho National Park. Amiskwi Lodge owners (Gary, Shorty, Doug) helped with heli access to the Upper Amiskwi drainage and kindly allowed us to stay at the lodge, rather than dig out the Amiskwi warden cabin from under 3 m of snow! This work couldn’t have been possible without the support and guidance of the Parks Canada Visitor Safety staff, particularly Marc Ledwidge and Brian Webster, who attended to our many questions and needs, from loaning us beacons and providing us with

snowmobile access to sites in class 2 and 3 avalanche terrain. We appreciate the help of Jon Pedlar and the Alberta Snow Survey crew by allowing us to ride along each month to access some of the more remote sites in our study area. Dan Rafla, Ron Leblanc, Blair Fyten and Ian Pengelly claimed the Cascade Corridor, tackling sites from Stoney to Windy cabins. Trevor Kinley was a keen monthly companion, always ready and willing to ski into the Simpson River site. The Wolverine Watch program was a source of more competent and enthusiastic volunteer help than we could handle: Sadie Parr was fast to her skis whenever we called and someone we could always rely on to help in the Golden area. Marg Gmoser was always there for helpful advice and eager to assist on day or overnight trips. Others assisting throughout the winter were Rachel Darvill, Bob Toothill, Tim Johnson, Earl Marsh, Jenn Redvers, Adam Ford, Heather Milligan and Barb Sobota. Tina Barzo helped us recruit and organize volunteers and shared a passion for getting citizen scientists involved with our research. The project is part of a contribution agreement between Parks Canada and the Miistakis Institute for the Rockies. Funding for the research comes from our project sponsors, Parks Canada, WTI-Montana State University, the Woodcock and Wilburforce Foundations. Additional funding was provided by Mountain Equipment Co-op (MEC), McLean Foundation, TD Foundation for the Environment, Alberta Sports, Recreation and Wildlife Foundation, Cameron Plewes, Patagonia Foundation, Alpine Club of Canada, Lake O’Hara Lodge and an anonymous donor.

Pro support was kindly provided by: Marmot Mountain Canada, Black Diamond Equipment, Outdoor Research, Scarpa boots, and Patagonia.

Literature cited Alberta Fish and Wildlife Division. 2008. Report on Alberta’s Endangered Species Conservation Committee. June 2006. Alberta SRD, FWD, Edmonton, AB. 44pp. Austin, M. 1998. Wolverine winter travel routes and response to transportation corridors in Kicking Horse Pass between Yoho and Banff National Parks. M.Environ. Des. Thesis. University of Calgary, Alberta. 40 pp. Banff-Bow Valley Study. 1996. Banff-Bow Valley: at the crossroads. Summary report for the Banff-Bow Valley Task Force. Canadian Heritage, Ottawa, Ontario. Bertch, B. 2003. Wolverine winter ecology and human disturbance in the Lake Louise and Yoho National Park areas. Parks Canada Species at Risk Recovery Fund Project # SARRF 2-14. Season 2 (2002-03) Final Report to Parks Canada Agency, Radium Hot Springs, B.C. COSEWIC 2003. COSEWIC assessment and update status report on the wolverine Gulo gulo in Canada. Committee on the Status of Endangered Wildlife in Canada. Ottawa. 41 pp. Copeland, J. 1996. Biology of the wolverine in central Idaho. Thesis, University of Idaho, Moscow, USA. Copeland, J. et al. 2010. The bioclimatic envelope of the wolverine (Gulo gulo):

do climatic constraints limit its geographic distribution? Can. J. Zool. 88: 233-246. Cushman, S. et al. 2006. Gene flow in complex landscapes: testing multiple hypotheses with causal modeling. American Naturalist 168, 486-499. Fisher, J., S. Bradbury, A. Fisher, and L. Nolan. 2009. Wolverines on the edge of Alberta’s Rockies. Alberta Research Council, Edmonton, Alberta. Foran, D.S., K.C. Crooks, and S.C. Minta. 1997. DNA-based analysis of hair to identify species and individuals for population research and monitoring. Wildlife Society Bulletin 25, 840-847. Ford, AT, AP Clevenger, K Rettie. 2010. Banff Wildlife Crossings, Trans-Canada Highway, Alberta – An international public-private partnership. Pages 157-172 in Safe passages: Highways, wildlife and habitat connectivity. J Beckmann, AP Clevenger, M Huijser, J Hilty (eds.). Island Press, Washington DC Garroway, C., J. Bowman and P. Wilson. 2011. Using a genetic network to parameterize a landscape resistance surface for fishers, Martes pennanti. Molecular Ecology (early view). Golder Associates. 2004. Screening report for the Trans-Canada Highway twinning project phase IIIB, Banff National Park. Report for Parks Canada, Banff National Park, Alberta. Gompper, M. et al. 2006. A comparison of noninvasive techniques to survey carnivore communities in Northeastern North America. Wildlife Society Bulletin 34: 1142–1151.

Krebs, J.A., E.C. Lofroth, J.Copeland, V. Banci, D. Cooley, H. Golden, A. Magoun, R. Mulders, and B.Shults. 2004. Synthesis of survival rates and causes of mortality in North American wolverines. Journal of Wildife Management 68: 493-502. Krebs, J.A. and D. Lewis. 2000. Wolverine ecology and habitat use in the North Columbia Mountains: progress report. In Proc. Conf. on the biology and management of species and habitats at risk. L.M. Darling (editor). Kamloops, B.C., Feb. 15–19, 1999. B.C. Min. Environ., Lands and Parks, Victoria, B.C., and Univ. Coll. Cariboo, Kamloops, B.C., pp. 695– 703. Koen, E.L., J.C. Ray, J. Bowman, F.N. Dawson, A.J. Magoun. 2008. Surveying and monitoring wolverines in Ontario and other lowland, boreal forest habitats: Recommendations and protocols. Ontario Ministry of Natural Resources, Northwest Science and Information. Thunder Bay, Ontario. NWSI Field Guide FG-06 94. Laliberte, A.S. and W.J. Ripple. 2004. Range contractions of North American carnivores and ungulates. Bioscience 54: 123-138. Lofroth, E.C., and J. Krebs. 2007. The abundance and distribution of wolverines in British Columbia. Journal of Wildlife Management 71: 2159-2169. Long, R. A., P. MacKay, W. J. Zielinski, and J. C. Ray, editors. 2008. Noninvasive survey methods for carnivores. Island Press, Washington D.C., USA. Magoun, A.J. et al. 2011. Integrating motion-detection cameras and hair snags for wolverine identification. Journal of Wildlife Management 75: 731–739.

Parks Canada. 1997. Yoho National Park management plan. Ministry of Canadian Heritage, Ottawa, Ontario. Parks Canada. 2007. Banff National Park management plan. July 2007 Amendment. Ministry of Canadian Heritage, Ottawa, Ontario. Schwartz, M.K., G. Luikart, R.S. Waples. 2007. Genetic monitoring as a promising tool for conservation and management. Trends in Ecology and Evolution 22: 25-33. Schwartz, M., Copeland, J. & Anderson, N. 2009. Wolverine gene flow across a narrow climatic niche. Ecology 90: 3222-3232. Suitor, M. 2005. Wolverine distribution and abundance in the Canadian Rocky Mountain Parks: Research priorities and techniques. Report for Parks Canada Western Service Centre, Calgary, Alberta. Weaver, J.L., Paquet, P.C., Ruggiero, L.F. 1996: Resilience and conservation of large carnivores in the Rocky Mountains. Conservation Biology 10: 964-976.