Report on the Plains Bison Reintroduction Pilot, 2017-2022
November 2022

The process garnered high levels of interest from a broad range of stakeholders, provincial and municipal governments, neighbouring land owners, and various user groups. Based on the results, Parks Canada developed a reintroduction plan, as well as a number of plans and assessments to guide and strengthen the reintroduction plan:

• Detailed Environmental Impact Assessment;
• Bison Excursion Prevention and Response Plan;
• Disease Risk Assessment; and
• Bison Reintroduction Health Monitoring Strategy.

The Reintroduction Plan for Plains Bison included an adaptive, five-year phased pilot that began with a small herd of bison released on the eastern slopes of Banff National Park. The second phase occurred 18 months later, when that same herd and its newly born calves were released into a remote 1200 km² reintroduction zone in the backcountry of the park (Figure 2).

Following the commitments made in the reintroduction plan, with the five-year pilot set to conclude, Parks Canada has prepared this report which will be shared for public, stakeholder, and Indigenous review and comment.

Overall, the goals and objectives of the pilot are considered to be achieved:

• The bison remained healthy and demonstrated a healthy growth rate with minimal mortalities (Table 1), they appear to have adapted quickly to the mountain habitat, and they remained, for the most part, within the Park.
• Negative impacts to other species were not observed, and there were no threats to human safety or property damage reported.
• Establishing and strengthening relationships occurred through the involvement of a number of Indigenous nations and stakeholders throughout the pilot.
• A variety of in-park and outreach and educational programs provided the opportunity to enrich visitor experience and understanding.

Following the commitments made in the reintroduction plan, with the five-year pilot set to conclude this year in 2022, Parks Canada has prepared its evaluation based on the eight evaluation factors outlined in the Reintroduction Plan:

• Visitor safety & visitor experience impacts;
• Indigenous groups & stakeholders - impacts & perspectives; (formerly neighbours & partners)
• Response to citizen engagement & stewardship programs;
• Bison home ranges, monitoring & responses;
• Mitigating effects of the bison project on other species;
• Bison population growth & population targets;
• Public awareness, education & appreciation; and
• Factors influencing bison movement & habitat selection.

These factors were developed based on input from and concerns raised by stakeholders, Indigenous nations, and the public as well as park objectives and priorities. The following is an in-depth assessment and evaluation of the five-year Plains Bison Reintroduction Pilot against each of those factors.

Visitor safety and experience impacts

The Banff bison were reintroduced in a remote backcountry area of the park that sees little human use. The reintroduction story was therefore communicated through several interpretive programs (Table 2), reaching nearly 14,000 people over the span of the five-year pilot.

While there were a few backcountry visitors who did see the bison during their wilderness experience, and those were mostly guides with clients, the existence of bison in the backcountry, did not appear to increase visitation to this part of the backcountry. Remote trail cameras (N=6) indicate annual human use (including Parks Canada staff) remained low, with fewer than 60 human events per year recorded in the heart of the reintroduction zone during the pilot period. ^{Footnote 1} This is well (~95%) below the target threshold of 100 events per month to maintain grizzly bear habitat security, as prescribed in BNP’s Management Plan. ^{Footnote 2}

Parks Canada did not receive any visitor or stakeholder complaints about bison inside the Park. Bison generally fled (up to 5km) from occasional hiking groups and neither fled nor approached groups on horseback (Figure 3). Outside the park, a lone bull did venture into horse camps, which caused concerns among campers. It was euthanized as it continued to wander east, as per the Bison Excursion Response Plan

Livestock grazing allotments begin ~20km beyond the current bison reintroduction zone, and extend to the public/private land boundary 50km east of the park (Figure 2). The livestock industry was concerned at the outset of the project that bison could carry and transmit diseases of concern (e.g., anthrax, brucellosis and bovine tuberculosis) to cattle if they frequented this area.

An extensive disease risk assessment conducted prior to the reintroduction concluded there was a low probability of this occurring ^{Footnote 5}. Nonetheless, in recognition that the economic consequences to the livestock industry would be significant if it did occur, an in-depth Bison Health Monitoring and Disease Response Plan ^{Footnote 6} was subsequently developed and all preventative measures were followed over the last 5 years, including:

• Sourcing animals from a certified disease-free herd (Elk Island National Park, as per Canadian Food Inspection Agency (CFIA) testing since the 1970s);
• Testing and quarantining the animals prior to translocation, and again prior to releasing them into the wild ^{Footnote 7};
• Testing 8 animals captured for radio collaring for brucellosis and other diseases after the 2018 release (all were negative) ^{Footnote 8};
• Evaluating any dead bison for signs of tuberculosis and other diseases (2 that had to be destroyed when they dispersed well beyond the reintroduction zone were necropsied by a licensed veterinarian and found to be in excellent health with no signs of tuberculosis or other diseases).

Other indicators of bison health were closely monitored with reassuring results: body condition for all founder animals averaged in the “good” to “very good” range through the pilot project (Figure 4) ^{Footnote 9}, fecal parasite counts were low ^{Footnote 10}, calving rates were very high (average 33% herd growth rate per year), and, despite intensive monitoring and frequent field and remote camera observations (radio collars maintained on at least 10% of the herd), no sick animals were detected and only two natural bison mortalities were recorded in the 5-year pilot (due to wolf predation of newborns in spring 2020 and 2021).

Although bison excursions (i.e. instances where bison venture beyond the reintroduction zone or the park boundary) were rare during the 5-year pilot project, lone bulls did disperse east of the reintroduction zone and the Park on four occasions (see Section 4). Three of these ventured onto livestock grazing allotments in the summers of 2018 and 2019 but did not mingle with cattle or damage any fences. Allotment holders were contacted immediately and PCA successfully recaptured or removed the bison within days, as per the BNP Bison Excursion Response Plan ^{Footnote 11}.

Response to citizen engagement and stewardship programs

Parks Canada exceeded its goal of increasing volunteer opportunities by 1,000 hours through seven backcountry volunteer programs that resulted in over 2,000 volunteer hours during fence construction of the soft release pasture (Figure 5), consulting on the overall project (2016), fortifying drift fencing on the periphery of the reintroduction zone (2018), helping to feed and care for the bison while they were held in the soft release pasture (2017 and 2018), assisting with public outreach (2018-2020), and analyzing data and writing reports (2021).

Bison home ranges, monitoring and responses

The support of some stakeholders for the five-year pilot reintroduction project was contingent on a commitment from Parks Canada to contain bison within the reintroduction zone and BNP as bison are not considered to be wildlife outside of the Park (Figure 1).

Strategies to encourage bison to anchor to this target area included starting with a small number of young (2-3-year-old) animals, holding them in a backcountry pasture for 18 months, radio collaring and monitoring at least 10% of the population, hazing them back into the reintroduction zone when required, and complementing the cliffs and steep ridges that surround much of the reintroduction zone with short drift fences constructed in key valleys (Figure 6).

By almost all measures, these strategies worked: the majority of bison (which, as of Fall 2022 is ~ 80 animals) have remained in the reintroduction zone and within 30 km of where they were released (Figure 6). Only four males dispersed outside of the zone and had to either be recaptured and relocated elsewhere (N=2) or were destroyed (N=2), as per the BNP Bison Excursion Response Plan. ^{Footnote 11}

Home range and movements

The rate of bison exploration diminished with time since release. The main bison population explored 257 km² of the 1200 km² reintroduction zone during the pilot period, the majority of which occurred over the first year (81.7%) ^{Footnote 12}. Explorations occurred more in summer (45.5%) and fall (32.3%) than in winter (7.3%) or spring (14.8%) ^{Footnote 13}.

The decline in exploration after the first year did not translate to less movement: daily movement remained steady between years (median of 1.9km/day) ^{Footnote 12} but large and rapid movements (e.g., ≥ 4.0 km in 2 hours) dropped by 51% after the first year, suggesting the animals did settle over time. ^{Footnote 13} This was further supported by a contraction in the size of their annual home range between the first and second years (133.2 km² to 92.9 km²). ^{Footnote 13}

The movements of the reintroduced bison have been highly asymmetrical (Figure 6): 97% of GPS collar positions occurred to the northwest, north, and northeast of the release site ^{Footnote 13}, and all explorations of the edge of the reintroduction zone occurred in these directions. Movements in all remaining directions account for only 3% of locations.

A number of hypotheses could explain these movements: many river valleys drain north or east in the study area, the more open and grassy (Southwest-facing slopes) of the reintroduction zone are always in a northeast direction, and the national park from which the animals were sourced (Elk Island) is in a northeast direction. Notwithstanding their asymmetrical pattern of exploration, the reintroduced Banff bison have developed a home range with predictable travel paths between preferred areas in the reintroduction zone, suggesting a learned fidelity for the area.

Responses to management interventions

It is important to note, however, that this fidelity and bison movements have been heavily influenced by management interventions, most notably two drift fences in the Panther and Red Deer valleys (Figure 7). These two drift fences have prevented north easterly bison movements outside of the reintroduction zone on 50 occasions since they were released, and the frequency of these visits has not diminished with time (Figure 7). Hazing the animals back into the reintroduction zone using low-stress herding techniques has also been necessary, albeit on fewer occasions (N=7) ^{Footnote 14} . These forays have also been in a northward direction (towards the Clearwater Valley from Divide Creek), and, because of the remoteness of the region, have required helicopters to get people on the ground to herd the animals in a timely manner.

Mitigating effects of the bison project on other species

Competition between bison and species that are hunted outside the Park (e.g., bighorn sheep and elk), and the potential impacts of bison drift fences on broader wildlife movement was a concern of Indigenous Groups and stakeholders. Mitigations were put in place to address these concerns in two ways: 1) by testing for and adopting a proven wildlife-friendly design before deploying drift fences to help contain bison; and 2) by searching for early evidence of bison directly impacting other species.

Wildlife-friendly drift fences

In an effort to encourage bison to anchor to the reintroduction zone, Parks Canada built 13 adjustable drift fences designed to encourage bison from leaving the reintroduction zone. Collectively they spanned 7.8 km across 13 pinch points on the periphery of the reintroduction zone in 2017 and 2018. Of these, two (Panther and Red Deer) have proven to be crucial to encouraging bison to stay in the reintroduction zone. Some of the others were ineffective or unnecessary and nine were removed over the last several years. Only six kilometers of drift fencing remains across four sites (Figure 6).

In order to arrive at the optimal design of these fences, Parks Canada team members rigorously tested several configurations and analyzed 6,104 wildlife-fence interactions with 34 remote cameras. The resulting design was a 150cm-high, five-wire fence that is permeable to wildlife but deflects bison (Figure 8) ^{Footnote 15} . There were no significant changes in wildlife activity, speed or through movements before versus after fence construction.

Monitoring and maintenance of these fences is ongoing. Located in remote, rugged terrain, they are prone to damage from fallen trees and seasonal flooding. This requires frequent maintenance checks by staff (every 2-3 weeks) who undertake multiday backcountry trips by foot, horseback and ski to access the fences.

Impacts of bison on other species

Parks Canada searched for evidence of bison directly impacting other wildlife. Long-term elk GPS radio collar data from the Ya Ha Tinda elk project ^{Footnote 16} was analyzed along with bison GPS radio collar data to assess whether there might be competitive (negative) or facultative (beneficial) interactions between the two species.

Elk and bison overlapped minimally in spring and summer (Figure 9), and neither the distribution of elk, nor how elk used and selected habitat changed after bison reintroduction. ^{Footnote 17} Furthermore, fewer than 50 elk currently use the bison reintroduction zone, making competition (which requires a resource to be limited), unlikely, especially at low bison densities.

A focused bighorn sheep study also began soon after bison were reintroduced. ^{Footnote 18} Fifteen bighorn sheep were radio collared throughout the bison reintroduction zone and another 25 collared elsewhere in BNP (196,103 locations). Bison and sheep were observed in close proximity in parts of the reintroduction zone, however home range maps showed low levels of spatial overlap between species.

Further analysis of habitat use revealed broad-scale resource separation, with sheep selecting higher elevations and steeper slopes than bison in all seasons (Figure 10). Using newly created sheep habitat models, the proportion of bison locations in higher quality sheep habitats was shown to fluctuate by season; the greatest proportions were seen in winter and spring as both species selected strongly for grassland habitat types.

Unlike more recent elk counts, bighorn sheep are relatively numerous in the area, with 364 individuals counted in the bison reintroduction zone during a larger aerial survey in 2020 ^{Footnote 19}. Though spatial overlap is low and the species are broadly separated by habitat selection, seasonal use of higher quality sheep habitat by bison could lead to range competition in some areas if bison numbers increase significantly.

Efforts to track wolf response to the initial reintroduction of bison were hampered by high wolf mortality due to trapping outside the park. ^{Footnote 20} However, two wolves were GPS-collared and survived for a few months before being trapped outside the Park in the first winter (2018/19) after bison were released. They approached the reintroduced animals repeatedly but their advances did not elicit corresponding movements from GPS-collared bison. Photos from Parks’ remote camera network show a similar lack of response from reintroduced bison to the presence of wolves. In a couple of instances, curious juvenile bison males actually chased wolves.

Bison reintroduced elsewhere have positively impacted other wildlife (and soils and vegetation) ^{Footnote 21}. Such ecosystem effects and the restoration of ecological integrity are also important goals for this project. Long term studies to quantify some of these effects (e.g. influence of bison grazing on breeding bird distribution and the structure of vegetation) are underway in BNP.

Bison population growth and population targets

Bison population growth during the pilot project averaged 33% per year and natural mortality was less than 1% per year. The removal of four dispersing males over the five-year pilot was also approximately 1% per year. As a result, the 16 translocated animals grew into a population of ~80 animals over the duration of the pilot.

Growth rates are expected to moderate somewhat as the predominantly young and female founding animals age and the sex ratio balances, but even the average growth rate for wild herds in North America (20%/yr.) would result in more than 200 animals within 8 years (Figure 11).

This kind of growth is encouraging for species recovery where hundreds or thousands of animals are needed to insulate against genetic drift, extreme weather, and other forces of extinction. ^{Footnote 22} However, the population (and range) of every modern free-roaming plains bison herd in North America is limited by surrounding development and is ultimately managed by people, either through Indigenous and non-Indigenous harvest (e.g. Pink Mountain, B.C.; Henry Mountains, Utah), frequent roundups, relocations, auctions and removals (e.g. Grasslands and Elk Island national parks), or both (e.g. Yellowstone National Park).

The BNP situation is no different: although opportunities to expand bison range exist within and outside the Park, they will ultimately be limited by outlying agriculture, other human developments, and active management.

Public awareness, education and appreciation

The story of bison reintroduction reached almost 120 million people (Table 3), in an effort aimed at expanding public appreciation for Park’s Canada’s role in conservation. This was achieved by transmitting the story from the backcountry, as it unfolded, in ways that delivered it to peoples’ doorsteps and devices.

This included 13 webisodes (e.g. Free and Thriving ^{Footnote 23}), digital storytelling (Bison Blog and associated social media posts), and over a dozen articles, and television and radio interviews in traditional media, all of which were received enthusiastically by the public. Collectively, an estimated 119.5 million people were reached with the Banff Bison story since the project began in 2017. This indicates a strong public awareness and appreciation of the bison pilot project amongst Canadians.

Factors influencing bison movement and habitat selection

A bison habitat assessment for the area was completed before the pilot began. Now, after having bison on the landscape for more than three years, we can compare predictions with actual habitat use. Assessment reasonably predicted winter bison use of low-elevation meadows and the bison’s preference for forests burned in the last 15 years. However, it poorly predicted their habitat use in summer. The animals spent more time at higher elevations, on steeper slopes (Figure 12), in shrubs, and further from water than expected. These trends were surprising and demonstrated a remarkable adaptability for animals that were translocated from a non-mountainous environment.

The bison have consistently returned to high elevations during each peak growing season, which could be explained by an analysis of bison scat that found forage at high elevations to be nutritionally superior and more palatable to forage found at lower elevations in the summer.

This is precisely what past research has shown for elk in the area ^{Footnote 24}. Exploiting this difference is critical in June, July and August when plant nutrients peak and bison can recover from a declining winter body condition, and gain nourishment for lactation. Low inter-year differences in dietary efficiency shows just how adaptable the bison are: despite coming from a non-mountainous environment, they honed in on the altitudinal distribution of summertime nutrients almost immediately.

An updated Resource Selection Function for bison in the reintroduction zone now replaces the old habitat model. ^{Footnote 12} Built from bison GPS collar data collected since the animals were released, it forms the basis for a new predictive bison movement model that projects where the animals would have gone if they had unimpeded access to areas beyond the target reintroduction zone (Figure 13). It accurately predicts their preferred areas of the reintroduction zone (e.g., middle Panther and lower Red Deer valleys) but also shows them strongly attracted to the nearby Ya Ha Tinda and Panther Corners areas outside the park (and to a lesser extent the Bow Valley in BNP). The model predicts that, without management interventions, individual bison would have exited the reintroduction zone and made their way to these areas within eight or nine months of release, significantly decreasing the time they explored (and presumably bonded to) the target reintroduction area.